- Kernel - mailweb.openeuler.org

[PATCH OLK-6.6] mm: Add sysctl to clear free list pages
by Yu Liao 04 Dec '23

04 Dec '23

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I4SK3S CVE: NA -------------------------------- This patch add sysctl to clear pages in free lists of each NUMA node. For each NUMA node, clear each page in the free list, these work is scheduled on a random CPU of the NUMA node. When kasan is enabled and the pages are free, the shadow memory will be filled with 0xFF, writing these free pages will cause UAF, so just disable KASAN for clear freelist. In the case of large memory, the clear freelist will hold zone lock for a long time. As a result, the process may be blocked unless clear freelist thread exit, and causing the system to be reset by the watchdog. Provide a mechanism to stop clear freelist threads when elapsed time exceeds cfp_timeout, which can be set by module_param(). Signed-off-by: Yu Liao <liaoyu15(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> --- .../admin-guide/kernel-parameters.txt | 3 + Documentation/admin-guide/sysctl/vm.rst | 13 ++ mm/Kconfig | 13 ++ mm/Makefile | 2 + mm/clear_freelist_page.c | 187 ++++++++++++++++++ 5 files changed, 218 insertions(+) create mode 100644 mm/clear_freelist_page.c diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 41644336e358..97c91370b01c 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -618,6 +618,9 @@ Also note the kernel might malfunction if you disable some critical bits. + clear_freelist + Enable clear_freelist feature. + clk_ignore_unused [CLK] Prevents the clock framework from automatically gating diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentation/admin-guide/sysctl/vm.rst index 45ba1f4dc004..cc15ea0fc406 100644 --- a/Documentation/admin-guide/sysctl/vm.rst +++ b/Documentation/admin-guide/sysctl/vm.rst @@ -25,6 +25,7 @@ files can be found in mm/swap.c. Currently, these files are in /proc/sys/vm: - admin_reserve_kbytes +- clear_freelist_pages - compact_memory - compaction_proactiveness - compact_unevictable_allowed @@ -106,6 +107,18 @@ On x86_64 this is about 128MB. Changing this takes effect whenever an application requests memory. +clear_freelist_pages +==================== + +Available only when CONFIG_CLEAR_FREELIST_PAGE is set. When 1 is written to the +file, all pages in free lists will be written with 0. + +Zone lock is held during clear_freelist_pages, if the execution time is too +long, RCU CPU Stall warnings will be print. For each NUMA node, +clear_freelist_pages is performed on a "random" CPU of the NUMA node. +The time consuming is related to the hardware. + + compact_memory ============== diff --git a/mm/Kconfig b/mm/Kconfig index ece4f2847e2b..e652d267e87e 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -1269,6 +1269,19 @@ config LOCK_MM_AND_FIND_VMA bool depends on !STACK_GROWSUP +config CLEAR_FREELIST_PAGE + bool "Support for clear free list pages" + depends on MMU && SYSCTL + default n + help + Say y here to enable the clear free list pages feature. When + writing to clear_freelist, trigger to clean up the free memory + of the buddy system. + + To enable this feature, kernel parameter "clear_freelist" also + needs to be added. + + source "mm/damon/Kconfig" endmenu diff --git a/mm/Makefile b/mm/Makefile index ec65984e2ade..1dc5d64a54f0 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -7,6 +7,7 @@ KASAN_SANITIZE_slab_common.o := n KASAN_SANITIZE_slab.o := n KASAN_SANITIZE_slub.o := n KCSAN_SANITIZE_kmemleak.o := n +KASAN_SANITIZE_clear_freelist_page.o := n # These produce frequent data race reports: most of them are due to races on # the same word but accesses to different bits of that word. Re-enable KCSAN @@ -138,3 +139,4 @@ obj-$(CONFIG_IO_MAPPING) += io-mapping.o obj-$(CONFIG_HAVE_BOOTMEM_INFO_NODE) += bootmem_info.o obj-$(CONFIG_GENERIC_IOREMAP) += ioremap.o obj-$(CONFIG_SHRINKER_DEBUG) += shrinker_debug.o +obj-$(CONFIG_CLEAR_FREELIST_PAGE) += clear_freelist_page.o diff --git a/mm/clear_freelist_page.c b/mm/clear_freelist_page.c new file mode 100644 index 000000000000..50b7ec918bfb --- /dev/null +++ b/mm/clear_freelist_page.c @@ -0,0 +1,187 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Support for clear free list pages. + */ + +#include <linux/mmzone.h> +#include <linux/mm_types.h> +#include <linux/mm.h> +#include <linux/sysctl.h> +#include <linux/highmem.h> +#include <linux/slab.h> +#include <linux/workqueue.h> +#include <linux/sched.h> +#include <linux/atomic.h> +#include <linux/nmi.h> +#include <linux/sched/clock.h> +#include <linux/module.h> + +#define CFP_DEFAULT_TIMEOUT 2000 +#define for_each_populated_zone_pgdat(pgdat, zone) \ + for (zone = pgdat->node_zones; \ + zone; \ + zone = next_pgdat_zone(zone)) \ + if (!populated_zone(zone)) \ + ; /* do nothing */ \ + else + +struct pgdat_entry { + struct pglist_data *pgdat; + struct work_struct work; +}; + +static DECLARE_WAIT_QUEUE_HEAD(clear_freelist_wait); +static DEFINE_MUTEX(clear_freelist_lock); +static atomic_t clear_freelist_workers; +static atomic_t clear_pages_num; +static ulong cfp_timeout_ms = CFP_DEFAULT_TIMEOUT; + +/* + * next_pgdat_zone - helper magic for for_each_populated_zone_pgdat() + */ +static struct zone *next_pgdat_zone(struct zone *zone) +{ + pg_data_t *pgdat = zone->zone_pgdat; + + if (zone < pgdat->node_zones + MAX_NR_ZONES - 1) + zone++; + else + zone = NULL; + return zone; +} + +static void clear_pgdat_freelist_pages(struct work_struct *work) +{ + struct pgdat_entry *entry = container_of(work, struct pgdat_entry, work); + u64 cfp_timeout_ns = cfp_timeout_ms * NSEC_PER_MSEC; + struct pglist_data *pgdat = entry->pgdat; + unsigned long flags, order, t; + struct page *page; + struct zone *zone; + u64 start, now; + + start = sched_clock(); + + for_each_populated_zone_pgdat(pgdat, zone) { + spin_lock_irqsave(&zone->lock, flags); + for_each_migratetype_order(order, t) { + list_for_each_entry(page, &zone->free_area[order].free_list[t], lru) { + now = sched_clock(); + if (unlikely(now - start > cfp_timeout_ns)) { + spin_unlock_irqrestore(&zone->lock, flags); + goto out; + } + +#ifdef CONFIG_KMAP_LOCAL + int i; + + /* Clear highmem by clear_highpage() */ + for (i = 0; i < (1 << order); i++) + clear_highpage(page + i); +#else + memset(page_address(page), 0, (1 << order) * PAGE_SIZE); +#endif + touch_nmi_watchdog(); + atomic_add(1 << order, &clear_pages_num); + } + } + spin_unlock_irqrestore(&zone->lock, flags); + + cond_resched(); + } + +out: + kfree(entry); + + if (atomic_dec_and_test(&clear_freelist_workers)) + wake_up(&clear_freelist_wait); +} + +static void init_clear_freelist_work(struct pglist_data *pgdat) +{ + struct pgdat_entry *entry; + + entry = kzalloc(sizeof(struct pgdat_entry), GFP_KERNEL); + if (!entry) + return; + + entry->pgdat = pgdat; + INIT_WORK(&entry->work, clear_pgdat_freelist_pages); + queue_work_node(pgdat->node_id, system_unbound_wq, &entry->work); +} + +static void clear_freelist_pages(void) +{ + struct pglist_data *pgdat; + + mutex_lock(&clear_freelist_lock); + drain_all_pages(NULL); + + for_each_online_pgdat(pgdat) { + atomic_inc(&clear_freelist_workers); + init_clear_freelist_work(pgdat); + } + + wait_event(clear_freelist_wait, atomic_read(&clear_freelist_workers) == 0); + + pr_debug("Cleared pages %d\nFree pages %lu\n", atomic_read(&clear_pages_num), + global_zone_page_state(NR_FREE_PAGES)); + atomic_set(&clear_pages_num, 0); + + mutex_unlock(&clear_freelist_lock); +} + +static int sysctl_clear_freelist_handler(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int ret; + int val; + + table->data = &val; + ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + + if (!ret && write) + clear_freelist_pages(); + + return ret; +} + +static struct ctl_table clear_freelist_table[] = { + { + .procname = "clear_freelist_pages", + .data = NULL, + .maxlen = sizeof(int), + .mode = 0200, + .proc_handler = &sysctl_clear_freelist_handler, + .extra1 = SYSCTL_ONE, + .extra2 = SYSCTL_ONE, + }, + { } +}; + +static struct ctl_table sys_ctl_table[] = { + { + .procname = "vm", + .mode = 0555, + .child = clear_freelist_table, + }, + { } +}; + +static bool clear_freelist_enabled; +static int __init setup_clear_freelist(char *str) +{ + clear_freelist_enabled = true; + return 1; +} +__setup("clear_freelist", setup_clear_freelist); + +static int __init clear_freelist_init(void) +{ + if (clear_freelist_enabled) + register_sysctl_table(sys_ctl_table); + + return 0; +} +module_init(clear_freelist_init); +module_param(cfp_timeout_ms, ulong, 0644); -- 2.33.0

2 1

[PATCH OLK-6.6 0/2] xfs: fix intent item leak during reovery
by Long Li 04 Dec '23

04 Dec '23

This patch set fix intent item leak during reovery when recovery intent fails. Long Li (2): xfs: factor out xfs_defer_pending_abort xfs: abort intent items when recovery intents fail fs/xfs/libxfs/xfs_defer.c | 28 ++++++++++++++++++---------- fs/xfs/libxfs/xfs_defer.h | 2 +- fs/xfs/xfs_log_recover.c | 2 +- 3 files changed, 20 insertions(+), 12 deletions(-) -- 2.31.1

2 3

[PATCH openEuler-22.03-LTS 0/6] backport Broadcom NIC driver patches
by Dapeng Yu 04 Dec '23

04 Dec '23

Backport the latest patches from upstream stable branch linux-5.10.y for Broadcom NIC driver. Florian Fainelli (2): net: bcmgenet: Remove phy_stop() from bcmgenet_netif_stop() net: bcmgenet: Restore phy_stop() depending upon suspend/close Somnath Kotur (2): bnxt_en: Query default VLAN before VNIC setup on a VF bnxt_en: Implement .set_port / .unset_port UDP tunnel callbacks Sreekanth Reddy (1): bnxt_en: Don't issue AP reset during ethtool's reset operation Tobias Heider (1): Add MODULE_FIRMWARE() for FIRMWARE_TG357766. drivers/net/ethernet/broadcom/bnxt/bnxt.c | 28 ++++++++++++++----- .../net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 2 +- .../net/ethernet/broadcom/genet/bcmgenet.c | 9 +++--- drivers/net/ethernet/broadcom/tg3.c | 1 + 4 files changed, 28 insertions(+), 12 deletions(-) -- 2.43.0

2 1

[PATCH openEuler-22.03-LTS 1/6] net: bcmgenet: Remove phy_stop() from bcmgenet_netif_stop()
by Dapeng Yu 04 Dec '23

04 Dec '23

From: Florian Fainelli <f.fainelli(a)gmail.com> stable inclusion from stable-v5.10.181 commit e92727ed9e8b6798efc1e7ae44bbe7c44c453670 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8L4MY CVE: NA backport: OLK-5.10, openEuler-22.03-LTS, openEuler-22.03-LTS-SP1, openEuler-22.03-LTS-SP2 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 93e0401e0fc0c54b0ac05b687cd135c2ac38187c ] The call to phy_stop() races with the later call to phy_disconnect(), resulting in concurrent phy_suspend() calls being run from different CPUs. The final call to phy_disconnect() ensures that the PHY is stopped and suspended, too. Fixes: c96e731c93ff ("net: bcmgenet: connect and disconnect from the PHY state machine") Signed-off-by: Florian Fainelli <f.fainelli(a)gmail.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Dapeng Yu <dapeng.yu(a)windriver.com> --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 6ee6c34f28f2..52c93701a9e4 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -3408,7 +3408,6 @@ static void bcmgenet_netif_stop(struct net_device *dev) /* Disable MAC transmit. TX DMA disabled must be done before this */ umac_enable_set(priv, CMD_TX_EN, false); - phy_stop(dev->phydev); bcmgenet_disable_rx_napi(priv); bcmgenet_intr_disable(priv); -- 2.43.0

1 5

[PATCH OLK-6.6] arm64: errata: add option to disable cache readunique prefetch on HIP08
by Yang Yingliang 04 Dec '23

04 Dec '23

From: Kai Shen <shenkai8(a)huawei.com> hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I3ZFV2 CVE: NA ----------------------------------------------------------- Random performance decreases appear on cases of Hackbench which test pipe or socket communication among multi-threads on Hisi HIP08 SoC. Cache sharing which caused by the change of the data layout and the cache readunique prefetch mechanism both lead to this problem. Readunique mechanism which may caused by store operation will invalid cachelines on other cores during data fetching stage which can cause cacheline invalidation happens frequently in a sharing data access situation. Disable cache readunique prefetch can trackle this problem. Test cases are like: for i in 20;do echo "--------pipe thread num=$i----------" for j in $(seq 1 10);do ./hackbench -pipe $i thread 1000 done done We disable readunique prefetch only in el2 for in el1 disabling readunique prefetch may cause panic due to lack of related priority which often be set in BIOS. Introduce CONFIG_HISILICON_ERRATUM_HIP08_RU_PREFETCH and disable RU prefetch using boot cmdline 'readunique_prefetch=off'. Signed-off-by: Kai Shen <shenkai8(a)huawei.com> Signed-off-by: Hanjun Guo <guohanjun(a)huawei.com> [XQ: adjusted context] Signed-off-by: Xie XiuQi <xiexiuqi(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Cheng Jian <cj.chengjian(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/Kconfig | 18 +++++++++++ arch/arm64/kernel/cpu_errata.c | 56 ++++++++++++++++++++++++++++++++++ arch/arm64/tools/cpucaps | 1 + 3 files changed, 75 insertions(+) diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 78f20e632712..bbe85ca456c9 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -1229,6 +1229,24 @@ config SOCIONEXT_SYNQUACER_PREITS If unsure, say Y. +config HISILICON_ERRATUM_HIP08_RU_PREFETCH + bool "HIP08 RU: HiSilicon HIP08 cache readunique might cause performance drop" + default y + help + The HiSilicon HIP08 cache readunique might compromise performance, + use cmdline "readunique_prefetch_disable" to disable RU prefetch. + + If unsure, say Y. + +config HISILICON_HIP08_RU_PREFETCH_DEFAULT_OFF + bool "HIP08 RU: disable HiSilicon HIP08 cache readunique by default" + depends on HISILICON_ERRATUM_HIP08_RU_PREFETCH + default n + help + Disable HiSilicon HIP08 cache readunique by default. + + If unsure, say N. + endmenu # "ARM errata workarounds via the alternatives framework" choice diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c index 5706e74c5578..6a2ae67e85f7 100644 --- a/arch/arm64/kernel/cpu_errata.c +++ b/arch/arm64/kernel/cpu_errata.c @@ -13,6 +13,11 @@ #include <asm/cpufeature.h> #include <asm/kvm_asm.h> #include <asm/smp_plat.h> +#ifdef CONFIG_HISILICON_ERRATUM_HIP08_RU_PREFETCH +#include <asm/ptrace.h> +#include <asm/sysreg.h> +#include <linux/smp.h> +#endif static bool __maybe_unused is_affected_midr_range(const struct arm64_cpu_capabilities *entry, int scope) @@ -121,6 +126,48 @@ cpu_enable_cache_maint_trap(const struct arm64_cpu_capabilities *__unused) sysreg_clear_set(sctlr_el1, SCTLR_EL1_UCI, 0); } +#ifdef CONFIG_HISILICON_ERRATUM_HIP08_RU_PREFETCH +# ifdef CONFIG_HISILICON_HIP08_RU_PREFETCH_DEFAULT_OFF +static bool readunique_prefetch_enabled; +# else +static bool readunique_prefetch_enabled = true; +# endif +static int __init readunique_prefetch_switch(char *data) +{ + if (!data) + return -EINVAL; + + if (strcmp(data, "off") == 0) + readunique_prefetch_enabled = false; + else if (strcmp(data, "on") == 0) + readunique_prefetch_enabled = true; + else + return -EINVAL; + + return 0; +} +early_param("readunique_prefetch", readunique_prefetch_switch); + +static bool +should_disable_hisi_hip08_ru_prefetch(const struct arm64_cpu_capabilities *entry, int unused) +{ + u64 el; + + if (readunique_prefetch_enabled) + return false; + + el = read_sysreg(CurrentEL); + return el == CurrentEL_EL2; +} + +#define CTLR_HISI_HIP08_RU_PREFETCH (1L << 40) +static void __maybe_unused +hisi_hip08_ru_prefetch_disable(const struct arm64_cpu_capabilities *__unused) +{ + sysreg_clear_set(S3_1_c15_c6_4, 0, CTLR_HISI_HIP08_RU_PREFETCH); +} +#endif + static DEFINE_RAW_SPINLOCK(reg_user_mask_modification); static void __maybe_unused cpu_clear_bf16_from_user_emulation(const struct arm64_cpu_capabilities *__unused) @@ -744,6 +791,15 @@ const struct arm64_cpu_capabilities arm64_errata[] = { .capability = ARM64_WORKAROUND_AMPERE_AC03_CPU_38, ERRATA_MIDR_ALL_VERSIONS(MIDR_AMPERE1), }, +#endif +#ifdef CONFIG_HISILICON_ERRATUM_HIP08_RU_PREFETCH + { + .desc = "HiSilicon HIP08 Cache Readunique Prefetch Disable", + .capability = ARM64_WORKAROUND_HISI_HIP08_RU_PREFETCH, + ERRATA_MIDR_ALL_VERSIONS(MIDR_HISI_TSV110), + .matches = should_disable_hisi_hip08_ru_prefetch, + .cpu_enable = hisi_hip08_ru_prefetch_disable, + }, #endif { } diff --git a/arch/arm64/tools/cpucaps b/arch/arm64/tools/cpucaps index dea3dc89234b..bbb97036074b 100644 --- a/arch/arm64/tools/cpucaps +++ b/arch/arm64/tools/cpucaps @@ -100,3 +100,4 @@ WORKAROUND_NVIDIA_CARMEL_CNP WORKAROUND_QCOM_FALKOR_E1003 WORKAROUND_REPEAT_TLBI WORKAROUND_SPECULATIVE_AT +WORKAROUND_HISI_HIP08_RU_PREFETCH -- 2.25.1

2 1

[PATCH OLK-5.10 0/3] scsi: scsi_device_gets returns failure
by Li Lingfeng 04 Dec '23

04 Dec '23

If scsi is compiled as a module, scsi_device_gets returns failure when the module is NULL. Li Lingfeng (2): scsi: don't fail if hostt->module is NULL scsi: fix kabi broken in struct Scsi_Host Zhong Jinghua (1): scsi: scsi_device_gets returns failure when the module is NULL. drivers/scsi/hosts.c | 3 +++ drivers/scsi/scsi.c | 6 +++++- include/scsi/scsi_host.h | 2 +- 3 files changed, 9 insertions(+), 2 deletions(-) -- 2.31.1

2 4

openEuler Kernel SIG双周例会
by openEuler conference 04 Dec '23

04 Dec '23

您好！ Kernel SIG 邀请您参加 2023-12-08 14:00 召开的WeLink会议(自动录制) 会议主题：openEuler Kernel SIG双周例会会议内容： 1. 进展update 2. 议题征集中（新增议题可回复邮件申报，也可直接填至会议看板）会议链接：https://bmeeting.huaweicloud.com:36443/#/j/981980862 会议纪要：https://etherpad.openeuler.org/p/Kernel-meetings 温馨提醒：建议接入会议后修改参会人的姓名，也可以使用您在gitee.com的ID 更多资讯尽在：https://openeuler.org/zh/ Hello! openEuler Kernel SIG invites you to attend the WeLink conference(auto recording) will be held at 2023-12-08 14:00, The subject of the conference is openEuler Kernel SIG双周例会, Summary: 1. 进展update 2. 议题征集中（新增议题可回复邮件申报，也可直接填至会议看板） You can join the meeting at https://bmeeting.huaweicloud.com:36443/#/j/981980862. Add topics at https://etherpad.openeuler.org/p/Kernel-meetings. Note: You are advised to change the participant name after joining the conference or use your ID at gitee.com. More information: https://openeuler.org/en/

1 0

[openEuler-1.0-LTS] net: prevent rewrite of msg_name in sock_sendmsg()
by Lu Wei 04 Dec '23

04 Dec '23

From: Jordan Rife <jrife(a)google.com> stable inclusion from stable-v4.19.297 commit 3d62f2577c7414e5b94e4baad6eb621165960f30 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I8EZY4 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 86a7e0b69bd5b812e48a20c66c2161744f3caa16 ] Callers of sock_sendmsg(), and similarly kernel_sendmsg(), in kernel space may observe their value of msg_name change in cases where BPF sendmsg hooks rewrite the send address. This has been confirmed to break NFS mounts running in UDP mode and has the potential to break other systems. This patch: 1) Creates a new function called __sock_sendmsg() with same logic as the old sock_sendmsg() function. 2) Replaces calls to sock_sendmsg() made by __sys_sendto() and __sys_sendmsg() with __sock_sendmsg() to avoid an unnecessary copy, as these system calls are already protected. 3) Modifies sock_sendmsg() so that it makes a copy of msg_name if present before passing it down the stack to insulate callers from changes to the send address. Link: https://lore.kernel.org/netdev/20230912013332.2048422-1-jrife@google.com/ Fixes: 1cedee13d25a ("bpf: Hooks for sys_sendmsg") Cc: stable(a)vger.kernel.org Reviewed-by: Willem de Bruijn <willemb(a)google.com> Signed-off-by: Jordan Rife <jrife(a)google.com> Reviewed-by: Simon Horman <horms(a)kernel.org> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <sashal(a)kernel.org> conflict: net/socket.c Signed-off-by: Lu Wei <luwei32(a)huawei.com> --- net/socket.c | 29 +++++++++++++++++++++++------ 1 file changed, 23 insertions(+), 6 deletions(-) diff --git a/net/socket.c b/net/socket.c index 3d3632fe4a34..0cc52ba0b0e2 100644 --- a/net/socket.c +++ b/net/socket.c @@ -637,6 +637,14 @@ void __sock_tx_timestamp(__u16 tsflags, __u8 *tx_flags) } EXPORT_SYMBOL(__sock_tx_timestamp); +static int __sock_sendmsg(struct socket *sock, struct msghdr *msg) +{ + int err = security_socket_sendmsg(sock, msg, + msg_data_left(msg)); + + return err ?: sock_sendmsg_nosec(sock, msg); +} + /** * sock_sendmsg - send a message through @sock * @sock: socket @@ -655,10 +663,19 @@ static inline int sock_sendmsg_nosec(struct socket *sock, struct msghdr *msg) int sock_sendmsg(struct socket *sock, struct msghdr *msg) { - int err = security_socket_sendmsg(sock, msg, - msg_data_left(msg)); + struct sockaddr_storage *save_addr = (struct sockaddr_storage *)msg->msg_name; + struct sockaddr_storage address; + int ret; - return err ?: sock_sendmsg_nosec(sock, msg); + if (msg->msg_name) { + memcpy(&address, msg->msg_name, msg->msg_namelen); + msg->msg_name = &address; + } + + ret = __sock_sendmsg(sock, msg); + msg->msg_name = save_addr; + + return ret; } EXPORT_SYMBOL(sock_sendmsg); @@ -963,7 +980,7 @@ static ssize_t sock_write_iter(struct kiocb *iocb, struct iov_iter *from) if (sock->type == SOCK_SEQPACKET) msg.msg_flags |= MSG_EOR; - res = sock_sendmsg(sock, &msg); + res = __sock_sendmsg(sock, &msg); *from = msg.msg_iter; return res; } @@ -1940,7 +1957,7 @@ int __sys_sendto(int fd, void __user *buff, size_t len, unsigned int flags, if (sock->file->f_flags & O_NONBLOCK) flags |= MSG_DONTWAIT; msg.msg_flags = flags; - err = sock_sendmsg(sock, &msg); + err = __sock_sendmsg(sock, &msg); out_put: fput_light(sock->file, fput_needed); @@ -2272,7 +2289,7 @@ static int ____sys_sendmsg(struct socket *sock, struct msghdr *msg_sys, err = sock_sendmsg_nosec(sock, msg_sys); goto out_freectl; } - err = sock_sendmsg(sock, msg_sys); + err = __sock_sendmsg(sock, msg_sys); /* * If this is sendmmsg() and sending to current destination address was * successful, remember it. -- 2.34.1

1 0

[PATCH OLK-5.10 0/3] scsi: scsi_device_gets returns failure
by Li Lingfeng 04 Dec '23

04 Dec '23

If scsi is compiled as a module, scsi_device_gets returns failure when the module is NULL. Li Lingfeng (2): scsi: don't fail if hostt->module is NULL scsi: fix kabi broken in struct Scsi_Host Zhong Jinghua (1): scsi: scsi_device_gets returns failure when the module is NULL. drivers/scsi/hosts.c | 3 +++ drivers/scsi/scsi.c | 6 +++++- include/scsi/scsi_host.h | 2 +- 3 files changed, 9 insertions(+), 2 deletions(-) -- 2.31.1

2 5

[PATCH OLK-6.6 0/5] xfs: fix some misc issue
by Long Li 04 Dec '23

04 Dec '23

Long Li (5): xfs: fix a UAF when inode item push xfs: fix a UAF in xfs_iflush_abort_clean xfs: fix dir3 block read verify fail during log recover xfs: don't verify agf length when log recovery xfs: xfs_trans_cancel() path must check for log shutdown fs/xfs/libxfs/xfs_alloc.c | 3 ++- fs/xfs/xfs_buf.c | 5 +++++ fs/xfs/xfs_buf_item_recover.c | 9 +++------ fs/xfs/xfs_inode_item.c | 8 +++++++- fs/xfs/xfs_trans.c | 2 +- 5 files changed, 18 insertions(+), 9 deletions(-) -- 2.31.1

1 5

[PATCH OLK-6.6 v2 0/3] Introduce NEON implementation of CRCT10DIF algorithm
by GUO Zihua 04 Dec '23

04 Dec '23

This patchset introduces ARM64 NEON implementation of CRC T10 DIF algorithm. This algorithm could be enabled by enabling CONFIG_CRYPTO_CRCT10DIF_ARM64_NEON, and the compiled KO is named crct10dif-neon.ko GUO Zihua (2): arm64/crypto: Introduce individual config for CRCT10DIF-NEON algorithm crypto: crct10dif-neon - fix use via crypto_shash_digest() YueHaibing (1): arm64/crypto: Accelerated CRC T10 DIF computation arch/arm64/crypto/Kconfig | 10 + arch/arm64/crypto/Makefile | 4 + arch/arm64/crypto/crct10dif-neon-asm_64.S | 752 ++++++++++++++++++++++ arch/arm64/crypto/crct10dif-neon_glue.c | 114 ++++ 4 files changed, 880 insertions(+) create mode 100644 arch/arm64/crypto/crct10dif-neon-asm_64.S create mode 100644 arch/arm64/crypto/crct10dif-neon_glue.c -- 2.34.1

2 4

[PATCH OLK-6.6] arm64: errata: add option to disable cache readunique prefetch on HIP08
by Yang Yingliang 04 Dec '23

04 Dec '23

From: Kai Shen <shenkai8(a)huawei.com> hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I3ZFV2 CVE: NA ----------------------------------------------------------- Random performance decreases appear on cases of Hackbench which test pipe or socket communication among multi-threads on Hisi HIP08 SoC. Cache sharing which caused by the change of the data layout and the cache readunique prefetch mechanism both lead to this problem. Readunique mechanism which may caused by store operation will invalid cachelines on other cores during data fetching stage which can cause cacheline invalidation happens frequently in a sharing data access situation. Disable cache readunique prefetch can trackle this problem. Test cases are like: for i in 20;do echo "--------pipe thread num=$i----------" for j in $(seq 1 10);do ./hackbench -pipe $i thread 1000 done done We disable readunique prefetch only in el2 for in el1 disabling readunique prefetch may cause panic due to lack of related priority which often be set in BIOS. Introduce CONFIG_HISILICON_ERRATUM_HIP08_RU_PREFETCH and disable RU prefetch using boot cmdline 'readunique_prefetch=off'. Signed-off-by: Kai Shen <shenkai8(a)huawei.com> Signed-off-by: Hanjun Guo <guohanjun(a)huawei.com> [XQ: adjusted context] Signed-off-by: Xie XiuQi <xiexiuqi(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Cheng Jian <cj.chengjian(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/Kconfig | 18 +++++++++++ arch/arm64/kernel/cpu_errata.c | 56 ++++++++++++++++++++++++++++++++++ arch/arm64/tools/cpucaps | 1 + 3 files changed, 75 insertions(+) diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 78f20e632712..bbe85ca456c9 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -1229,6 +1229,24 @@ config SOCIONEXT_SYNQUACER_PREITS If unsure, say Y. +config HISILICON_ERRATUM_HIP08_RU_PREFETCH + bool "HIP08 RU: HiSilicon HIP08 cache readunique might cause performance drop" + default y + help + The HiSilicon HIP08 cache readunique might compromise performance, + use cmdline "readunique_prefetch_disable" to disable RU prefetch. + + If unsure, say Y. + +config HISILICON_HIP08_RU_PREFETCH_DEFAULT_OFF + bool "HIP08 RU: disable HiSilicon HIP08 cache readunique by default" + depends on HISILICON_ERRATUM_HIP08_RU_PREFETCH + default n + help + Disable HiSilicon HIP08 cache readunique by default. + + If unsure, say N. + endmenu # "ARM errata workarounds via the alternatives framework" choice diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c index 5706e74c5578..6a2ae67e85f7 100644 --- a/arch/arm64/kernel/cpu_errata.c +++ b/arch/arm64/kernel/cpu_errata.c @@ -13,6 +13,11 @@ #include <asm/cpufeature.h> #include <asm/kvm_asm.h> #include <asm/smp_plat.h> +#ifdef CONFIG_HISILICON_ERRATUM_HIP08_RU_PREFETCH +#include <asm/ptrace.h> +#include <asm/sysreg.h> +#include <linux/smp.h> +#endif static bool __maybe_unused is_affected_midr_range(const struct arm64_cpu_capabilities *entry, int scope) @@ -121,6 +126,48 @@ cpu_enable_cache_maint_trap(const struct arm64_cpu_capabilities *__unused) sysreg_clear_set(sctlr_el1, SCTLR_EL1_UCI, 0); } +#ifdef CONFIG_HISILICON_ERRATUM_HIP08_RU_PREFETCH +# ifdef CONFIG_HISILICON_HIP08_RU_PREFETCH_DEFAULT_OFF +static bool readunique_prefetch_enabled; +# else +static bool readunique_prefetch_enabled = true; +# endif +static int __init readunique_prefetch_switch(char *data) +{ + if (!data) + return -EINVAL; + + if (strcmp(data, "off") == 0) + readunique_prefetch_enabled = false; + else if (strcmp(data, "on") == 0) + readunique_prefetch_enabled = true; + else + return -EINVAL; + + return 0; +} +early_param("readunique_prefetch", readunique_prefetch_switch); + +static bool +should_disable_hisi_hip08_ru_prefetch(const struct arm64_cpu_capabilities *entry, int unused) +{ + u64 el; + + if (readunique_prefetch_enabled) + return false; + + el = read_sysreg(CurrentEL); + return el == CurrentEL_EL2; +} + +#define CTLR_HISI_HIP08_RU_PREFETCH (1L << 40) +static void __maybe_unused +hisi_hip08_ru_prefetch_disable(const struct arm64_cpu_capabilities *__unused) +{ + sysreg_clear_set(S3_1_c15_c6_4, 0, CTLR_HISI_HIP08_RU_PREFETCH); +} +#endif + static DEFINE_RAW_SPINLOCK(reg_user_mask_modification); static void __maybe_unused cpu_clear_bf16_from_user_emulation(const struct arm64_cpu_capabilities *__unused) @@ -744,6 +791,15 @@ const struct arm64_cpu_capabilities arm64_errata[] = { .capability = ARM64_WORKAROUND_AMPERE_AC03_CPU_38, ERRATA_MIDR_ALL_VERSIONS(MIDR_AMPERE1), }, +#endif +#ifdef CONFIG_HISILICON_ERRATUM_HIP08_RU_PREFETCH + { + .desc = "HiSilicon HIP08 Cache Readunique Prefetch Disable", + .capability = ARM64_WORKAROUND_HISI_HIP08_RU_PREFETCH, + ERRATA_MIDR_ALL_VERSIONS(MIDR_HISI_TSV110), + .matches = should_disable_hisi_hip08_ru_prefetch, + .cpu_enable = hisi_hip08_ru_prefetch_disable, + }, #endif { } diff --git a/arch/arm64/tools/cpucaps b/arch/arm64/tools/cpucaps index dea3dc89234b..bbb97036074b 100644 --- a/arch/arm64/tools/cpucaps +++ b/arch/arm64/tools/cpucaps @@ -100,3 +100,4 @@ WORKAROUND_NVIDIA_CARMEL_CNP WORKAROUND_QCOM_FALKOR_E1003 WORKAROUND_REPEAT_TLBI WORKAROUND_SPECULATIVE_AT +WORKAROUND_HISI_HIP08_RU_PREFETCH -- 2.25.1

1 0

[PATCH OLK-6.6 v2 0/4] add redis sockmap sample code
by Zhengchao Shao 04 Dec '23

04 Dec '23

add redis sockmap sample code. Liu Jian (4): cgroup: make cgroup_bpf_prog_attach work when cgroup2 is not mounted net: let sockops can use bpf_get_current_comm() net: add local_skb parameter to identify local tcp connection tools: add sample sockmap code for redis include/linux/cgroup.h | 1 + include/linux/filter.h | 3 + include/linux/skbuff.h | 5 + include/uapi/linux/bpf.h | 1 + kernel/bpf/cgroup.c | 8 +- kernel/cgroup/cgroup.c | 26 +++ net/Kconfig | 5 + net/core/filter.c | 13 ++ net/ipv4/tcp_input.c | 6 +- net/ipv4/tcp_output.c | 9 + tools/include/uapi/linux/bpf.h | 1 + tools/netacc/Makefile | 24 +++ tools/netacc/bpf_sockmap.h | 167 +++++++++++++++++++ tools/netacc/net-acc | 35 ++++ tools/netacc/netacc.c | 296 +++++++++++++++++++++++++++++++++ tools/netacc/netaccsockmap.c | 176 ++++++++++++++++++++ 16 files changed, 771 insertions(+), 5 deletions(-) create mode 100644 tools/netacc/Makefile create mode 100644 tools/netacc/bpf_sockmap.h create mode 100755 tools/netacc/net-acc create mode 100644 tools/netacc/netacc.c create mode 100644 tools/netacc/netaccsockmap.c -- 2.34.1

2 5

[PATCH OLK-6.6 0/2] xfs: fix intent item leak during reovery
by Long Li 04 Dec '23

04 Dec '23

Long Li (2): xfs: factor out xfs_defer_pending_abort xfs: abort intent items when recovery intents fail fs/xfs/libxfs/xfs_defer.c | 28 ++++++++++++++++++---------- fs/xfs/libxfs/xfs_defer.h | 2 +- fs/xfs/xfs_log_recover.c | 2 +- 3 files changed, 20 insertions(+), 12 deletions(-) -- 2.31.1

1 0

[PATCH OLK-5.10] vdpa: add vmstate header file
by Jiang Dongxu 04 Dec '23

04 Dec '23

From: jiangdongxu <jiangdongxu1(a)huawei.com> virt inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I86ITO ---------------------------------------------------------------------- Add the vmstate.h header file to define the structure of vdpa device status to achieve live migration of cross-vendor vdpa devices. Signed-off-by: jiangdongxu <jiangdongxu1(a)huawei.com> --- include/linux/vdpa_vmstate.h | 182 +++++++++++++++++++++++++++++++++++ 1 file changed, 182 insertions(+) create mode 100644 include/linux/vdpa_vmstate.h diff --git a/include/linux/vdpa_vmstate.h b/include/linux/vdpa_vmstate.h new file mode 100644 index 000000000000..0e577a9605f1 --- /dev/null +++ b/include/linux/vdpa_vmstate.h @@ -0,0 +1,182 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + * + * Description: vdpa vmstate header file + * Author: jiangdongxu + * Create: 2023-12-3 + * Note: + * History: 2023-12-3: Create file + */ + +#ifndef VDPA_VMSTATE_H +#define VDPA_VMSTATE_H + +#include <linux/virtio_net.h> +#include <linux/virtio_scsi.h> + + +#define VIRTIO_MIG_STATE_TYPE_DEVICE 0 +#define VIRTIO_MIG_STATE_TYPE_VQ 1 +#define VIRTIO_MIG_STATE_TYPE_CONFIG 2 +#define VIRTIO_MIG_STATE_TYPE_FEATURE 3 +#define VIRTIO_MIG_STATE_TYPE_PLATFORM 4 +#define VIRTIO_MIG_STATE_TYPE_VENDOR 255 + +#define VIRTIO_MIG_DEVICE_T_COMMON 0 +#define VIRTIO_MIG_DEVICE_T_NET 1 +#define VIRTIO_MIG_DEVICE_T_BLK 2 +#define VIRTIO_MIG_DEVICE_T_SCSI 8 + +#define VIRTIO_MIG_CONFIG_T_NET 1 +#define VIRTIO_MIG_CONFIG_T_BLK 2 +#define VIRTIO_MIG_CONFIG_T_SCSI 8 + +#define MAC_LEN 6 + +struct virtio_mig_state_header { + le32 type; + le32 len; +}; + +struct virtio_mig_dev_common_data { + le32 vendor_id; + le32 device_id; + le32 device_features_l; + le32 device_features_h; + le32 driver_features_l; + le32 driver_features_h; + le32 status; + le32 generation; + le32 msix_en; +}; + +struct virtio_mig_dev_common_state { + struct virtio_mig_state_header hdr; + struct virtio_mig_dev_common_data data; +}; + +struct virtio_mig_vq_split_state { + le16 avail_index; + le16 used_index; +}; +struct virtio_mig_vq_packed_state { + le16 avail_wrapped : 1; + le16 avail_index : 15; + le16 used_wrapped : 1; + le16 used_index : 15; +}; + +struct virtio_mig_per_vq_data { + le32 qsize; + + u8 qenabled; + le16 msix_vector; + + le32 desc_l; + le32 desc_h; + le32 avail_l; + le32 avail_h; + le32 used_l; + le32 used_h; + union { + struct virtio_mig_vq_split_state split; + struct virtio_mig_vq_packed_state packed; + } version; +}; + +/* vq state */ +struct virtio_mig_vq_state { + struct virtio_mig_state_header hdr; + + le16 msix_config; + le16 valid_queues; + + le16 num_queues; + + struct virtio_mig_per_vq_data vq_state[]; +}; + +/* config space */ +struct virtio_mig_config_state { + struct virtio_mig_state_header hdr; + union { + struct virtio_net_config net; + struct virtio_blk_config blk; + struct virtio_scsi_config scsi; + } dev; +}; + +struct virtio_mig_cfg_blk_features { + +}; + +struct virtio_mig_cfg_scsi_features { + +}; + +struct virtio_mig_cfg_net_ctrl_guest_offloads { + struct virtio_mig_state_header hdr; + le64 offloads; + le64 reserved; +}; + +struct virtio_mig_cfg_net_ctrl_mq_vq_pairs { + struct virtio_mig_state_header hdr; + le16 cur_virtqueue_pairs; +}; + +struct virtio_mig_cfg_net_ctrl_mac_table { + struct virtio_mig_state_header hdr; + le16 num_unicast; + /* TODO: need to be implemented later */ + // u8 unicast_macs[][6]; + le16 num_multicast; + /* TODO: need to be implemented later */ + // u8 multicast_macs[][6]; +}; + +struct virtio_mig_cfg_net_ctrl_vlan { + struct virtio_mig_state_header hdr; + le32 vlans[128]; +}; + +struct virtio_mig_cfg_net_data { + le32 nfeatures; + struct virtio_mig_cfg_net_ctrl_guest_offloads offloads; + struct virtio_mig_cfg_net_ctrl_mq_vq_pairs mq_pairs; + struct virtio_mig_cfg_net_ctrl_mac_table mac_table; + struct virtio_mig_cfg_net_ctrl_vlan vlan_table; +}; + +struct virtio_mig_cfg_net_features { + struct virtio_mig_state_header hdr; + struct virtio_mig_cfg_net_data data; +}; + +/* feature */ +struct virtio_mig_feat_state { + union { + struct virtio_mig_cfg_net_features net; + struct virtio_mig_cfg_blk_features blk; + struct virtio_mig_cfg_scsi_features scsi; + }; +}; + +struct vdpa_mig_state { + struct virtio_mig_dev_common_state dev_state; + struct virtio_mig_config_state cfg_state; + struct virtio_mig_feat_state feat_state; + struct virtio_mig_vq_state vq_state; +}; + +#endif /* VDPA_VMSTATE_H */ -- 2.27.0

2 1

[PATCH OLK-6.6 0/2] improve gettimeofday() performance in user space
by Yang Yingliang 04 Dec '23

04 Dec '23

Don't trap kernel to read the counter register, do this in userspace to improve performance. Yang Yingliang (2): arm64: arch_timer: Disable CNTVCT_EL0 trap if workaround is enabled vdso: do cntvct workaround in the VDSO arch/arm64/include/asm/vdso/gettimeofday.h | 30 ++++++++++++++++++++++ drivers/clocksource/arm_arch_timer.c | 20 ++++++++++++--- include/linux/clocksource.h | 2 ++ include/vdso/datapage.h | 6 +++-- kernel/time/vsyscall.c | 8 ++++++ 5 files changed, 61 insertions(+), 5 deletions(-) -- 2.25.1

3 4

[PATCH V4 OLK-5.10] vhost-vdpa: add reset state params to indicate reset level
by Jiang Dongxu 04 Dec '23

04 Dec '23

From: jiangdongxu <jiangdongxu1(a)huawei.com> virt inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I86ITO ---------------------------------------------------------------------- When vdpa hardware is used, some hardware initialization may be required. Currently, qemu connects to vdpa devices through the vhost framework. Since qemu opens the vhost device, the vdpa device cannot sense the action of qemu opening, which may cause the hardware status to be incorrect. Add the interface parameter state to the vdpa reset interface, which respectively identifies the reset when the device is turned on/off and the virtio reset issued by the virtual machine. Signed-off-by: jiangdongxu <jiangdongxu1(a)huawei.com> --- drivers/vdpa/ifcvf/ifcvf_main.c | 2 +- drivers/vdpa/mlx5/net/mlx5_vnet.c | 2 +- drivers/vdpa/vdpa_sim/vdpa_sim.c | 2 +- drivers/vdpa/virtio_pci/vp_vdpa.c | 2 +- drivers/vhost/vdpa.c | 10 +++++----- drivers/virtio/virtio_vdpa.c | 2 +- include/linux/vdpa.h | 16 +++++++++++++--- 7 files changed, 23 insertions(+), 13 deletions(-) diff --git a/drivers/vdpa/ifcvf/ifcvf_main.c b/drivers/vdpa/ifcvf/ifcvf_main.c index 862f4591fff2..2a4991464cfd 100644 --- a/drivers/vdpa/ifcvf/ifcvf_main.c +++ b/drivers/vdpa/ifcvf/ifcvf_main.c @@ -233,7 +233,7 @@ static void ifcvf_vdpa_set_status(struct vdpa_device *vdpa_dev, u8 status) ifcvf_set_status(vf, status); } -static int ifcvf_vdpa_reset(struct vdpa_device *vdpa_dev) +static int ifcvf_vdpa_reset(struct vdpa_device *vdpa_dev, int state) { struct ifcvf_adapter *adapter; struct ifcvf_hw *vf; diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c index 76e7b4bab9d3..c676bb70a453 100644 --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c @@ -1800,7 +1800,7 @@ static void mlx5_vdpa_set_status(struct vdpa_device *vdev, u8 status) ndev->mvdev.status |= VIRTIO_CONFIG_S_FAILED; } -static int mlx5_vdpa_reset(struct vdpa_device *vdev) +static int mlx5_vdpa_reset(struct vdpa_device *vdev, int state) { struct mlx5_vdpa_dev *mvdev = to_mvdev(vdev); struct mlx5_vdpa_net *ndev = to_mlx5_vdpa_ndev(mvdev); diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c index c141f08b8274..be925462922d 100644 --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c @@ -454,7 +454,7 @@ static void vdpasim_set_status(struct vdpa_device *vdpa, u8 status) spin_unlock(&vdpasim->lock); } -static int vdpasim_reset(struct vdpa_device *vdpa) +static int vdpasim_reset(struct vdpa_device *vdpa, int state) { struct vdpasim *vdpasim = vdpa_to_sim(vdpa); diff --git a/drivers/vdpa/virtio_pci/vp_vdpa.c b/drivers/vdpa/virtio_pci/vp_vdpa.c index f4e375b1d903..bc0bdcc7bcea 100644 --- a/drivers/vdpa/virtio_pci/vp_vdpa.c +++ b/drivers/vdpa/virtio_pci/vp_vdpa.c @@ -220,7 +220,7 @@ static void vp_vdpa_set_status(struct vdpa_device *vdpa, u8 status) vp_modern_set_status(mdev, status); } -static int vp_vdpa_reset(struct vdpa_device *vdpa) +static int vp_vdpa_reset(struct vdpa_device *vdpa, int state) { struct vp_vdpa *vp_vdpa = vdpa_to_vp(vdpa); struct virtio_pci_modern_device *mdev = vp_vdpa_to_mdev(vp_vdpa); diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index eec8027dfc4f..eed51d004531 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -212,13 +212,13 @@ static void vhost_vdpa_unsetup_vq_irq(struct vhost_vdpa *v, u16 qid) irq_bypass_unregister_producer(&vq->call_ctx.producer); } -static int vhost_vdpa_reset(struct vhost_vdpa *v) +static int vhost_vdpa_reset(struct vhost_vdpa *v, int state) { struct vdpa_device *vdpa = v->vdpa; v->in_batch = 0; - return vdpa_reset(vdpa); + return vdpa_reset(vdpa, state); } static long vhost_vdpa_bind_mm(struct vhost_vdpa *v) @@ -297,7 +297,7 @@ static long vhost_vdpa_set_status(struct vhost_vdpa *v, u8 __user *statusp) vhost_vdpa_unsetup_vq_irq(v, i); if (status == 0) { - ret = vdpa_reset(vdpa); + ret = vdpa_reset(vdpa, VDPA_DEV_RESET_VIRTIO); if (ret) return ret; } else @@ -1496,7 +1496,7 @@ static int vhost_vdpa_open(struct inode *inode, struct file *filep) return r; nvqs = v->nvqs; - r = vhost_vdpa_reset(v); + r = vhost_vdpa_reset(v, VDPA_DEV_RESET_OPEN); if (r) goto err; @@ -1542,7 +1542,7 @@ static int vhost_vdpa_release(struct inode *inode, struct file *filep) mutex_lock(&d->mutex); filep->private_data = NULL; vhost_vdpa_clean_irq(v); - vhost_vdpa_reset(v); + vhost_vdpa_reset(v, VDPA_DEV_RESET_CLOSE); vhost_dev_stop(&v->vdev); vhost_vdpa_unbind_mm(v); vhost_vdpa_config_put(v); diff --git a/drivers/virtio/virtio_vdpa.c b/drivers/virtio/virtio_vdpa.c index 46c71653f508..8126a9be22cd 100644 --- a/drivers/virtio/virtio_vdpa.c +++ b/drivers/virtio/virtio_vdpa.c @@ -99,7 +99,7 @@ static void virtio_vdpa_reset(struct virtio_device *vdev) { struct vdpa_device *vdpa = vd_get_vdpa(vdev); - vdpa_reset(vdpa); + vdpa_reset(vdpa, VDPA_DEV_RESET_VIRTIO); } static bool virtio_vdpa_notify(struct virtqueue *vq) diff --git a/include/linux/vdpa.h b/include/linux/vdpa.h index 83df2380eeb9..ef53829c7a7a 100644 --- a/include/linux/vdpa.h +++ b/include/linux/vdpa.h @@ -120,6 +120,12 @@ struct vdpa_map_file { u64 offset; }; +enum vdpa_reset_state { + VDPA_DEV_RESET_VIRTIO = 0, + VDPA_DEV_RESET_OPEN = 1, + VDPA_DEV_RESET_CLOSE = 2, +}; + /** * struct vdpa_config_ops - operations for configuring a vDPA device. * Note: vDPA device drivers are required to implement all of the @@ -218,6 +224,10 @@ struct vdpa_map_file { * @status: virtio device status * @reset: Reset device * @vdev: vdpa device + * @state: state for reset + * VDPA_DEV_RESET_VIRTIO for virtio reset + * VDPA_DEV_RESET_OPEN for vhost-vdpa device open + * VDPA_DEV_RESET_CLOSE for vhost-vdpa device close * Returns integer: success (0) or error (< 0) * @suspend: Suspend the device (optional) * @vdev: vdpa device @@ -359,7 +369,7 @@ struct vdpa_config_ops { u32 (*get_vendor_id)(struct vdpa_device *vdev); u8 (*get_status)(struct vdpa_device *vdev); void (*set_status)(struct vdpa_device *vdev, u8 status); - int (*reset)(struct vdpa_device *vdev); + int (*reset)(struct vdpa_device *vdev, int state); int (*suspend)(struct vdpa_device *vdev); int (*resume)(struct vdpa_device *vdev); size_t (*get_config_size)(struct vdpa_device *vdev); @@ -482,14 +492,14 @@ static inline struct device *vdpa_get_dma_dev(struct vdpa_device *vdev) return vdev->dma_dev; } -static inline int vdpa_reset(struct vdpa_device *vdev) +static inline int vdpa_reset(struct vdpa_device *vdev, int state) { const struct vdpa_config_ops *ops = vdev->config; int ret; down_write(&vdev->cf_lock); vdev->features_valid = false; - ret = ops->reset(vdev); + ret = ops->reset(vdev, state); up_write(&vdev->cf_lock); return ret; } -- 2.27.0

2 1

[PATCH OLK-6.6 000/525] Backport 6.6.3 LTS Patches
by Zheng Zengkai 03 Dec '23

03 Dec '23

Backport linux-6.6.3 LTS patches from upstream. git cherry-pick v6.6.2..v6.6.3~1 -s No conflicts. Build and boot test for arm64 & x86 passed. Adrian Hunter (1): perf intel-pt: Fix async branch flags Ajay Singh (1): wifi: wilc1000: use vmm_table as array in wilc struct Al Viro (1): gfs2: fix an oops in gfs2_permission Alain Volmat (1): dmaengine: stm32-mdma: correct desc prep when channel running Alex Deucher (4): drm/amdgpu/smu13: drop compute workload workaround drm/amdgpu: don't use pci_is_thunderbolt_attached() drm/amdgpu: add a retry for IP discovery init drm/amdgpu: don't use ATRM for external devices Alex Spataru (1): ALSA: hda/realtek: Add quirk for ASUS UX7602ZM Alexander Sverdlin (1): net: dsa: lan9303: consequently nested-lock physical MDIO Alexandre Ghiti (1): drivers: perf: Check find_first_bit() return value Alvin Lee (2): drm/amd/display: Blank phantom OTG before enabling drm/amd/display: Don't lock phantom pipe on disabling Amir Goldstein (1): ima: annotate iint mutex to avoid lockdep false positive warnings Anastasia Belova (1): cifs: spnego: add ';' in HOST_KEY_LEN Andreas Gruenbacher (2): gfs2: Silence "suspicious RCU usage in gfs2_permission" warning gfs2: don't withdraw if init_threads() got interrupted Andrew Jones (1): RISC-V: hwprobe: Fix vDSO SIGSEGV Andrew Lunn (1): net: ethtool: Fix documentation of ethtool_sprintf() Andrey Konovalov (1): media: qcom: camss: Fix csid-gen2 for test pattern generator Andrii Nakryiko (4): bpf: Ensure proper register state printing for cond jumps bpf: handle ldimm64 properly in check_cfg() bpf: fix precision backtracking instruction iteration bpf: fix control-flow graph checking in privileged mode Arnd Bergmann (1): drm/i915/mtl: avoid stringop-overflow warning Arseniy Krasnov (1): vsock: read from socket's error queue Axel Lin (1): i2c: sun6i-p2wi: Prevent potential division by zero Baochen Qiang (2): wifi: ath12k: fix possible out-of-bound read in ath12k_htt_pull_ppdu_stats() wifi: ath12k: fix possible out-of-bound write in ath12k_wmi_ext_hal_reg_caps() Baokun Li (1): ext4: fix race between writepages and remount Bartosz Pawlowski (2): PCI: Extract ATS disabling to a helper function PCI: Disable ATS for specific Intel IPU E2000 devices Baruch Siach (2): net: stmmac: fix rx budget limit check net: stmmac: avoid rx queue overrun Bas Nieuwenhuizen (1): drm/amd/pm: Handle non-terminated overdrive commands. Basavaraj Natikar (1): xhci: Enable RPM on controllers that support low-power states Bean Huo (1): mmc: Add quirk MMC_QUIRK_BROKEN_CACHE_FLUSH for Micron eMMC Q2J54A Ben Wolsieffer (1): scripts/gdb/vmalloc: disable on no-MMU Benjamin Bara (2): kernel/reboot: emergency_restart: Set correct system_state i2c: core: Run atomic i2c xfer when !preemptible Biju Das (1): dt-bindings: timer: renesas,rz-mtu3: Fix overflow/underflow interrupt names Billy Tsai (1): i3c: master: mipi-i3c-hci: Fix a kernel panic for accessing DAT_data. Bjorn Helgaas (1): PCI: Use FIELD_GET() in Sapphire RX 5600 XT Pulse quirk Bob Peterson (1): gfs2: ignore negated quota changes Brian Foster (1): ext4: fix racy may inline data check in dio write Brian Geffon (2): PM: hibernate: Use __get_safe_page() rather than touching the list PM: hibernate: Clean up sync_read handling in snapshot_write_next() Bryan O'Donoghue (7): media: qcom: camss: Fix pm_domain_on sequence in probe media: qcom: camss: Fix vfe_get() error jump media: qcom: camss: Fix VFE-17x vfe_disable_output() media: qcom: camss: Fix VFE-480 vfe_disable_output() media: qcom: camss: Fix missing vfe_lite clocks check media: qcom: camss: Fix set CSI2_RX_CFG1_VC_MODE when VC is greater than 3 media: qcom: camss: Fix invalid clock enable bit disjunction Catalin Marinas (1): rcu: kmemleak: Ignore kmemleak false positives when RCU-freeing objects Cezary Rojewski (1): ALSA: hda: Fix possible null-ptr-deref when assigning a stream Chaitanya Kumar Borah (1): drm/i915/mtl: Support HBR3 rate with C10 phy and eDP in MTL Chandradeep Dey (1): ALSA: hda/realtek - Enable internal speaker of ASUS K6500ZC Chandrakanth patil (1): scsi: megaraid_sas: Increase register read retry rount from 3 to 30 for selected registers Chen Yu (1): tools/power/turbostat: Enable the C-state Pre-wake printing Christian König (3): drm/amdgpu: fix error handling in amdgpu_vm_init drm/amdgpu: fix error handling in amdgpu_bo_list_get() drm/amdgpu: lower CS errors to debug severity Christian Marangi (3): net: sfp: add quirk for Fiberstone GPON-ONU-34-20BI cpufreq: stats: Fix buffer overflow detection in trans_stats() leds: trigger: netdev: Move size check in set_device_name Christoph Hellwig (2): riscv: split cache ops out of dma-noncoherent.c blk-mq: make sure active queue usage is held for bio_integrity_prep() Chuck Lever (2): svcrdma: Drop connection after an RDMA Read error NFSD: Update nfsd_cache_append() to use xdr_stream ChunHao Lin (2): r8169: fix network lost after resume on DASH systems r8169: add handling DASH when DASH is disabled Dan Carpenter (5): pwm: Fix double shift bug SUNRPC: Add an IS_ERR() check back to where it was vhost-vdpa: fix use after free in vhost_vdpa_probe() netfilter: nf_tables: fix pointer math issue in nft_byteorder_eval() mmc: vub300: fix an error code Dan Williams (1): cxl/port: Fix delete_endpoint() vs parent unregistration race Darren Hart (1): sbsa_gwdt: Calculate timeout with 64-bit math Dave Chinner (1): xfs: recovery should not clear di_flushiter unconditionally David (Ming Qiang) Wu (1): drm/amdgpu: not to save bo in the case of RAS err_event_athub David Arcari (1): thermal: intel: powerclamp: fix mismatch in get function for max_idle David Howells (1): cifs: Fix encryption of cleared, but unset rq_iter data buffers David Woodhouse (3): hvc/xen: fix console unplug hvc/xen: fix error path in xen_hvc_init() to always register frontend driver hvc/xen: fix event channel handling for secondary consoles Deepak Gupta (1): riscv: VMAP_STACK overflow detection thread-safe Denis Arefev (1): srcu: Fix srcu_struct node grpmask overflow on 64-bit systems Dmitry Antipov (5): wifi: plfxlc: fix clang-specific fortify warning wifi: mac80211_hwsim: fix clang-specific fortify warning wifi: ath9k: fix clang-specific fortify warnings wifi: ath10k: fix clang-specific fortify warning wifi: mt76: fix clang-specific fortify warnings Dominique Martinet (1): 9p: v9fs_listxattr: fix %s null argument warning Douglas Anderson (2): wifi: ath10k: Don't touch the CE interrupt registers after power up kgdb: Flush console before entering kgdb on panic Dust Li (1): net/mlx5e: fix double free of encap_header Ekaterina Esina (1): cifs: fix check of rc in function generate_smb3signingkey Eric Biggers (1): quota: explicitly forbid quota files from being encrypted Eric Dumazet (6): net: annotate data-races around sk->sk_tx_queue_mapping net: annotate data-races around sk->sk_dst_pending_confirm ipvlan: add ipvlan_route_v6_outbound() helper ptp: annotate data-race around q->head and q->tail bonding: stop the device in bond_setup_by_slave() af_unix: fix use-after-free in unix_stream_read_actor() Eymen Yigit (1): ALSA: hda/realtek: Enable Mute LED on HP 255 G8 Fangzhi Zuo (1): drm/amd/display: Fix DSC not Enabled on Direct MST Sink Felix Kuehling (1): drm/amdgpu: Fix possible null pointer dereference Filipe Manana (1): btrfs: abort transaction on generation mismatch when marking eb as dirty Finn Thain (1): sched/core: Optimize in_task() and in_interrupt() a bit Florent Revest (1): mm: make PR_MDWE_REFUSE_EXEC_GAIN an unsigned long Frank Li (6): i3c: master: svc: fix race condition in ibi work thread i3c: master: svc: fix wrong data return when IBI happen during start frame i3c: master: svc: fix ibi may not return mandatory data byte i3c: master: svc: fix check wrong status register in irq handler i3c: master: svc: fix SDA keep low when polling IBIWON timeout happen i3c: master: svc: fix random hot join failure since timeout error Frederic Weisbecker (2): srcu: Only accelerate on enqueue time workqueue: Provide one lock class key per work_on_cpu() callsite Gabe Teeger (1): drm/amd/display: Add Null check for DPP resource Gaosheng Cui (4): apparmor: Fix kernel-doc warnings in apparmor/audit.c apparmor: Fix kernel-doc warnings in apparmor/lib.c apparmor: Fix kernel-doc warnings in apparmor/resource.c apparmor: Fix kernel-doc warnings in apparmor/policy.c Gavin Li (1): net/mlx5e: fix double free of encap_header in update funcs Geliang Tang (1): mptcp: add validity check for sending RM_ADDR Gerhard Engleder (1): tsnep: Fix tsnep_request_irq() format-overflow warning Gregory Greenman (1): wifi: iwlwifi: mvm: fix size check for fw_link_id Guan Wentao (1): Bluetooth: btusb: Add 0bda:b85b for Fn-Link RTL8852BE Gustavo A. R. Silva (2): clk: socfpga: Fix undefined behavior bug in struct stratix10_clock_data clk: visconti: Fix undefined behavior bug in struct visconti_pll_provider Haitao Shan (1): KVM: x86: Fix lapic timer interrupt lost after loading a snapshot. Hans Verkuil (2): media: vivid: avoid integer overflow media: ipu-bridge: increase sensor_name size Hans de Goede (2): gpiolib: acpi: Add a ignore interrupt quirk for Peaq C1010 ASoC: Intel: soc-acpi-cht: Add Lenovo Yoga Tab 3 Pro YT3-X90 quirk Hao Jia (1): sched/core: Fix RQCF_ACT_SKIP leak Hao Sun (1): bpf: Fix check_stack_write_fixed_off() to correctly spill imm Harald Freudenberger (1): s390/ap: fix AP bus crash on early config change callback invocation Hardik Gajjar (1): usb: gadget: f_ncm: Always set current gadget in ncm_bind() Harish Kasiviswanathan (1): drm/amdkfd: ratelimited SQ interrupt messages Harshit Mogalapalli (1): i915/perf: Fix NULL deref bugs with drm_dbg() calls Harshitha Prem (1): wifi: ath12k: Ignore fragments from uninitialized peer in dp Heiko Carstens (3): s390/mm: add missing arch_set_page_dat() call to vmem_crst_alloc() s390/mm: add missing arch_set_page_dat() call to gmap allocations s390/cmma: fix detection of DAT pages Heiner Kallweit (3): PCI/ASPM: Fix L1 substate handling in aspm_attr_store_common() i2c: i801: fix potential race in i801_block_transaction_byte_by_byte Revert "net: r8169: Disable multicast filter for RTL8168H and RTL8107E" Helge Deller (8): fbdev: stifb: Make the STI next font pointer a 32-bit signed offset parisc/agp: Use 64-bit LE values in SBA IOMMU PDIR table parisc/pdc: Add width field to struct pdc_model parisc/power: Add power soft-off when running on qemu parisc: Prevent booting 64-bit kernels on PA1.x machines parisc/pgtable: Do not drop upper 5 address bits of physical address parisc/power: Fix power soft-off when running on qemu parisc: fix mmap_base calculation when stack grows upwards Herve Codina (2): of: address: Fix address translation when address-size is greater than 2 genirq/generic_chip: Make irq_remove_generic_chip() irqdomain aware Hyeongtak Ji (1): mm/damon/core.c: avoid unintentional filtering out of schemes Ilkka Koskinen (1): perf: arm_cspmu: Reject events meant for other PMUs Ilpo Järvinen (17): RDMA/hfi1: Use FIELD_GET() to extract Link Width PCI: tegra194: Use FIELD_GET()/FIELD_PREP() with Link Width fields PCI: mvebu: Use FIELD_PREP() with Link Width atm: iphase: Do PCI error checks on own line PCI: Do error check on own line to split long "if" conditions PCI: Use FIELD_GET() to extract Link Width media: cobalt: Use FIELD_GET() to extract Link Width selftests/resctrl: Fix uninitialized .sa_flags selftests/resctrl: Remove duplicate feature check from CMT test selftests/resctrl: Move _GNU_SOURCE define into Makefile selftests/resctrl: Refactor feature check to use resource and feature name selftests/resctrl: Fix feature checks selftests/resctrl: Reduce failures due to outliers in MBA/MBM tests selftests/resctrl: Remove bw_report and bm_type from main() selftests/resctrl: Simplify span lifetime selftests/resctrl: Make benchmark command const and build it with pointers selftests/resctrl: Extend signal handler coverage to unmount on receiving signal Ingo Rohloff (1): wifi: mt76: mt7921e: Support MT7992 IP in Xiaomi Redmibook 15 Pro (2023) Itamar Gozlan (1): Revert "net/mlx5: DR, Supporting inline WQE when possible" Jacky Bai (1): clocksource/drivers/timer-imx-gpt: Fix potential memory leak Jaegeuk Kim (3): f2fs: do not return EFSCORRUPTED, but try to run online repair f2fs: set the default compress_level on ioctl f2fs: split initial and dynamic conditions for extent_cache Jamie Lentin (1): hid: lenovo: Resend all settings on reset_resume for compact keyboards Jan Kara (1): ext4: properly sync file size update after O_SYNC direct IO Jan Kiszka (2): net: ti: icssg-prueth: Add missing icss_iep_put to error path net: ti: icssg-prueth: Fix error cleanup on failing pruss_request_mem_region Jani Nikula (4): drm/msm/dp: skip validity check for DP CTS EDID checksum drm/mediatek/dp: fix memory leak on ->get_edid callback audio detection drm/mediatek/dp: fix memory leak on ->get_edid callback error path drm: bridge: it66121: ->get_edid callback must not return err pointers Jarkko Nikula (3): mfd: intel-lpss: Add Intel Lunar Lake-M PCI IDs i3c: mipi-i3c-hci: Fix out of bounds access in hci_dma_irq_handler i2c: i801: Add support for Intel Birch Stream SoC Jarkko Sakkinen (1): KEYS: trusted: Rollback init_trusted() consistently Jens Axboe (1): io_uring/fdinfo: remove need for sqpoll lock for thread/pid retrieval Jesse Zhang (1): drm/amdkfd: Fix shift out-of-bounds issue Jian Shen (2): net: hns3: fix add VLAN fail issue net: hns3: fix incorrect capability bit display for copper port Jianbo Liu (1): net/mlx5e: Don't modify the peer sent-to-vport rules for IPSec offload Jijie Shao (2): net: hns3: fix VF reset fail issue net: hns3: fix VF wrong speed and duplex issue Jim Harris (2): cxl/region: Do not try to cleanup after cxl_region_setup_targets() fails cxl/region: Fix x1 root-decoder granularity calculations Jinghao Jia (2): samples/bpf: syscall_tp_user: Rename num_progs into nr_tests samples/bpf: syscall_tp_user: Fix array out-of-bound access Jiri Kosina (2): HID: Add quirk for Dell Pro Wireless Keyboard and Mouse KM5221W Revert "HID: logitech-dj: Add support for a new lightspeed receiver iteration" Joel Fernandes (Google) (2): rcu/tree: Defer setting of jiffies during stall reset rcutorture: Fix stuttering races and other issues Johan Hovold (8): wifi: ath11k: fix temperature event locking wifi: ath11k: fix dfs radar event locking wifi: ath11k: fix htt pktlog locking wifi: ath11k: fix gtk offload status event locking wifi: ath12k: fix htt mlo-offset event locking wifi: ath12k: fix dfs-radar and temperature event locking mfd: qcom-spmi-pmic: Fix reference leaks in revid helper mfd: qcom-spmi-pmic: Fix revid implementation Johannes Weiner (1): sched: psi: fix unprivileged polling against cgroups John Clark (1): arm64: dts: rockchip: Add NanoPC T6 PCIe e-key support John David Anglin (1): parisc: Add nop instructions after TLB inserts John Johansen (4): apparmor: combine common_audit_data and apparmor_audit_data apparmor: rename audit_data->label to audit_data->subj_label apparmor: pass cred through to audit info. apparmor: Fix regression in mount mediation John Stultz (1): locking/ww_mutex/test: Fix potential workqueue corruption Johnathan Mantey (1): Revert ncsi: Propagate carrier gain/loss events to the NCSI controller Jonathan Denose (1): ACPI: EC: Add quirk for HP 250 G7 Notebook PC Josef Bacik (1): btrfs: don't arbitrarily slow down delalloc if we're committing Josh Poimboeuf (2): x86/retpoline: Make sure there are no unconverted return thunks due to KCSAN x86/srso: Move retbleed IBPB check into existing 'has_microcode' code block Joshua Yeong (1): i3c: master: cdns: Fix reading status register Juergen Gross (2): xen/events: avoid using info_for_irq() in xen_send_IPI_one() xen/events: fix delayed eoi list handling Juntong Deng (3): fs/jfs: Add check for negative db_l2nbperpage fs/jfs: Add validity check for db_maxag and db_agpref gfs2: Fix slab-use-after-free in gfs2_qd_dealloc Kailang Yang (1): ALSA: hda/realtek - Add Dell ALC295 to pin fall back table Kathiravan Thirumoorthy (3): clk: qcom: ipq8074: drop the CLK_SET_RATE_PARENT flag from PLL clocks clk: qcom: ipq6018: drop the CLK_SET_RATE_PARENT flag from PLL clocks firmware: qcom_scm: use 64-bit calling convention only when client is 64-bit Kees Cook (2): gcc-plugins: randstruct: Only warn about true flexible arrays randstruct: Fix gcc-plugin performance mode to stay in group Kemeng Shi (4): ext4: correct offset of gdb backup in non meta_bg group to update_backups ext4: correct return value of ext4_convert_meta_bg ext4: remove gdb backup copy for meta bg in setup_new_flex_group_blocks ext4: add missed brelse in update_backups Kent Overstreet (1): lib/generic-radix-tree.c: Don't overflow in peek() Klaus Kudielka (1): net: phylink: initialize carrier state at creation Koichiro Den (2): x86/apic/msi: Fix misconfigured non-maskable MSI quirk iommufd: Fix missing update of domains_itree after splitting iopt_area Konrad Dybcio (3): dt-bindings: phy: qcom,snps-eusb2-repeater: Add magic tuning overrides phy: qualcomm: phy-qcom-eusb2-repeater: Use regmap_fields phy: qualcomm: phy-qcom-eusb2-repeater: Zero out untouched tuning regs Krister Johansen (2): proc: sysctl: prevent aliased sysctls from getting passed to init watchdog: move softlockup_panic back to early_param Krzysztof Kozlowski (2): dt-bindings: serial: fix regex pattern for matching serial node children ASoC: codecs: wsa-macro: fix uninitialized stack variables with name prefix Kumar Kartikeya Dwivedi (1): bpf: Detect IP == ksym.end as part of BPF program Kunwu Chan (1): drm/i915: Fix potential spectre vulnerability Laurentiu Tudor (1): arm64: dts: ls208xa: use a pseudo-bus to constrain usb dma size Lewis Huang (1): drm/amd/display: Change the DMCUB mailbox memory location from FB to inbox Lin.Cao (1): drm/amd: check num of link levels when update pcie param Linkui Xiao (1): netfilter: nf_conntrack_bridge: initialize err to 0 Linus Walleij (5): gpiolib: of: Add quirk for mt2701-cs42448 ASoC sound net: ethernet: cortina: Fix max RX frame define net: ethernet: cortina: Handle large frames net: ethernet: cortina: Fix MTU max setting mtd: cfi_cmdset_0001: Byte swap OTP info Longfang Liu (1): crypto: hisilicon/qm - prevent soft lockup in receive loop Lu Hongfei (1): soc: qcom: pmic: Fix resource leaks in a device_for_each_child_node() loop Lu Jialin (1): crypto: pcrypt - Fix hungtask for PADATA_RESET Lukas Wunner (2): PCI/sysfs: Protect driver's D3cold preference from user space PCI: Lengthen reset delay for VideoPropulsion Torrent QN16e card Ma Jun (1): drm/amd/pm: Fix error of MACO flag setting code Ma Ke (5): wifi: ath12k: mhi: fix potential memory leak in ath12k_mhi_register() drm/panel: fix a possible null pointer dereference drm/panel/panel-tpo-tpg110: fix a possible null pointer dereference drm/radeon: fix a possible null pointer dereference drm/amdgpu/vkms: fix a possible null pointer dereference Maciej S. Szmigiero (1): KVM: x86: Ignore MSR_AMD64_TW_CFG access Maher Sanalla (1): net/mlx5: Free used cpus mask when an IRQ is released Mahmoud Adam (1): nfsd: fix file memleak on client_opens_release Manas Ghandat (2): jfs: fix array-index-out-of-bounds in dbFindLeaf jfs: fix array-index-out-of-bounds in diAlloc Manivannan Sadhasivam (2): scsi: ufs: qcom: Update PHY settings only when scaling to higher gears PCI: qcom-ep: Add dedicated callback for writing to DBI2 registers Marco Elver (1): 9p/trans_fd: Annotate data-racy writes to file::f_flags Maria Yu (1): arm64: module: Fix PLT counting when CONFIG_RANDOMIZE_BASE=n Mario Limonciello (6): drm/amd: Update `update_pcie_parameters` functions to use uint8_t arguments drm/amd: Fix UBSAN array-index-out-of-bounds for SMU7 drm/amd: Fix UBSAN array-index-out-of-bounds for Polaris and Tonga drm/amd: Disable PP_PCIE_DPM_MASK when dynamic speed switching not supported x86/PCI: Avoid PME from D3hot/D3cold for AMD Rembrandt and Phoenix USB4 drm/amd/display: fix a NULL pointer dereference in amdgpu_dm_i2c_xfer() Marios Makassikis (1): ksmbd: fix recursive locking in vfs helpers Mark Brown (1): regmap: Ensure range selector registers are updated after cache sync Mark Hasemeyer (1): spi: Fix null dereference on suspend Masami Hiramatsu (Google) (1): tracing: fprobe-event: Fix to check tracepoint event and return Masum Reza (1): Bluetooth: btusb: Add RTW8852BE device 13d3:3570 to device tables Matus Malych (1): ALSA: hda/realtek: Enable Mute LED on HP 255 G10 Max Kellermann (1): ext4: apply umask if ACL support is disabled Maíra Canal (1): pmdomain: bcm: bcm2835-power: check if the ASB register is equal to enable Michal Suchanek (1): integrity: powerpc: Do not select CA_MACHINE_KEYRING Mika Westerberg (1): thunderbolt: Apply USB 3.x bandwidth quirk only in software connection manager Mike Rapoport (IBM) (1): x86/mm: Drop the 4 MB restriction on minimal NUMA node memory size Mikhail Khvainitski (1): HID: lenovo: Detect quirk-free fw on cptkbd and stop applying workaround Mikulas Patocka (3): dm crypt: account large pages in cc->n_allocated_pages dm-bufio: fix no-sleep mode dm-verity: don't use blocking calls from tasklets Mimi Zohar (1): ima: detect changes to the backing overlay file Minda Chen (1): riscv: Using TOOLCHAIN_HAS_ZIHINTPAUSE marco replace zihintpause Miri Korenblit (1): wifi: iwlwifi: Use FW rate for non-data frames Muhammad Ahmed (1): drm/amd/display: enable dsc_clk even if dsc_pg disabled Muhammad Usama Anjum (1): tty/sysrq: replace smp_processor_id() with get_cpu() Nam Cao (3): riscv: provide riscv-specific is_trap_insn() riscv: put interrupt entries into .irqentry.text riscv: kprobes: allow writing to x0 Namjae Jeon (2): ksmbd: handle malformed smb1 message ksmbd: fix slab out of bounds write in smb_inherit_dacl() Naohiro Aota (1): btrfs: zoned: wait for data BG to be finished on direct IO allocation Naomi Chu (1): scsi: ufs: core: Expand MCQ queue slot to DeviceQueueDepth + 1 Nathan Chancellor (2): arm64: Restrict CPU_BIG_ENDIAN to GNU as or LLVM IAS 15.x or newer LoongArch: Mark __percpu functions as always inline Neil Armstrong (1): usb: ucsi: glink: use the connector orientation GPIO to provide switch events Nicholas Kazlauskas (1): drm/amd/display: Guard against invalid RPTR/WPTR being set Nicholas Piggin (1): powerpc/perf: Fix disabling BHRB and instruction sampling Nicolas Saenz Julienne (1): KVM: x86: hyper-v: Don't auto-enable stimer on write from user-space Nirmoy Das (2): drm/i915/tc: Fix -Wformat-truncation in intel_tc_port_init drm/i915: Flush WC GGTT only on required platforms Nitin Yadav (1): mmc: sdhci_am654: fix start loop index for TAP value parsing Ojaswin Mujoo (1): ext4: mark buffer new if it is unwritten to avoid stale data exposure Olga Kornievskaia (2): NFSv4.1: fix handling NFS4ERR_DELAY when testing for session trunking NFSv4.1: fix SP4_MACH_CRED protection for pnfs IO Olli Asikainen (1): platform/x86: thinkpad_acpi: Add battery quirk for Thinkpad X120e Ondrej Jirman (1): drm/panel: st7703: Pick different reset sequence Ondrej Mosnacek (2): lsm: fix default return value for vm_enough_memory lsm: fix default return value for inode_getsecctx Pablo Neira Ayuso (3): netfilter: nf_tables: bogus ENOENT when destroying element which does not exist netfilter: nf_tables: remove catchall element in GC sync path netfilter: nf_tables: split async and sync catchall in two functions Paolo Abeni (3): mptcp: deal with large GSO size mptcp: fix setsockopt(IP_TOS) subflow locking selftests: mptcp: fix fastclose with csum failure Paul E. McKenney (1): torture: Make torture_hrtimeout_ns() take an hrtimer mode parameter Paul Hsieh (1): drm/amd/display: Clear dpcd_sink_ext_caps if not set Paul Moore (2): audit: don't take task_lock() in audit_exe_compare() code path audit: don't WARN_ON_ONCE(!current->mm) in audit_exe_compare() Paulo Alcantara (4): smb: client: fix use-after-free bug in cifs_debug_data_proc_show() smb: client: fix use-after-free in smb2_query_info_compound() smb: client: fix potential deadlock when releasing mids smb: client: fix mount when dns_resolver key is not available Pavel Krasavin (1): tty: serial: meson: fix hard LOCKUP on crtscts mode Pengfei Li (1): pmdomain: imx: Make imx pgc power domain also set the fwnode Peter Wang (1): scsi: ufs: core: Fix racing issue between ufshcd_mcq_abort() and ISR Peter Zijlstra (1): perf/core: Fix cpuctx refcounting Petr Tesarik (2): swiotlb: do not free decrypted pages if dynamic swiotlb: fix out-of-bounds TLB allocations with CONFIG_SWIOTLB_DYNAMIC Philipp Stanner (6): string.h: add array-wrappers for (v)memdup_user() kernel: kexec: copy user-array safely kernel: watch_queue: copy user-array safely drm_lease.c: copy user-array safely drm: vmwgfx_surface.c: copy user-array safely i2c: dev: copy userspace array safely Pierre-Louis Bossart (1): soundwire: dmi-quirks: update HP Omen match Ping-Ke Shih (1): wifi: mac80211: don't return unset power in ieee80211_get_tx_power() Pu Wen (1): x86/cpu/hygon: Fix the CPU topology evaluation for real Qu Huang (1): drm/amdgpu: Fix a null pointer access when the smc_rreg pointer is NULL Quinn Tran (1): scsi: qla2xxx: Fix system crash due to bad pointer access Rahul Rameshbabu (7): net/mlx5: Decouple PHC .adjtime and .adjphase implementations net/mlx5e: Avoid referencing skb after free-ing in drop path of mlx5e_sq_xmit_wqe net/mlx5e: Track xmit submission to PTP WQ after populating metadata map net/mlx5e: Update doorbell for port timestamping CQ before the software counter net/mlx5: Increase size of irq name buffer net/mlx5e: Check return value of snprintf writing to fw_version buffer net/mlx5e: Check return value of snprintf writing to fw_version buffer for representors Rajeshwar R Shinde (1): media: gspca: cpia1: shift-out-of-bounds in set_flicker Raju Lakkaraju (1): net: sfp: add quirk for FS's 2.5G copper SFP Ran Xiaokai (1): cpu/hotplug: Don't offline the last non-isolated CPU Rander Wang (1): ASoC: SOF: ipc4: handle EXCEPTION_CAUGHT notification from firmware Ranjan Kumar (1): scsi: mpt3sas: Fix loop logic Ricardo Cañuelo (1): selftests/lkdtm: Disable CONFIG_UBSAN_TRAP in test config Richard Fitzgerald (4): ASoC: soc-card: Add storage for PCI SSID ASoC: SOF: Pass PCI SSID to machine driver ASoC: Intel: sof_sdw: Copy PCI SSID to struct snd_soc_card ASoC: cs35l56: Use PCI SSID as the firmware UID Rick Edgecombe (1): x86/shstk: Delay signal entry SSP write until after user accesses Rik van Riel (1): smp,csd: Throw an error if a CSD lock is stuck for too long Robert Marko (1): Revert "i2c: pxa: move to generic GPIO recovery" Roger Pau Monne (1): acpi/processor: sanitize _OSC/_PDC capabilities for Xen dom0 Roman Gushchin (1): mm: kmem: drop __GFP_NOFAIL when allocating objcg vectors Ronald Wahl (1): clocksource/drivers/timer-atmel-tcb: Fix initialization on SAM9 hardware Rong Chen (1): mmc: meson-gx: Remove setting of CMD_CFG_ERROR Roxana Nicolescu (1): crypto: x86/sha - load modules based on CPU features Saeed Mahameed (1): net/mlx5e: Reduce the size of icosq_str Sakari Ailus (2): media: ccs: Fix driver quirk struct documentation media: ccs: Correctly initialise try compose rectangle Samson Tam (1): drm/amd/display: fix num_ways overflow error Sanjuán García, Jorge (1): mcb: fix error handling for different scenarios when parsing Saravana Kannan (1): driver core: Release all resources during unbind before updating device links Sean Young (2): media: lirc: drop trailing space from scancode transmit media: sharp: fix sharp encoding SeongJae Park (9): mm/damon/sysfs: remove requested targets when online-commit inputs mm/damon/sysfs: update monitoring target regions for online input commit mm/damon/lru_sort: avoid divide-by-zero in hot threshold calculation mm/damon/ops-common: avoid divide-by-zero during region hotness calculation mm/damon: implement a function for max nr_accesses safe calculation mm/damon/core: avoid divide-by-zero during monitoring results update mm/damon/sysfs-schemes: handle tried region directory allocation failure mm/damon/sysfs-schemes: handle tried regions sysfs directory allocation failure mm/damon/sysfs: check error from damon_sysfs_update_target() Shannon Nelson (2): pds_core: use correct index to mask irq pds_core: fix up some format-truncation complaints Shigeru Yoshida (2): tty: Fix uninit-value access in ppp_sync_receive() tipc: Fix kernel-infoleak due to uninitialized TLV value Shiju Jose (1): ACPI: APEI: Fix AER info corruption when error status data has multiple sections Shuai Xue (1): perf/core: Bail out early if the request AUX area is out of bound Shung-Hsi Yu (1): bpf: Fix precision tracking for BPF_ALU | BPF_TO_BE | BPF_END Shyam Prasad N (4): cifs: reconnect helper should set reconnect for the right channel cifs: force interface update before a fresh session setup cifs: do not reset chan_max if multichannel is not supported at mount cifs: do not pass cifs_sb when trying to add channels Sieng-Piaw Liew (1): atl1c: Work around the DMA RX overflow issue Song Shuai (2): riscv: mm: Update the comment of CONFIG_PAGE_OFFSET riscv: correct pt_level name via pgtable_l5/4_enabled Srinivas Kandagatla (2): ASoC: soc-dai: add flag to mute and unmute stream during trigger ASoC: codecs: wsa883x: make use of new mute_unmute_on_trigger flag Stanislav Fomichev (1): net: set SOCK_RCU_FREE before inserting socket into hashtable Stanley Chang (1): usb: dwc3: core: configure TX/RX threshold for DWC3_IP Stanley.Yang (1): drm/amdgpu: Fix potential null pointer derefernce Stefan Binding (1): ALSA: hda/realtek: Add quirks for HP Laptops Stefan Roesch (1): mm: fix for negative counter: nr_file_hugepages Stefano Garzarella (1): vdpa_sim_blk: allocate the buffer zeroed Steve French (4): smb3: fix creating FIFOs when mounting with "sfu" mount option smb3: fix touch -h of symlink smb3: allow dumping session and tcon id to improve stats analysis and debugging smb3: fix caching of ctime on setxattr Steven Rostedt (Google) (1): tracing: Have the user copy of synthetic event address use correct context Su Hui (1): f2fs: avoid format-overflow warning Sui Jingfeng (1): drm/gma500: Fix call trace when psb_gem_mm_init() fails Sumit Garg (1): KEYS: trusted: tee: Refactor register SHM usage Sven Auhagen (1): net: mvneta: fix calls to page_pool_get_stats Takashi Iwai (2): media: imon: fix access to invalid resource for the second interface ALSA: info: Fix potential deadlock at disconnection Tam Nguyen (1): i2c: designware: Disable TX_EMPTY irq while waiting for block length byte Tao Su (1): KVM: x86: Clear bit12 of ICR after APIC-write VM-exit Tao Zhou (1): drm/amdgpu: update retry times for psp vmbx wait Tianci Yin (1): drm/amd/display: Enable fast plane updates on DCN3.2 and above Tiezhu Yang (1): selftests/clone3: Fix broken test under !CONFIG_TIME_NS Tim Huang (1): drm/amdgpu: fix GRBM read timeout when do mes_self_test Tomeu Vizoso (1): pmdomain: amlogic: Fix mask for the second NNA mem PD domain Tony Lindgren (1): ASoC: ti: omap-mcbsp: Fix runtime PM underflow warnings Trevor Wu (1): ASoC: mediatek: mt8188-mt6359: support dynamic pinctrl Trond Myklebust (1): SUNRPC: ECONNRESET might require a rebind Tyrel Datwyler (1): scsi: ibmvfc: Remove BUG_ON in the case of an empty event pool Tzung-Bi Shih (1): platform/chrome: kunit: initialize lock for fake ec_dev Uwe Kleine-König (4): PCI: keystone: Don't discard .remove() callback PCI: keystone: Don't discard .probe() callback PCI: kirin: Don't discard .remove() callback PCI: exynos: Don't discard .remove() callback Vasily Khoruzhick (1): ACPI: FPDT: properly handle invalid FPDT subtables Victor Shih (3): mmc: sdhci-pci-gli: GL9755: Mask the replay timer timeout of AER mmc: sdhci-pci-gli: A workaround to allow GL9750 to enter ASPM L1.2 mmc: sdhci-pci-gli: GL9750: Mask the replay timer timeout of AER Vignesh Viswanathan (5): arm64: dts: qcom: ipq6018: Fix hwlock index for SMEM arm64: dts: qcom: ipq5332: Fix hwlock index for SMEM arm64: dts: qcom: ipq8074: Fix hwlock index for SMEM arm64: dts: qcom: ipq9574: Fix hwlock index for SMEM arm64: dts: qcom: ipq6018: Fix tcsr_mutex register size Vikash Garodia (4): media: venus: hfi: add checks to perform sanity on queue pointers media: venus: hfi_parser: Add check to keep the number of codecs within range media: venus: hfi: fix the check to handle session buffer requirement media: venus: hfi: add checks to handle capabilities from firmware Ville Syrjälä (3): drm/edid: Fixup h/vsync_end instead of h/vtotal powercap: intel_rapl: Downgrade BIOS locked limits pr_warn() to pr_debug() drm/i915: Bump GLK CDCLK frequency when driving multiple pipes Vincent Whitchurch (1): ARM: 9320/1: fix stack depot IRQ stack filter Vitaly Prosyak (1): drm/amdgpu: fix software pci_unplug on some chips Vlad Buslov (2): net/mlx5e: Fix pedit endianness macvlan: Don't propagate promisc change to lower dev in passthru Wang Jianjian (1): ext4: no need to generate from free list in mballoc Wayne Lin (1): drm/amd/display: Avoid NULL dereference of timing generator Wenchao Hao (1): scsi: libfc: Fix potential NULL pointer dereference in fc_lport_ptp_setup() Wenjing Liu (2): drm/amd/display: add seamless pipe topology transition check drm/amd/display: use full update for clip size increase of large plane source Werner Sembach (1): ACPI: resource: Do IRQ override on TongFang GMxXGxx Wesley Cheng (1): usb: host: xhci: Avoid XHCI resume delay if SSUSB device is not present Willem de Bruijn (1): ppp: limit MRU to 64K Wolfram Sang (1): i2c: fix memleak in i2c_new_client_device() Xiaogang Chen (1): drm/amdkfd: Fix a race condition of vram buffer unref in svm code Xin Long (1): net: sched: do not offload flows with a helper in act_ct Yi Yang (4): tty: vcc: Add check for kstrdup() in vcc_probe() mtd: rawnand: tegra: add missing check for platform_get_irq() mtd: rawnand: intel: check return value of devm_kasprintf() mtd: rawnand: meson: check return value of devm_kasprintf() Yihang Li (1): scsi: hisi_sas: Set debugfs_dir pointer to NULL after removing debugfs Yonglong Liu (3): net: hns3: add barrier in vf mailbox reply process net: hns3: fix out-of-bounds access may occur when coalesce info is read via debugfs net: hns3: fix variable may not initialized problem in hns3_init_mac_addr() Yoshihiro Shimoda (3): PCI: dwc: Add dw_pcie_link_set_max_link_width() PCI: dwc: Add missing PCI_EXP_LNKCAP_MLW handling misc: pci_endpoint_test: Add Device ID for R-Car S4-8 PCIe controller Yu Kuai (1): md: don't rely on 'mddev->pers' to be set in mddev_suspend() Yuezhang Mo (1): exfat: support handle zero-size directory Zhang Rui (1): tools/power/turbostat: Fix a knl bug Zhang Shurong (1): iio: adc: stm32-adc: harden against NULL pointer deref in stm32_adc_probe() Zhang Yi (2): ext4: make sure allocate pending entry not fail ext4: correct the start block of counting reserved clusters ZhengHan Wang (1): Bluetooth: Fix double free in hci_conn_cleanup Zhiguo Niu (2): f2fs: fix error path of __f2fs_build_free_nids f2fs: fix error handling of __get_node_page Zhihao Cheng (1): jbd2: fix potential data lost in recovering journal raced with synchronizing fs bdev Zi Yan (5): fs: use nth_page() in place of direct struct page manipulation mips: use nth_page() in place of direct struct page manipulation mm/cma: use nth_page() in place of direct struct page manipulation mm/hugetlb: use nth_page() in place of direct struct page manipulation mm/memory_hotplug: use pfn math in place of direct struct page manipulation Ziwei Xiao (1): gve: Fixes for napi_poll when budget is 0 Zongmin Zhou (1): drm/qxl: prevent memory leak baozhu.liu (1): drm/komeda: drop all currently held locks if deadlock happens felix (1): SUNRPC: Fix RPC client cleaned up the freed pipefs dentries youwan Wang (1): Bluetooth: btusb: Add date->evt_skb is NULL check zhenwei pi (1): virtio-blk: fix implicit overflow on virtio_max_dma_size zhujun2 (1): selftests/efivarfs: create-read: fix a resource leak .../admin-guide/kernel-parameters.txt | 7 + .../phy/qcom,snps-eusb2-repeater.yaml | 21 ++ .../devicetree/bindings/serial/serial.yaml | 2 +- .../bindings/timer/renesas,rz-mtu3.yaml | 38 ++-- Documentation/i2c/busses/i2c-i801.rst | 1 + arch/arm/include/asm/exception.h | 4 - arch/arm64/Kconfig | 2 + .../arm64/boot/dts/freescale/fsl-ls208xa.dtsi | 46 ++-- arch/arm64/boot/dts/qcom/ipq5332.dtsi | 2 +- arch/arm64/boot/dts/qcom/ipq6018.dtsi | 4 +- arch/arm64/boot/dts/qcom/ipq8074.dtsi | 2 +- arch/arm64/boot/dts/qcom/ipq9574.dtsi | 2 +- .../boot/dts/rockchip/rk3588-nanopc-t6.dts | 28 +++ arch/arm64/kernel/module-plts.c | 6 - arch/loongarch/include/asm/percpu.h | 10 +- arch/mips/mm/cache.c | 2 +- arch/parisc/Kconfig | 6 +- arch/parisc/include/asm/elf.h | 10 +- arch/parisc/include/asm/processor.h | 2 + arch/parisc/include/uapi/asm/pdc.h | 1 + arch/parisc/kernel/entry.S | 88 +++++--- arch/parisc/kernel/head.S | 5 +- arch/parisc/kernel/sys_parisc.c | 2 +- arch/powerpc/perf/core-book3s.c | 5 +- arch/riscv/include/asm/asm-prototypes.h | 1 - arch/riscv/include/asm/asm.h | 22 ++ arch/riscv/include/asm/hwprobe.h | 5 + arch/riscv/include/asm/page.h | 4 +- arch/riscv/include/asm/thread_info.h | 3 - arch/riscv/include/asm/vdso/processor.h | 2 +- arch/riscv/kernel/asm-offsets.c | 1 + arch/riscv/kernel/entry.S | 72 ++----- arch/riscv/kernel/probes/simulate-insn.c | 2 +- arch/riscv/kernel/probes/uprobes.c | 6 + arch/riscv/kernel/traps.c | 36 +--- arch/riscv/kernel/vdso/hwprobe.c | 2 +- arch/riscv/mm/Makefile | 1 + arch/riscv/mm/cache-ops.c | 17 ++ arch/riscv/mm/dma-noncoherent.c | 15 -- arch/riscv/mm/ptdump.c | 3 + arch/s390/mm/gmap.c | 24 ++- arch/s390/mm/page-states.c | 6 +- arch/s390/mm/pgalloc.c | 1 + arch/s390/mm/vmem.c | 8 +- arch/x86/crypto/sha1_ssse3_glue.c | 12 ++ arch/x86/crypto/sha256_ssse3_glue.c | 12 ++ arch/x86/include/asm/acpi.h | 14 ++ arch/x86/include/asm/kvm-x86-ops.h | 1 + arch/x86/include/asm/kvm_host.h | 1 + arch/x86/include/asm/msr-index.h | 1 + arch/x86/include/asm/numa.h | 7 - arch/x86/include/asm/xen/hypervisor.h | 9 + arch/x86/kernel/apic/msi.c | 8 +- arch/x86/kernel/cpu/bugs.c | 4 +- arch/x86/kernel/cpu/hygon.c | 8 +- arch/x86/kernel/signal_64.c | 6 +- arch/x86/kvm/hyperv.c | 10 +- arch/x86/kvm/lapic.c | 30 +-- arch/x86/kvm/vmx/vmx.c | 4 +- arch/x86/kvm/x86.c | 2 + arch/x86/mm/numa.c | 7 - arch/x86/pci/fixup.c | 59 ++++++ block/blk-mq.c | 75 +++---- crypto/pcrypt.c | 4 + drivers/acpi/acpi_fpdt.c | 45 +++- drivers/acpi/apei/ghes.c | 23 +- drivers/acpi/ec.c | 10 + drivers/acpi/resource.c | 12 ++ drivers/atm/iphase.c | 20 +- drivers/base/dd.c | 2 +- drivers/base/regmap/regcache.c | 30 +++ drivers/block/virtio_blk.c | 4 +- drivers/bluetooth/btusb.c | 7 + drivers/char/agp/parisc-agp.c | 16 +- drivers/clk/qcom/gcc-ipq6018.c | 6 - drivers/clk/qcom/gcc-ipq8074.c | 6 - drivers/clk/socfpga/stratix10-clk.h | 4 +- drivers/clk/visconti/pll.h | 4 +- drivers/clocksource/timer-atmel-tcb.c | 1 + drivers/clocksource/timer-imx-gpt.c | 18 +- drivers/cpufreq/cpufreq_stats.c | 14 +- drivers/crypto/hisilicon/qm.c | 2 + drivers/cxl/core/port.c | 34 +-- drivers/cxl/core/region.c | 23 +- drivers/dma/stm32-mdma.c | 4 +- drivers/firmware/qcom_scm.c | 7 + drivers/gpio/gpiolib-acpi.c | 20 ++ drivers/gpio/gpiolib-of.c | 4 + drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c | 5 + drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 6 + drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 13 +- drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 23 +- drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c | 16 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 9 +- drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 7 + drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c | 2 + drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 35 +-- drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c | 5 +- drivers/gpu/drm/amd/amdgpu/psp_v13_0.c | 5 +- .../gpu/drm/amd/amdkfd/kfd_int_process_v10.c | 6 +- .../gpu/drm/amd/amdkfd/kfd_int_process_v11.c | 6 +- .../gpu/drm/amd/amdkfd/kfd_int_process_v9.c | 6 +- drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 13 +- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 24 ++- .../display/amdgpu_dm/amdgpu_dm_mst_types.c | 29 ++- drivers/gpu/drm/amd/display/dc/core/dc.c | 78 +++---- .../gpu/drm/amd/display/dc/core/dc_resource.c | 2 +- .../gpu/drm/amd/display/dc/core/dc_stream.c | 4 +- drivers/gpu/drm/amd/display/dc/dc.h | 5 + drivers/gpu/drm/amd/display/dc/dc_types.h | 1 + .../amd/display/dc/dcn10/dcn10_hw_sequencer.c | 3 +- .../drm/amd/display/dc/dcn20/dcn20_hwseq.c | 10 +- .../drm/amd/display/dc/dcn32/dcn32_hwseq.c | 108 +++++++++- .../drm/amd/display/dc/dcn32/dcn32_hwseq.h | 9 + .../gpu/drm/amd/display/dc/dcn32/dcn32_init.c | 2 + .../gpu/drm/amd/display/dc/inc/hw_sequencer.h | 8 + .../drm/amd/display/dc/link/link_detection.c | 3 + drivers/gpu/drm/amd/display/dmub/dmub_srv.h | 22 +- .../gpu/drm/amd/display/dmub/src/dmub_srv.c | 50 ++++- drivers/gpu/drm/amd/include/pptable.h | 4 +- drivers/gpu/drm/amd/pm/amdgpu_pm.c | 8 +- .../drm/amd/pm/powerplay/hwmgr/pptable_v1_0.h | 16 +- .../drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c | 4 +- drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c | 2 +- drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h | 2 +- drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h | 4 +- .../gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c | 4 +- .../amd/pm/swsmu/smu11/sienna_cichlid_ppt.c | 10 +- .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c | 9 +- .../drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 40 +--- .../drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c | 9 +- .../display/komeda/komeda_pipeline_state.c | 9 +- drivers/gpu/drm/bridge/ite-it66121.c | 4 +- drivers/gpu/drm/drm_edid.c | 18 +- drivers/gpu/drm/drm_lease.c | 4 +- drivers/gpu/drm/gma500/psb_drv.h | 1 + drivers/gpu/drm/gma500/psb_irq.c | 5 + drivers/gpu/drm/i915/display/intel_cdclk.c | 12 ++ drivers/gpu/drm/i915/display/intel_dp.c | 2 +- drivers/gpu/drm/i915/display/intel_tc.c | 11 +- drivers/gpu/drm/i915/gem/i915_gem_context.c | 1 + drivers/gpu/drm/i915/gt/intel_ggtt.c | 35 ++- drivers/gpu/drm/i915/gt/intel_rc6.c | 16 +- drivers/gpu/drm/i915/i915_perf.c | 15 +- drivers/gpu/drm/mediatek/mtk_dp.c | 6 +- drivers/gpu/drm/msm/dp/dp_panel.c | 21 +- drivers/gpu/drm/panel/panel-arm-versatile.c | 2 + drivers/gpu/drm/panel/panel-sitronix-st7703.c | 25 +-- drivers/gpu/drm/panel/panel-tpo-tpg110.c | 2 + drivers/gpu/drm/qxl/qxl_display.c | 3 + drivers/gpu/drm/radeon/radeon_connectors.c | 2 + drivers/gpu/drm/vmwgfx/vmwgfx_surface.c | 4 +- drivers/hid/hid-ids.h | 2 +- drivers/hid/hid-lenovo.c | 118 +++++++---- drivers/hid/hid-logitech-dj.c | 11 +- drivers/hid/hid-quirks.c | 1 + drivers/i2c/busses/Kconfig | 1 + drivers/i2c/busses/i2c-designware-master.c | 19 +- drivers/i2c/busses/i2c-i801.c | 22 +- drivers/i2c/busses/i2c-pxa.c | 76 ++++++- drivers/i2c/busses/i2c-sun6i-p2wi.c | 5 + drivers/i2c/i2c-core-base.c | 13 +- drivers/i2c/i2c-core.h | 2 +- drivers/i2c/i2c-dev.c | 4 +- drivers/i3c/master/i3c-master-cdns.c | 6 +- drivers/i3c/master/mipi-i3c-hci/dat_v1.c | 29 ++- drivers/i3c/master/mipi-i3c-hci/dma.c | 2 +- drivers/i3c/master/svc-i3c-master.c | 54 ++++- drivers/iio/adc/stm32-adc-core.c | 9 +- drivers/infiniband/hw/hfi1/pcie.c | 9 +- drivers/iommu/iommufd/io_pagetable.c | 10 + drivers/leds/trigger/ledtrig-netdev.c | 6 +- drivers/mcb/mcb-core.c | 1 + drivers/mcb/mcb-parse.c | 2 +- drivers/md/dm-bufio.c | 87 +++++--- drivers/md/dm-crypt.c | 15 +- drivers/md/dm-verity-fec.c | 4 +- drivers/md/dm-verity-target.c | 23 +- drivers/md/dm-verity.h | 2 +- drivers/md/md.c | 2 +- drivers/media/i2c/ccs/ccs-core.c | 2 +- drivers/media/i2c/ccs/ccs-quirk.h | 4 +- drivers/media/pci/cobalt/cobalt-driver.c | 11 +- .../platform/qcom/camss/camss-csid-gen2.c | 11 +- .../qcom/camss/camss-csiphy-3ph-1-0.c | 2 +- .../media/platform/qcom/camss/camss-vfe-170.c | 22 +- .../media/platform/qcom/camss/camss-vfe-480.c | 22 +- drivers/media/platform/qcom/camss/camss-vfe.c | 5 +- drivers/media/platform/qcom/camss/camss.c | 12 +- drivers/media/platform/qcom/venus/hfi_msgs.c | 2 +- .../media/platform/qcom/venus/hfi_parser.c | 15 ++ drivers/media/platform/qcom/venus/hfi_venus.c | 10 + drivers/media/rc/imon.c | 6 + drivers/media/rc/ir-sharp-decoder.c | 8 +- drivers/media/rc/lirc_dev.c | 6 +- .../media/test-drivers/vivid/vivid-rds-gen.c | 2 +- drivers/media/usb/gspca/cpia1.c | 3 + drivers/mfd/intel-lpss-pci.c | 13 ++ drivers/mfd/qcom-spmi-pmic.c | 101 ++++++--- drivers/misc/pci_endpoint_test.c | 4 + drivers/mmc/core/block.c | 4 +- drivers/mmc/core/card.h | 4 + drivers/mmc/core/mmc.c | 8 +- drivers/mmc/core/quirks.h | 7 +- drivers/mmc/host/meson-gx-mmc.c | 1 - drivers/mmc/host/sdhci-pci-gli.c | 30 +++ drivers/mmc/host/sdhci_am654.c | 2 +- drivers/mmc/host/vub300.c | 1 + drivers/mtd/chips/cfi_cmdset_0001.c | 20 +- drivers/mtd/nand/raw/intel-nand-controller.c | 10 + drivers/mtd/nand/raw/meson_nand.c | 3 + drivers/mtd/nand/raw/tegra_nand.c | 4 + drivers/net/bonding/bond_main.c | 6 + drivers/net/dsa/lan9303_mdio.c | 4 +- drivers/net/ethernet/amd/pds_core/adminq.c | 2 +- drivers/net/ethernet/amd/pds_core/core.h | 2 +- drivers/net/ethernet/amd/pds_core/dev.c | 8 +- drivers/net/ethernet/amd/pds_core/devlink.c | 2 +- drivers/net/ethernet/atheros/atl1c/atl1c.h | 3 - .../net/ethernet/atheros/atl1c/atl1c_main.c | 67 ++---- drivers/net/ethernet/cortina/gemini.c | 45 ++-- drivers/net/ethernet/cortina/gemini.h | 4 +- drivers/net/ethernet/engleder/tsnep.h | 2 +- drivers/net/ethernet/engleder/tsnep_main.c | 12 +- drivers/net/ethernet/google/gve/gve_main.c | 8 +- drivers/net/ethernet/google/gve/gve_rx.c | 4 - drivers/net/ethernet/google/gve/gve_tx.c | 4 - .../ethernet/hisilicon/hns3/hns3_debugfs.c | 9 +- .../net/ethernet/hisilicon/hns3/hns3_enet.c | 2 +- .../hisilicon/hns3/hns3pf/hclge_main.c | 33 ++- .../hisilicon/hns3/hns3vf/hclgevf_main.c | 25 ++- .../hisilicon/hns3/hns3vf/hclgevf_main.h | 1 + .../hisilicon/hns3/hns3vf/hclgevf_mbx.c | 7 + drivers/net/ethernet/marvell/mvneta.c | 28 ++- .../net/ethernet/mellanox/mlx5/core/en/ptp.c | 20 +- .../mellanox/mlx5/core/en/reporter_rx.c | 4 +- .../ethernet/mellanox/mlx5/core/en/tc_tun.c | 30 ++- .../ethernet/mellanox/mlx5/core/en_ethtool.c | 13 +- .../net/ethernet/mellanox/mlx5/core/en_rep.c | 12 +- .../net/ethernet/mellanox/mlx5/core/en_tc.c | 60 +++--- .../net/ethernet/mellanox/mlx5/core/en_tx.c | 4 +- drivers/net/ethernet/mellanox/mlx5/core/eq.c | 25 ++- .../mellanox/mlx5/core/eswitch_offloads.c | 3 +- .../mellanox/mlx5/core/irq_affinity.c | 42 ---- .../ethernet/mellanox/mlx5/core/lib/clock.c | 7 +- .../net/ethernet/mellanox/mlx5/core/pci_irq.c | 6 +- .../net/ethernet/mellanox/mlx5/core/pci_irq.h | 3 + .../mellanox/mlx5/core/steering/dr_send.c | 115 ++-------- drivers/net/ethernet/realtek/r8169_main.c | 46 ++-- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 3 +- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 15 +- drivers/net/ipvlan/ipvlan_core.c | 41 ++-- drivers/net/macvlan.c | 2 +- drivers/net/phy/phylink.c | 1 + drivers/net/phy/sfp.c | 8 + drivers/net/ppp/ppp_synctty.c | 6 +- drivers/net/wireless/ath/ath10k/debug.c | 2 +- drivers/net/wireless/ath/ath10k/snoc.c | 18 +- drivers/net/wireless/ath/ath11k/dp_rx.c | 8 +- drivers/net/wireless/ath/ath11k/wmi.c | 19 +- drivers/net/wireless/ath/ath12k/dp.c | 1 + drivers/net/wireless/ath/ath12k/dp_rx.c | 33 ++- drivers/net/wireless/ath/ath12k/mhi.c | 11 +- drivers/net/wireless/ath/ath12k/peer.h | 3 + drivers/net/wireless/ath/ath12k/wmi.c | 17 +- drivers/net/wireless/ath/ath9k/debug.c | 2 +- .../net/wireless/ath/ath9k/htc_drv_debug.c | 2 +- drivers/net/wireless/intel/iwlwifi/mvm/link.c | 4 +- drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 14 +- .../net/wireless/mediatek/mt76/mt7915/main.c | 2 +- .../net/wireless/mediatek/mt76/mt7921/pci.c | 2 + .../net/wireless/mediatek/mt76/mt792x_core.c | 2 +- .../net/wireless/mediatek/mt76/mt7996/main.c | 2 +- .../net/wireless/microchip/wilc1000/wlan.c | 2 +- drivers/net/wireless/purelifi/plfxlc/mac.c | 2 +- drivers/net/wireless/virtual/mac80211_hwsim.c | 2 +- drivers/of/address.c | 30 ++- drivers/parisc/power.c | 16 +- drivers/pci/controller/dwc/pci-exynos.c | 4 +- drivers/pci/controller/dwc/pci-keystone.c | 8 +- drivers/pci/controller/dwc/pcie-designware.c | 93 ++++---- drivers/pci/controller/dwc/pcie-kirin.c | 4 +- drivers/pci/controller/dwc/pcie-qcom-ep.c | 17 ++ drivers/pci/controller/dwc/pcie-tegra194.c | 9 +- drivers/pci/controller/pci-mvebu.c | 2 +- drivers/pci/pci-acpi.c | 2 +- drivers/pci/pci-sysfs.c | 10 +- drivers/pci/pci.c | 22 +- drivers/pci/pcie/aer.c | 10 + drivers/pci/pcie/aspm.c | 2 + drivers/pci/probe.c | 6 +- drivers/pci/quirks.c | 53 ++++- drivers/perf/arm_cspmu/arm_cspmu.c | 3 + drivers/perf/riscv_pmu_sbi.c | 5 + .../phy/qualcomm/phy-qcom-eusb2-repeater.c | 131 +++++++++--- drivers/platform/chrome/cros_ec_proto_test.c | 1 + drivers/platform/x86/thinkpad_acpi.c | 1 + drivers/pmdomain/amlogic/meson-ee-pwrc.c | 2 +- drivers/pmdomain/bcm/bcm2835-power.c | 2 +- drivers/pmdomain/imx/gpc.c | 1 + drivers/powercap/intel_rapl_common.c | 2 +- drivers/ptp/ptp_chardev.c | 3 +- drivers/ptp/ptp_clock.c | 5 +- drivers/ptp/ptp_private.h | 8 +- drivers/ptp/ptp_sysfs.c | 3 +- drivers/s390/crypto/ap_bus.c | 4 + drivers/scsi/hisi_sas/hisi_sas_v3_hw.c | 13 +- drivers/scsi/ibmvscsi/ibmvfc.c | 124 ++++++++++- drivers/scsi/libfc/fc_lport.c | 6 + drivers/scsi/megaraid/megaraid_sas_base.c | 4 +- drivers/scsi/mpt3sas/mpt3sas_base.c | 4 +- drivers/scsi/qla2xxx/qla_os.c | 12 +- drivers/soc/qcom/pmic_glink_altmode.c | 30 ++- drivers/soundwire/dmi-quirks.c | 2 +- drivers/spi/spi.c | 56 +++-- drivers/thermal/intel/intel_powerclamp.c | 2 +- drivers/thunderbolt/quirks.c | 3 + drivers/tty/hvc/hvc_xen.c | 39 +++- drivers/tty/serial/meson_uart.c | 14 +- drivers/tty/sysrq.c | 3 +- drivers/tty/vcc.c | 16 +- drivers/ufs/core/ufs-mcq.c | 5 +- drivers/ufs/core/ufshcd.c | 3 +- drivers/ufs/host/ufs-qcom.c | 9 +- drivers/usb/dwc3/core.c | 160 ++++++++++---- drivers/usb/dwc3/core.h | 13 ++ drivers/usb/gadget/function/f_ncm.c | 27 +-- drivers/usb/host/xhci-pci.c | 4 +- drivers/usb/host/xhci.c | 12 +- drivers/usb/typec/ucsi/ucsi_glink.c | 54 ++++- drivers/vdpa/vdpa_sim/vdpa_sim_blk.c | 4 +- drivers/vhost/vdpa.c | 1 - drivers/watchdog/sbsa_gwdt.c | 4 +- drivers/xen/events/events_base.c | 16 +- drivers/xen/pcpu.c | 22 ++ fs/9p/xattr.c | 5 +- fs/btrfs/block-group.c | 4 +- fs/btrfs/ctree.c | 109 +++++----- fs/btrfs/ctree.h | 11 +- fs/btrfs/delalloc-space.c | 3 - fs/btrfs/delayed-inode.c | 2 +- fs/btrfs/dev-replace.c | 2 +- fs/btrfs/dir-item.c | 8 +- fs/btrfs/disk-io.c | 13 +- fs/btrfs/disk-io.h | 3 +- fs/btrfs/extent-tree.c | 36 ++-- fs/btrfs/file-item.c | 17 +- fs/btrfs/file.c | 34 +-- fs/btrfs/free-space-cache.c | 6 +- fs/btrfs/free-space-tree.c | 17 +- fs/btrfs/inode-item.c | 16 +- fs/btrfs/inode.c | 17 +- fs/btrfs/ioctl.c | 4 +- fs/btrfs/qgroup.c | 14 +- fs/btrfs/relocation.c | 10 +- fs/btrfs/root-tree.c | 4 +- fs/btrfs/tests/extent-buffer-tests.c | 6 +- fs/btrfs/tests/inode-tests.c | 12 +- fs/btrfs/tree-log.c | 12 +- fs/btrfs/uuid-tree.c | 6 +- fs/btrfs/volumes.c | 10 +- fs/btrfs/xattr.c | 8 +- fs/exfat/namei.c | 29 ++- fs/ext4/acl.h | 5 + fs/ext4/ext4.h | 3 +- fs/ext4/extents_status.c | 127 +++++++---- fs/ext4/file.c | 169 +++++++-------- fs/ext4/inode.c | 14 +- fs/ext4/mballoc.c | 39 +--- fs/ext4/resize.c | 23 +- fs/ext4/super.c | 14 ++ fs/f2fs/compress.c | 2 +- fs/f2fs/extent_cache.c | 53 ++--- fs/f2fs/file.c | 9 + fs/f2fs/node.c | 18 +- fs/f2fs/xattr.c | 20 +- fs/gfs2/inode.c | 14 +- fs/gfs2/ops_fstype.c | 4 +- fs/gfs2/quota.c | 11 + fs/gfs2/super.c | 12 +- fs/hugetlbfs/inode.c | 4 +- fs/jbd2/recovery.c | 8 + fs/jfs/jfs_dmap.c | 23 +- fs/jfs/jfs_imap.c | 5 +- fs/nfs/nfs4proc.c | 12 +- fs/nfsd/nfs4state.c | 2 +- fs/nfsd/nfscache.c | 23 +- fs/overlayfs/super.c | 2 +- fs/proc/proc_sysctl.c | 8 +- fs/quota/dquot.c | 14 ++ fs/smb/client/cached_dir.c | 84 +++++--- fs/smb/client/cifs_debug.c | 6 + fs/smb/client/cifs_ioctl.h | 6 + fs/smb/client/cifs_spnego.c | 4 +- fs/smb/client/cifsfs.c | 1 + fs/smb/client/cifsglob.h | 12 +- fs/smb/client/cifspdu.h | 2 +- fs/smb/client/cifsproto.h | 9 +- fs/smb/client/connect.c | 17 +- fs/smb/client/dfs.c | 18 +- fs/smb/client/fs_context.h | 1 + fs/smb/client/inode.c | 4 + fs/smb/client/ioctl.c | 25 +++ fs/smb/client/namespace.c | 17 +- fs/smb/client/sess.c | 13 +- fs/smb/client/smb2misc.c | 2 +- fs/smb/client/smb2ops.c | 8 +- fs/smb/client/smb2transport.c | 5 +- fs/smb/client/transport.c | 11 +- fs/smb/client/xattr.c | 5 +- fs/smb/server/smb_common.c | 11 + fs/smb/server/smbacl.c | 29 ++- fs/smb/server/vfs.c | 23 +- fs/xfs/xfs_inode_item_recover.c | 32 +-- include/acpi/ghes.h | 4 + include/linux/bpf.h | 8 +- include/linux/damon.h | 7 + include/linux/ethtool.h | 4 +- include/linux/f2fs_fs.h | 1 + include/linux/generic-radix-tree.h | 7 + include/linux/irq.h | 26 +-- include/linux/lsm_hook_defs.h | 4 +- include/linux/mmc/card.h | 2 + include/linux/msi.h | 6 - include/linux/pci_ids.h | 2 + include/linux/perf_event.h | 13 +- include/linux/preempt.h | 15 +- include/linux/pwm.h | 4 +- include/linux/socket.h | 1 + include/linux/spi/spi.h | 1 + include/linux/string.h | 40 ++++ include/linux/sunrpc/clnt.h | 1 + include/linux/sysctl.h | 6 + include/linux/torture.h | 3 +- include/linux/workqueue.h | 46 +++- include/media/ipu-bridge.h | 2 +- include/net/netfilter/nf_tables.h | 4 +- include/net/sock.h | 26 ++- include/net/tc_act/tc_ct.h | 9 + include/sound/soc-acpi.h | 7 + include/sound/soc-card.h | 37 ++++ include/sound/soc-dai.h | 1 + include/sound/soc.h | 11 + include/sound/sof.h | 8 + include/uapi/linux/prctl.h | 2 +- include/uapi/linux/vm_sockets.h | 17 ++ include/video/sticore.h | 2 +- init/Makefile | 1 + init/main.c | 4 + io_uring/fdinfo.c | 9 +- io_uring/sqpoll.c | 12 +- kernel/audit_watch.c | 9 +- kernel/bpf/core.c | 6 +- kernel/bpf/verifier.c | 83 +++++--- kernel/cgroup/cgroup.c | 12 -- kernel/cpu.c | 11 +- kernel/debug/debug_core.c | 3 + kernel/dma/swiotlb.c | 28 ++- kernel/events/core.c | 17 ++ kernel/events/ring_buffer.c | 6 + kernel/irq/debugfs.c | 1 - kernel/irq/generic-chip.c | 25 ++- kernel/irq/msi.c | 12 +- kernel/kexec.c | 2 +- kernel/locking/test-ww_mutex.c | 20 +- kernel/padata.c | 2 +- kernel/power/snapshot.c | 16 +- kernel/rcu/srcutree.c | 10 +- kernel/rcu/tree.c | 21 ++ kernel/rcu/tree.h | 4 + kernel/rcu/tree_stall.h | 20 +- kernel/reboot.c | 1 + kernel/sched/core.c | 5 +- kernel/smp.c | 13 +- kernel/torture.c | 58 ++--- kernel/trace/trace_events_synth.c | 2 +- kernel/trace/trace_fprobe.c | 9 +- kernel/watch_queue.c | 2 +- kernel/watchdog.c | 7 + kernel/workqueue.c | 20 +- lib/generic-radix-tree.c | 17 +- mm/cma.c | 2 +- mm/damon/core.c | 12 +- mm/damon/lru_sort.c | 4 +- mm/damon/ops-common.c | 5 +- mm/damon/sysfs-schemes.c | 5 + mm/damon/sysfs.c | 89 ++++---- mm/huge_memory.c | 16 +- mm/hugetlb.c | 2 +- mm/memcontrol.c | 3 +- mm/memory_hotplug.c | 2 +- mm/util.c | 10 + net/9p/client.c | 2 +- net/9p/trans_fd.c | 13 +- net/bluetooth/hci_conn.c | 6 +- net/bluetooth/hci_sysfs.c | 23 +- net/bridge/netfilter/nf_conntrack_bridge.c | 2 +- net/core/sock.c | 2 +- net/ipv4/inet_hashtables.c | 2 +- net/ipv4/tcp_output.c | 2 +- net/mac80211/cfg.c | 4 + net/mptcp/pm_netlink.c | 5 +- net/mptcp/protocol.c | 4 + net/mptcp/sockopt.c | 3 + net/ncsi/ncsi-aen.c | 5 - net/netfilter/nf_tables_api.c | 58 +++-- net/netfilter/nft_byteorder.c | 5 +- net/netfilter/nft_meta.c | 2 +- net/sched/act_ct.c | 3 + net/sunrpc/clnt.c | 7 +- net/sunrpc/rpcb_clnt.c | 4 + net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 3 +- net/tipc/netlink_compat.c | 1 + net/unix/af_unix.c | 9 +- net/vmw_vsock/af_vsock.c | 6 + samples/bpf/syscall_tp_user.c | 45 ++-- scripts/Makefile.vmlinux | 1 + scripts/gcc-plugins/randomize_layout_plugin.c | 21 +- scripts/gdb/linux/constants.py.in | 1 + scripts/gdb/linux/vmalloc.py | 8 +- security/apparmor/apparmorfs.c | 11 +- security/apparmor/audit.c | 71 ++++--- security/apparmor/capability.c | 29 +-- security/apparmor/domain.c | 97 ++++++--- security/apparmor/file.c | 199 +++++++++++------- security/apparmor/include/audit.h | 37 ++-- security/apparmor/include/capability.h | 3 +- security/apparmor/include/file.h | 17 +- security/apparmor/include/ipc.h | 4 +- security/apparmor/include/mount.h | 26 ++- security/apparmor/include/net.h | 19 +- security/apparmor/include/perms.h | 4 +- security/apparmor/include/policy.h | 9 +- security/apparmor/include/resource.h | 3 +- security/apparmor/include/task.h | 3 +- security/apparmor/ipc.c | 49 +++-- security/apparmor/lib.c | 50 ++--- security/apparmor/lsm.c | 113 ++++++---- security/apparmor/mount.c | 166 +++++++++------ security/apparmor/net.c | 61 +++--- security/apparmor/policy.c | 73 ++++--- security/apparmor/policy_unpack.c | 29 +-- security/apparmor/resource.c | 54 +++-- security/apparmor/task.c | 58 ++--- security/integrity/Kconfig | 2 - security/integrity/iint.c | 48 ++++- security/integrity/ima/ima_api.c | 5 + security/integrity/ima/ima_main.c | 16 +- security/integrity/integrity.h | 2 + security/keys/trusted-keys/trusted_core.c | 20 +- security/keys/trusted-keys/trusted_tee.c | 64 ++---- sound/core/info.c | 21 +- sound/hda/hdac_stream.c | 6 +- sound/pci/hda/patch_realtek.c | 26 +-- sound/soc/codecs/cs35l56.c | 11 + sound/soc/codecs/lpass-wsa-macro.c | 3 + sound/soc/codecs/wsa883x.c | 7 +- sound/soc/intel/boards/sof_sdw.c | 6 + .../intel/common/soc-acpi-intel-cht-match.c | 43 ++++ sound/soc/mediatek/mt8188/mt8188-mt6359.c | 21 ++ sound/soc/soc-dai.c | 7 + sound/soc/soc-pcm.c | 12 +- sound/soc/sof/ipc4.c | 3 + sound/soc/sof/sof-audio.c | 7 + sound/soc/sof/sof-pci-dev.c | 8 + sound/soc/ti/omap-mcbsp.c | 6 +- tools/include/uapi/linux/prctl.h | 2 +- tools/perf/util/intel-pt.c | 2 + tools/power/x86/turbostat/turbostat.c | 3 +- tools/testing/cxl/test/cxl.c | 2 +- .../selftests/bpf/progs/verifier_loops1.c | 9 +- tools/testing/selftests/bpf/verifier/calls.c | 6 +- .../testing/selftests/bpf/verifier/ld_imm64.c | 8 +- tools/testing/selftests/clone3/clone3.c | 7 +- .../testing/selftests/efivarfs/create-read.c | 2 + tools/testing/selftests/lkdtm/config | 1 - tools/testing/selftests/lkdtm/tests.txt | 2 +- .../testing/selftests/net/mptcp/mptcp_join.sh | 2 +- tools/testing/selftests/resctrl/Makefile | 2 +- tools/testing/selftests/resctrl/cache.c | 5 +- tools/testing/selftests/resctrl/cat_test.c | 21 +- tools/testing/selftests/resctrl/cmt_test.c | 37 ++-- tools/testing/selftests/resctrl/mba_test.c | 6 +- tools/testing/selftests/resctrl/mbm_test.c | 9 +- tools/testing/selftests/resctrl/resctrl.h | 23 +- .../testing/selftests/resctrl/resctrl_tests.c | 132 +++++++----- tools/testing/selftests/resctrl/resctrl_val.c | 36 ++-- tools/testing/selftests/resctrl/resctrlfs.c | 69 +++--- 590 files changed, 6023 insertions(+), 3333 deletions(-) create mode 100644 arch/riscv/mm/cache-ops.c -- 2.20.1

1 525

[PATCH OLK-6.6 v3 0/2] ipvlan: Introduce l2e mode
by Zhengchao Shao 02 Dec '23

02 Dec '23

Inherited features of olk5.10 version. Keefe LIU (2): ipvlan: Introduce l2e mode ipvlan: Introduce local xmit queue for l2e mode drivers/net/ipvlan/ipvlan.h | 7 + drivers/net/ipvlan/ipvlan_core.c | 222 +++++++++++++++++++++++++++++++ drivers/net/ipvlan/ipvlan_main.c | 85 +++++++++++- include/uapi/linux/if_link.h | 1 + 4 files changed, 314 insertions(+), 1 deletion(-) -- 2.34.1

2 3

[PATCH OLK-6.6 v2 0/6] ipvlan: Introduce l2e mode
by Zhengchao Shao 02 Dec '23

02 Dec '23

Inherited features of olk5.10 version. Keefe LIU (4): ipvlan: Introduce l2e mode ipvlan: check the value of module parameter ipvlan_default_mode ipvlan: Introduce local xmit queue for l2e mode ipvlan: limit loop_qlen ranges Keefe Liu (1): ipvlan: limit loop_delay ranges Lu Wei (1): ipvlan: Modify the value of ipvlan modes drivers/net/ipvlan/ipvlan.h | 7 + drivers/net/ipvlan/ipvlan_core.c | 222 +++++++++++++++++++++++++++++++ drivers/net/ipvlan/ipvlan_main.c | 85 +++++++++++- include/uapi/linux/if_link.h | 1 + 4 files changed, 314 insertions(+), 1 deletion(-) -- 2.34.1

2 7

[PATCH OLK-6.6 0/4] add redis sockmap sample code
by Zhengchao Shao 02 Dec '23

02 Dec '23

add redis sockmap sample code. Liu Jian (4): cgroup: make cgroup_bpf_prog_attach work when cgroup2 is not mounted net: let sockops can use bpf_get_current_comm() net: add local_skb parameter to identify local tcp connection tools: add sample sockmap code for redis include/linux/cgroup.h | 1 + include/linux/filter.h | 1 + include/linux/skbuff.h | 4 +- include/uapi/linux/bpf.h | 1 + kernel/bpf/cgroup.c | 8 +- kernel/cgroup/cgroup.c | 22 +++ net/core/filter.c | 9 + net/ipv4/tcp_input.c | 4 +- net/ipv4/tcp_output.c | 4 + tools/include/uapi/linux/bpf.h | 1 + tools/netacc/Makefile | 24 +++ tools/netacc/bpf_sockmap.h | 167 +++++++++++++++++++ tools/netacc/net-acc | 35 ++++ tools/netacc/netacc.c | 296 +++++++++++++++++++++++++++++++++ tools/netacc/netaccsockmap.c | 176 ++++++++++++++++++++ 15 files changed, 747 insertions(+), 6 deletions(-) create mode 100644 tools/netacc/Makefile create mode 100644 tools/netacc/bpf_sockmap.h create mode 100755 tools/netacc/net-acc create mode 100644 tools/netacc/netacc.c create mode 100644 tools/netacc/netaccsockmap.c -- 2.34.1

1 4

[PATCH OLK-6.6 0/5] LoongArch: add old BPI compatibility
by Hongchen Zhang 02 Dec '23

02 Dec '23

add old BPI compatibility for LoongArch machines. Hongchen Zhang (1): LoongArch: use arch specific phys_to_dma liuyun (3): LoongArch: add kernel setvirtmap for runtime LoongArch: Old BPI compatibility LoongArch: Fix virtual machine startup error yangqiming (1): LoongArch: Fixed EIOINTC structure members arch/loongarch/Kconfig | 1 + arch/loongarch/include/asm/addrspace.h | 1 + arch/loongarch/include/asm/efi.h | 1 + arch/loongarch/include/asm/irq.h | 1 + arch/loongarch/include/asm/loongarch.h | 1 + arch/loongarch/kernel/Makefile | 1 + arch/loongarch/kernel/acpi.c | 7 +- arch/loongarch/kernel/dma.c | 27 ++ arch/loongarch/kernel/efi.c | 175 ++++++++- arch/loongarch/kernel/env.c | 6 + arch/loongarch/kernel/irq.c | 25 +- arch/loongarch/kernel/legacy_boot.c | 488 +++++++++++++++++++++++++ arch/loongarch/kernel/legacy_boot.h | 90 +++++ arch/loongarch/kernel/mem.c | 26 +- arch/loongarch/kernel/numa.c | 39 +- arch/loongarch/kernel/reset.c | 3 +- arch/loongarch/kernel/setup.c | 18 +- arch/loongarch/kernel/smp.c | 6 +- arch/loongarch/pci/acpi.c | 148 +++++++- drivers/firmware/efi/Makefile | 1 + drivers/irqchip/irq-loongarch-cpu.c | 7 +- drivers/irqchip/irq-loongson-eiointc.c | 44 ++- drivers/irqchip/irq-loongson-pch-pic.c | 5 + 23 files changed, 1081 insertions(+), 40 deletions(-) create mode 100644 arch/loongarch/kernel/legacy_boot.c create mode 100644 arch/loongarch/kernel/legacy_boot.h -- 2.33.0

2 6

[PATCH OLK-6.6 0/5] LoongArch: add old BPI compatibility
by Hongchen Zhang 02 Dec '23

02 Dec '23

add old BPI compatibility for LoongArch machines. Hongchen Zhang (1): LoongArch: use arch specific phys_to_dma liuyun (3): LoongArch: add kernel setvirtmap for runtime LoongArch: Old BPI compatibility LoongArch: Fix virtual machine startup error yangqiming (1): LoongArch: Fixed EIOINTC structure members arch/loongarch/Kconfig | 1 + arch/loongarch/include/asm/addrspace.h | 1 + arch/loongarch/include/asm/efi.h | 1 + arch/loongarch/include/asm/irq.h | 1 + arch/loongarch/include/asm/loongarch.h | 1 + arch/loongarch/kernel/Makefile | 1 + arch/loongarch/kernel/acpi.c | 7 +- arch/loongarch/kernel/dma.c | 27 ++ arch/loongarch/kernel/efi.c | 175 ++++++++- arch/loongarch/kernel/env.c | 6 + arch/loongarch/kernel/irq.c | 25 +- arch/loongarch/kernel/legacy_boot.c | 488 +++++++++++++++++++++++++ arch/loongarch/kernel/legacy_boot.h | 90 +++++ arch/loongarch/kernel/mem.c | 26 +- arch/loongarch/kernel/numa.c | 39 +- arch/loongarch/kernel/reset.c | 3 +- arch/loongarch/kernel/setup.c | 18 +- arch/loongarch/kernel/smp.c | 6 +- arch/loongarch/pci/acpi.c | 148 +++++++- drivers/firmware/efi/Makefile | 1 + drivers/irqchip/irq-loongarch-cpu.c | 7 +- drivers/irqchip/irq-loongson-eiointc.c | 44 ++- drivers/irqchip/irq-loongson-pch-pic.c | 5 + 23 files changed, 1081 insertions(+), 40 deletions(-) create mode 100644 arch/loongarch/kernel/legacy_boot.c create mode 100644 arch/loongarch/kernel/legacy_boot.h -- 2.33.0

2 6

[PATCH openEuler-22.03-LTS-SP1 0/6] backport Broadcom ethernet driver patch from upstream stable branch linux-5.10.y
by Dapeng Yu 01 Dec '23

01 Dec '23

Backport the latest patches from upstream stable branch linux-5.10.y for Broadcom ethernet driver. Florian Fainelli (2): net: bcmgenet: Remove phy_stop() from bcmgenet_netif_stop() net: bcmgenet: Restore phy_stop() depending upon suspend/close Somnath Kotur (2): bnxt_en: Query default VLAN before VNIC setup on a VF bnxt_en: Implement .set_port / .unset_port UDP tunnel callbacks Sreekanth Reddy (1): bnxt_en: Don't issue AP reset during ethtool's reset operation Tobias Heider (1): Add MODULE_FIRMWARE() for FIRMWARE_TG357766. drivers/net/ethernet/broadcom/bnxt/bnxt.c | 28 ++++++++++++++----- .../net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 2 +- .../net/ethernet/broadcom/genet/bcmgenet.c | 9 +++--- drivers/net/ethernet/broadcom/tg3.c | 1 + 4 files changed, 28 insertions(+), 12 deletions(-) -- 2.43.0

2 8

[PATCH openEuler-22.03-LTS-SP2] Add MODULE_FIRMWARE() for FIRMWARE_TG357766.
by Dapeng Yu 01 Dec '23

01 Dec '23

From: Tobias Heider <me(a)tobhe.de> stable inclusion from stable-v5.10.188 commit c377451012cec94f01b6c698409fec865e55dd3f category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8L4MY CVE: NA backport: OLK-5.10, openEuler-22.03-LTS, openEuler-22.03-LTS-SP1, openEuler-22.03-LTS-SP2 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 046f753da6143ee16452966915087ec8b0de3c70 ] Fixes a bug where on the M1 mac mini initramfs-tools fails to include the necessary firmware into the initrd. Fixes: c4dab50697ff ("tg3: Download 57766 EEE service patch firmware") Signed-off-by: Tobias Heider <me(a)tobhe.de> Reviewed-by: Michael Chan <michael.chan(a)broadcom.com> Link: https://lore.kernel.org/r/ZJt7LKzjdz8+dClx@tobhe.de Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Dapeng Yu <dapeng.yu(a)windriver.com> --- drivers/net/ethernet/broadcom/tg3.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 911c5fcdaf93..7a8f7bc40553 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -224,6 +224,7 @@ MODULE_AUTHOR("David S. Miller (davem(a)redhat.com) and Jeff Garzik (jgarzik@pobox MODULE_DESCRIPTION("Broadcom Tigon3 ethernet driver"); MODULE_LICENSE("GPL"); MODULE_FIRMWARE(FIRMWARE_TG3); +MODULE_FIRMWARE(FIRMWARE_TG357766); MODULE_FIRMWARE(FIRMWARE_TG3TSO); MODULE_FIRMWARE(FIRMWARE_TG3TSO5); -- 2.43.0

1 0

[PATCH OLK-5.10] Add MODULE_FIRMWARE() for FIRMWARE_TG357766.
by Dapeng Yu 01 Dec '23

01 Dec '23

From: Tobias Heider <me(a)tobhe.de> stable inclusion from stable-v5.10.188 commit c377451012cec94f01b6c698409fec865e55dd3f category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8L4MY CVE: NA backport: OLK-5.10, openEuler-22.03-LTS, openEuler-22.03-LTS-SP1, openEuler-22.03-LTS-SP2 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 046f753da6143ee16452966915087ec8b0de3c70 ] Fixes a bug where on the M1 mac mini initramfs-tools fails to include the necessary firmware into the initrd. Fixes: c4dab50697ff ("tg3: Download 57766 EEE service patch firmware") Signed-off-by: Tobias Heider <me(a)tobhe.de> Reviewed-by: Michael Chan <michael.chan(a)broadcom.com> Link: https://lore.kernel.org/r/ZJt7LKzjdz8+dClx@tobhe.de Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Dapeng Yu <dapeng.yu(a)windriver.com> --- drivers/net/ethernet/broadcom/tg3.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 911c5fcdaf93..7a8f7bc40553 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -224,6 +224,7 @@ MODULE_AUTHOR("David S. Miller (davem(a)redhat.com) and Jeff Garzik (jgarzik@pobox MODULE_DESCRIPTION("Broadcom Tigon3 ethernet driver"); MODULE_LICENSE("GPL"); MODULE_FIRMWARE(FIRMWARE_TG3); +MODULE_FIRMWARE(FIRMWARE_TG357766); MODULE_FIRMWARE(FIRMWARE_TG3TSO); MODULE_FIRMWARE(FIRMWARE_TG3TSO5); -- 2.43.0

2 1

[PATCH OLK-6.6 0/4] add redis sockmap sample code
by Zhengchao Shao 01 Dec '23

01 Dec '23

add redis sockmap sample code. Liu Jian (4): cgroup: make cgroup_bpf_prog_attach work when cgroup2 is not mounted net: let sockops can use bpf_get_current_comm() net: add local_skb parameter to identify local tcp connection tools: add sample sockmap code for redis include/linux/cgroup.h | 1 + include/linux/filter.h | 1 + include/linux/skbuff.h | 4 +- include/uapi/linux/bpf.h | 1 + kernel/bpf/cgroup.c | 8 +- kernel/cgroup/cgroup.c | 22 +++ net/core/filter.c | 9 + net/ipv4/tcp_input.c | 4 +- net/ipv4/tcp_output.c | 4 + tools/include/uapi/linux/bpf.h | 1 + tools/netacc/Makefile | 24 +++ tools/netacc/bpf_sockmap.h | 167 +++++++++++++++++++ tools/netacc/net-acc | 35 ++++ tools/netacc/netacc.c | 296 +++++++++++++++++++++++++++++++++ tools/netacc/netaccsockmap.c | 176 ++++++++++++++++++++ 15 files changed, 747 insertions(+), 6 deletions(-) create mode 100644 tools/netacc/Makefile create mode 100644 tools/netacc/bpf_sockmap.h create mode 100755 tools/netacc/net-acc create mode 100644 tools/netacc/netacc.c create mode 100644 tools/netacc/netaccsockmap.c -- 2.34.1

2 6

[PATCH v2 OLK-5.10] config: enable COBFIG_ARM64_BRBE for arm64
by Junhao He 01 Dec '23

01 Dec '23

driver inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8EC9K CVE: NA -------------------------------- The BRBE feature depend on ARMV8 PMUv3 hardware support for branch stack sampling. Signed-off-by: Junhao He <hejunhao3(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 0200cc78ac7b..2258c56af469 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -6179,7 +6179,7 @@ CONFIG_QCOM_L3_PMU=y CONFIG_THUNDERX2_PMU=m CONFIG_XGENE_PMU=y CONFIG_ARM_SPE_PMU=y -# CONFIG_ARM64_BRBE is not set +CONFIG_ARM64_BRBE=y CONFIG_HISI_PMU=m CONFIG_HISI_PCIE_PMU=m CONFIG_HNS3_PMU=m -- 2.33.0

2 1

[PATCH OLK-5.10 1/1] config: enable COBFIG_ARM64_BRBE for arm64
by Junhao He 01 Dec '23

01 Dec '23

driver inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8EC9K CVE: NA -------------------------------- The BRBE feature depend on ARMV8 PMUv3 hardware support for branch stack sampling. Signed-off-by: Junhao He <hejunhao3(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 0200cc78ac7b..2258c56af469 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -6179,7 +6179,7 @@ CONFIG_QCOM_L3_PMU=y CONFIG_THUNDERX2_PMU=m CONFIG_XGENE_PMU=y CONFIG_ARM_SPE_PMU=y -# CONFIG_ARM64_BRBE is not set +CONFIG_ARM64_BRBE=y CONFIG_HISI_PMU=m CONFIG_HISI_PCIE_PMU=m CONFIG_HNS3_PMU=m -- 2.33.0

2 1

[PATCH OLK-5.10 0/4] fix ksmbd to release the ones allocated for async work
by Zizhi Wo 01 Dec '23

01 Dec '23

Fix memory leak and typo. What's more, add release_async_work() to delete asynchronous work. Dawei Li (1): ksmbd: fix typo, syncronous->synchronous Hangyu Hua (1): ksmbd: fix possible memory leak in smb2_lock() Jakob Koschel (1): ksmbd: replace usage of found with dedicated list iterator variable Namjae Jeon (1): ksmbd: delete asynchronous work from list fs/ksmbd/connection.c | 12 ++++---- fs/ksmbd/ksmbd_work.h | 2 +- fs/ksmbd/smb2pdu.c | 66 +++++++++++++++++++++++-------------------- fs/ksmbd/smb2pdu.h | 1 + fs/ksmbd/vfs_cache.c | 5 ++-- 5 files changed, 45 insertions(+), 41 deletions(-) -- 2.39.2

2 5

[PATCH OLK-5.10 0/9] crypto: hisilicon - fix the process to obtain capability register value
by Zhiqi Song 01 Dec '23

01 Dec '23

From: JiangShui Yang <yangjiangshui(a)h-partners.com> This seires patch modify the process to obtain the value of capablity registers. Pre-store the valid values of them. JiangShui Yang (4): Revert "crypto: hisilicon/qm - save capability registers in qm init process" Revert "crypto: hisilicon/hpre - save capability registers in probe process" Revert "crypto: hisilicon/sec2 - save capability registers in probe process" Revert "crypto: hisilicon/zip - save capability registers in probe process" Weili Qian (1): crypto: hisilicon/qm - check function qp num before alg register Zhiqi Song (4): crypto: hisilicon/qm - save capability registers in qm init process crypto: hisilicon/hpre - save capability registers in probe process crypto: hisilicon/sec2 - save capability registers in probe process crypto: hisilicon/zip - save capability registers in probe process drivers/crypto/hisilicon/hpre/hpre_crypto.c | 25 ++++- drivers/crypto/hisilicon/hpre/hpre_main.c | 92 ++++++++++------ drivers/crypto/hisilicon/qm.c | 112 ++++++++++---------- drivers/crypto/hisilicon/sec2/sec.h | 2 +- drivers/crypto/hisilicon/sec2/sec_crypto.c | 32 +++++- drivers/crypto/hisilicon/sec2/sec_main.c | 66 +++++++----- drivers/crypto/hisilicon/zip/zip_crypto.c | 23 +++- drivers/crypto/hisilicon/zip/zip_main.c | 80 ++++++++------ include/linux/hisi_acc_qm.h | 25 ++++- 9 files changed, 302 insertions(+), 155 deletions(-) -- 2.33.0

2 10

[PATCH OLK-5.10 0/4] fix ksmbd to release the ones allocated for async work
by Zizhi Wo 01 Dec '23

01 Dec '23

Fix memory leak and typo. What's more, add release_async_work() to delete asynchronous work. Dawei Li (1): ksmbd: fix typo, syncronous->synchronous Hangyu Hua (1): ksmbd: fix possible memory leak in smb2_lock() Jakob Koschel (1): ksmbd: replace usage of found with dedicated list iterator variable Namjae Jeon (1): ksmbd: delete asynchronous work from list fs/ksmbd/connection.c | 12 ++++---- fs/ksmbd/ksmbd_work.h | 2 +- fs/ksmbd/smb2pdu.c | 66 +++++++++++++++++++++++-------------------- fs/ksmbd/smb2pdu.h | 1 + fs/ksmbd/vfs_cache.c | 5 ++-- 5 files changed, 45 insertions(+), 41 deletions(-) -- 2.39.2

1 4

[PATCH openEuler-1.0-LTS] nvme: retain split access workaround for capability reads
by Li Lingfeng 01 Dec '23

01 Dec '23

From: Ard Biesheuvel <ard.biesheuvel(a)linaro.org> mainline inclusion from mainline-v5.4-rc4 commit 3a8ecc935efabdad106b5e06d07b150c394b4465 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8L166 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… --------------------------- Commit 7fd8930f26be4 "nvme: add a common helper to read Identify Controller data" has re-introduced an issue that we have attempted to work around in the past, in commit a310acd7a7ea ("NVMe: use split lo_hi_{read,write}q"). The problem is that some PCIe NVMe controllers do not implement 64-bit outbound accesses correctly, which is why the commit above switched to using lo_hi_[read|write]q for all 64-bit BAR accesses occuring in the code. In the mean time, the NVMe subsystem has been refactored, and now calls into the PCIe support layer for NVMe via a .reg_read64() method, which fails to use lo_hi_readq(), and thus reintroduces the problem that the workaround above aimed to address. Given that, at the moment, .reg_read64() is only used to read the capability register [which is known to tolerate split reads], let's switch .reg_read64() to lo_hi_readq() as well. This fixes a boot issue on some ARM boxes with NVMe behind a Synopsys DesignWare PCIe host controller. Fixes: 7fd8930f26be4 ("nvme: add a common helper to read Identify Controller data") Signed-off-by: Ard Biesheuvel <ard.biesheuvel(a)linaro.org> Signed-off-by: Sagi Grimberg <sagi(a)grimberg.me> Signed-off-by: Li Lingfeng <lilingfeng3(a)huawei.com> --- drivers/nvme/host/pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index f66c0062736e..85d10d69afe8 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -2428,7 +2428,7 @@ static int nvme_pci_reg_write32(struct nvme_ctrl *ctrl, u32 off, u32 val) static int nvme_pci_reg_read64(struct nvme_ctrl *ctrl, u32 off, u64 *val) { - *val = readq(to_nvme_dev(ctrl)->bar + off); + *val = lo_hi_readq(to_nvme_dev(ctrl)->bar + off); return 0; } -- 2.31.1

2 1

[PATCH openEuler-1.0-LTS] vfs: fix readahead(2) on block devices
by Jinjiang Tu 01 Dec '23

01 Dec '23

From: Reuben Hawkins <reubenhwk(a)gmail.com> stable inclusion from stable-4.19.299 commit 70409dcd485a244b9f508aa6cdb21ac899a85dc1 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8L190 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… ------------------------------------------- [ Upstream commit 7116c0af4b8414b2f19fdb366eea213cbd9d91c2 ] Readahead was factored to call generic_fadvise. That refactor added an S_ISREG restriction which broke readahead on block devices. In addition to S_ISREG, this change checks S_ISBLK to fix block device readahead. There is no change in behavior with any file type besides block devices in this change. Fixes: 3d8f7615319b ("vfs: implement readahead(2) using POSIX_FADV_WILLNEED") Signed-off-by: Reuben Hawkins <reubenhwk(a)gmail.com> Link: https://lore.kernel.org/r/20231003015704.2415-1-reubenhwk@gmail.com Reviewed-by: Amir Goldstein <amir73il(a)gmail.com> Signed-off-by: Christian Brauner <brauner(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> --- mm/readahead.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/mm/readahead.c b/mm/readahead.c index 7a21199c6227..7d78f86fcf6f 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -659,7 +659,8 @@ ssize_t ksys_readahead(int fd, loff_t offset, size_t count) */ ret = -EINVAL; if (!f.file->f_mapping || !f.file->f_mapping->a_ops || - !S_ISREG(file_inode(f.file)->i_mode)) + (!S_ISREG(file_inode(f.file)->i_mode) && + !S_ISBLK(file_inode(f.file)->i_mode))) goto out; ret = vfs_fadvise(f.file, offset, count, POSIX_FADV_WILLNEED); -- 2.25.1

2 1

[PATCH openEuler-20.03-LTS-SP3] Add feature: eNFS - nfs multipath to improve performance and reliability
by mingqian218472 01 Dec '23

01 Dec '23

From: zhangmingqian <zhangmingqian.zhang(a)huawei.com> driver inclusion category: feature bugzilla: https://gitee.com/src-openeuler/kernel/issues/I8KVU1 --------------------------------- Currently, the NFS client can use only one server IP address at a single mount point. As a result, the hardware capability of multiple storage nodes and NICs cannot be fully utilized. In multiple financial sites, the performance cannot meet service requirements. In addition, when a single link is faulty, services are suspended. The reliability problem needs to be solved. OpenEuler-based commercial OS vendors hope that the eNFS feature will be integrated into 20.03 SP4 to resolve performance and reliability problems. When user mount one NFS share, can input localaddrs/remoteaddrs these two optional Parameters to use eNFS multipath. If these optional parameters are not used, NFS will behave as before. For example, mount -t nfs -o [localaddrs=127.17.0.1-127.17.0.4],[remoteaddrs=127.17.1.1-127.17.1.4] xx.xx.xx.xx:/test /mnt/test Changes in eNFS are as follows: 1. patch 0001: At the NFS layer, the eNFS registration function is called back when the mount command parses parameters. The eNFS parses and saves the IP address list entered by users. 2. patch 0002: At the sunrpc layer, the eNFS registration function is called back When the NFS uses sunrpc to create rpc_clnt, the eNFS combines the IP address list entered for mount to generate multiple xprts. When the I/O times out, the callback function of the eNFS is called back so that the eNFS switches to an available link for retry. 3. patch 0003: The eNFS module registers the interface for parsing the mount command. During the mount process, the NFS invokes the eNFS interface to enable the eNFS to parse the mounting parameters of UltraPath. The eNFS module saves the mounting parameters to the context of nfs_client. 4. patch 0004: When the NFS invokes the SunRPC to create rpc_clnt, the eNFS interface is called back. The eNFS creates multiple xprts based on the output IP address list. When NFS V3 I/Os are delivered, eNFS distributes I/Os to available links based on the link status, improving performance through load balancing. 5. patch 0005: When sending I/Os from the SunRPC module to the NFS server times out, the SunRPC module calls back the eNFS module to reselect a link. The eNFS module distributes I/Os to other available links, preventing service interruption caused by a single link failure. 6. patch 0006: The eNFS compilation option and makefile are added. By default, the eNFS compilation is not performed. Signed-off-by: mingqian218472 <zhangmingqian.zhang(a)huawei.com> --- ...enfs_registe_and_handle_mount_option.patch | 794 +++++++ ...nd_create_multipath_then_dispatch_IO.patch | 832 ++++++++ ...add_enfs_module_for_nfs_mount_option.patch | 1208 +++++++++++ ...dd_enfs_module_for_sunrpc_multipatch.patch | 1860 +++++++++++++++++ ...le_for_sunrpc_failover_and_configure.patch | 1600 ++++++++++++++ 0006-add_enfs_compile_option.patch | 70 + 6 files changed, 6364 insertions(+) create mode 100644 0001-nfs_add_api_to_support_enfs_registe_and_handle_mount_option.patch create mode 100644 0002-sunrpc_add_api_to_support_enfs_registe_and_create_multipath_then_dispatch_IO.patch create mode 100644 0003-add_enfs_module_for_nfs_mount_option.patch create mode 100644 0004-add_enfs_module_for_sunrpc_multipatch.patch create mode 100644 0005-add_enfs_module_for_sunrpc_failover_and_configure.patch create mode 100644 0006-add_enfs_compile_option.patch diff --git a/0001-nfs_add_api_to_support_enfs_registe_and_handle_mount_option.patch b/0001-nfs_add_api_to_support_enfs_registe_and_handle_mount_option.patch new file mode 100644 index 0000000..f9e1803 --- /dev/null +++ b/0001-nfs_add_api_to_support_enfs_registe_and_handle_mount_option.patch @@ -0,0 +1,794 @@ +diff --git a/fs/nfs/client.c b/fs/nfs/client.c +index 7d02dc522..b1194a3fb 100644 +--- a/fs/nfs/client.c ++++ b/fs/nfs/client.c +@@ -48,7 +48,7 @@ + #include "callback.h" + #include "delegation.h" + #include "iostat.h" +-#include "internal.h" ++#include "enfs_adapter.h" + #include "fscache.h" + #include "pnfs.h" + #include "nfs.h" +@@ -255,6 +255,7 @@ void nfs_free_client(struct nfs_client *clp) + put_nfs_version(clp->cl_nfs_mod); + kfree(clp->cl_hostname); + kfree(clp->cl_acceptor); ++ nfs_free_multi_path_client(clp); + kfree(clp); + } + EXPORT_SYMBOL_GPL(nfs_free_client); +@@ -324,11 +325,26 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat + continue; + /* Match the full socket address */ + if (!rpc_cmp_addr_port(sap, clap)) ++#if IS_ENABLED(CONFIG_ENFS) ++ { ++ if (data->enfs_option != NULL) ++ continue; ++ else { ++ if (IS_ERR(clp->cl_rpcclient) || ++ !rpc_clnt_xprt_switch_has_addr( ++ clp->cl_rpcclient, sap)) ++ continue; ++ } ++ } ++#else + /* Match all xprt_switch full socket addresses */ + if (IS_ERR(clp->cl_rpcclient) || + !rpc_clnt_xprt_switch_has_addr(clp->cl_rpcclient, + sap)) + continue; ++#endif ++ if (!nfs_multipath_client_match(clp, data)) ++ continue; + + refcount_inc(&clp->cl_count); + return clp; +@@ -512,6 +528,9 @@ int nfs_create_rpc_client(struct nfs_client *clp, + .program = &nfs_program, + .version = clp->rpc_ops->version, + .authflavor = flavor, ++#if IS_ENABLED(CONFIG_ENFS) ++ .multipath_option = cl_init->enfs_option, ++#endif + }; + + if (test_bit(NFS_CS_DISCRTRY, &clp->cl_flags)) +@@ -634,6 +653,13 @@ struct nfs_client *nfs_init_client(struct nfs_client *clp, + /* the client is already initialised */ + if (clp->cl_cons_state == NFS_CS_READY) + return clp; ++ error = nfs_create_multi_path_client(clp, cl_init); ++ if (error < 0) { ++ dprintk("%s: create failed.%d!\n", __func__, error); ++ nfs_put_client(clp); ++ clp = ERR_PTR(error); ++ return clp; ++ } + + /* + * Create a client RPC handle for doing FSSTAT with UNIX auth only +@@ -666,6 +692,9 @@ static int nfs_init_server(struct nfs_server *server, + .net = data->net, + .timeparms = &timeparms, + .init_flags = (1UL << NFS_CS_REUSEPORT), ++#if IS_ENABLED(CONFIG_ENFS) ++ .enfs_option = data->enfs_option, ++#endif + }; + struct nfs_client *clp; + int error; +diff --git a/fs/nfs/enfs_adapter.c b/fs/nfs/enfs_adapter.c +new file mode 100644 +index 000000000..e0f3841c1 +--- /dev/null ++++ b/fs/nfs/enfs_adapter.c +@@ -0,0 +1,276 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Client-side ENFS adapter. ++ * ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++#include <linux/types.h> ++#include <linux/sunrpc/clnt.h> ++#include <linux/nfs.h> ++#include <linux/nfs4.h> ++#include <linux/nfs3.h> ++#include <linux/nfs_fs.h> ++#include <linux/nfs_fs_sb.h> ++#include <linux/sunrpc/sched.h> ++#include <linux/nfs_iostat.h> ++#include "enfs_adapter.h" ++#include "iostat.h" ++ ++struct enfs_adapter_ops __rcu *enfs_adapter; ++ ++int enfs_adapter_register(struct enfs_adapter_ops *ops) ++{ ++ struct enfs_adapter_ops *old; ++ ++ old = cmpxchg((struct enfs_adapter_ops **)&enfs_adapter, NULL, ops); ++ if (old == NULL || old == ops) ++ return 0; ++ pr_err("regist %s ops %p failed. old %p\n", __func__, ops, old); ++ return -EPERM; ++} ++EXPORT_SYMBOL_GPL(enfs_adapter_register); ++ ++int enfs_adapter_unregister(struct enfs_adapter_ops *ops) ++{ ++ struct enfs_adapter_ops *old; ++ ++ old = cmpxchg((struct enfs_adapter_ops **)&enfs_adapter, ops, NULL); ++ if (old == ops || old == NULL) ++ return 0; ++ pr_err("unregist %s ops %p failed. old %p\n", __func__, ops, old); ++ return -EPERM; ++} ++EXPORT_SYMBOL_GPL(enfs_adapter_unregister); ++ ++struct enfs_adapter_ops *nfs_multipath_router_get(void) ++{ ++ struct enfs_adapter_ops *ops; ++ ++ rcu_read_lock(); ++ ops = rcu_dereference(enfs_adapter); ++ if (ops == NULL) { ++ rcu_read_unlock(); ++ return NULL; ++ } ++ if (!try_module_get(ops->owner)) ++ ops = NULL; ++ rcu_read_unlock(); ++ return ops; ++} ++ ++void nfs_multipath_router_put(struct enfs_adapter_ops *ops) ++{ ++ if (ops) ++ module_put(ops->owner); ++} ++ ++bool is_valid_option(enum nfs_multi_path_options option) ++{ ++ if (option < REMOTEADDR || option >= INVALID_OPTION) { ++ pr_warn("%s: ENFS invalid option %d\n", __func__, option); ++ return false; ++ } ++ ++ return true; ++} ++ ++int enfs_parse_mount_options(enum nfs_multi_path_options option, char *str, ++ struct nfs_parsed_mount_data *mnt) ++{ ++ int rc; ++ struct enfs_adapter_ops *ops; ++ ++ ops = nfs_multipath_router_get(); ++ if ((ops == NULL) || (ops->parse_mount_options == NULL) || ++ !is_valid_option(option)) { ++ nfs_multipath_router_put(ops); ++ dfprintk(MOUNT, ++ "NFS: parsing nfs mount option enfs not load[%s]\n" ++ , __func__); ++ return -EOPNOTSUPP; ++ } ++ // nfs_multipath_parse_options ++ dfprintk(MOUNT, "NFS: parsing nfs mount option '%s' type: %d[%s]\n" ++ , str, option, __func__); ++ rc = ops->parse_mount_options(option, str, &mnt->enfs_option, mnt->net); ++ nfs_multipath_router_put(ops); ++ return rc; ++} ++ ++void enfs_free_mount_options(struct nfs_parsed_mount_data *data) ++{ ++ struct enfs_adapter_ops *ops; ++ ++ if (data->enfs_option == NULL) ++ return; ++ ++ ops = nfs_multipath_router_get(); ++ if ((ops == NULL) || (ops->free_mount_options == NULL)) { ++ nfs_multipath_router_put(ops); ++ return; ++ } ++ ops->free_mount_options((void *)&data->enfs_option); ++ nfs_multipath_router_put(ops); ++} ++ ++int nfs_create_multi_path_client(struct nfs_client *client, ++ const struct nfs_client_initdata *cl_init) ++{ ++ int ret = 0; ++ struct enfs_adapter_ops *ops; ++ ++ if (cl_init->enfs_option == NULL) ++ return 0; ++ ++ ops = nfs_multipath_router_get(); ++ if (ops != NULL && ops->client_info_init != NULL) ++ ret = ops->client_info_init( ++ (void *)&client->cl_multipath_data, cl_init); ++ nfs_multipath_router_put(ops); ++ ++ return ret; ++} ++EXPORT_SYMBOL_GPL(nfs_create_multi_path_client); ++ ++void nfs_free_multi_path_client(struct nfs_client *clp) ++{ ++ struct enfs_adapter_ops *ops; ++ ++ if (clp->cl_multipath_data == NULL) ++ return; ++ ++ ops = nfs_multipath_router_get(); ++ if (ops != NULL && ops->client_info_free != NULL) ++ ops->client_info_free(clp->cl_multipath_data); ++ nfs_multipath_router_put(ops); ++} ++ ++int nfs_multipath_client_match(struct nfs_client *clp, ++ const struct nfs_client_initdata *sap) ++{ ++ bool ret = true; ++ struct enfs_adapter_ops *ops; ++ ++ pr_info("%s src %p dst %p\n.", __func__, ++ clp->cl_multipath_data, sap->enfs_option); ++ ++ if (clp->cl_multipath_data == NULL && sap->enfs_option == NULL) ++ return true; ++ ++ if ((clp->cl_multipath_data == NULL && sap->enfs_option) || ++ (clp->cl_multipath_data && sap->enfs_option == NULL)) { ++ pr_err("not match client src %p dst %p\n.", ++ clp->cl_multipath_data, sap->enfs_option); ++ return false; ++ } ++ ++ ops = nfs_multipath_router_get(); ++ if (ops != NULL && ops->client_info_match != NULL) ++ ret = ops->client_info_match(clp->cl_multipath_data, ++ sap->enfs_option); ++ nfs_multipath_router_put(ops); ++ ++ return ret; ++} ++ ++int nfs4_multipath_client_match(struct nfs_client *src, struct nfs_client *dst) ++{ ++ int ret = true; ++ struct enfs_adapter_ops *ops; ++ ++ if (src->cl_multipath_data == NULL && dst->cl_multipath_data == NULL) ++ return true; ++ ++ if (src->cl_multipath_data == NULL || dst->cl_multipath_data == NULL) ++ return false; ++ ++ ops = nfs_multipath_router_get(); ++ if (ops != NULL && ops->nfs4_client_info_match != NULL) ++ ret = ops->nfs4_client_info_match(src->cl_multipath_data, ++ src->cl_multipath_data); ++ nfs_multipath_router_put(ops); ++ ++ return ret; ++} ++EXPORT_SYMBOL_GPL(nfs4_multipath_client_match); ++ ++void nfs_multipath_show_client_info(struct seq_file *mount_option, ++ struct nfs_server *server) ++{ ++ struct enfs_adapter_ops *ops; ++ ++ if (mount_option == NULL || server == NULL || ++ server->client == NULL || ++ server->nfs_client->cl_multipath_data == NULL) ++ return; ++ ++ ops = nfs_multipath_router_get(); ++ if (ops != NULL && ops->client_info_show != NULL) ++ ops->client_info_show(mount_option, server); ++ nfs_multipath_router_put(ops); ++} ++ ++int nfs_remount_iplist(struct nfs_client *nfs_client, void *data) ++{ ++ int ret = 0; ++ struct enfs_adapter_ops *ops; ++ struct nfs_parsed_mount_data *parsed_data = ++ (struct nfs_parsed_mount_data *)data; ++ ++ if (!parsed_data->enfs_option) ++ return 0; ++ ++ if (nfs_client == NULL || nfs_client->cl_rpcclient == NULL) ++ return 0; ++ ++ ops = nfs_multipath_router_get(); ++ if (ops != NULL && ops->remount_ip_list != NULL) ++ ret = ops->remount_ip_list(nfs_client, ++ parsed_data->enfs_option); ++ nfs_multipath_router_put(ops); ++ return ret; ++} ++EXPORT_SYMBOL_GPL(nfs_remount_iplist); ++ ++/* ++ * Error-check and convert a string of mount options from ++ * user space into a data structure. The whole mount string ++ * is processed; bad options are skipped as they are encountered. ++ * If there were no errors, return 1; otherwise return zero(0) ++ */ ++int enfs_check_mount_parse_info(char *p, int token, ++ struct nfs_parsed_mount_data *mnt, const substring_t *args) ++{ ++ char *string; ++ int rc; ++ ++ string = match_strdup(args); ++ if (string == NULL) { ++ pr_info("NFS: not enough memory to parse option\n"); ++ return 0; ++ } ++ rc = enfs_parse_mount_options(get_nfs_multi_path_opt(token), ++ string, mnt); ++ ++ kfree(string); ++ switch (rc) { ++ case 0: ++ return 1; ++ case -ENOMEM: ++ pr_info("NFS: not enough memory to parse option\n"); ++ return 0; ++ case -ENOSPC: ++ pr_info("NFS: param is more than supported limit: %d\n", rc); ++ return 0; ++ case -EINVAL: ++ pr_info("NFS: bad IP address specified: %s\n", p); ++ return 0; ++ case -ENOTSUPP: ++ pr_info("NFS: bad IP address specified: %s\n", p); ++ return 0; ++ case -EOPNOTSUPP: ++ pr_info("NFS: bad IP address specified: %s\n", p); ++ return 0; ++ } ++ return 1; ++} +diff --git a/fs/nfs/enfs_adapter.h b/fs/nfs/enfs_adapter.h +new file mode 100644 +index 000000000..98b4e3292 +--- /dev/null ++++ b/fs/nfs/enfs_adapter.h +@@ -0,0 +1,116 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Client-side ENFS adapt header. ++ * ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++#ifndef _NFS_MULTIPATH_H_ ++#define _NFS_MULTIPATH_H_ ++ ++#include <linux/parser.h> ++#include "internal.h" ++ ++#if IS_ENABLED(CONFIG_ENFS) ++enum nfs_multi_path_options { ++ REMOTEADDR, ++ LOCALADDR, ++ REMOTEDNSNAME, ++ REMOUNTREMOTEADDR, ++ REMOUNTLOCALADDR, ++ INVALID_OPTION ++}; ++ ++ ++struct enfs_adapter_ops { ++ const char *name; ++ struct module *owner; ++ int (*parse_mount_options)(enum nfs_multi_path_options option, ++ char *str, void **enfs_option, struct net *net_ns); ++ ++ void (*free_mount_options)(void **data); ++ ++ int (*client_info_init)(void **data, ++ const struct nfs_client_initdata *cl_init); ++ void (*client_info_free)(void *data); ++ int (*client_info_match)(void *src, void *dst); ++ int (*nfs4_client_info_match)(void *src, void *dst); ++ void (*client_info_show)(struct seq_file *mount_option, void *data); ++ int (*remount_ip_list)(struct nfs_client *nfs_client, ++ void *enfs_option); ++}; ++ ++int enfs_parse_mount_options(enum nfs_multi_path_options option, char *str, ++ struct nfs_parsed_mount_data *mnt); ++void enfs_free_mount_options(struct nfs_parsed_mount_data *data); ++int nfs_create_multi_path_client(struct nfs_client *client, ++ const struct nfs_client_initdata *cl_init); ++void nfs_free_multi_path_client(struct nfs_client *clp); ++int nfs_multipath_client_match(struct nfs_client *clp, ++ const struct nfs_client_initdata *sap); ++int nfs4_multipath_client_match(struct nfs_client *src, struct nfs_client *dst); ++void nfs_multipath_show_client_info(struct seq_file *mount_option, ++ struct nfs_server *server); ++int enfs_adapter_register(struct enfs_adapter_ops *ops); ++int enfs_adapter_unregister(struct enfs_adapter_ops *ops); ++int nfs_remount_iplist(struct nfs_client *nfs_client, void *data); ++int nfs4_create_multi_path(struct nfs_server *server, ++ struct nfs_parsed_mount_data *data, ++ const struct rpc_timeout *timeparms); ++int enfs_check_mount_parse_info(char *p, int token, ++ struct nfs_parsed_mount_data *mnt, const substring_t *args); ++ ++#else ++static inline ++void nfs_free_multi_path_client(struct nfs_client *clp) ++{ ++ ++} ++ ++static inline ++int nfs_multipath_client_match(struct nfs_client *clp, ++ const struct nfs_client_initdata *sap) ++{ ++ return 1; ++} ++ ++static inline ++int nfs_create_multi_path_client(struct nfs_client *client, ++ const struct nfs_client_initdata *cl_init) ++{ ++ return 0; ++} ++ ++static inline ++void nfs_multipath_show_client_info(struct seq_file *mount_option, ++ struct nfs_server *server) ++{ ++ ++} ++ ++static inline ++int nfs4_multipath_client_match(struct nfs_client *src, ++ struct nfs_client *dst) ++{ ++ return 1; ++} ++ ++static inline ++void enfs_free_mount_options(struct nfs_parsed_mount_data *data) ++{ ++ ++} ++ ++static inline ++int enfs_check_mount_parse_info(char *p, int token, ++ struct nfs_parsed_mount_data *mnt, const substring_t *args) ++{ ++ return 1; ++} ++ ++static inline ++int nfs_remount_iplist(struct nfs_client *nfs_client, void *data) ++{ ++ return 0; ++} ++#endif // CONFIG_ENFS ++#endif // _NFS_MULTIPATH_H_ +diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h +index 0ce5a9064..84ac82dbb 100644 +--- a/fs/nfs/internal.h ++++ b/fs/nfs/internal.h +@@ -93,6 +93,9 @@ struct nfs_client_initdata { + u32 minorversion; + struct net *net; + const struct rpc_timeout *timeparms; ++#if IS_ENABLED(CONFIG_ENFS) ++ void *enfs_option; /* struct multipath_mount_options * */ ++#endif + }; + + /* +@@ -135,6 +138,9 @@ struct nfs_parsed_mount_data { + + struct security_mnt_opts lsm_opts; + struct net *net; ++#if IS_ENABLED(CONFIG_ENFS) ++ void *enfs_option; /* struct multipath_mount_options * */ ++#endif + }; + + /* mount_clnt.c */ +@@ -430,6 +436,10 @@ extern void nfs_sb_deactive(struct super_block *sb); + extern int nfs_client_for_each_server(struct nfs_client *clp, + int (*fn)(struct nfs_server *, void *), + void *data); ++#if IS_ENABLED(CONFIG_ENFS) ++extern enum nfs_multi_path_options get_nfs_multi_path_opt(int token); ++#endif ++ + /* io.c */ + extern void nfs_start_io_read(struct inode *inode); + extern void nfs_end_io_read(struct inode *inode); +diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c +index 1350ea673..f97646b98 100644 +--- a/fs/nfs/nfs4client.c ++++ b/fs/nfs/nfs4client.c +@@ -10,7 +10,7 @@ + #include <linux/sunrpc/xprt.h> + #include <linux/sunrpc/bc_xprt.h> + #include <linux/sunrpc/rpc_pipe_fs.h> +-#include "internal.h" ++#include "enfs_adapter.h" + #include "callback.h" + #include "delegation.h" + #include "nfs4session.h" +@@ -225,6 +225,14 @@ struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *cl_init) + __set_bit(NFS_CS_DISCRTRY, &clp->cl_flags); + __set_bit(NFS_CS_NO_RETRANS_TIMEOUT, &clp->cl_flags); + ++ err = nfs_create_multi_path_client(clp, cl_init); ++ if (err < 0) { ++ dprintk("%s: create failed.%d\n", __func__, err); ++ nfs_put_client(clp); ++ clp = ERR_PTR(err); ++ return clp; ++ } ++ + /* + * Set up the connection to the server before we add add to the + * global list. +@@ -529,6 +537,9 @@ static int nfs4_match_client(struct nfs_client *pos, struct nfs_client *new, + if (!nfs4_match_client_owner_id(pos, new)) + return 1; + ++ if (!nfs4_multipath_client_match(pos, new)) ++ return 1; ++ + return 0; + } + +@@ -860,7 +871,7 @@ static int nfs4_set_client(struct nfs_server *server, + const size_t addrlen, + const char *ip_addr, + int proto, const struct rpc_timeout *timeparms, +- u32 minorversion, struct net *net) ++ u32 minorversion, struct net *net, void *enfs_option) + { + struct nfs_client_initdata cl_init = { + .hostname = hostname, +@@ -872,6 +883,9 @@ static int nfs4_set_client(struct nfs_server *server, + .minorversion = minorversion, + .net = net, + .timeparms = timeparms, ++#if IS_ENABLED(CONFIG_ENFS) ++ .enfs_option = enfs_option, ++#endif + }; + struct nfs_client *clp; + +@@ -1050,6 +1064,7 @@ static int nfs4_init_server(struct nfs_server *server, + { + struct rpc_timeout timeparms; + int error; ++ void *enfs_option = NULL; + + nfs_init_timeout_values(&timeparms, data->nfs_server.protocol, + data->timeo, data->retrans); +@@ -1067,6 +1082,10 @@ static int nfs4_init_server(struct nfs_server *server, + else + data->selected_flavor = RPC_AUTH_UNIX; + ++#if IS_ENABLED(CONFIG_ENFS) ++ enfs_option = data->enfs_option; ++#endif ++ + /* Get a client record */ + error = nfs4_set_client(server, + data->nfs_server.hostname, +@@ -1076,7 +1095,7 @@ static int nfs4_init_server(struct nfs_server *server, + data->nfs_server.protocol, + &timeparms, + data->minorversion, +- data->net); ++ data->net, enfs_option); + if (error < 0) + return error; + +@@ -1161,7 +1180,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, + XPRT_TRANSPORT_RDMA, + parent_server->client->cl_timeout, + parent_client->cl_mvops->minor_version, +- parent_client->cl_net); ++ parent_client->cl_net, NULL); + if (!error) + goto init_server; + #endif /* IS_ENABLED(CONFIG_SUNRPC_XPRT_RDMA) */ +@@ -1174,7 +1193,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, + XPRT_TRANSPORT_TCP, + parent_server->client->cl_timeout, + parent_client->cl_mvops->minor_version, +- parent_client->cl_net); ++ parent_client->cl_net, NULL); + if (error < 0) + goto error; + +@@ -1269,7 +1288,7 @@ int nfs4_update_server(struct nfs_server *server, const char *hostname, + set_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status); + error = nfs4_set_client(server, hostname, sap, salen, buf, + clp->cl_proto, clnt->cl_timeout, +- clp->cl_minorversion, net); ++ clp->cl_minorversion, net, NULL); + clear_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status); + if (error != 0) { + nfs_server_insert_lists(server); +diff --git a/fs/nfs/super.c b/fs/nfs/super.c +index a05e1eb2c..168da9700 100644 +--- a/fs/nfs/super.c ++++ b/fs/nfs/super.c +@@ -51,7 +51,6 @@ + #include <linux/netdevice.h> + #include <linux/nfs_xdr.h> + #include <linux/magic.h> +-#include <linux/parser.h> + #include <linux/nsproxy.h> + #include <linux/rcupdate.h> + +@@ -61,7 +60,7 @@ + #include "callback.h" + #include "delegation.h" + #include "iostat.h" +-#include "internal.h" ++#include "enfs_adapter.h" + #include "fscache.h" + #include "nfs4session.h" + #include "pnfs.h" +@@ -113,6 +112,12 @@ enum { + + /* Special mount options */ + Opt_userspace, Opt_deprecated, Opt_sloppy, ++#if IS_ENABLED(CONFIG_ENFS) ++ Opt_remote_iplist, ++ Opt_local_iplist, ++ Opt_remote_dnslist, ++ Opt_enfs_info, ++#endif + + Opt_err + }; +@@ -183,6 +188,13 @@ static const match_table_t nfs_mount_option_tokens = { + { Opt_fscache_uniq, "fsc=%s" }, + { Opt_local_lock, "local_lock=%s" }, + ++#if IS_ENABLED(CONFIG_ENFS) ++ { Opt_remote_iplist, "remoteaddrs=%s" }, ++ { Opt_local_iplist, "localaddrs=%s" }, ++ { Opt_remote_dnslist, "remotednsname=%s" }, ++ { Opt_enfs_info, "enfs_info=%s" }, ++#endif ++ + /* The following needs to be listed after all other options */ + { Opt_nfsvers, "v%s" }, + +@@ -365,6 +377,21 @@ static struct shrinker acl_shrinker = { + .seeks = DEFAULT_SEEKS, + }; + ++#if IS_ENABLED(CONFIG_ENFS) ++enum nfs_multi_path_options get_nfs_multi_path_opt(int token) ++{ ++ switch (token) { ++ case Opt_remote_iplist: ++ return REMOUNTREMOTEADDR; ++ case Opt_local_iplist: ++ return REMOUNTLOCALADDR; ++ case Opt_remote_dnslist: ++ return REMOTEDNSNAME; ++ } ++ return INVALID_OPTION; ++} ++#endif ++ + /* + * Register the NFS filesystems + */ +@@ -758,6 +785,9 @@ int nfs_show_options(struct seq_file *m, struct dentry *root) + seq_printf(m, ",addr=%s", + rpc_peeraddr2str(nfss->nfs_client->cl_rpcclient, + RPC_DISPLAY_ADDR)); ++ ++ nfs_multipath_show_client_info(m, nfss); ++ + rcu_read_unlock(); + + return 0; +@@ -853,6 +883,8 @@ int nfs_show_stats(struct seq_file *m, struct dentry *root) + seq_puts(m, root->d_sb->s_flags & SB_NODIRATIME ? ",nodiratime" : ""); + nfs_show_mount_options(m, nfss, 1); + ++ nfs_multipath_show_client_info(m, nfss); ++ + seq_printf(m, "\n\tage:\t%lu", (jiffies - nfss->mount_time) / HZ); + + show_implementation_id(m, nfss); +@@ -977,6 +1009,7 @@ static void nfs_free_parsed_mount_data(struct nfs_parsed_mount_data *data) + kfree(data->nfs_server.export_path); + kfree(data->nfs_server.hostname); + kfree(data->fscache_uniq); ++ enfs_free_mount_options(data); + security_free_mnt_opts(&data->lsm_opts); + kfree(data); + } +@@ -1641,7 +1674,6 @@ static int nfs_parse_mount_options(char *raw, + return 0; + }; + break; +- + /* + * Special options + */ +@@ -1654,7 +1686,18 @@ static int nfs_parse_mount_options(char *raw, + dfprintk(MOUNT, "NFS: ignoring mount option " + "'%s'\n", p); + break; +- ++#if IS_ENABLED(CONFIG_ENFS) ++ case Opt_remote_iplist: ++ case Opt_local_iplist: ++ case Opt_remote_dnslist: ++ rc = enfs_check_mount_parse_info(p, ++ token, mnt, args); ++ if (rc != 1) ++ return rc; ++ break; ++ case Opt_enfs_info: ++ break; ++#endif + default: + invalid_option = 1; + dfprintk(MOUNT, "NFS: unrecognized mount option " +@@ -2335,6 +2378,10 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data) + if (!nfs_parse_mount_options((char *)options, data)) + goto out; + ++ error = nfs_remount_iplist(nfss->nfs_client, data); ++ if (error) ++ goto out; ++ + /* + * noac is a special case. It implies -o sync, but that's not + * necessarily reflected in the mtab options. do_remount_sb +@@ -2347,6 +2394,8 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data) + /* compare new mount options with old ones */ + error = nfs_compare_remount_data(nfss, data); + out: ++ /* release remount option member */ ++ enfs_free_mount_options(data); + nfs_free_parsed_mount_data(data); + return error; + } +diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h +index 7023ae64e..2c19678af 100644 +--- a/include/linux/nfs_fs_sb.h ++++ b/include/linux/nfs_fs_sb.h +@@ -123,6 +123,11 @@ struct nfs_client { + + struct net *cl_net; + struct list_head pending_cb_stateids; ++ ++#if IS_ENABLED(CONFIG_ENFS) ++ /* multi path private structure (struct multipath_client_info *) */ ++ void *cl_multipath_data; ++#endif + }; + + /* diff --git a/0002-sunrpc_add_api_to_support_enfs_registe_and_create_multipath_then_dispatch_IO.patch b/0002-sunrpc_add_api_to_support_enfs_registe_and_create_multipath_then_dispatch_IO.patch new file mode 100644 index 0000000..98af7d8 --- /dev/null +++ b/0002-sunrpc_add_api_to_support_enfs_registe_and_create_multipath_then_dispatch_IO.patch @@ -0,0 +1,832 @@ +diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h +index 8aa865bce..0425aaabf 100644 +--- a/include/linux/sunrpc/clnt.h ++++ b/include/linux/sunrpc/clnt.h +@@ -70,6 +70,10 @@ struct rpc_clnt { + struct dentry *cl_debugfs; /* debugfs directory */ + #endif + struct rpc_xprt_iter cl_xpi; ++ ++#if IS_ENABLED(CONFIG_ENFS) ++ bool cl_enfs; ++#endif + }; + + /* +@@ -124,6 +128,9 @@ struct rpc_create_args { + unsigned long flags; + char *client_name; + struct svc_xprt *bc_xprt; /* NFSv4.1 backchannel */ ++#if IS_ENABLED(CONFIG_ENFS) ++ void *multipath_option; ++#endif + }; + + struct rpc_add_xprt_test { +@@ -221,6 +228,13 @@ bool rpc_clnt_xprt_switch_has_addr(struct rpc_clnt *clnt, + const struct sockaddr *sap); + void rpc_cleanup_clids(void); + ++#if IS_ENABLED(CONFIG_ENFS) ++int ++rpc_clnt_test_xprt(struct rpc_clnt *clnt, struct rpc_xprt *xprt, ++ const struct rpc_call_ops *ops, void *data, int flags); ++struct rpc_xprt *rpc_task_get_next_xprt(struct rpc_clnt *clnt); ++#endif /* CONFIG_ENFS */ ++ + static inline int rpc_reply_expected(struct rpc_task *task) + { + return (task->tk_msg.rpc_proc != NULL) && +diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h +index ad2e243f3..124f5a0fa 100644 +--- a/include/linux/sunrpc/sched.h ++++ b/include/linux/sunrpc/sched.h +@@ -90,6 +90,9 @@ struct rpc_task { + tk_garb_retry : 2, + tk_cred_retry : 2, + tk_rebind_retry : 2; ++#if IS_ENABLED(CONFIG_ENFS) ++ unsigned long tk_major_timeo; /* major timeout ticks */ ++#endif + }; + + typedef void (*rpc_action)(struct rpc_task *); +@@ -118,6 +121,9 @@ struct rpc_task_setup { + */ + #define RPC_TASK_ASYNC 0x0001 /* is an async task */ + #define RPC_TASK_SWAPPER 0x0002 /* is swapping in/out */ ++#if IS_ENABLED(CONFIG_ENFS) ++#define RPC_TASK_FIXED 0x0004 /* detect xprt status task */ ++#endif + #define RPC_CALL_MAJORSEEN 0x0020 /* major timeout seen */ + #define RPC_TASK_ROOTCREDS 0x0040 /* force root creds */ + #define RPC_TASK_DYNAMIC 0x0080 /* task was kmalloc'ed */ +@@ -257,6 +263,9 @@ void rpc_destroy_mempool(void); + extern struct workqueue_struct *rpciod_workqueue; + extern struct workqueue_struct *xprtiod_workqueue; + void rpc_prepare_task(struct rpc_task *task); ++#if IS_ENABLED(CONFIG_ENFS) ++void rpc_init_task_retry_counters(struct rpc_task *task); ++#endif + + static inline int rpc_wait_for_completion_task(struct rpc_task *task) + { +diff --git a/include/linux/sunrpc/sunrpc_enfs_adapter.h b/include/linux/sunrpc/sunrpc_enfs_adapter.h +new file mode 100644 +index 000000000..cdd7fa699 +--- /dev/null ++++ b/include/linux/sunrpc/sunrpc_enfs_adapter.h +@@ -0,0 +1,135 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* Client-side SUNRPC ENFS adapter header. ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++#ifndef _SUNRPC_ENFS_ADAPTER_H_ ++#define _SUNRPC_ENFS_ADAPTER_H_ ++#include <linux/sunrpc/clnt.h> ++ ++#if IS_ENABLED(CONFIG_ENFS) ++ ++static inline void rpc_xps_nactive_add_one(struct rpc_xprt_switch *xps) ++{ ++ xps->xps_nactive--; ++} ++ ++static inline void rpc_xps_nactive_sub_one(struct rpc_xprt_switch *xps) ++{ ++ xps->xps_nactive--; ++} ++ ++struct rpc_xprt *rpc_task_get_xprt ++(struct rpc_clnt *clnt, struct rpc_xprt *xprt); ++ ++struct rpc_multipath_ops { ++ struct module *owner; ++ void (*create_clnt)(struct rpc_create_args *args, ++ struct rpc_clnt *clnt); ++ void (*releas_clnt)(struct rpc_clnt *clnt); ++ void (*create_xprt)(struct rpc_xprt *xprt); ++ void (*destroy_xprt)(struct rpc_xprt *xprt); ++ void (*xprt_iostat)(struct rpc_task *task); ++ void (*failover_handle)(struct rpc_task *task); ++ bool (*task_need_call_start_again)(struct rpc_task *task); ++ void (*adjust_task_timeout)(struct rpc_task *task, void *condition); ++ void (*init_task_req)(struct rpc_task *task, struct rpc_rqst *req); ++ bool (*prepare_transmit)(struct rpc_task *task); ++ void (*set_transport)(struct rpc_task *task, struct rpc_clnt *clnt); ++}; ++ ++extern struct rpc_multipath_ops __rcu *multipath_ops; ++void rpc_init_task_retry_counters(struct rpc_task *task); ++int rpc_multipath_ops_register(struct rpc_multipath_ops *ops); ++int rpc_multipath_ops_unregister(struct rpc_multipath_ops *ops); ++struct rpc_multipath_ops *rpc_multipath_ops_get(void); ++void rpc_multipath_ops_put(struct rpc_multipath_ops *ops); ++void rpc_task_release_xprt(struct rpc_clnt *clnt, struct rpc_xprt *xprt); ++void rpc_multipath_ops_create_clnt(struct rpc_create_args *args, ++ struct rpc_clnt *clnt); ++void rpc_multipath_ops_releas_clnt(struct rpc_clnt *clnt); ++bool rpc_multipath_ops_create_xprt(struct rpc_xprt *xprt); ++void rpc_multipath_ops_destroy_xprt(struct rpc_xprt *xprt); ++void rpc_multipath_ops_xprt_iostat(struct rpc_task *task); ++void rpc_multipath_ops_failover_handle(struct rpc_task *task); ++bool rpc_multipath_ops_task_need_call_start_again(struct rpc_task *task); ++void rpc_multipath_ops_adjust_task_timeout(struct rpc_task *task, ++ void *condition); ++void rpc_multipath_ops_init_task_req(struct rpc_task *task, ++ struct rpc_rqst *req); ++bool rpc_multipath_ops_prepare_transmit(struct rpc_task *task); ++void rpc_multipath_ops_set_transport(struct rpc_task *task, ++ struct rpc_clnt *clnt); ++#else ++static inline struct rpc_xprt *rpc_task_get_xprt(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt) ++{ ++ return NULL; ++} ++ ++static inline void rpc_task_release_xprt(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt) ++{ ++} ++ ++static inline void rpc_xps_nactive_add_one(struct rpc_xprt_switch *xps) ++{ ++} ++ ++static inline void rpc_xps_nactive_sub_one(struct rpc_xprt_switch *xps) ++{ ++} ++ ++static inline void rpc_multipath_ops_create_clnt ++(struct rpc_create_args *args, struct rpc_clnt *clnt) ++{ ++} ++ ++static inline void rpc_multipath_ops_releas_clnt(struct rpc_clnt *clnt) ++{ ++} ++ ++static inline bool rpc_multipath_ops_create_xprt(struct rpc_xprt *xprt) ++{ ++ return false; ++} ++ ++static inline void rpc_multipath_ops_destroy_xprt(struct rpc_xprt *xprt) ++{ ++} ++ ++static inline void rpc_multipath_ops_xprt_iostat(struct rpc_task *task) ++{ ++} ++ ++static inline void rpc_multipath_ops_failover_handle(struct rpc_task *task) ++{ ++} ++ ++static inline ++bool rpc_multipath_ops_task_need_call_start_again(struct rpc_task *task) ++{ ++ return false; ++} ++ ++static inline void ++rpc_multipath_ops_adjust_task_timeout(struct rpc_task *task, void *condition) ++{ ++} ++ ++static inline void ++rpc_multipath_ops_init_task_req(struct rpc_task *task, struct rpc_rqst *req) ++{ ++} ++ ++static inline bool rpc_multipath_ops_prepare_transmit(struct rpc_task *task) ++{ ++ return false; ++} ++ ++static inline void ++rpc_multipath_ops_set_transport(struct rpc_task *task, struct rpc_clnt *clnt) ++{ ++} ++ ++#endif ++#endif // _SUNRPC_ENFS_ADAPTER_H_ +diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h +index ccfacca1e..2e47b3577 100644 +--- a/include/linux/sunrpc/xprt.h ++++ b/include/linux/sunrpc/xprt.h +@@ -279,6 +279,10 @@ struct rpc_xprt { + atomic_t inject_disconnect; + #endif + struct rcu_head rcu; ++#if IS_ENABLED(CONFIG_ENFS) ++ atomic_long_t queuelen; ++ void *multipath_context; ++#endif + }; + + #if defined(CONFIG_SUNRPC_BACKCHANNEL) +diff --git a/include/linux/sunrpc/xprtmultipath.h b/include/linux/sunrpc/xprtmultipath.h +index af1257c03..5677a46ee 100644 +--- a/include/linux/sunrpc/xprtmultipath.h ++++ b/include/linux/sunrpc/xprtmultipath.h +@@ -22,6 +22,11 @@ struct rpc_xprt_switch { + const struct rpc_xprt_iter_ops *xps_iter_ops; + + struct rcu_head xps_rcu; ++#if IS_ENABLED(CONFIG_ENFS) ++ unsigned int xps_nactive; ++ atomic_long_t xps_queuelen; ++ unsigned long xps_tmp_time; ++#endif + }; + + struct rpc_xprt_iter { +@@ -69,4 +74,8 @@ extern struct rpc_xprt *xprt_iter_get_next(struct rpc_xprt_iter *xpi); + + extern bool rpc_xprt_switch_has_addr(struct rpc_xprt_switch *xps, + const struct sockaddr *sap); ++#if IS_ENABLED(CONFIG_ENFS) ++extern void xprt_switch_add_xprt_locked(struct rpc_xprt_switch *xps, ++ struct rpc_xprt *xprt); ++#endif + #endif +diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c +index 0fc540b0d..ca10bc61a 100644 +--- a/net/sunrpc/clnt.c ++++ b/net/sunrpc/clnt.c +@@ -37,6 +37,7 @@ + #include <linux/sunrpc/rpc_pipe_fs.h> + #include <linux/sunrpc/metrics.h> + #include <linux/sunrpc/bc_xprt.h> ++#include <linux/sunrpc/sunrpc_enfs_adapter.h> + #include <trace/events/sunrpc.h> + + #include "sunrpc.h" +@@ -490,6 +491,8 @@ static struct rpc_clnt *rpc_create_xprt(struct rpc_create_args *args, + } + } + ++ rpc_multipath_ops_create_clnt(args, clnt); ++ + clnt->cl_softrtry = 1; + if (args->flags & RPC_CLNT_CREATE_HARDRTRY) + clnt->cl_softrtry = 0; +@@ -869,6 +872,8 @@ void rpc_shutdown_client(struct rpc_clnt *clnt) + list_empty(&clnt->cl_tasks), 1*HZ); + } + ++ rpc_multipath_ops_releas_clnt(clnt); ++ + rpc_release_client(clnt); + } + EXPORT_SYMBOL_GPL(rpc_shutdown_client); +@@ -981,6 +986,12 @@ void rpc_task_release_transport(struct rpc_task *task) + + if (xprt) { + task->tk_xprt = NULL; ++#if IS_ENABLED(CONFIG_ENFS) ++ if (task->tk_client) { ++ rpc_task_release_xprt(task->tk_client, xprt); ++ return; ++ } ++#endif + xprt_put(xprt); + } + } +@@ -990,6 +1001,10 @@ void rpc_task_release_client(struct rpc_task *task) + { + struct rpc_clnt *clnt = task->tk_client; + ++#if IS_ENABLED(CONFIG_ENFS) ++ rpc_task_release_transport(task); ++#endif ++ + if (clnt != NULL) { + /* Remove from client task list */ + spin_lock(&clnt->cl_lock); +@@ -999,14 +1014,31 @@ void rpc_task_release_client(struct rpc_task *task) + + rpc_release_client(clnt); + } +- rpc_task_release_transport(task); ++ ++ if (!IS_ENABLED(CONFIG_ENFS)) ++ rpc_task_release_transport(task); ++ + } + ++#if IS_ENABLED(CONFIG_ENFS) ++struct rpc_xprt * ++rpc_task_get_next_xprt(struct rpc_clnt *clnt) ++{ ++ return rpc_task_get_xprt(clnt, xprt_iter_get_next(&clnt->cl_xpi)); ++} ++EXPORT_SYMBOL_GPL(rpc_task_get_next_xprt); ++#endif ++ + static + void rpc_task_set_transport(struct rpc_task *task, struct rpc_clnt *clnt) + { ++ rpc_multipath_ops_set_transport(task, clnt); + if (!task->tk_xprt) ++#if IS_ENABLED(CONFIG_ENFS) ++ task->tk_xprt = rpc_task_get_next_xprt(clnt); ++#else + task->tk_xprt = xprt_iter_get_next(&clnt->cl_xpi); ++#endif + } + + static +@@ -1597,6 +1629,14 @@ call_reserveresult(struct rpc_task *task) + return; + case -EIO: /* probably a shutdown */ + break; ++#if IS_ENABLED(CONFIG_ENFS) ++ case -ETIMEDOUT: /* woken up; restart */ ++ if (rpc_multipath_ops_task_need_call_start_again(task)) { ++ rpc_task_release_transport(task); ++ task->tk_action = call_start; ++ return; ++ } ++#endif + default: + printk(KERN_ERR "%s: unrecognized error %d, exiting\n", + __func__, status); +@@ -1962,6 +2002,10 @@ call_transmit(struct rpc_task *task) + return; + if (!xprt_prepare_transmit(task)) + return; ++ ++ if (rpc_multipath_ops_prepare_transmit(task)) ++ return; ++ + task->tk_action = call_transmit_status; + /* Encode here so that rpcsec_gss can use correct sequence number. */ + if (rpc_task_need_encode(task)) { +@@ -2277,6 +2321,7 @@ call_timeout(struct rpc_task *task) + + retry: + task->tk_action = call_bind; ++ rpc_multipath_ops_failover_handle(task); + task->tk_status = 0; + } + +@@ -2798,8 +2843,8 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt, + xprt->ops->set_connect_timeout(xprt, + connect_timeout, + reconnect_timeout); +- +- rpc_xprt_switch_set_roundrobin(xps); ++ if (!clnt->cl_enfs) ++ rpc_xprt_switch_set_roundrobin(xps); + if (setup) { + ret = setup(clnt, xps, xprt, data); + if (ret != 0) +@@ -2961,3 +3006,30 @@ rpc_clnt_swap_deactivate(struct rpc_clnt *clnt) + } + EXPORT_SYMBOL_GPL(rpc_clnt_swap_deactivate); + #endif /* CONFIG_SUNRPC_SWAP */ ++ ++#if IS_ENABLED(CONFIG_ENFS) ++/* rpc_clnt_test_xprt - Test and add a new transport to a rpc_clnt ++ * @clnt: pointer to struct rpc_clnt ++ * @xprt: pointer struct rpc_xprt ++ * @ops: async operation ++ */ ++int ++rpc_clnt_test_xprt(struct rpc_clnt *clnt, struct rpc_xprt *xprt, ++ const struct rpc_call_ops *ops, void *data, int flags) ++{ ++ struct rpc_cred *cred; ++ struct rpc_task *task; ++ ++ cred = authnull_ops.lookup_cred(NULL, NULL, 0); ++ task = rpc_call_null_helper(clnt, xprt, cred, ++ RPC_TASK_SOFT | RPC_TASK_SOFTCONN | flags, ++ ops, data); ++ put_rpccred(cred); ++ if (IS_ERR(task)) ++ return PTR_ERR(task); ++ ++ rpc_put_task(task); ++ return 1; ++} ++EXPORT_SYMBOL_GPL(rpc_clnt_test_xprt); ++#endif +diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c +index a873c92a4..2254fea0e 100644 +--- a/net/sunrpc/sched.c ++++ b/net/sunrpc/sched.c +@@ -20,7 +20,7 @@ + #include <linux/mutex.h> + #include <linux/freezer.h> + +-#include <linux/sunrpc/clnt.h> ++#include <linux/sunrpc/sunrpc_enfs_adapter.h> + + #include "sunrpc.h" + +@@ -962,7 +962,12 @@ static void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *ta + /* Initialize workqueue for async tasks */ + task->tk_workqueue = task_setup_data->workqueue; + ++#if IS_ENABLED(CONFIG_ENFS) ++ task->tk_xprt = rpc_task_get_xprt(task_setup_data->rpc_client, ++ xprt_get(task_setup_data->rpc_xprt)); ++#else + task->tk_xprt = xprt_get(task_setup_data->rpc_xprt); ++#endif + + if (task->tk_ops->rpc_call_prepare != NULL) + task->tk_action = rpc_prepare_task; +diff --git a/net/sunrpc/sunrpc_enfs_adapter.c b/net/sunrpc/sunrpc_enfs_adapter.c +new file mode 100644 +index 000000000..106ad7309 +--- /dev/null ++++ b/net/sunrpc/sunrpc_enfs_adapter.c +@@ -0,0 +1,225 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* Client-side SUNRPC ENFS adapter header. ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++#include <linux/sunrpc/sunrpc_enfs_adapter.h> ++ ++struct rpc_multipath_ops __rcu *multipath_ops; ++ ++void rpc_init_task_retry_counters(struct rpc_task *task) ++{ ++ /* Initialize retry counters */ ++ task->tk_garb_retry = 2; ++ task->tk_cred_retry = 2; ++ task->tk_rebind_retry = 2; ++} ++EXPORT_SYMBOL_GPL(rpc_init_task_retry_counters); ++ ++struct rpc_xprt * ++rpc_task_get_xprt(struct rpc_clnt *clnt, struct rpc_xprt *xprt) ++{ ++ struct rpc_xprt_switch *xps; ++ ++ if (!xprt) ++ return NULL; ++ rcu_read_lock(); ++ xps = rcu_dereference(clnt->cl_xpi.xpi_xpswitch); ++ atomic_long_inc(&xps->xps_queuelen); ++ rcu_read_unlock(); ++ atomic_long_inc(&xprt->queuelen); ++ ++ return xprt; ++} ++ ++int rpc_multipath_ops_register(struct rpc_multipath_ops *ops) ++{ ++ struct rpc_multipath_ops *old; ++ ++ old = cmpxchg((struct rpc_multipath_ops **)&multipath_ops, NULL, ops); ++ if (!old || old == ops) ++ return 0; ++ pr_err("regist rpc_multipath ops %p fail. old %p\n", ops, old); ++ return -EPERM; ++} ++EXPORT_SYMBOL_GPL(rpc_multipath_ops_register); ++ ++int rpc_multipath_ops_unregister(struct rpc_multipath_ops *ops) ++{ ++ struct rpc_multipath_ops *old; ++ ++ old = cmpxchg((struct rpc_multipath_ops **)&multipath_ops, ops, NULL); ++ if (!old || old == ops) ++ return 0; ++ pr_err("regist rpc_multipath ops %p fail. old %p\n", ops, old); ++ return -EPERM; ++} ++EXPORT_SYMBOL_GPL(rpc_multipath_ops_unregister); ++ ++struct rpc_multipath_ops *rpc_multipath_ops_get(void) ++{ ++ struct rpc_multipath_ops *ops; ++ ++ rcu_read_lock(); ++ ops = rcu_dereference(multipath_ops); ++ if (!ops) { ++ rcu_read_unlock(); ++ return NULL; ++ } ++ if (!try_module_get(ops->owner)) ++ ops = NULL; ++ rcu_read_unlock(); ++ return ops; ++} ++EXPORT_SYMBOL_GPL(rpc_multipath_ops_get); ++ ++void rpc_multipath_ops_put(struct rpc_multipath_ops *ops) ++{ ++ if (ops) ++ module_put(ops->owner); ++} ++EXPORT_SYMBOL_GPL(rpc_multipath_ops_put); ++ ++void rpc_task_release_xprt(struct rpc_clnt *clnt, struct rpc_xprt *xprt) ++{ ++ struct rpc_xprt_switch *xps; ++ ++ atomic_long_dec(&xprt->queuelen); ++ rcu_read_lock(); ++ xps = rcu_dereference(clnt->cl_xpi.xpi_xpswitch); ++ atomic_long_dec(&xps->xps_queuelen); ++ rcu_read_unlock(); ++ ++ xprt_put(xprt); ++} ++ ++void rpc_multipath_ops_create_clnt(struct rpc_create_args *args, ++ struct rpc_clnt *clnt) ++{ ++ struct rpc_multipath_ops *mops; ++ ++ if (args->multipath_option) { ++ mops = rpc_multipath_ops_get(); ++ if (mops && mops->create_clnt) ++ mops->create_clnt(args, clnt); ++ rpc_multipath_ops_put(mops); ++ } ++} ++ ++void rpc_multipath_ops_releas_clnt(struct rpc_clnt *clnt) ++{ ++ struct rpc_multipath_ops *mops; ++ ++ mops = rpc_multipath_ops_get(); ++ if (mops && mops->releas_clnt) ++ mops->releas_clnt(clnt); ++ ++ rpc_multipath_ops_put(mops); ++} ++ ++bool rpc_multipath_ops_create_xprt(struct rpc_xprt *xprt) ++{ ++ struct rpc_multipath_ops *mops = NULL; ++ ++ mops = rpc_multipath_ops_get(); ++ if (mops && mops->create_xprt) { ++ mops->create_xprt(xprt); ++ if (!xprt->multipath_context) { ++ rpc_multipath_ops_put(mops); ++ return true; ++ } ++ } ++ rpc_multipath_ops_put(mops); ++ return false; ++} ++ ++void rpc_multipath_ops_destroy_xprt(struct rpc_xprt *xprt) ++{ ++ struct rpc_multipath_ops *mops; ++ ++ if (xprt->multipath_context) { ++ mops = rpc_multipath_ops_get(); ++ if (mops && mops->destroy_xprt) ++ mops->destroy_xprt(xprt); ++ rpc_multipath_ops_put(mops); ++ } ++} ++ ++void rpc_multipath_ops_xprt_iostat(struct rpc_task *task) ++{ ++ struct rpc_multipath_ops *mops; ++ ++ mops = rpc_multipath_ops_get(); ++ if (task->tk_client && mops && mops->xprt_iostat) ++ mops->xprt_iostat(task); ++ rpc_multipath_ops_put(mops); ++} ++ ++void rpc_multipath_ops_failover_handle(struct rpc_task *task) ++{ ++ struct rpc_multipath_ops *mpath_ops = NULL; ++ ++ mpath_ops = rpc_multipath_ops_get(); ++ if (mpath_ops && mpath_ops->failover_handle) ++ mpath_ops->failover_handle(task); ++ rpc_multipath_ops_put(mpath_ops); ++} ++ ++bool rpc_multipath_ops_task_need_call_start_again(struct rpc_task *task) ++{ ++ struct rpc_multipath_ops *mpath_ops = NULL; ++ bool ret = false; ++ ++ mpath_ops = rpc_multipath_ops_get(); ++ if (mpath_ops && mpath_ops->task_need_call_start_again) ++ ret = mpath_ops->task_need_call_start_again(task); ++ rpc_multipath_ops_put(mpath_ops); ++ return ret; ++} ++ ++void rpc_multipath_ops_adjust_task_timeout(struct rpc_task *task, ++ void *condition) ++{ ++ struct rpc_multipath_ops *mops = NULL; ++ ++ mops = rpc_multipath_ops_get(); ++ if (mops && mops->adjust_task_timeout) ++ mops->adjust_task_timeout(task, NULL); ++ rpc_multipath_ops_put(mops); ++} ++ ++void rpc_multipath_ops_init_task_req(struct rpc_task *task, ++ struct rpc_rqst *req) ++{ ++ struct rpc_multipath_ops *mops = NULL; ++ ++ mops = rpc_multipath_ops_get(); ++ if (mops && mops->init_task_req) ++ mops->init_task_req(task, req); ++ rpc_multipath_ops_put(mops); ++} ++ ++bool rpc_multipath_ops_prepare_transmit(struct rpc_task *task) ++{ ++ struct rpc_multipath_ops *mops = NULL; ++ ++ mops = rpc_multipath_ops_get(); ++ if (mops && mops->prepare_transmit) { ++ if (!(mops->prepare_transmit(task))) { ++ rpc_multipath_ops_put(mops); ++ return true; ++ } ++ } ++ rpc_multipath_ops_put(mops); ++ return false; ++} ++ ++void rpc_multipath_ops_set_transport(struct rpc_task *task, ++ struct rpc_clnt *clnt) ++{ ++ struct rpc_multipath_ops *mops = NULL; ++ ++ mops = rpc_multipath_ops_get(); ++ if (mops && mops->set_transport) ++ mops->set_transport(task, clnt); ++ rpc_multipath_ops_put(mops); ++} +diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c +index c912bf20f..c2b63b3d5 100644 +--- a/net/sunrpc/xprt.c ++++ b/net/sunrpc/xprt.c +@@ -48,6 +48,7 @@ + #include <linux/sunrpc/clnt.h> + #include <linux/sunrpc/metrics.h> + #include <linux/sunrpc/bc_xprt.h> ++#include <linux/sunrpc/sunrpc_enfs_adapter.h> + #include <linux/rcupdate.h> + + #include <trace/events/sunrpc.h> +@@ -259,6 +260,9 @@ int xprt_reserve_xprt(struct rpc_xprt *xprt, struct rpc_task *task) + dprintk("RPC: %5u failed to lock transport %p\n", + task->tk_pid, xprt); + task->tk_timeout = 0; ++ ++ rpc_multipath_ops_adjust_task_timeout(task, NULL); ++ + task->tk_status = -EAGAIN; + if (req == NULL) + priority = RPC_PRIORITY_LOW; +@@ -560,6 +564,9 @@ void xprt_wait_for_buffer_space(struct rpc_task *task, rpc_action action) + struct rpc_xprt *xprt = req->rq_xprt; + + task->tk_timeout = RPC_IS_SOFT(task) ? req->rq_timeout : 0; ++ ++ rpc_multipath_ops_adjust_task_timeout(task, NULL); ++ + rpc_sleep_on(&xprt->pending, task, action); + } + EXPORT_SYMBOL_GPL(xprt_wait_for_buffer_space); +@@ -1347,6 +1354,9 @@ xprt_request_init(struct rpc_task *task) + req->rq_rcv_buf.buflen = 0; + req->rq_release_snd_buf = NULL; + xprt_reset_majortimeo(req); ++ ++ rpc_multipath_ops_init_task_req(task, req); ++ + dprintk("RPC: %5u reserved req %p xid %08x\n", task->tk_pid, + req, ntohl(req->rq_xid)); + } +@@ -1427,6 +1437,9 @@ void xprt_release(struct rpc_task *task) + task->tk_ops->rpc_count_stats(task, task->tk_calldata); + else if (task->tk_client) + rpc_count_iostats(task, task->tk_client->cl_metrics); ++ ++ rpc_multipath_ops_xprt_iostat(task); ++ + spin_lock(&xprt->recv_lock); + if (!list_empty(&req->rq_list)) { + list_del_init(&req->rq_list); +@@ -1455,6 +1468,7 @@ void xprt_release(struct rpc_task *task) + else + xprt_free_bc_request(req); + } ++EXPORT_SYMBOL_GPL(xprt_release); + + static void xprt_init(struct rpc_xprt *xprt, struct net *net) + { +@@ -1528,6 +1542,10 @@ struct rpc_xprt *xprt_create_transport(struct xprt_create *args) + return ERR_PTR(-ENOMEM); + } + ++if (rpc_multipath_ops_create_xprt(xprt)) { ++ xprt_destroy(xprt); ++ return ERR_PTR(-ENOMEM); ++} + rpc_xprt_debugfs_register(xprt); + + dprintk("RPC: created transport %p with %u slots\n", xprt, +@@ -1547,6 +1565,9 @@ static void xprt_destroy_cb(struct work_struct *work) + rpc_destroy_wait_queue(&xprt->sending); + rpc_destroy_wait_queue(&xprt->backlog); + kfree(xprt->servername); ++ ++ rpc_multipath_ops_destroy_xprt(xprt); ++ + /* + * Tear down transport state and free the rpc_xprt + */ +diff --git a/net/sunrpc/xprtmultipath.c b/net/sunrpc/xprtmultipath.c +index 6ebaa58b4..f6a004ee7 100644 +--- a/net/sunrpc/xprtmultipath.c ++++ b/net/sunrpc/xprtmultipath.c +@@ -18,6 +18,7 @@ + #include <linux/sunrpc/xprt.h> + #include <linux/sunrpc/addr.h> + #include <linux/sunrpc/xprtmultipath.h> ++#include <linux/sunrpc/sunrpc_enfs_adapter.h> + + typedef struct rpc_xprt *(*xprt_switch_find_xprt_t)(struct list_head *head, + const struct rpc_xprt *cur); +@@ -26,8 +27,8 @@ static const struct rpc_xprt_iter_ops rpc_xprt_iter_singular; + static const struct rpc_xprt_iter_ops rpc_xprt_iter_roundrobin; + static const struct rpc_xprt_iter_ops rpc_xprt_iter_listall; + +-static void xprt_switch_add_xprt_locked(struct rpc_xprt_switch *xps, +- struct rpc_xprt *xprt) ++void xprt_switch_add_xprt_locked(struct rpc_xprt_switch *xps, ++ struct rpc_xprt *xprt) + { + if (unlikely(xprt_get(xprt) == NULL)) + return; +@@ -36,7 +37,9 @@ static void xprt_switch_add_xprt_locked(struct rpc_xprt_switch *xps, + if (xps->xps_nxprts == 0) + xps->xps_net = xprt->xprt_net; + xps->xps_nxprts++; ++ rpc_xps_nactive_add_one(xps); + } ++EXPORT_SYMBOL(xprt_switch_add_xprt_locked); + + /** + * rpc_xprt_switch_add_xprt - Add a new rpc_xprt to an rpc_xprt_switch +@@ -63,6 +66,7 @@ static void xprt_switch_remove_xprt_locked(struct rpc_xprt_switch *xps, + if (unlikely(xprt == NULL)) + return; + xps->xps_nxprts--; ++ rpc_xps_nactive_sub_one(xps); + if (xps->xps_nxprts == 0) + xps->xps_net = NULL; + smp_wmb(); +@@ -84,7 +88,7 @@ void rpc_xprt_switch_remove_xprt(struct rpc_xprt_switch *xps, + spin_unlock(&xps->xps_lock); + xprt_put(xprt); + } +- ++EXPORT_SYMBOL(rpc_xprt_switch_remove_xprt); + /** + * xprt_switch_alloc - Allocate a new struct rpc_xprt_switch + * @xprt: pointer to struct rpc_xprt +@@ -103,6 +107,10 @@ struct rpc_xprt_switch *xprt_switch_alloc(struct rpc_xprt *xprt, + spin_lock_init(&xps->xps_lock); + kref_init(&xps->xps_kref); + xps->xps_nxprts = 0; ++#if IS_ENABLED(CONFIG_ENFS) ++ xps->xps_nactive = 0; ++ atomic_long_set(&xps->xps_queuelen, 0); ++#endif + INIT_LIST_HEAD(&xps->xps_xprt_list); + xps->xps_iter_ops = &rpc_xprt_iter_singular; + xprt_switch_add_xprt_locked(xps, xprt); +@@ -148,6 +156,7 @@ struct rpc_xprt_switch *xprt_switch_get(struct rpc_xprt_switch *xps) + return xps; + return NULL; + } ++EXPORT_SYMBOL(xprt_switch_get); + + /** + * xprt_switch_put - Release a reference to a rpc_xprt_switch +@@ -160,6 +169,7 @@ void xprt_switch_put(struct rpc_xprt_switch *xps) + if (xps != NULL) + kref_put(&xps->xps_kref, xprt_switch_free); + } ++EXPORT_SYMBOL(xprt_switch_put); + + /** + * rpc_xprt_switch_set_roundrobin - Set a round-robin policy on rpc_xprt_switch diff --git a/0003-add_enfs_module_for_nfs_mount_option.patch b/0003-add_enfs_module_for_nfs_mount_option.patch new file mode 100644 index 0000000..7461fb6 --- /dev/null +++ b/0003-add_enfs_module_for_nfs_mount_option.patch @@ -0,0 +1,1208 @@ +diff --git a/fs/nfs/enfs/Makefile b/fs/nfs/enfs/Makefile +new file mode 100644 +index 000000000..6e83eb23c +--- /dev/null ++++ b/fs/nfs/enfs/Makefile +@@ -0,0 +1,18 @@ ++obj-m += enfs.o ++ ++#EXTRA_CFLAGS += -I$(PWD)/.. ++ ++enfs-y := enfs_init.o ++enfs-y += enfs_config.o ++enfs-y += mgmt_init.o ++enfs-y += enfs_multipath_client.o ++enfs-y += enfs_multipath_parse.o ++enfs-y += failover_path.o ++enfs-y += failover_time.o ++enfs-y += enfs_roundrobin.o ++enfs-y += enfs_multipath.o ++enfs-y += enfs_path.o ++enfs-y += enfs_proc.o ++enfs-y += enfs_remount.o ++enfs-y += pm_ping.o ++enfs-y += pm_state.o +diff --git a/fs/nfs/enfs/enfs.h b/fs/nfs/enfs/enfs.h +new file mode 100644 +index 000000000..be3d95220 +--- /dev/null ++++ b/fs/nfs/enfs/enfs.h +@@ -0,0 +1,62 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Client-side ENFS multipath adapt header. ++ * ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++ ++#ifndef _ENFS_H_ ++#define _ENFS_H_ ++#include <linux/atomic.h> ++#include <linux/nfs.h> ++#include <linux/nfs4.h> ++#include <linux/nfs3.h> ++#include <linux/nfs_fs.h> ++#include <linux/nfs_fs_sb.h> ++#include "../enfs_adapter.h" ++ ++#define IP_ADDRESS_LEN_MAX 64 ++#define MAX_IP_PAIR_PER_MOUNT 8 ++#define MAX_IP_INDEX (MAX_IP_PAIR_PER_MOUNT) ++#define MAX_SUPPORTED_LOCAL_IP_COUNT 8 ++#define MAX_SUPPORTED_REMOTE_IP_COUNT 32 ++#define MAX_DNS_NAME_LEN 512 ++#define MAX_DNS_SUPPORTED 2 ++#define EXTEND_CMD_MAX_BUF_LEN 65356 ++ ++ ++struct nfs_ip_list { ++ int count; ++ struct sockaddr_storage address[MAX_SUPPORTED_REMOTE_IP_COUNT]; ++ size_t addrlen[MAX_SUPPORTED_REMOTE_IP_COUNT]; ++}; ++ ++struct NFS_ROUTE_DNS_S { ++ char dnsname[MAX_DNS_NAME_LEN]; // valid only if dnsExist is true ++}; ++ ++struct NFS_ROUTE_DNS_INFO_S { ++ int dnsNameCount; // Count of DNS name in the list ++ // valid only if dnsExist is true ++ struct NFS_ROUTE_DNS_S routeRemoteDnsList[MAX_DNS_SUPPORTED]; ++}; ++ ++struct rpc_iostats; ++struct enfs_xprt_context { ++ struct sockaddr_storage srcaddr; ++ struct rpc_iostats *stats; ++ bool main; ++ atomic_t path_state; ++ atomic_t path_check_state; ++}; ++ ++static inline bool enfs_is_main_xprt(struct rpc_xprt *xprt) ++{ ++ struct enfs_xprt_context *ctx = xprt->multipath_context; ++ ++ if (!ctx) ++ return false; ++ return ctx->main; ++} ++ ++#endif +diff --git a/fs/nfs/enfs/enfs_init.c b/fs/nfs/enfs/enfs_init.c +new file mode 100644 +index 000000000..4b5560819 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_init.c +@@ -0,0 +1,98 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Client-side ENFS adapter. ++ * ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++#include <linux/module.h> ++#include <linux/sunrpc/sched.h> ++#include <linux/sunrpc/clnt.h> ++#include <linux/nfs.h> ++#include <linux/nfs4.h> ++#include <linux/nfs3.h> ++#include <linux/nfs_fs.h> ++#include <linux/nfs_fs_sb.h> ++#include "enfs.h" ++#include "enfs_multipath_parse.h" ++#include "enfs_multipath_client.h" ++#include "enfs_remount.h" ++#include "init.h" ++#include "enfs_log.h" ++#include "enfs_multipath.h" ++#include "mgmt_init.h" ++ ++struct enfs_adapter_ops enfs_adapter = { ++ .name = "enfs", ++ .owner = THIS_MODULE, ++ .parse_mount_options = nfs_multipath_parse_options, ++ .free_mount_options = nfs_multipath_free_options, ++ .client_info_init = nfs_multipath_client_info_init, ++ .client_info_free = nfs_multipath_client_info_free, ++ .client_info_match = nfs_multipath_client_info_match, ++ .client_info_show = nfs_multipath_client_info_show, ++ .remount_ip_list = enfs_remount_iplist, ++}; ++ ++int32_t enfs_init(void) ++{ ++ int err; ++ ++ err = enfs_multipath_init(); ++ if (err) { ++ enfs_log_error("init multipath failed.\n"); ++ goto out; ++ } ++ ++ err = mgmt_init(); ++ if (err != 0) { ++ enfs_log_error("init mgmt failed.\n"); ++ goto out_tp_exit; ++ } ++ ++ return 0; ++ ++out_tp_exit: ++ enfs_multipath_exit(); ++out: ++ return err; ++} ++ ++void enfs_fini(void) ++{ ++ mgmt_fini(); ++ ++ enfs_multipath_exit(); ++} ++ ++static int __init init_enfs(void) ++{ ++ int ret; ++ ++ ret = enfs_adapter_register(&enfs_adapter); ++ if (ret) { ++ pr_err("regist enfs_adapter fail. ret %d\n", ret); ++ return -1; ++ } ++ ++ ret = enfs_init(); ++ if (ret) { ++ enfs_adapter_unregister(&enfs_adapter); ++ return -1; ++ } ++ ++ return 0; ++} ++ ++static void __exit exit_enfs(void) ++{ ++ enfs_fini(); ++ enfs_adapter_unregister(&enfs_adapter); ++} ++ ++MODULE_LICENSE("GPL"); ++MODULE_AUTHOR("Huawei Tech. Co., Ltd."); ++MODULE_DESCRIPTION("Nfs client router"); ++MODULE_VERSION("1.0"); ++ ++module_init(init_enfs); ++module_exit(exit_enfs); +diff --git a/fs/nfs/enfs/enfs_multipath_client.c b/fs/nfs/enfs/enfs_multipath_client.c +new file mode 100644 +index 000000000..63c02898a +--- /dev/null ++++ b/fs/nfs/enfs/enfs_multipath_client.c +@@ -0,0 +1,340 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Client-side ENFS adapter. ++ * ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++#include <linux/types.h> ++#include <linux/nfs.h> ++#include <linux/nfs4.h> ++#include <linux/nfs_fs.h> ++#include <linux/nfs_fs_sb.h> ++#include <linux/proc_fs.h> ++#include <linux/seq_file.h> ++#include <linux/sunrpc/clnt.h> ++#include <linux/sunrpc/addr.h> ++#include "enfs_multipath_client.h" ++#include "enfs_multipath_parse.h" ++ ++int ++nfs_multipath_client_mount_info_init(struct multipath_client_info *client_info, ++ const struct nfs_client_initdata *client_init_data) ++{ ++ struct multipath_mount_options *mount_options = ++ (struct multipath_mount_options *)client_init_data->enfs_option; ++ ++ if (mount_options->local_ip_list) { ++ client_info->local_ip_list = ++ kzalloc(sizeof(struct nfs_ip_list), GFP_KERNEL); ++ ++ if (!client_info->local_ip_list) ++ return -ENOMEM; ++ ++ memcpy(client_info->local_ip_list, mount_options->local_ip_list, ++ sizeof(struct nfs_ip_list)); ++ } ++ ++ if (mount_options->remote_ip_list) { ++ ++ client_info->remote_ip_list = ++ kzalloc(sizeof(struct nfs_ip_list), GFP_KERNEL); ++ ++ if (!client_info->remote_ip_list) { ++ kfree(client_info->local_ip_list); ++ client_info->local_ip_list = NULL; ++ return -ENOMEM; ++ } ++ memcpy(client_info->remote_ip_list, ++ mount_options->remote_ip_list, ++ sizeof(struct nfs_ip_list)); ++ } ++ ++ if (mount_options->pRemoteDnsInfo) { ++ client_info->pRemoteDnsInfo = ++ kzalloc(sizeof(struct NFS_ROUTE_DNS_INFO_S), GFP_KERNEL); ++ ++ if (!client_info->pRemoteDnsInfo) { ++ kfree(client_info->local_ip_list); ++ client_info->local_ip_list = NULL; ++ kfree(client_info->remote_ip_list); ++ client_info->remote_ip_list = NULL; ++ return -ENOMEM; ++ } ++ memcpy(client_info->pRemoteDnsInfo, ++ mount_options->pRemoteDnsInfo, ++ sizeof(struct NFS_ROUTE_DNS_INFO_S)); ++ } ++ return 0; ++} ++ ++void nfs_multipath_client_info_free_work(struct work_struct *work) ++{ ++ ++ struct multipath_client_info *clp_info; ++ ++ if (work == NULL) ++ return; ++ ++ clp_info = container_of(work, struct multipath_client_info, work); ++ ++ if (clp_info->local_ip_list != NULL) { ++ kfree(clp_info->local_ip_list); ++ clp_info->local_ip_list = NULL; ++ } ++ if (clp_info->remote_ip_list != NULL) { ++ kfree(clp_info->remote_ip_list); ++ clp_info->remote_ip_list = NULL; ++ } ++ kfree(clp_info); ++} ++ ++void nfs_multipath_client_info_free(void *data) ++{ ++ struct multipath_client_info *clp_info = ++ (struct multipath_client_info *)data; ++ ++ if (clp_info == NULL) ++ return; ++ pr_info("free client info %p.\n", clp_info); ++ INIT_WORK(&clp_info->work, nfs_multipath_client_info_free_work); ++ schedule_work(&clp_info->work); ++} ++ ++int nfs_multipath_client_info_init(void **data, ++ const struct nfs_client_initdata *cl_init) ++{ ++ int rc; ++ struct multipath_client_info *info; ++ struct multipath_client_info **enfs_info; ++ /* no multi path info, no need do multipath init */ ++ if (cl_init->enfs_option == NULL) ++ return 0; ++ enfs_info = (struct multipath_client_info **)data; ++ if (enfs_info == NULL) ++ return -EINVAL; ++ ++ if (*enfs_info == NULL) ++ *enfs_info = kzalloc(sizeof(struct multipath_client_info), ++ GFP_KERNEL); ++ ++ if (*enfs_info == NULL) ++ return -ENOMEM; ++ ++ info = (struct multipath_client_info *)*enfs_info; ++ pr_info("init client info %p.\n", info); ++ rc = nfs_multipath_client_mount_info_init(info, cl_init); ++ if (rc) { ++ nfs_multipath_client_info_free((void *)info); ++ return rc; ++ } ++ return rc; ++} ++ ++bool nfs_multipath_ip_list_info_match(const struct nfs_ip_list *ip_list_src, ++ const struct nfs_ip_list *ip_list_dst) ++{ ++ int i; ++ int j; ++ bool is_find; ++ /* if both are equal or NULL, then return true. */ ++ if (ip_list_src == ip_list_dst) ++ return true; ++ ++ if ((ip_list_src == NULL || ip_list_dst == NULL)) ++ return false; ++ ++ if (ip_list_src->count != ip_list_dst->count) ++ return false; ++ ++ for (i = 0; i < ip_list_src->count; i++) { ++ is_find = false; ++ for (j = 0; j < ip_list_src->count; j++) { ++ if (rpc_cmp_addr_port( ++ (const struct sockaddr *) ++ &ip_list_src->address[i], ++ (const struct sockaddr *) ++ &ip_list_dst->address[j]) ++ ) { ++ is_find = true; ++ break; ++ } ++ } ++ if (is_find == false) ++ return false; ++ } ++ return true; ++} ++ ++int ++nfs_multipath_dns_list_info_match( ++ const struct NFS_ROUTE_DNS_INFO_S *pRemoteDnsInfoSrc, ++ const struct NFS_ROUTE_DNS_INFO_S *pRemoteDnsInfoDst) ++{ ++ int i; ++ ++ /* if both are equal or NULL, then return true. */ ++ if (pRemoteDnsInfoSrc == pRemoteDnsInfoDst) ++ return true; ++ ++ if ((pRemoteDnsInfoSrc == NULL || pRemoteDnsInfoDst == NULL)) ++ return false; ++ ++ if (pRemoteDnsInfoSrc->dnsNameCount != pRemoteDnsInfoDst->dnsNameCount) ++ return false; ++ ++ for (i = 0; i < pRemoteDnsInfoSrc->dnsNameCount; i++) { ++ if (!strcmp(pRemoteDnsInfoSrc->routeRemoteDnsList[i].dnsname, ++ pRemoteDnsInfoDst->routeRemoteDnsList[i].dnsname)) ++ return false; ++ } ++ return true; ++} ++ ++int nfs_multipath_client_info_match(void *src, void *dst) ++{ ++ int ret = true; ++ ++ struct multipath_client_info *src_info; ++ struct multipath_mount_options *dst_info; ++ ++ src_info = (struct multipath_client_info *)src; ++ dst_info = (struct multipath_mount_options *)dst; ++ pr_info("try match client .\n"); ++ ret = nfs_multipath_ip_list_info_match(src_info->local_ip_list, ++ dst_info->local_ip_list); ++ if (ret == false) { ++ pr_err("local_ip not match.\n"); ++ return ret; ++ } ++ ++ ret = nfs_multipath_ip_list_info_match(src_info->remote_ip_list, ++ dst_info->remote_ip_list); ++ if (ret == false) { ++ pr_err("remote_ip not match.\n"); ++ return ret; ++ } ++ ++ ret = nfs_multipath_dns_list_info_match(src_info->pRemoteDnsInfo, ++ dst_info->pRemoteDnsInfo); ++ if (ret == false) { ++ pr_err("dns not match.\n"); ++ return ret; ++ } ++ pr_info("try match client ret %d.\n", ret); ++ return ret; ++} ++ ++void nfs_multipath_print_ip_info(struct seq_file *mount_option, ++ struct nfs_ip_list *ip_list, ++ const char *type) ++{ ++ char buf[IP_ADDRESS_LEN_MAX + 1]; ++ int len = 0; ++ int i = 0; ++ ++ seq_printf(mount_option, ",%s=", type); ++ for (i = 0; i < ip_list->count; i++) { ++ len = rpc_ntop((struct sockaddr *)&ip_list->address[i], ++ buf, IP_ADDRESS_LEN_MAX); ++ if (len > 0 && len < IP_ADDRESS_LEN_MAX) ++ buf[len] = '\0'; ++ ++ if (i == 0) ++ seq_printf(mount_option, "%s", buf); ++ else ++ seq_printf(mount_option, "~%s", buf); ++ dfprintk(MOUNT, ++ "NFS: show nfs mount option type:%s %s [%s]\n", ++ type, buf, __func__); ++ } ++} ++ ++void nfs_multipath_print_dns_info(struct seq_file *mount_option, ++ struct NFS_ROUTE_DNS_INFO_S *pRemoteDnsInfo, ++ const char *type) ++{ ++ int i = 0; ++ ++ seq_printf(mount_option, ",%s=", type); ++ for (i = 0; i < pRemoteDnsInfo->dnsNameCount; i++) { ++ if (i == 0) ++ seq_printf(mount_option, ++ "[%s", pRemoteDnsInfo->routeRemoteDnsList[i].dnsname); ++ else if (i == pRemoteDnsInfo->dnsNameCount - 1) ++ seq_printf(mount_option, ",%s]", ++ pRemoteDnsInfo->routeRemoteDnsList[i].dnsname); ++ else ++ seq_printf(mount_option, ++ ",%s", pRemoteDnsInfo->routeRemoteDnsList[i].dnsname); ++ } ++} ++ ++ ++static void multipath_print_sockaddr(struct seq_file *seq, ++ struct sockaddr *addr) ++{ ++ switch (addr->sa_family) { ++ case AF_INET: { ++ struct sockaddr_in *sin = (struct sockaddr_in *)addr; ++ ++ seq_printf(seq, "%pI4", &sin->sin_addr); ++ return; ++ } ++ case AF_INET6: { ++ struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)addr; ++ ++ seq_printf(seq, "%pI6", &sin6->sin6_addr); ++ return; ++ } ++ default: ++ break; ++ } ++ pr_err("unsupport family:%d\n", addr->sa_family); ++} ++ ++static void multipath_print_enfs_info(struct seq_file *seq, ++ struct nfs_server *server) ++{ ++ struct sockaddr_storage peeraddr; ++ struct rpc_clnt *next = server->client; ++ ++ rpc_peeraddr(server->client, ++ (struct sockaddr *)&peeraddr, sizeof(peeraddr)); ++ seq_puts(seq, ",enfs_info="); ++ multipath_print_sockaddr(seq, (struct sockaddr *)&peeraddr); ++ ++ while (next->cl_parent) { ++ if (next == next->cl_parent) ++ break; ++ next = next->cl_parent; ++ } ++ seq_printf(seq, "_%u", next->cl_clid); ++} ++ ++void nfs_multipath_client_info_show(struct seq_file *mount_option, void *data) ++{ ++ struct nfs_server *server = data; ++ struct multipath_client_info *client_info = ++ server->nfs_client->cl_multipath_data; ++ ++ dfprintk(MOUNT, "NFS: show nfs mount option[%s]\n", __func__); ++ if ((client_info->remote_ip_list) && ++ (client_info->remote_ip_list->count > 0)) ++ nfs_multipath_print_ip_info(mount_option, ++ client_info->remote_ip_list, ++ "remoteaddrs"); ++ ++ if ((client_info->local_ip_list) && ++ (client_info->local_ip_list->count > 0)) ++ nfs_multipath_print_ip_info(mount_option, ++ client_info->local_ip_list, ++ "localaddrs"); ++ ++ if ((client_info->pRemoteDnsInfo) && ++ (client_info->pRemoteDnsInfo->dnsNameCount > 0)) ++ nfs_multipath_print_dns_info(mount_option, ++ client_info->pRemoteDnsInfo, ++ "remotednsname"); ++ ++ multipath_print_enfs_info(mount_option, server); ++} +diff --git a/fs/nfs/enfs/enfs_multipath_client.h b/fs/nfs/enfs/enfs_multipath_client.h +new file mode 100644 +index 000000000..208f72606 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_multipath_client.h +@@ -0,0 +1,26 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Client-side ENFS adapter. ++ * ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++#ifndef _ENFS_MULTIPATH_CLIENT_H_ ++#define _ENFS_MULTIPATH_CLIENT_H_ ++ ++#include "enfs.h" ++ ++struct multipath_client_info { ++ struct work_struct work; ++ struct nfs_ip_list *remote_ip_list; ++ struct nfs_ip_list *local_ip_list; ++ struct NFS_ROUTE_DNS_INFO_S *pRemoteDnsInfo; ++ s64 client_id; ++}; ++ ++int nfs_multipath_client_info_init(void **data, ++ const struct nfs_client_initdata *cl_init); ++void nfs_multipath_client_info_free(void *data); ++int nfs_multipath_client_info_match(void *src, void *dst); ++void nfs_multipath_client_info_show(struct seq_file *mount_option, void *data); ++ ++#endif +diff --git a/fs/nfs/enfs/enfs_multipath_parse.c b/fs/nfs/enfs/enfs_multipath_parse.c +new file mode 100644 +index 000000000..868782b0a +--- /dev/null ++++ b/fs/nfs/enfs/enfs_multipath_parse.c +@@ -0,0 +1,600 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Client-side ENFS adapter. ++ * ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++#include <linux/types.h> ++#include <linux/nfs.h> ++#include <linux/nfs4.h> ++#include <linux/nfs_fs.h> ++#include <linux/nfs_fs_sb.h> ++#include <linux/kern_levels.h> ++#include <linux/sunrpc/addr.h> ++#include "enfs_multipath_parse.h" ++#include "enfs_log.h" ++ ++#define NFSDBG_FACILITY NFSDBG_CLIENT ++ ++void nfs_multipath_parse_ip_ipv6_add(struct sockaddr_in6 *sin6, int add_num) ++{ ++ int i; ++ ++ pr_info("NFS: before %08x%08x%08x%08x add_num: %d[%s]\n", ++ ntohl(sin6->sin6_addr.in6_u.u6_addr32[0]), ++ ntohl(sin6->sin6_addr.in6_u.u6_addr32[1]), ++ ntohl(sin6->sin6_addr.in6_u.u6_addr32[2]), ++ ntohl(sin6->sin6_addr.in6_u.u6_addr32[3]), ++ add_num, __func__); ++ for (i = 0; i < add_num; i++) { ++ sin6->sin6_addr.in6_u.u6_addr32[3] = ++ htonl(ntohl(sin6->sin6_addr.in6_u.u6_addr32[3]) + 1); ++ ++ if (sin6->sin6_addr.in6_u.u6_addr32[3] != 0) ++ continue; ++ ++ sin6->sin6_addr.in6_u.u6_addr32[2] = ++ htonl(ntohl(sin6->sin6_addr.in6_u.u6_addr32[2]) + 1); ++ ++ if (sin6->sin6_addr.in6_u.u6_addr32[2] != 0) ++ continue; ++ ++ sin6->sin6_addr.in6_u.u6_addr32[1] = ++ htonl(ntohl(sin6->sin6_addr.in6_u.u6_addr32[1]) + 1); ++ ++ if (sin6->sin6_addr.in6_u.u6_addr32[1] != 0) ++ continue; ++ ++ sin6->sin6_addr.in6_u.u6_addr32[0] = ++ htonl(ntohl(sin6->sin6_addr.in6_u.u6_addr32[0]) + 1); ++ ++ if (sin6->sin6_addr.in6_u.u6_addr32[0] != 0) ++ continue; ++ } ++ ++ return; ++ ++} ++ ++static int nfs_multipath_parse_ip_range(struct net *net_ns, const char *cursor, ++ struct nfs_ip_list *ip_list, enum nfs_multi_path_options type) ++{ ++ struct sockaddr_storage addr; ++ struct sockaddr_storage tmp_addr; ++ int i; ++ size_t len; ++ int add_num = 1; ++ bool duplicate_flag = false; ++ bool is_complete = false; ++ struct sockaddr_in *sin4; ++ struct sockaddr_in6 *sin6; ++ ++ pr_info("NFS: parsing nfs mount option '%s' type: %d[%s]\n", ++ cursor, type, __func__); ++ len = rpc_pton(net_ns, cursor, strlen(cursor), ++ (struct sockaddr *)&addr, sizeof(addr)); ++ if (!len) ++ return -EINVAL; ++ ++ if (addr.ss_family != ip_list->address[ip_list->count - 1].ss_family) { ++ pr_info("NFS: %s parsing nfs mount option type: %d fail.\n", ++ __func__, type); ++ return -EINVAL; ++ } ++ ++ if (rpc_cmp_addr((const struct sockaddr *) ++ &ip_list->address[ip_list->count - 1], ++ (const struct sockaddr *)&addr)) { ++ ++ pr_info("range ip is same ip.\n"); ++ return 0; ++ ++ } ++ ++ while (true) { ++ ++ tmp_addr = ip_list->address[ip_list->count - 1]; ++ ++ switch (addr.ss_family) { ++ case AF_INET: ++ sin4 = (struct sockaddr_in *)&tmp_addr; ++ ++ sin4->sin_addr.s_addr = ++ htonl(ntohl(sin4->sin_addr.s_addr) + add_num); ++ ++ pr_info("NFS: mount option ip%08x type: %d ipcont %d [%s]\n", ++ ntohl(sin4->sin_addr.s_addr), ++ type, ip_list->count, __func__); ++ break; ++ case AF_INET6: ++ sin6 = (struct sockaddr_in6 *)&tmp_addr; ++ nfs_multipath_parse_ip_ipv6_add(sin6, add_num); ++ pr_info("NFS: mount option ip %08x%08x%08x%08x type: %d ipcont %d [%s]\n", ++ ntohl(sin6->sin6_addr.in6_u.u6_addr32[0]), ++ ntohl(sin6->sin6_addr.in6_u.u6_addr32[1]), ++ ntohl(sin6->sin6_addr.in6_u.u6_addr32[2]), ++ ntohl(sin6->sin6_addr.in6_u.u6_addr32[3]), ++ type, ip_list->count, __func__); ++ break; ++ // return -EOPNOTSUPP; ++ default: ++ return -EOPNOTSUPP; ++ } ++ ++ if (rpc_cmp_addr((const struct sockaddr *)&tmp_addr, ++ (const struct sockaddr *)&addr)) { ++ is_complete = true; ++ } ++ // delete duplicate ip, continuosly repeat, skip it ++ for (i = 0; i < ip_list->count; i++) { ++ duplicate_flag = false; ++ if (rpc_cmp_addr((const struct sockaddr *) ++ &ip_list->address[i], ++ (const struct sockaddr *)&tmp_addr)) { ++ add_num++; ++ duplicate_flag = true; ++ break; ++ } ++ } ++ ++ if (duplicate_flag == false) { ++ pr_info("this ip not duplicate;"); ++ add_num = 1; ++ // if not repeat but omit limit return false ++ if ((type == LOCALADDR && ++ ip_list->count >= MAX_SUPPORTED_LOCAL_IP_COUNT) || ++ (type == REMOTEADDR && ++ ip_list->count >= MAX_SUPPORTED_REMOTE_IP_COUNT)) { ++ ++ pr_info("[MULTIPATH:%s] iplist for type %d reached %d, more than supported limit %d\n", ++ __func__, type, ip_list->count, ++ type == LOCALADDR ? ++ MAX_SUPPORTED_LOCAL_IP_COUNT : ++ MAX_SUPPORTED_REMOTE_IP_COUNT); ++ ip_list->count = 0; ++ return -ENOSPC; ++ } ++ ip_list->address[ip_list->count] = tmp_addr; ++ ++ ip_list->addrlen[ip_list->count] = ++ ip_list->addrlen[ip_list->count - 1]; ++ ++ ip_list->count += 1; ++ } ++ if (is_complete == true) ++ break; ++ } ++ return 0; ++} ++ ++int nfs_multipath_parse_ip_list_inter(struct nfs_ip_list *ip_list, ++ struct net *net_ns, ++ char *cursor, enum nfs_multi_path_options type) ++{ ++ int i = 0; ++ struct sockaddr_storage addr; ++ struct sockaddr_storage swap; ++ int len; ++ ++ pr_info("NFS: parsing nfs mount option '%s' type: %d[%s]\n", ++ cursor, type, __func__); ++ ++ len = rpc_pton(net_ns, cursor, ++ strlen(cursor), ++ (struct sockaddr *)&addr, sizeof(addr)); ++ if (!len) ++ return -EINVAL; ++ ++ // check repeated ip ++ for (i = 0; i < ip_list->count; i++) { ++ if (rpc_cmp_addr((const struct sockaddr *) ++ &ip_list->address[i], ++ (const struct sockaddr *)&addr)) { ++ ++ pr_info("NFS: mount option '%s' type:%d index %d same as before index %d [%s]\n", ++ cursor, type, ip_list->count, i, __func__); ++ // prevent this ip is beginning ++ // if repeated take it to the end of list ++ swap = ip_list->address[i]; ++ ++ ip_list->address[i] = ++ ip_list->address[ip_list->count-1]; ++ ++ ip_list->address[ip_list->count-1] = swap; ++ return 0; ++ } ++ } ++ // if not repeated, check exceed limit ++ if ((type == LOCALADDR && ++ ip_list->count >= MAX_SUPPORTED_LOCAL_IP_COUNT) || ++ (type == REMOTEADDR && ++ ip_list->count >= MAX_SUPPORTED_REMOTE_IP_COUNT)) { ++ ++ pr_info("[MULTIPATH:%s] iplist for type %d reached %d, more than supported limit %d\n", ++ __func__, type, ip_list->count, ++ type == LOCALADDR ? ++ MAX_SUPPORTED_LOCAL_IP_COUNT : ++ MAX_SUPPORTED_REMOTE_IP_COUNT); ++ ++ ip_list->count = 0; ++ return -ENOSPC; ++ } ++ ip_list->address[ip_list->count] = addr; ++ ip_list->addrlen[ip_list->count] = len; ++ ip_list->count++; ++ ++ return 0; ++} ++ ++char *nfs_multipath_parse_ip_list_get_cursor(char **buf_to_parse, bool *single) ++{ ++ char *cursor = NULL; ++ const char *single_sep = strchr(*buf_to_parse, '~'); ++ const char *range_sep = strchr(*buf_to_parse, '-'); ++ ++ *single = true; ++ if (range_sep) { ++ if (range_sep > single_sep) { // A-B or A~B-C ++ if (single_sep == NULL) { // A-B ++ cursor = strsep(buf_to_parse, "-"); ++ if (cursor) ++ *single = false; ++ } else// A~B-C ++ cursor = strsep(buf_to_parse, "~"); ++ } else { // A-B~C ++ cursor = strsep(buf_to_parse, "-"); ++ if (cursor) ++ *single = false; ++ } ++ } else { // A~B~C ++ cursor = strsep(buf_to_parse, "~"); ++ } ++ return cursor; ++} ++ ++bool nfs_multipath_parse_param_check(enum nfs_multi_path_options type, ++ struct multipath_mount_options *options) ++{ ++ if (type == REMOUNTREMOTEADDR && options->remote_ip_list->count != 0) { ++ memset(options->remote_ip_list, 0, sizeof(struct nfs_ip_list)); ++ return true; ++ } ++ if (type == REMOUNTLOCALADDR && options->local_ip_list->count != 0) { ++ memset(options->local_ip_list, 0, sizeof(struct nfs_ip_list)); ++ return true; ++ } ++ if ((type == REMOTEADDR || type == REMOTEDNSNAME) && ++ options->pRemoteDnsInfo->dnsNameCount != 0) { ++ ++ pr_info("[MULTIPATH:%s] parse for %d ,already have dns\n", ++ __func__, type); ++ return false; ++ } else if ((type == REMOTEADDR || type == REMOTEDNSNAME) && ++ options->remote_ip_list->count != 0) { ++ ++ pr_info("[MULTIPATH:%s] parse for %d ,already have iplist\n", ++ __func__, type); ++ return false; ++ } ++ return true; ++} ++ ++int nfs_multipath_parse_ip_list(char *buffer, struct net *net_ns, ++ struct multipath_mount_options *options, ++ enum nfs_multi_path_options type) ++{ ++ char *buf_to_parse = NULL; ++ bool prev_range = false; ++ int ret = 0; ++ char *cursor = NULL; ++ bool single = true; ++ struct nfs_ip_list *ip_list_tmp = NULL; ++ ++ if (!nfs_multipath_parse_param_check(type, options)) ++ return -ENOTSUPP; ++ ++ if (type == REMOUNTREMOTEADDR) ++ type = REMOTEADDR; ++ ++ if (type == REMOUNTLOCALADDR) ++ type = LOCALADDR; ++ ++ if (type == LOCALADDR) ++ ip_list_tmp = options->local_ip_list; ++ else ++ ip_list_tmp = options->remote_ip_list; ++ ++ pr_info("NFS: parsing nfs mount option '%s' type: %d[%s]\n", ++ buffer, type, __func__); ++ ++ buf_to_parse = buffer; ++ while (buf_to_parse != NULL) { ++ cursor = ++ nfs_multipath_parse_ip_list_get_cursor(&buf_to_parse, &single); ++ if (!cursor) ++ break; ++ ++ if (single == false && prev_range == true) { ++ pr_info("NFS: mount option type: %d fail. Multiple Range.[%s]\n", ++ type, __func__); ++ ++ ret = -EINVAL; ++ goto out; ++ } ++ ++ if (prev_range == false) { ++ ret = nfs_multipath_parse_ip_list_inter(ip_list_tmp, ++ net_ns, cursor, type); ++ if (ret) ++ goto out; ++ if (single == false) ++ prev_range = true; ++ } else { ++ ret = nfs_multipath_parse_ip_range(net_ns, cursor, ++ ip_list_tmp, type); ++ if (ret != 0) ++ goto out; ++ prev_range = false; ++ } ++ } ++ ++out: ++ if (ret) ++ memset(ip_list_tmp, 0, sizeof(struct nfs_ip_list)); ++ ++ return ret; ++} ++ ++int nfs_multipath_parse_dns_list(char *buffer, struct net *net_ns, ++ struct multipath_mount_options *options) ++{ ++ struct NFS_ROUTE_DNS_INFO_S *dns_name_list_tmp = NULL; ++ char *cursor = NULL; ++ char *bufToParse; ++ ++ if (!nfs_multipath_parse_param_check(REMOTEDNSNAME, options)) ++ return -ENOTSUPP; ++ ++ pr_info("[MULTIPATH:%s] buffer %s\n", __func__, buffer); ++ // freed in nfs_free_parsed_mount_data ++ dns_name_list_tmp = kmalloc(sizeof(struct NFS_ROUTE_DNS_INFO_S), ++ GFP_KERNEL); ++ if (!dns_name_list_tmp) ++ return -ENOMEM; ++ ++ dns_name_list_tmp->dnsNameCount = 0; ++ bufToParse = buffer; ++ while (bufToParse) { ++ if (dns_name_list_tmp->dnsNameCount >= MAX_DNS_SUPPORTED) { ++ pr_err("%s: dnsname for %s reached %d,more than supported limit %d\n", ++ __func__, cursor, ++ dns_name_list_tmp->dnsNameCount, ++ MAX_DNS_SUPPORTED); ++ dns_name_list_tmp->dnsNameCount = 0; ++ return -ENOSPC; ++ } ++ cursor = strsep(&bufToParse, "~"); ++ if (!cursor) ++ break; ++ ++ strcpy(dns_name_list_tmp->routeRemoteDnsList ++ [dns_name_list_tmp->dnsNameCount].dnsname, ++ cursor); ++ dns_name_list_tmp->dnsNameCount++; ++ } ++ if (dns_name_list_tmp->dnsNameCount == 0) ++ return -EINVAL; ++ options->pRemoteDnsInfo = dns_name_list_tmp; ++ return 0; ++} ++ ++int nfs_multipath_parse_options_check_ipv4_valid(struct sockaddr_in *addr) ++{ ++ if (addr->sin_addr.s_addr == 0 || addr->sin_addr.s_addr == 0xffffffff) ++ return -EINVAL; ++ return 0; ++} ++ ++int nfs_multipath_parse_options_check_ipv6_valid(struct sockaddr_in6 *addr) ++{ ++ if (addr->sin6_addr.in6_u.u6_addr32[0] == 0 && ++ addr->sin6_addr.in6_u.u6_addr32[1] == 0 && ++ addr->sin6_addr.in6_u.u6_addr32[2] == 0 && ++ addr->sin6_addr.in6_u.u6_addr32[3] == 0) ++ return -EINVAL; ++ ++ if (addr->sin6_addr.in6_u.u6_addr32[0] == 0xffffffff && ++ addr->sin6_addr.in6_u.u6_addr32[1] == 0xffffffff && ++ addr->sin6_addr.in6_u.u6_addr32[2] == 0xffffffff && ++ addr->sin6_addr.in6_u.u6_addr32[3] == 0xffffffff) ++ return -EINVAL; ++ return 0; ++} ++ ++int nfs_multipath_parse_options_check_ip_valid(struct sockaddr_storage *address) ++{ ++ int rc = 0; ++ ++ if (address->ss_family == AF_INET) ++ rc = nfs_multipath_parse_options_check_ipv4_valid( ++ (struct sockaddr_in *)address); ++ else if (address->ss_family == AF_INET6) ++ rc = nfs_multipath_parse_options_check_ipv6_valid( ++ (struct sockaddr_in6 *)address); ++ else ++ rc = -EINVAL; ++ ++ return rc; ++} ++ ++int nfs_multipath_parse_options_check_valid( ++ struct multipath_mount_options *options) ++{ ++ int rc; ++ int i; ++ ++ if (options == NULL) ++ return 0; ++ ++ for (i = 0; i < options->local_ip_list->count; i++) { ++ rc = nfs_multipath_parse_options_check_ip_valid( ++ &options->local_ip_list->address[i]); ++ if (rc != 0) ++ return rc; ++ } ++ ++ for (i = 0; i < options->remote_ip_list->count; i++) { ++ rc = nfs_multipath_parse_options_check_ip_valid( ++ &options->remote_ip_list->address[i]); ++ if (rc != 0) ++ return rc; ++ } ++ ++ return 0; ++} ++int nfs_multipath_parse_options_check_duplicate( ++ struct multipath_mount_options *options) ++{ ++ int i; ++ int j; ++ ++ if (options == NULL || ++ options->local_ip_list->count == 0 || ++ options->remote_ip_list->count == 0) ++ ++ return 0; ++ ++ for (i = 0; i < options->local_ip_list->count; i++) { ++ for (j = 0; j < options->remote_ip_list->count; j++) { ++ if (rpc_cmp_addr((const struct sockaddr *) ++ &options->local_ip_list->address[i], ++ (const struct sockaddr *) ++ &options->remote_ip_list->address[j])) ++ return -ENOTSUPP; ++ } ++ } ++ return 0; ++} ++ ++int nfs_multipath_parse_options_check(struct multipath_mount_options *options) ++{ ++ int rc = 0; ++ ++ rc = nfs_multipath_parse_options_check_valid(options); ++ ++ if (rc != 0) { ++ pr_err("has invaild ip.\n"); ++ return rc; ++ } ++ ++ rc = nfs_multipath_parse_options_check_duplicate(options); ++ if (rc != 0) ++ return rc; ++ return rc; ++} ++ ++int nfs_multipath_alloc_options(void **enfs_option) ++{ ++ struct multipath_mount_options *options = NULL; ++ ++ options = kzalloc(sizeof(struct multipath_mount_options), GFP_KERNEL); ++ ++ if (options == NULL) ++ return -ENOMEM; ++ ++ options->local_ip_list = ++ kzalloc(sizeof(struct nfs_ip_list), GFP_KERNEL); ++ if (options->local_ip_list == NULL) { ++ kfree(options); ++ return -ENOMEM; ++ } ++ ++ options->remote_ip_list = ++ kzalloc(sizeof(struct nfs_ip_list), GFP_KERNEL); ++ if (options->remote_ip_list == NULL) { ++ kfree(options->local_ip_list); ++ kfree(options); ++ return -ENOMEM; ++ } ++ ++ options->pRemoteDnsInfo = kzalloc(sizeof(struct NFS_ROUTE_DNS_INFO_S), ++ GFP_KERNEL); ++ if (options->pRemoteDnsInfo == NULL) { ++ kfree(options->remote_ip_list); ++ kfree(options->local_ip_list); ++ kfree(options); ++ return -ENOMEM; ++ } ++ ++ *enfs_option = options; ++ return 0; ++} ++ ++int nfs_multipath_parse_options(enum nfs_multi_path_options type, ++ char *str, void **enfs_option, struct net *net_ns) ++{ ++ int rc; ++ struct multipath_mount_options *options = NULL; ++ ++ if ((str == NULL) || (enfs_option == NULL) || (net_ns == NULL)) ++ return -EINVAL; ++ ++ if (*enfs_option == NULL) { ++ rc = nfs_multipath_alloc_options(enfs_option); ++ if (rc != 0) { ++ enfs_log_error( ++ "alloc enfs_options failed! errno:%d\n", rc); ++ return rc; ++ } ++ } ++ ++ options = (struct multipath_mount_options *)*enfs_option; ++ ++ if (type == LOCALADDR || type == REMOUNTLOCALADDR || ++ type == REMOTEADDR || type == REMOUNTREMOTEADDR) { ++ rc = nfs_multipath_parse_ip_list(str, net_ns, options, type); ++ } else if (type == REMOTEDNSNAME) { ++ /* alloc and release need to modify */ ++ rc = nfs_multipath_parse_dns_list(str, net_ns, options); ++ } else { ++ rc = -EOPNOTSUPP; ++ } ++ ++ // after parsing cmd, need checking local and remote ++ // IP is same. if not means illegal cmd ++ if (rc == 0) ++ rc = nfs_multipath_parse_options_check_duplicate(options); ++ ++ if (rc == 0) ++ rc = nfs_multipath_parse_options_check(options); ++ ++ return rc; ++} ++ ++void nfs_multipath_free_options(void **enfs_option) ++{ ++ struct multipath_mount_options *options; ++ ++ if (enfs_option == NULL || *enfs_option == NULL) ++ return; ++ ++ options = (struct multipath_mount_options *)*enfs_option; ++ ++ if (options->remote_ip_list != NULL) { ++ kfree(options->remote_ip_list); ++ options->remote_ip_list = NULL; ++ } ++ ++ if (options->local_ip_list != NULL) { ++ kfree(options->local_ip_list); ++ options->local_ip_list = NULL; ++ } ++ ++ if (options->pRemoteDnsInfo != NULL) { ++ kfree(options->pRemoteDnsInfo); ++ options->pRemoteDnsInfo = NULL; ++ } ++ ++ kfree(options); ++ *enfs_option = NULL; ++} +diff --git a/fs/nfs/enfs/enfs_multipath_parse.h b/fs/nfs/enfs/enfs_multipath_parse.h +new file mode 100644 +index 000000000..d5a36b2eb +--- /dev/null ++++ b/fs/nfs/enfs/enfs_multipath_parse.h +@@ -0,0 +1,22 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Client-side ENFS adapter. ++ * ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++#ifndef _ENFS_MULTIPATH_PARSE_H_ ++#define _ENFS_MULTIPATH_PARSE_H_ ++ ++#include "enfs.h" ++ ++struct multipath_mount_options { ++ struct nfs_ip_list *remote_ip_list; ++ struct nfs_ip_list *local_ip_list; ++ struct NFS_ROUTE_DNS_INFO_S *pRemoteDnsInfo; ++}; ++ ++int nfs_multipath_parse_options(enum nfs_multi_path_options type, ++ char *str, void **enfs_option, struct net *net_ns); ++void nfs_multipath_free_options(void **enfs_option); ++ ++#endif diff --git a/0004-add_enfs_module_for_sunrpc_multipatch.patch b/0004-add_enfs_module_for_sunrpc_multipatch.patch new file mode 100644 index 0000000..eb9acf8 --- /dev/null +++ b/0004-add_enfs_module_for_sunrpc_multipatch.patch @@ -0,0 +1,1860 @@ +diff --git a/fs/nfs/enfs/enfs_multipath.c b/fs/nfs/enfs/enfs_multipath.c +new file mode 100644 +index 000000000..d49dc91e1 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_multipath.c +@@ -0,0 +1,654 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ */ ++#include "enfs_multipath.h" ++#include <linux/in.h> ++#include <linux/in6.h> ++#include <linux/kallsyms.h> ++#include <linux/mm.h> ++#include <linux/module.h> ++#include <linux/mount.h> ++#include <linux/namei.h> ++#include <linux/rcupdate.h> ++#include <linux/slab.h> ++#include <linux/socket.h> ++#include <linux/atomic.h> ++#include <linux/sunrpc/addr.h> ++#include <linux/sunrpc/bc_xprt.h> ++#include <linux/sunrpc/clnt.h> ++#include <linux/sunrpc/metrics.h> ++#include <linux/sunrpc/rpc_pipe_fs.h> ++#include <linux/sunrpc/xprt.h> ++#include <linux/sunrpc/xprtmultipath.h> ++#include <linux/types.h> ++#include <linux/un.h> ++#include <linux/utsname.h> ++#include <linux/workqueue.h> ++#include <trace/events/sunrpc.h> ++#include <linux/sunrpc/sunrpc_enfs_adapter.h> ++ ++#include "enfs.h" ++#include "enfs_log.h" ++#include "enfs_multipath_parse.h" ++#include "enfs_path.h" ++#include "enfs_proc.h" ++#include "enfs_remount.h" ++#include "enfs_roundrobin.h" ++#include "failover_path.h" ++#include "failover_time.h" ++#include "pm_ping.h" ++#include "pm_state.h" ++ ++struct xprt_attach_callback_data { ++ atomic_t *conditon; ++ wait_queue_head_t *waitq; ++}; ++ ++struct xprt_attach_info { ++ struct sockaddr_storage *localAddress; ++ struct sockaddr_storage *remoteAddress; ++ struct rpc_xprt *xprt; ++ struct xprt_attach_callback_data *data; ++}; ++ ++static DECLARE_WAIT_QUEUE_HEAD(path_attach_wait_queue); ++ ++/** ++ * set socket port. ++ * @ns: need transform to network byte order ++ */ ++static void sockaddr_set_port(struct sockaddr *addr, __be16 port, bool ns) ++{ ++ switch (addr->sa_family) { ++ case AF_INET: ++ ((struct sockaddr_in *)addr)->sin_port = ++ ns ? htons(port) : port; ++ break; ++ case AF_INET6: ++ ((struct sockaddr_in6 *)addr)->sin6_port = ++ ns ? htons(port) : port; ++ break; ++ } ++} ++ ++static __be16 get_rpc_clnt_port(struct rpc_clnt *clnt) ++{ ++ struct sockaddr_storage ss; ++ struct sockaddr *addr = (struct sockaddr *)&ss; ++ ++ rpc_peeraddr(clnt, (struct sockaddr *)&ss, sizeof(ss)); ++ switch (addr->sa_family) { ++ case AF_INET: ++ return ((struct sockaddr_in *)addr)->sin_port; ++ ++ case AF_INET6: ++ return ((struct sockaddr_in6 *)addr)->sin6_port; ++ ++ default: ++ enfs_log_error("not support family:%d.\n", addr->sa_family); ++ return -1; ++ } ++} ++ ++static int sockaddr_ip_to_str(struct sockaddr *addr, char *buf, int len) ++{ ++ switch (addr->sa_family) { ++ case AF_INET: { ++ struct sockaddr_in *sin = (struct sockaddr_in *)addr; ++ ++ snprintf(buf, len, "%pI4", &sin->sin_addr); ++ return 0; ++ } ++ case AF_INET6: { ++ struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)addr; ++ ++ snprintf(buf, len, "%pI6", &sin6->sin6_addr); ++ return 0; ++ } ++ default: ++ break; ++ } ++ return 1; ++} ++ ++void print_enfs_multipath_addr(struct sockaddr *local, struct sockaddr *remote) ++{ ++ char buf1[128]; ++ char buf2[128]; ++ ++ sockaddr_ip_to_str(local, buf1, sizeof(buf1)); ++ sockaddr_ip_to_str(remote, buf2, sizeof(buf2)); ++ ++ pr_info("local:%s remote:%s\n", buf1, buf2); ++} ++ ++static int enfs_servername(char *servername, unsigned long long len, ++ struct rpc_create_args *args) ++{ ++ struct sockaddr_un *sun = (struct sockaddr_un *)args->address; ++ struct sockaddr_in *sin = (struct sockaddr_in *)args->address; ++ struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)args->address; ++ ++ servername[0] = '\0'; ++ switch (args->address->sa_family) { ++ case AF_LOCAL: ++ snprintf(servername, len, "%s", sun->sun_path); ++ break; ++ case AF_INET: ++ snprintf(servername, len, "%pI4", &sin->sin_addr.s_addr); ++ break; ++ case AF_INET6: ++ snprintf(servername, len, "%pI6", &sin6->sin6_addr); ++ break; ++ default: ++ pr_info("invalid family:%d\n", ++ args->address->sa_family); ++ return -EINVAL; ++ } ++ return 0; ++} ++ ++static void pm_xprt_ping_callback(void *data) ++{ ++ struct xprt_attach_callback_data *attach_callback_data = ++ (struct xprt_attach_callback_data *)data; ++ ++ atomic_dec(attach_callback_data->conditon); ++ wake_up(attach_callback_data->waitq); ++} ++ ++static ++int enfs_add_xprt_setup(struct rpc_clnt *clnt, struct rpc_xprt_switch *xps, ++ struct rpc_xprt *xprt, void *data) ++{ ++ int ret; ++ struct enfs_xprt_context *ctx; ++ struct xprt_attach_info *attach_info = data; ++ struct sockaddr_storage *srcaddr = attach_info->localAddress; ++ ++ ctx = (struct enfs_xprt_context *)xprt->multipath_context; ++ ctx->stats = rpc_alloc_iostats(clnt); ++ ctx->main = false; ++ ctx->srcaddr = *srcaddr; ++ pm_set_path_state(xprt, PM_STATE_INIT); ++ pm_ping_set_path_check_state(xprt, PM_CHECK_INIT); ++ ++ attach_info->xprt = xprt; ++ xprt_get(xprt); ++ ++ ret = pm_ping_rpc_test_xprt_with_callback(clnt, xprt, ++ pm_xprt_ping_callback, attach_info->data); ++ if (ret != 1) { ++ enfs_log_error("Failed to add ping task.\n"); ++ atomic_dec(attach_info->data->conditon); ++ } ++ ++ // so that rpc_clnt_add_xprt ++ // does not call rpc_xprt_switch_add_xprt ++ return 1; ++} ++ ++int enfs_configure_xprt_to_clnt(struct xprt_create *xprtargs, ++ struct rpc_clnt *clnt, ++ struct xprt_attach_info *attach_info) ++{ ++ int err = 0; ++ __be16 port; ++ ++ xprtargs->srcaddr = (struct sockaddr *)attach_info->localAddress; ++ xprtargs->dstaddr = (struct sockaddr *)attach_info->remoteAddress; ++ ++ port = get_rpc_clnt_port(clnt); ++ sockaddr_set_port((struct sockaddr *)attach_info->remoteAddress, port, ++ false); ++ print_enfs_multipath_addr((struct sockaddr *)attach_info->localAddress, ++ (struct sockaddr *)attach_info->remoteAddress); ++ ++ err = rpc_clnt_add_xprt(clnt, xprtargs, ++ enfs_add_xprt_setup, ++ attach_info); ++ if (err != 1) { ++ enfs_log_error("clnt add xprt err:%d\n", err); ++ return err; ++ } ++ return 0; ++} ++ ++// Calculate the greatest common divisor of two numbers ++static int enfs_cal_gcd(int num1, int num2) ++{ ++ if (num2 == 0) ++ return num1; ++ return enfs_cal_gcd(num2, num1 % num2); ++} ++ ++bool enfs_cmp_addrs(struct sockaddr_storage *srcaddr, ++ struct sockaddr_storage *dstaddr, ++ struct sockaddr_storage *localAddress, ++ struct sockaddr_storage *remoteAddress) ++{ ++ if (rpc_cmp_addr((struct sockaddr *)srcaddr, ++ (struct sockaddr *)localAddress)) { ++ ++ if (rpc_cmp_addr((struct sockaddr *)dstaddr, ++ (struct sockaddr *)remoteAddress)) ++ return true; ++ ++ } ++ ++ return false; ++} ++ ++bool enfs_xprt_addrs_is_same(struct rpc_xprt *xprt, ++ struct sockaddr_storage *localAddress, ++ struct sockaddr_storage *remoteAddress) ++{ ++ struct enfs_xprt_context *xprt_local_info = NULL; ++ struct sockaddr_storage *srcaddr = NULL; ++ struct sockaddr_storage *dstaddr = NULL; ++ ++ if (xprt == NULL) ++ return true; ++ xprt_local_info = (struct enfs_xprt_context *)xprt->multipath_context; ++ srcaddr = &xprt_local_info->srcaddr; ++ dstaddr = &xprt->addr; ++ ++ return enfs_cmp_addrs(srcaddr, dstaddr, localAddress, remoteAddress); ++} ++ ++bool enfs_already_have_xprt(struct rpc_clnt *clnt, ++ struct sockaddr_storage *localAddress, ++ struct sockaddr_storage *remoteAddress) ++{ ++ struct rpc_xprt *pos = NULL; ++ struct rpc_xprt_switch *xps = NULL; ++ ++ rcu_read_lock(); ++ xps = xprt_switch_get(rcu_dereference(clnt->cl_xpi.xpi_xpswitch)); ++ if (xps == NULL) { ++ rcu_read_unlock(); ++ return false; ++ } ++ list_for_each_entry_rcu(pos, &xps->xps_xprt_list, xprt_switch) { ++ if (enfs_xprt_addrs_is_same(pos, localAddress, remoteAddress)) { ++ xprt_switch_put(xps); ++ rcu_read_unlock(); ++ return true; ++ } ++ } ++ xprt_switch_put(xps); ++ rcu_read_unlock(); ++ return false; ++} ++ ++static void enfs_xprt_switch_add_xprt(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt) ++{ ++ struct rpc_xprt_switch *xps; ++ ++ rcu_read_lock(); ++ xps = rcu_dereference(clnt->cl_xpi.xpi_xpswitch); ++ spin_lock(&xps->xps_lock); ++ if (xps->xps_net == xprt->xprt_net || xps->xps_net == NULL) ++ xprt_switch_add_xprt_locked(xps, xprt); ++ spin_unlock(&xps->xps_lock); ++ rcu_read_unlock(); ++} ++ ++static void enfs_add_xprts_to_clnt(struct rpc_clnt *clnt, ++ struct xprt_attach_info *attach_infos, ++ int total_combinations) ++{ ++ struct rpc_xprt *xprt; ++ enum pm_path_state state; ++ int i; ++ int link_count = 0; ++ ++ for (i = 0; i < total_combinations; i++) { ++ xprt = attach_infos[i].xprt; ++ ++ if (xprt == NULL) ++ continue; ++ ++ state = pm_get_path_state(xprt); ++ ++ if (link_count < MAX_XPRT_NUM_PER_CLIENT && ++ state == PM_STATE_NORMAL) { ++ ++ enfs_xprt_switch_add_xprt(clnt, xprt); ++ link_count++; ++ continue; ++ ++ } ++ ++ xprt_put(xprt); ++ } ++} ++ ++static void enfs_combine_addr(struct xprt_create *xprtargs, ++ struct rpc_clnt *clnt, ++ struct nfs_ip_list *local, ++ struct nfs_ip_list *remote) ++{ ++ int i; ++ int err; ++ int local_index; ++ int remote_index; ++ int link_count = 0; ++ int local_total = local->count; ++ int remote_total = remote->count; ++ int local_remote_total_lcm; ++ int total_combinations = local_total * remote_total; ++ struct xprt_attach_info *attach_infos; ++ atomic_t wait_queue_condition; ++ ++ struct xprt_attach_callback_data attach_callback_data = { ++ &wait_queue_condition, &path_attach_wait_queue}; ++ ++ atomic_set(&wait_queue_condition, total_combinations); ++ ++ pr_info("local count:%d remote count:%d\n", local_total, remote_total); ++ if (local_total == 0 || remote_total == 0) { ++ pr_err("no ip list is present.\n"); ++ return; ++ } ++ ++ attach_infos = kzalloc( ++ (total_combinations) * sizeof(struct xprt_attach_info), ++ GFP_KERNEL); ++ if (attach_infos == NULL) { ++ enfs_log_error("Failed to kzalloc memory\n"); ++ return; ++ } ++ // Calculate the least common multiple of local_total and remote_total ++ local_remote_total_lcm = total_combinations / ++ enfs_cal_gcd(local_total, remote_total); ++ ++ // It needs to be offset one for lcm times of ++ // cycle so that all possible link setup method would be traversed ++ for (i = 0; i < total_combinations; i++) { ++ local_index = i % local_total; ++ remote_index = (i + link_count / local_remote_total_lcm) % ++ remote_total; ++ ++ if (enfs_already_have_xprt(clnt, ++ &local->address[local_index], ++ &remote->address[remote_index])) { ++ ++ atomic_dec(&wait_queue_condition); ++ link_count++; ++ continue; ++ ++ } ++ ++ attach_infos[i].localAddress = &local->address[local_index]; ++ attach_infos[i].remoteAddress = &remote->address[remote_index]; ++ attach_infos[i].data = &attach_callback_data; ++ ++ err = enfs_configure_xprt_to_clnt(xprtargs, ++ clnt, &attach_infos[i]); ++ if (err) { ++ pr_err("add xprt ippair err:%d\n", err); ++ atomic_dec(&wait_queue_condition); ++ } ++ link_count++; ++ } ++ ++ wait_event(path_attach_wait_queue, ++ atomic_read(&wait_queue_condition) == 0); ++ ++ enfs_add_xprts_to_clnt(clnt, attach_infos, total_combinations); ++ ++ kfree(attach_infos); ++} ++ ++void enfs_xprt_ippair_create(struct xprt_create *xprtargs, ++ struct rpc_clnt *clnt, ++ void *data) ++{ ++ struct multipath_mount_options *mopt = ++ (struct multipath_mount_options *)data; ++ ++ if (mopt == NULL) { ++ pr_err("ip list is NULL.\n"); ++ return; ++ } ++ ++ enfs_combine_addr(xprtargs, clnt, ++ mopt->local_ip_list, ++ mopt->remote_ip_list); ++ enfs_lb_set_policy(clnt); ++} ++ ++struct xprts_options_and_clnt { ++ struct rpc_create_args *args; ++ struct rpc_clnt *clnt; ++ void *data; ++}; ++ ++static void set_clnt_enfs_flag(struct rpc_clnt *clnt) ++{ ++ clnt->cl_enfs = true; ++} ++ ++int enfs_config_xprt_create_args(struct xprt_create *xprtargs, ++ struct rpc_create_args *args, ++ char *servername, size_t length) ++{ ++ int errno = 0; ++ ++ xprtargs->ident = args->protocol; ++ xprtargs->net = args->net; ++ xprtargs->addrlen = args->addrsize; ++ xprtargs->servername = args->servername; ++ ++ if (args->flags & RPC_CLNT_CREATE_INFINITE_SLOTS) ++ xprtargs->flags |= XPRT_CREATE_INFINITE_SLOTS; ++ if (args->flags & RPC_CLNT_CREATE_NO_IDLE_TIMEOUT) ++ xprtargs->flags |= XPRT_CREATE_NO_IDLE_TIMEOUT; ++ ++ if (xprtargs->servername == NULL) { ++ errno = enfs_servername(servername, length, args); ++ if (errno) ++ return errno; ++ xprtargs->servername = servername; ++ } ++ ++ return 0; ++} ++ ++int enfs_multipath_create_thread(void *data) ++{ ++ int errno; ++ char servername[48]; ++ struct xprts_options_and_clnt *create_args = ++ (struct xprts_options_and_clnt *)data; ++ struct multipath_mount_options *mount_options = ++ (struct multipath_mount_options *) ++ create_args->args->multipath_option; ++ struct xprt_create xprtargs; ++ ++ memset(&xprtargs, 0, sizeof(struct xprt_create)); ++ ++ if (mount_options == NULL) { ++ enfs_log_error("enfs: local and remot addr empty\n"); ++ return -EINVAL; ++ } ++ ++ errno = enfs_config_xprt_create_args(&xprtargs, ++ create_args->args, servername, ++ sizeof(servername)); ++ if (errno) { ++ enfs_log_error("enfs: create xprt failed! errno:%d\n", ++ errno); ++ return errno; ++ } ++ ++ //mount : localaddrs or remoteaddrs is empty ++ if (mount_options->local_ip_list->count == 0) { ++ errno = rpc_localaddr(create_args->clnt, ++ (struct sockaddr *) ++ &mount_options->local_ip_list->address[0], ++ sizeof(struct sockaddr_storage)); ++ if (errno) { ++ enfs_log_error("enfs: get clnt srcaddr errno:%d\n", ++ errno); ++ return errno; ++ } ++ mount_options->local_ip_list->count = 1; ++ } ++ ++ if (mount_options->remote_ip_list->count == 0) { ++ errno = rpc_peeraddr(create_args->clnt, ++ (struct sockaddr *) ++ &mount_options->remote_ip_list->address[0], ++ sizeof(struct sockaddr_storage)); ++ if (errno == 0) { ++ enfs_log_error("enfs: get clnt dstaddr errno:%d\n", ++ errno); ++ return errno; ++ } ++ mount_options->remote_ip_list->count = 1; ++ } ++ ++ errno = enfs_proc_create_clnt(create_args->clnt); ++ if (errno != 0) ++ pr_err("create clnt proc failed.\n"); ++ ++ set_clnt_enfs_flag(create_args->clnt); ++ enfs_xprt_ippair_create(&xprtargs, create_args->clnt, mount_options); ++ ++ kfree(create_args->args); ++ kfree(data); ++ return 0; ++} ++ ++static int set_main_xprt_ctx(struct rpc_clnt *clnt, struct rpc_xprt *xprt, ++ struct sockaddr_storage *srcaddr) ++{ ++ struct enfs_xprt_context *ctx = xprt->multipath_context; ++ ++ if (!ctx) { ++ enfs_log_error("main xprt not multipath ctx.\n"); ++ return -1; ++ } ++ ++ ctx->main = true; ++ ctx->stats = rpc_alloc_iostats(clnt); ++ ctx->srcaddr = *srcaddr; ++ pm_set_path_state(xprt, PM_STATE_NORMAL); ++ pm_ping_set_path_check_state(xprt, PM_CHECK_INIT); ++ ++ return 0; ++} ++ ++static int alloc_main_xprt_multicontext(struct rpc_create_args *args, ++ struct rpc_clnt *clnt) ++{ ++ int err; ++ struct sockaddr_storage srcaddr; ++ ++ // avoid main xprt multicontex local addr is empty. ++ err = rpc_localaddr(clnt, (struct sockaddr *)&srcaddr, sizeof(srcaddr)); ++ if (err) { ++ enfs_log_error("get clnt localaddr err:%d\n", err); ++ return err; ++ } ++ ++ err = set_main_xprt_ctx(clnt, clnt->cl_xprt, &srcaddr); ++ if (err) ++ enfs_log_error("alloc main xprt failed.\n"); ++ ++ return err; ++} ++ ++void enfs_create_multi_xprt(struct rpc_create_args *args, struct rpc_clnt *clnt) ++{ ++ // struct task_struct *th; ++ struct xprts_options_and_clnt *thargs; ++ struct rpc_create_args *cargs; ++ int err; ++ ++ enfs_log_info("%s %p\n", __func__, clnt); ++ ++ cargs = kmalloc(sizeof(struct rpc_create_args), GFP_KERNEL); ++ if (cargs == NULL) ++ return; ++ ++ *cargs = *args; ++ ++ thargs = kmalloc(sizeof(struct xprts_options_and_clnt), GFP_KERNEL); ++ if (thargs == NULL) { ++ kfree(cargs); ++ return; ++ } ++ ++ alloc_main_xprt_multicontext(args, clnt); ++ ++ thargs->args = cargs; ++ thargs->clnt = clnt; ++ thargs->data = args->multipath_option; ++ ++ err = enfs_multipath_create_thread(thargs); ++ ++ if (err != 0) { ++ kfree(cargs); ++ kfree(thargs); ++ } ++} ++ ++static void enfs_create_xprt_ctx(struct rpc_xprt *xprt) ++{ ++ int err; ++ ++ err = enfs_alloc_xprt_ctx(xprt); ++ if (err) ++ enfs_log_error("alloc xprt failed.\n"); ++} ++ ++static struct rpc_multipath_ops ops = { ++ .owner = THIS_MODULE, ++ .create_clnt = enfs_create_multi_xprt, ++ .releas_clnt = enfs_proc_delete_clnt, ++ .create_xprt = enfs_create_xprt_ctx, ++ .destroy_xprt = enfs_free_xprt_ctx, ++ .xprt_iostat = enfs_count_iostat, ++ .failover_handle = failover_handle, ++ .task_need_call_start_again = failover_task_need_call_start_again, ++ .adjust_task_timeout = failover_adjust_task_timeout, ++ .init_task_req = failover_init_task_req, ++ .prepare_transmit = failover_prepare_transmit, ++}; ++ ++int enfs_multipath_init(void) ++{ ++ int err; ++ ++ enfs_log_info("multipath init.\n"); ++ ++ err = pm_ping_init(); ++ if (err != 0) { ++ enfs_log_error("pm ping init err:%d\n", err); ++ return err; ++ } ++ ++ err = enfs_proc_init(); ++ if (err != 0) { ++ enfs_log_error("enfs proc init err:%d\n", err); ++ pm_ping_fini(); ++ return err; ++ } ++ ++ rpc_multipath_ops_register(&ops); ++ ++ return 0; ++} ++ ++void enfs_multipath_exit(void) ++{ ++ enfs_log_info("multipath exit.\n"); ++ rpc_multipath_ops_unregister(&ops); ++ enfs_proc_exit(); ++ pm_ping_fini(); ++} +diff --git a/fs/nfs/enfs/enfs_multipath.h b/fs/nfs/enfs/enfs_multipath.h +new file mode 100644 +index 000000000..3c58b0c4c +--- /dev/null ++++ b/fs/nfs/enfs/enfs_multipath.h +@@ -0,0 +1,23 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: enfs multipath ++ * Create: 2023-07-31 ++ */ ++ ++#ifndef ENFS_MULTIPATH_H ++#define ENFS_MULTIPATH_H ++#include <linux/sunrpc/clnt.h> ++ ++#define MAX_XPRT_NUM_PER_CLIENT 32 ++ ++int enfs_multipath_init(void); ++void enfs_multipath_exit(void); ++void enfs_xprt_ippair_create(struct xprt_create *xprtargs, ++ struct rpc_clnt *clnt, void *data); ++int enfs_config_xprt_create_args(struct xprt_create *xprtargs, ++ struct rpc_create_args *args, ++ char *servername, size_t length); ++void print_enfs_multipath_addr(struct sockaddr *local, struct sockaddr *remote); ++ ++#endif // ENFS_MULTIPATH_H +diff --git a/fs/nfs/enfs/enfs_path.c b/fs/nfs/enfs/enfs_path.c +new file mode 100644 +index 000000000..7355f8c2f +--- /dev/null ++++ b/fs/nfs/enfs/enfs_path.c +@@ -0,0 +1,47 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ */ ++ ++#include <linux/sunrpc/metrics.h> ++#include <linux/sunrpc/xprt.h> ++ ++#include "enfs.h" ++#include "enfs_log.h" ++#include "enfs_path.h" ++ ++// only create ctx in this function ++// alloc iostat memory in create_clnt ++int enfs_alloc_xprt_ctx(struct rpc_xprt *xprt) ++{ ++ struct enfs_xprt_context *ctx; ++ ++ if (!xprt) { ++ enfs_log_error("invalid xprt pointer.\n"); ++ return -EINVAL; ++ } ++ ++ ctx = kzalloc(sizeof(struct enfs_xprt_context), GFP_KERNEL); ++ if (!ctx) { ++ enfs_log_error("add xprt test failed.\n"); ++ return -ENOMEM; ++ } ++ ++ xprt->multipath_context = (void *)ctx; ++ return 0; ++} ++ ++// free multi_context and iostat memory ++void enfs_free_xprt_ctx(struct rpc_xprt *xprt) ++{ ++ struct enfs_xprt_context *ctx = xprt->multipath_context; ++ ++ if (ctx) { ++ if (ctx->stats) { ++ rpc_free_iostats(ctx->stats); ++ ctx->stats = NULL; ++ } ++ kfree(xprt->multipath_context); ++ xprt->multipath_context = NULL; ++ } ++} +diff --git a/fs/nfs/enfs/enfs_path.h b/fs/nfs/enfs/enfs_path.h +new file mode 100644 +index 000000000..97b1ef373 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_path.h +@@ -0,0 +1,12 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ */ ++ ++#ifndef ENFS_PATH_H ++#define ENFS_PATH_H ++ ++int enfs_alloc_xprt_ctx(struct rpc_xprt *xprt); ++void enfs_free_xprt_ctx(struct rpc_xprt *xprt); ++ ++#endif // ENFS_PATH_H +diff --git a/fs/nfs/enfs/enfs_proc.c b/fs/nfs/enfs/enfs_proc.c +new file mode 100644 +index 000000000..53fa1a076 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_proc.c +@@ -0,0 +1,545 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ */ ++#include <linux/module.h> ++#include <linux/proc_fs.h> ++#include <linux/seq_file.h> ++#include <linux/spinlock.h> ++#include <linux/sunrpc/clnt.h> ++#include <linux/sunrpc/metrics.h> ++#include <linux/sunrpc/xprtsock.h> ++#include <net/netns/generic.h> ++ ++#include "../../../net/sunrpc/netns.h" ++ ++#include "enfs.h" ++#include "enfs_log.h" ++#include "enfs_proc.h" ++#include "enfs_multipath.h" ++#include "pm_state.h" ++ ++#define ENFS_PROC_DIR "enfs" ++#define ENFS_PROC_PATH_STATUS_LEN 256 ++ ++static struct proc_dir_entry *enfs_proc_parent; ++ ++void ++enfs_iterate_each_rpc_clnt(int (*fn)(struct rpc_clnt *clnt, void *data), ++ void *data) ++{ ++ struct net *net; ++ struct sunrpc_net *sn; ++ struct rpc_clnt *clnt; ++ ++ rcu_read_lock(); ++ for_each_net_rcu(net) { ++ sn = net_generic(net, sunrpc_net_id); ++ if (sn == NULL) ++ continue; ++ spin_lock(&sn->rpc_client_lock); ++ list_for_each_entry(clnt, &sn->all_clients, cl_clients) { ++ fn(clnt, data); ++ } ++ spin_unlock(&sn->rpc_client_lock); ++ } ++ rcu_read_unlock(); ++} ++ ++struct proc_dir_entry *enfs_get_proc_parent(void) ++{ ++ return enfs_proc_parent; ++} ++ ++static int sockaddr_ip_to_str(struct sockaddr *addr, char *buf, int len) ++{ ++ switch (addr->sa_family) { ++ case AF_INET: { ++ struct sockaddr_in *sin = (struct sockaddr_in *)addr; ++ ++ snprintf(buf, len, "%pI4", &sin->sin_addr); ++ return 0; ++ } ++ case AF_INET6: { ++ struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)addr; ++ ++ snprintf(buf, len, "%pI6", &sin6->sin6_addr); ++ return 0; ++ } ++ default: ++ break; ++ } ++ return 1; ++} ++ ++static bool should_print(const char *name) ++{ ++ int i; ++ static const char * const proc_names[] = { ++ "READ", ++ "WRITE", ++ }; ++ ++ if (name == NULL) ++ return false; ++ ++ for (i = 0; i < ARRAY_SIZE(proc_names); i++) { ++ if (strcmp(name, proc_names[i]) == 0) ++ return true; ++ } ++ return false; ++} ++ ++struct enfs_xprt_iter { ++ unsigned int id; ++ struct seq_file *seq; ++ unsigned int max_addrs_length; ++}; ++ ++static int debug_show_xprt(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt, ++ void *data) ++{ ++ struct enfs_xprt_context *ctx = NULL; ++ ++ if (xprt->multipath_context) ++ ctx = xprt->multipath_context; ++ ++ pr_info(" xprt:%p ctx:%p main:%d queue_len:%lu.\n", xprt, ++ xprt->multipath_context, ++ ctx ? ctx->main : false, ++ atomic_long_read(&xprt->queuelen)); ++ return 0; ++} ++ ++static int debug_show_clnt(struct rpc_clnt *clnt, void *data) ++{ ++ pr_info(" clnt %d addr:%p enfs:%d\n", ++ clnt->cl_clid, clnt, ++ clnt->cl_enfs); ++ rpc_clnt_iterate_for_each_xprt(clnt, debug_show_xprt, NULL); ++ return 0; ++} ++ ++static void debug_print_all_xprt(void) ++{ ++ enfs_iterate_each_rpc_clnt(debug_show_clnt, NULL); ++} ++ ++static ++void enfs_proc_format_xprt_addr_display(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt, ++ char *local_name_buf, ++ int local_name_buf_len, ++ char *remote_name_buf, ++ int remote_name_buf_len) ++{ ++ int err; ++ struct sockaddr_storage srcaddr; ++ struct enfs_xprt_context *ctx; ++ ++ ctx = (struct enfs_xprt_context *)xprt->multipath_context; ++ ++ sockaddr_ip_to_str((struct sockaddr *)&xprt->addr, ++ remote_name_buf, remote_name_buf_len); ++ ++ // get local address depend one main or not ++ if (enfs_is_main_xprt(xprt)) { ++ err = rpc_localaddr(clnt, (struct sockaddr *)&srcaddr, ++ sizeof(srcaddr)); ++ if (err != 0) ++ (void)snprintf(local_name_buf, ++ local_name_buf_len, "Unknown"); ++ else ++ sockaddr_ip_to_str((struct sockaddr *)&srcaddr, ++ local_name_buf, ++ local_name_buf_len); ++ } else { ++ sockaddr_ip_to_str((struct sockaddr *)&ctx->srcaddr, ++ local_name_buf, ++ local_name_buf_len); ++ } ++} ++ ++static int enfs_show_xprt_stats(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt, ++ void *data) ++{ ++ unsigned int op; ++ unsigned int maxproc = clnt->cl_maxproc; ++ struct enfs_xprt_iter *iter = (struct enfs_xprt_iter *)data; ++ struct enfs_xprt_context *ctx; ++ char local_name[INET6_ADDRSTRLEN]; ++ char remote_name[INET6_ADDRSTRLEN]; ++ ++ if (!xprt->multipath_context) ++ return 0; ++ ++ ctx = (struct enfs_xprt_context *)xprt->multipath_context; ++ ++ enfs_proc_format_xprt_addr_display(clnt, xprt, local_name, ++ sizeof(local_name), ++ remote_name, sizeof(remote_name)); ++ ++ seq_printf(iter->seq, "%-6u%-*s%-*s", iter->id, ++ iter->max_addrs_length + 4, ++ local_name, ++ iter->max_addrs_length + 4, ++ remote_name); ++ ++ iter->id++; ++ ++ for (op = 0; op < maxproc; op++) { ++ if (!should_print(clnt->cl_procinfo[op].p_name)) ++ continue; ++ ++ seq_printf(iter->seq, "%-22lu%-22Lu%-22Lu", ++ ctx->stats[op].om_ops, ++ ctx->stats[op].om_ops == 0 ? 0 : ++ ktime_to_ms(ctx->stats[op].om_rtt) / ++ ctx->stats[op].om_ops, ++ ctx->stats[op].om_ops == 0 ? 0 : ++ ktime_to_ms(ctx->stats[op].om_execute) / ++ ctx->stats[op].om_ops); ++ } ++ seq_puts(iter->seq, "\n"); ++ return 0; ++} ++ ++static int rpc_proc_show_path_status(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt, ++ void *data) ++{ ++ struct enfs_xprt_iter *iter = (struct enfs_xprt_iter *)data; ++ struct enfs_xprt_context *ctx = NULL; ++ char local_name[INET6_ADDRSTRLEN] = {0}; ++ char remote_name[INET6_ADDRSTRLEN] = {0}; ++ char multiapth_status[ENFS_PROC_PATH_STATUS_LEN] = {0}; ++ char xprt_status[ENFS_PROC_PATH_STATUS_LEN] = {0}; ++ ++ if (!xprt->multipath_context) { ++ enfs_log_debug("multipath_context is null.\n"); ++ return 0; ++ } ++ ctx = (struct enfs_xprt_context *)xprt->multipath_context; ++ ++ enfs_proc_format_xprt_addr_display(clnt, xprt, ++ local_name, ++ sizeof(local_name), ++ remote_name, sizeof(remote_name)); ++ ++ pm_get_path_state_desc(xprt, ++ multiapth_status, ++ ENFS_PROC_PATH_STATUS_LEN); ++ ++ pm_get_xprt_state_desc(xprt, ++ xprt_status, ++ ENFS_PROC_PATH_STATUS_LEN); ++ ++ seq_printf(iter->seq, "%-6u%-*s%-*s%-12s%-12s\n", ++ iter->id, iter->max_addrs_length + 4, ++ local_name, iter->max_addrs_length + 4, ++ remote_name, multiapth_status, ++ xprt_status); ++ iter->id++; ++ return 0; ++} ++ ++static int enfs_get_max_addrs_length(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt, ++ void *data) ++{ ++ struct enfs_xprt_iter *iter = (struct enfs_xprt_iter *)data; ++ char local_name[INET6_ADDRSTRLEN]; ++ char remote_name[INET6_ADDRSTRLEN]; ++ ++ enfs_proc_format_xprt_addr_display(clnt, xprt, ++ local_name, sizeof(local_name), ++ remote_name, sizeof(remote_name)); ++ ++ if (iter->max_addrs_length < strlen(local_name)) ++ iter->max_addrs_length = strlen(local_name); ++ ++ if (iter->max_addrs_length < strlen(remote_name)) ++ iter->max_addrs_length = strlen(remote_name); ++ ++ return 0; ++} ++ ++static int rpc_proc_clnt_showpath(struct seq_file *seq, void *v) ++{ ++ struct rpc_clnt *clnt = seq->private; ++ struct enfs_xprt_iter iter; ++ ++ iter.seq = seq; ++ iter.id = 0; ++ iter.max_addrs_length = 0; ++ ++ rpc_clnt_iterate_for_each_xprt(clnt, ++ enfs_get_max_addrs_length, ++ (void *)&iter); ++ ++ seq_printf(seq, "%-6s%-*s%-*s%-12s%-12s\n", "id", ++ iter.max_addrs_length + 4, ++ "local_addr", ++ iter.max_addrs_length + 4, ++ "remote_addr", ++ "path_state", ++ "xprt_state"); ++ ++ rpc_clnt_iterate_for_each_xprt(clnt, ++ rpc_proc_show_path_status, ++ (void *)&iter); ++ return 0; ++} ++ ++static int enfs_rpc_proc_show(struct seq_file *seq, void *v) ++{ ++ struct rpc_clnt *clnt = seq->private; ++ struct enfs_xprt_iter iter; ++ ++ iter.seq = seq; ++ iter.id = 0; ++ iter.max_addrs_length = 0; ++ ++ debug_print_all_xprt(); ++ pr_info("enfs proc clnt:%p\n", clnt); ++ ++ rpc_clnt_iterate_for_each_xprt(clnt, ++ enfs_get_max_addrs_length, ++ (void *)&iter); ++ ++ seq_printf(seq, "%-6s%-*s%-*s%-22s%-22s%-22s%-22s%-22s%-22s\n", "id", ++ iter.max_addrs_length + 4, "local_addr", ++ iter.max_addrs_length + 4, ++ "remote_addr", "r_count", ++ "r_rtt", "r_exec", "w_count", "w_rtt", "w_exec"); ++ ++ // rpc_clnt_show_stats(seq, clnt); ++ rpc_clnt_iterate_for_each_xprt(clnt, ++ enfs_show_xprt_stats, ++ (void *)&iter); ++ return 0; ++} ++ ++static int rpc_proc_open(struct inode *inode, struct file *file) ++{ ++ struct rpc_clnt *clnt = PDE_DATA(inode); ++ ++ pr_info("%s %p\n", __func__, clnt); ++ return single_open(file, enfs_rpc_proc_show, clnt); ++} ++ ++static int enfs_reset_xprt_stats(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt, ++ void *data) ++{ ++ unsigned int op; ++ struct enfs_xprt_context *ctx; ++ unsigned int maxproc = clnt->cl_maxproc; ++ struct rpc_iostats stats = {0}; ++ ++ if (!xprt->multipath_context) ++ return 0; ++ ctx = (struct enfs_xprt_context *)xprt->multipath_context; ++ ++ for (op = 0; op < maxproc; op++) { ++ spin_lock(&ctx->stats[op].om_lock); ++ ctx->stats[op] = stats; ++ spin_unlock(&ctx->stats[op].om_lock); ++ } ++ return 0; ++} ++ ++static void trim_newline_ch(char *str, int len) ++{ ++ int i; ++ ++ for (i = 0; str[i] != '\0' && i < len; i++) { ++ if (str[i] == '\n') ++ str[i] = '\0'; ++ } ++} ++ ++static ssize_t enfs_proc_write(struct file *file, ++ const char __user *user_buf, ++ size_t len, ++ loff_t *offset) ++{ ++ char buffer[128]; ++ struct rpc_clnt *clnt = ++ ((struct seq_file *)file->private_data)->private; ++ ++ if (len >= sizeof(buffer)) ++ return -E2BIG; ++ ++ if (copy_from_user(buffer, user_buf, len) != 0) ++ return -EFAULT; ++ ++ buffer[len] = '\0'; ++ trim_newline_ch(buffer, len); ++ if (strcmp(buffer, "reset") != 0) ++ return -EINVAL; ++ ++ rpc_clnt_iterate_for_each_xprt(clnt, enfs_reset_xprt_stats, NULL); ++ return len; ++} ++ ++static int rpc_proc_show_path(struct inode *inode, struct file *file) ++{ ++ struct rpc_clnt *clnt = PDE_DATA(inode); ++ ++ return single_open(file, rpc_proc_clnt_showpath, clnt); ++} ++ ++static const struct file_operations rpc_proc_fops = { ++ .owner = THIS_MODULE, ++ .open = rpc_proc_open, ++ .read = seq_read, ++ .llseek = seq_lseek, ++ .release = single_release, ++ .write = enfs_proc_write, ++}; ++ ++static const struct file_operations rpc_show_path_fops = { ++ .owner = THIS_MODULE, ++ .open = rpc_proc_show_path, ++ .read = seq_read, ++ .llseek = seq_lseek, ++ .release = single_release, ++}; ++ ++static int clnt_proc_name(struct rpc_clnt *clnt, char *buf, int len) ++{ ++ int ret; ++ ++ ret = snprintf(buf, len, "%s_%u", ++ rpc_peeraddr2str(clnt, RPC_DISPLAY_ADDR), ++ clnt->cl_clid); ++ if (ret > len) ++ return -E2BIG; ++ return 0; ++} ++ ++static int enfs_proc_create_file(struct rpc_clnt *clnt) ++{ ++ int err; ++ char buf[128]; ++ ++ struct proc_dir_entry *clnt_entry; ++ struct proc_dir_entry *stat_entry; ++ ++ err = clnt_proc_name(clnt, buf, sizeof(buf)); ++ if (err) ++ return err; ++ ++ clnt_entry = proc_mkdir(buf, enfs_proc_parent); ++ if (clnt_entry == NULL) ++ return -EINVAL; ++ ++ stat_entry = proc_create_data("stat", ++ 0, clnt_entry, ++ &rpc_proc_fops, clnt); ++ ++ if (stat_entry == NULL) ++ return -EINVAL; ++ ++ stat_entry = proc_create_data("path", ++ 0, clnt_entry, ++ &rpc_show_path_fops, clnt); ++ ++ if (stat_entry == NULL) ++ return -EINVAL; ++ ++ return 0; ++} ++ ++void enfs_count_iostat(struct rpc_task *task) ++{ ++ struct enfs_xprt_context *ctx = task->tk_xprt->multipath_context; ++ ++ if (!ctx || !ctx->stats) ++ return; ++ rpc_count_iostats(task, ctx->stats); ++} ++ ++static void enfs_proc_delete_file(struct rpc_clnt *clnt) ++{ ++ int err; ++ char buf[128]; ++ ++ err = clnt_proc_name(clnt, buf, sizeof(buf)); ++ if (err) { ++ pr_err("gen clnt name failed.\n"); ++ return; ++ } ++ remove_proc_subtree(buf, enfs_proc_parent); ++} ++ ++// create proc file "/porc/enfs/[mount_ip]_[id]/stat" ++int enfs_proc_create_clnt(struct rpc_clnt *clnt) ++{ ++ int err; ++ ++ err = enfs_proc_create_file(clnt); ++ if (err) { ++ pr_err("create client %d\n", err); ++ return err; ++ } ++ ++ return 0; ++} ++ ++void enfs_proc_delete_clnt(struct rpc_clnt *clnt) ++{ ++ if (clnt->cl_enfs) ++ enfs_proc_delete_file(clnt); ++} ++ ++static int enfs_proc_create_parent(void) ++{ ++ enfs_proc_parent = proc_mkdir(ENFS_PROC_DIR, NULL); ++ ++ if (enfs_proc_parent == NULL) { ++ pr_err("Enfs create proc dir err\n"); ++ return -ENOMEM; ++ } ++ return 0; ++} ++ ++static void enfs_proc_delete_parent(void) ++{ ++ remove_proc_entry(ENFS_PROC_DIR, NULL); ++} ++ ++static int enfs_proc_init_create_clnt(struct rpc_clnt *clnt, void *data) ++{ ++ if (clnt->cl_enfs) ++ enfs_proc_create_file(clnt); ++ return 0; ++} ++ ++static int enfs_proc_destroy_clnt(struct rpc_clnt *clnt, void *data) ++{ ++ if (clnt->cl_enfs) ++ enfs_proc_delete_file(clnt); ++ return 0; ++} ++ ++int enfs_proc_init(void) ++{ ++ int err; ++ ++ err = enfs_proc_create_parent(); ++ if (err) ++ return err; ++ ++ enfs_iterate_each_rpc_clnt(enfs_proc_init_create_clnt, NULL); ++ return 0; ++} ++ ++void enfs_proc_exit(void) ++{ ++ enfs_iterate_each_rpc_clnt(enfs_proc_destroy_clnt, NULL); ++ enfs_proc_delete_parent(); ++} +diff --git a/fs/nfs/enfs/enfs_proc.h b/fs/nfs/enfs/enfs_proc.h +new file mode 100644 +index 000000000..321951031 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_proc.h +@@ -0,0 +1,21 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Client-side ENFS PROC. ++ * ++ * Copyright (c) 2023. Huawei Technologies Co., Ltd. All rights reserved. ++ */ ++#ifndef ENFS_PROC_H ++#define ENFS_PROC_H ++ ++struct rpc_clnt; ++struct rpc_task; ++struct proc_dir_entry; ++ ++int enfs_proc_init(void); ++void enfs_proc_exit(void); ++struct proc_dir_entry *enfs_get_proc_parent(void); ++int enfs_proc_create_clnt(struct rpc_clnt *clnt); ++void enfs_proc_delete_clnt(struct rpc_clnt *clnt); ++void enfs_count_iostat(struct rpc_task *task); ++ ++#endif +diff --git a/fs/nfs/enfs/enfs_remount.c b/fs/nfs/enfs/enfs_remount.c +new file mode 100644 +index 000000000..3b7d6a03d +--- /dev/null ++++ b/fs/nfs/enfs/enfs_remount.c +@@ -0,0 +1,220 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: remount ip source file ++ * Create: 2023-08-12 ++ */ ++#include "enfs_remount.h" ++ ++#include <linux/string.h> ++#include <linux/in.h> ++#include <linux/in6.h> ++#include <linux/sunrpc/clnt.h> ++#include <linux/spinlock.h> ++#include <linux/sunrpc/addr.h> ++#include <linux/sunrpc/metrics.h> ++#include <linux/sunrpc/xprtmultipath.h> ++#include <linux/sunrpc/xprtsock.h> ++#include <linux/sunrpc/xprt.h> ++#include <linux/smp.h> ++#include <linux/delay.h> ++ ++#include "enfs.h" ++#include "enfs_log.h" ++#include "enfs_multipath.h" ++#include "enfs_multipath_parse.h" ++#include "enfs_path.h" ++#include "enfs_proc.h" ++#include "enfs_multipath_client.h" ++ ++static bool enfs_rpc_xprt_switch_need_delete_addr( ++ struct multipath_mount_options *enfs_option, ++ struct sockaddr *dstaddr, struct sockaddr *srcaddr) ++{ ++ int i; ++ bool find_same_ip = false; ++ int32_t local_total; ++ int32_t remote_total; ++ ++ local_total = enfs_option->local_ip_list->count; ++ remote_total = enfs_option->remote_ip_list->count; ++ if (local_total == 0 || remote_total == 0) { ++ pr_err("no ip list is present.\n"); ++ return false; ++ } ++ ++ for (i = 0; i < local_total; i++) { ++ find_same_ip = ++ rpc_cmp_addr((struct sockaddr *) ++ &enfs_option->local_ip_list->address[i], ++ srcaddr); ++ if (find_same_ip) ++ break; ++ } ++ ++ if (find_same_ip == false) ++ return true; ++ ++ find_same_ip = false; ++ for (i = 0; i < remote_total; i++) { ++ find_same_ip = ++ rpc_cmp_addr((struct sockaddr *) ++ &enfs_option->remote_ip_list->address[i], ++ dstaddr); ++ if (find_same_ip) ++ break; ++ } ++ ++ if (find_same_ip == false) ++ return true; ++ ++ return false; ++} ++ ++// Used in rcu_lock ++static bool enfs_delete_xprt_from_switch(struct rpc_xprt *xprt, ++ void *enfs_option, ++ struct rpc_xprt_switch *xps) ++{ ++ struct enfs_xprt_context *ctx = NULL; ++ struct multipath_mount_options *mopt = ++ (struct multipath_mount_options *)enfs_option; ++ ++ if (enfs_is_main_xprt(xprt)) ++ return true; ++ ++ ctx = (struct enfs_xprt_context *)xprt->multipath_context; ++ if (enfs_rpc_xprt_switch_need_delete_addr(mopt, ++ (struct sockaddr *)&xprt->addr, ++ (struct sockaddr *)&ctx->srcaddr)) { ++ ++ print_enfs_multipath_addr((struct sockaddr *)&ctx->srcaddr, ++ (struct sockaddr *)&xprt->addr); ++ rpc_xprt_switch_remove_xprt(xps, xprt); ++ return true; ++ } ++ ++ return false; ++} ++ ++void enfs_clnt_delete_obsolete_xprts(struct nfs_client *nfs_client, ++ void *enfs_option) ++{ ++ int xprt_count = 0; ++ struct rpc_xprt *pos = NULL; ++ struct rpc_xprt_switch *xps = NULL; ++ ++ rcu_read_lock(); ++ xps = xprt_switch_get( ++ rcu_dereference( ++ nfs_client->cl_rpcclient->cl_xpi.xpi_xpswitch)); ++ if (xps == NULL) { ++ rcu_read_unlock(); ++ xprt_switch_put(xps); ++ return; ++ } ++ list_for_each_entry_rcu(pos, &xps->xps_xprt_list, xprt_switch) { ++ if (xprt_count < MAX_XPRT_NUM_PER_CLIENT) { ++ if (enfs_delete_xprt_from_switch( ++ pos, enfs_option, xps) == false) ++ xprt_count++; ++ } else ++ rpc_xprt_switch_remove_xprt(xps, pos); ++ } ++ rcu_read_unlock(); ++ xprt_switch_put(xps); ++} ++ ++int enfs_remount_iplist(struct nfs_client *nfs_client, void *enfs_option) ++{ ++ int errno = 0; ++ char servername[48]; ++ struct multipath_mount_options *remount_lists = ++ (struct multipath_mount_options *)enfs_option; ++ struct multipath_client_info *client_info = ++ (struct multipath_client_info *)nfs_client->cl_multipath_data; ++ struct xprt_create xprtargs; ++ struct rpc_create_args args = { ++ .protocol = nfs_client->cl_proto, ++ .net = nfs_client->cl_net, ++ .addrsize = nfs_client->cl_addrlen, ++ .servername = nfs_client->cl_hostname, ++ }; ++ ++ memset(&xprtargs, 0, sizeof(struct xprt_create)); ++ ++ //mount is not use multipath ++ if (client_info == NULL || enfs_option == NULL) { ++ enfs_log_error( ++ "mount information or remount information is empty.\n"); ++ return -EINVAL; ++ } ++ ++ //remount : localaddrs and remoteaddrs are empty ++ if (remount_lists->local_ip_list->count == 0 && ++ remount_lists->remote_ip_list->count == 0) { ++ enfs_log_info("remount local_ip_list and remote_ip_list are NULL\n"); ++ return 0; ++ } ++ ++ errno = enfs_config_xprt_create_args(&xprtargs, ++ &args, servername, sizeof(servername)); ++ ++ if (errno) { ++ enfs_log_error("config_xprt_create failed! errno:%d\n", errno); ++ return errno; ++ } ++ ++ if (remount_lists->local_ip_list->count == 0) { ++ if (client_info->local_ip_list->count == 0) { ++ errno = rpc_localaddr(nfs_client->cl_rpcclient, ++ (struct sockaddr *) ++ &remount_lists->local_ip_list->address[0], ++ sizeof(struct sockaddr_storage)); ++ if (errno) { ++ enfs_log_error("get clnt srcaddr errno:%d\n", ++ errno); ++ return errno; ++ } ++ remount_lists->local_ip_list->count = 1; ++ } else ++ memcpy(remount_lists->local_ip_list, ++ client_info->local_ip_list, ++ sizeof(struct nfs_ip_list)); ++ } ++ ++ if (remount_lists->remote_ip_list->count == 0) { ++ if (client_info->remote_ip_list->count == 0) { ++ errno = rpc_peeraddr(nfs_client->cl_rpcclient, ++ (struct sockaddr *) ++ &remount_lists->remote_ip_list->address[0], ++ sizeof(struct sockaddr_storage)); ++ if (errno == 0) { ++ enfs_log_error("get clnt dstaddr errno:%d\n", ++ errno); ++ return errno; ++ } ++ remount_lists->remote_ip_list->count = 1; ++ } else ++ memcpy(remount_lists->remote_ip_list, ++ client_info->remote_ip_list, ++ sizeof(struct nfs_ip_list)); ++ } ++ ++ enfs_log_info("Remount creating new links...\n"); ++ enfs_xprt_ippair_create(&xprtargs, ++ nfs_client->cl_rpcclient, ++ remount_lists); ++ ++ enfs_log_info("Remount deleting obsolete links...\n"); ++ enfs_clnt_delete_obsolete_xprts(nfs_client, remount_lists); ++ ++ memcpy(client_info->local_ip_list, ++ remount_lists->local_ip_list, ++ sizeof(struct nfs_ip_list)); ++ memcpy(client_info->remote_ip_list, ++ remount_lists->remote_ip_list, ++ sizeof(struct nfs_ip_list)); ++ ++ return 0; ++} +diff --git a/fs/nfs/enfs/enfs_remount.h b/fs/nfs/enfs/enfs_remount.h +new file mode 100644 +index 000000000..c7a1dcbb6 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_remount.h +@@ -0,0 +1,14 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: remount ip header file ++ * Create: 2023-08-12 ++ */ ++#ifndef _ENFS_REMOUNT_ ++#define _ENFS_REMOUNT_ ++#include <linux/string.h> ++#include "enfs.h" ++ ++int enfs_remount_iplist(struct nfs_client *nfs_client, void *enfs_option); ++ ++#endif +diff --git a/fs/nfs/enfs/enfs_roundrobin.c b/fs/nfs/enfs/enfs_roundrobin.c +new file mode 100644 +index 000000000..4e4eda784 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_roundrobin.c +@@ -0,0 +1,255 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ */ ++#include <linux/spinlock.h> ++#include <linux/module.h> ++#include <linux/printk.h> ++#include <linux/kref.h> ++#include <linux/rculist.h> ++#include <linux/types.h> ++#include <linux/sunrpc/xprt.h> ++#include <linux/sunrpc/clnt.h> ++#include <linux/sunrpc/xprtmultipath.h> ++#include "enfs_roundrobin.h" ++ ++#include "enfs.h" ++#include "enfs_config.h" ++#include "pm_state.h" ++ ++typedef struct rpc_xprt *(*enfs_xprt_switch_find_xprt_t)( ++ struct rpc_xprt_switch *xps, const struct rpc_xprt *cur); ++static const struct rpc_xprt_iter_ops enfs_xprt_iter_roundrobin; ++static const struct rpc_xprt_iter_ops enfs_xprt_iter_singular; ++ ++static bool enfs_xprt_is_active(struct rpc_xprt *xprt) ++{ ++ enum pm_path_state state; ++ ++ if (kref_read(&xprt->kref) <= 0) ++ return false; ++ ++ state = pm_get_path_state(xprt); ++ if (state == PM_STATE_NORMAL) ++ return true; ++ ++ return false; ++} ++ ++static struct rpc_xprt *enfs_lb_set_cursor_xprt( ++ struct rpc_xprt_switch *xps, struct rpc_xprt **cursor, ++ enfs_xprt_switch_find_xprt_t find_next) ++{ ++ struct rpc_xprt *pos; ++ struct rpc_xprt *old; ++ ++ old = smp_load_acquire(cursor); /* read latest cursor */ ++ pos = find_next(xps, old); ++ smp_store_release(cursor, pos); /* let cursor point to pos */ ++ return pos; ++} ++ ++static ++struct rpc_xprt *enfs_lb_find_next_entry_roundrobin( ++ struct rpc_xprt_switch *xps, const struct rpc_xprt *cur) ++{ ++ struct rpc_xprt *pos; ++ struct rpc_xprt *prev = NULL; ++ bool found = false; ++ struct rpc_xprt *min_queuelen_xprt = NULL; ++ unsigned long pos_xprt_queuelen; ++ unsigned long min_xprt_queuelen = 0; ++ ++ unsigned long xps_queuelen = atomic_long_read(&xps->xps_queuelen); ++ // delete origin xprt ++ unsigned int multipath_nactive = READ_ONCE(xps->xps_nactive) - 1; ++ ++ list_for_each_entry_rcu(pos, &xps->xps_xprt_list, xprt_switch) { ++ if (enfs_is_main_xprt(pos) || !enfs_xprt_is_active(pos)) { ++ prev = pos; ++ continue; ++ } ++ ++ pos_xprt_queuelen = atomic_long_read(&pos->queuelen); ++ if (min_queuelen_xprt == NULL || ++ pos_xprt_queuelen < min_xprt_queuelen) { ++ ++ min_queuelen_xprt = pos; ++ min_xprt_queuelen = pos_xprt_queuelen; ++ } ++ ++ if (cur == prev) ++ found = true; ++ ++ if (found && pos_xprt_queuelen * ++ multipath_nactive <= xps_queuelen) ++ return pos; ++ prev = pos; ++ }; ++ ++ return min_queuelen_xprt; ++} ++ ++struct rpc_xprt *enfs_lb_switch_find_first_active_xprt( ++ struct rpc_xprt_switch *xps) ++{ ++ struct rpc_xprt *pos; ++ ++ list_for_each_entry_rcu(pos, &xps->xps_xprt_list, xprt_switch) { ++ if (enfs_xprt_is_active(pos)) ++ return pos; ++ }; ++ return NULL; ++} ++ ++struct rpc_xprt *enfs_lb_switch_get_main_xprt(struct rpc_xprt_switch *xps) ++{ ++ return list_first_or_null_rcu(&xps->xps_xprt_list, ++ struct rpc_xprt, xprt_switch); ++} ++ ++static struct rpc_xprt *enfs_lb_switch_get_next_xprt_roundrobin( ++ struct rpc_xprt_switch *xps, const struct rpc_xprt *cur) ++{ ++ struct rpc_xprt *xprt; ++ ++ // disable multipath ++ if (enfs_get_config_multipath_state()) ++ return enfs_lb_switch_get_main_xprt(xps); ++ ++ xprt = enfs_lb_find_next_entry_roundrobin(xps, cur); ++ if (xprt != NULL) ++ return xprt; ++ ++ return enfs_lb_switch_get_main_xprt(xps); ++} ++ ++static ++struct rpc_xprt *enfs_lb_iter_next_entry_roundrobin(struct rpc_xprt_iter *xpi) ++{ ++ struct rpc_xprt_switch *xps = rcu_dereference(xpi->xpi_xpswitch); ++ ++ if (xps == NULL) ++ return NULL; ++ ++ return enfs_lb_set_cursor_xprt(xps, &xpi->xpi_cursor, ++ enfs_lb_switch_get_next_xprt_roundrobin); ++} ++ ++static ++struct rpc_xprt *enfs_lb_switch_find_singular_entry( ++ struct rpc_xprt_switch *xps, const struct rpc_xprt *cur) ++{ ++ struct rpc_xprt *pos; ++ bool found = false; ++ ++ list_for_each_entry_rcu(pos, &xps->xps_xprt_list, xprt_switch) { ++ if (cur == pos) ++ found = true; ++ ++ if (found && enfs_xprt_is_active(pos)) ++ return pos; ++ } ++ return NULL; ++} ++ ++struct rpc_xprt *enfs_lb_get_singular_xprt( ++ struct rpc_xprt_switch *xps, const struct rpc_xprt *cur) ++{ ++ struct rpc_xprt *xprt; ++ ++ if (xps == NULL) ++ return NULL; ++ ++ // disable multipath ++ if (enfs_get_config_multipath_state()) ++ return enfs_lb_switch_get_main_xprt(xps); ++ ++ if (cur == NULL || xps->xps_nxprts < 2) ++ return enfs_lb_switch_find_first_active_xprt(xps); ++ ++ xprt = enfs_lb_switch_find_singular_entry(xps, cur); ++ if (!xprt) ++ return enfs_lb_switch_get_main_xprt(xps); ++ ++ return xprt; ++} ++ ++static ++struct rpc_xprt *enfs_lb_iter_next_entry_sigular(struct rpc_xprt_iter *xpi) ++{ ++ struct rpc_xprt_switch *xps = rcu_dereference(xpi->xpi_xpswitch); ++ ++ if (xps == NULL) ++ return NULL; ++ ++ return enfs_lb_set_cursor_xprt(xps, &xpi->xpi_cursor, ++ enfs_lb_get_singular_xprt); ++} ++ ++static void enfs_lb_iter_default_rewind(struct rpc_xprt_iter *xpi) ++{ ++ WRITE_ONCE(xpi->xpi_cursor, NULL); ++} ++ ++static void enfs_lb_switch_set_roundrobin(struct rpc_clnt *clnt) ++{ ++ struct rpc_xprt_switch *xps; ++ ++ rcu_read_lock(); ++ xps = rcu_dereference(clnt->cl_xpi.xpi_xpswitch); ++ rcu_read_unlock(); ++ if (clnt->cl_vers == 3) { ++ ++ if (READ_ONCE(xps->xps_iter_ops) != &enfs_xprt_iter_roundrobin) ++ WRITE_ONCE(xps->xps_iter_ops, ++ &enfs_xprt_iter_roundrobin); ++ ++ return; ++ } ++ if (READ_ONCE(xps->xps_iter_ops) != &enfs_xprt_iter_singular) ++ WRITE_ONCE(xps->xps_iter_ops, &enfs_xprt_iter_singular); ++} ++ ++static ++struct rpc_xprt *enfs_lb_switch_find_current(struct list_head *head, ++ const struct rpc_xprt *cur) ++{ ++ struct rpc_xprt *pos; ++ ++ list_for_each_entry_rcu(pos, head, xprt_switch) { ++ if (cur == pos) ++ return pos; ++ } ++ return NULL; ++} ++ ++static struct rpc_xprt *enfs_lb_iter_current_entry(struct rpc_xprt_iter *xpi) ++{ ++ struct rpc_xprt_switch *xps = rcu_dereference(xpi->xpi_xpswitch); ++ struct list_head *head; ++ ++ if (xps == NULL) ++ return NULL; ++ head = &xps->xps_xprt_list; ++ if (xpi->xpi_cursor == NULL || xps->xps_nxprts < 2) ++ return enfs_lb_switch_get_main_xprt(xps); ++ return enfs_lb_switch_find_current(head, xpi->xpi_cursor); ++} ++ ++void enfs_lb_set_policy(struct rpc_clnt *clnt) ++{ ++ enfs_lb_switch_set_roundrobin(clnt); ++} ++ ++static const struct rpc_xprt_iter_ops enfs_xprt_iter_roundrobin = { ++ .xpi_rewind = enfs_lb_iter_default_rewind, ++ .xpi_xprt = enfs_lb_iter_current_entry, ++ .xpi_next = enfs_lb_iter_next_entry_roundrobin, ++}; ++ ++static const struct rpc_xprt_iter_ops enfs_xprt_iter_singular = { ++ .xpi_rewind = enfs_lb_iter_default_rewind, ++ .xpi_xprt = enfs_lb_iter_current_entry, ++ .xpi_next = enfs_lb_iter_next_entry_sigular, ++}; +diff --git a/fs/nfs/enfs/enfs_roundrobin.h b/fs/nfs/enfs/enfs_roundrobin.h +new file mode 100644 +index 000000000..b72b088a6 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_roundrobin.h +@@ -0,0 +1,9 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ */ ++#ifndef ENFS_ROUNDROBIN_H ++#define ENFS_ROUNDROBIN_H ++ ++void enfs_lb_set_policy(struct rpc_clnt *clnt); ++#endif diff --git a/0005-add_enfs_module_for_sunrpc_failover_and_configure.patch b/0005-add_enfs_module_for_sunrpc_failover_and_configure.patch new file mode 100644 index 0000000..9f3beea --- /dev/null +++ b/0005-add_enfs_module_for_sunrpc_failover_and_configure.patch @@ -0,0 +1,1600 @@ +diff --git a/fs/nfs/enfs/enfs_config.c b/fs/nfs/enfs/enfs_config.c +new file mode 100644 +index 000000000..a0ca93114 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_config.c +@@ -0,0 +1,378 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ */ ++#include <linux/cdev.h> ++#include <linux/errno.h> ++#include <linux/fcntl.h> ++#include <linux/fs.h> ++#include <linux/kernel.h> ++#include <linux/kthread.h> ++#include <linux/slab.h> ++#include <linux/string.h> ++#include <linux/uaccess.h> ++#include <linux/delay.h> ++ ++#include "enfs_errcode.h" ++#include "enfs_log.h" ++#include "enfs_config.h" ++ ++#define MAX_FILE_SIZE 8192 ++#define STRING_BUF_SIZE 128 ++#define CONFIG_FILE_PATH "/etc/enfs/config.ini" ++#define ENFS_NOTIFY_FILE_PERIOD 1000UL ++ ++#define MAX_PATH_DETECT_INTERVAL 300 ++#define MIN_PATH_DETECT_INTERVAL 5 ++#define MAX_PATH_DETECT_TIMEOUT 60 ++#define MIN_PATH_DETECT_TIMEOUT 1 ++#define MAX_MULTIPATH_TIMEOUT 60 ++#define MIN_MULTIPATH_TIMEOUT 0 ++#define MAX_MULTIPATH_STATE ENFS_MULTIPATH_DISABLE ++#define MIN_MULTIPATH_STATE ENFS_MULTIPATH_ENABLE ++ ++#define DEFAULT_PATH_DETECT_INTERVAL 10 ++#define DEFAULT_PATH_DETECT_TIMEOUT 5 ++#define DEFAULT_MULTIPATH_TIMEOUT 0 ++#define DEFAULT_MULTIPATH_STATE ENFS_MULTIPATH_ENABLE ++#define DEFAULT_LOADBALANCE_MODE ENFS_LOADBALANCE_RR ++ ++typedef int (*check_and_assign_func)(char *, char *, int, int); ++ ++struct enfs_config_info { ++ int path_detect_interval; ++ int path_detect_timeout; ++ int multipath_timeout; ++ int loadbalance_mode; ++ int multipath_state; ++}; ++ ++struct check_and_assign_value { ++ char *field_name; ++ check_and_assign_func func; ++ int min_value; ++ int max_value; ++}; ++ ++static struct enfs_config_info g_enfs_config_info; ++static struct timespec64 modify_time; ++static struct task_struct *thread; ++ ++static int enfs_check_config_value(char *value, int min_value, int max_value) ++{ ++ unsigned long num_value; ++ int ret; ++ ++ ret = kstrtol(value, 10, &num_value); ++ if (ret != 0) { ++ enfs_log_error("Failed to convert string to int\n"); ++ return -EINVAL; ++ } ++ ++ if (num_value < min_value || num_value > max_value) ++ return -EINVAL; ++ ++ return num_value; ++} ++ ++static int enfs_check_and_assign_int_value(char *field_name, char *value, ++ int min_value, int max_value) ++{ ++ int int_value = enfs_check_config_value(value, min_value, max_value); ++ ++ if (int_value < 0) ++ return -EINVAL; ++ ++ if (strcmp(field_name, "path_detect_interval") == 0) { ++ g_enfs_config_info.path_detect_interval = int_value; ++ return ENFS_RET_OK; ++ } ++ if (strcmp(field_name, "path_detect_timeout") == 0) { ++ g_enfs_config_info.path_detect_timeout = int_value; ++ return ENFS_RET_OK; ++ } ++ if (strcmp(field_name, "multipath_timeout") == 0) { ++ g_enfs_config_info.multipath_timeout = int_value; ++ return ENFS_RET_OK; ++ } ++ if (strcmp(field_name, "multipath_disable") == 0) { ++ g_enfs_config_info.multipath_state = int_value; ++ return ENFS_RET_OK; ++ } ++ return -EINVAL; ++} ++ ++static int enfs_check_and_assign_loadbalance_mode(char *field_name, ++ char *value, ++ int min_value, ++ int max_value) ++{ ++ if (value == NULL) ++ return -EINVAL; ++ ++ if (strcmp(field_name, "multipath_select_policy") == 0) { ++ if (strcmp(value, "roundrobin") == 0) { ++ g_enfs_config_info.loadbalance_mode ++ = ENFS_LOADBALANCE_RR; ++ return ENFS_RET_OK; ++ } ++ } ++ return -EINVAL; ++} ++ ++static const struct check_and_assign_value g_check_and_assign_value[] = { ++ {"path_detect_interval", enfs_check_and_assign_int_value, ++ MIN_PATH_DETECT_INTERVAL, MAX_PATH_DETECT_INTERVAL}, ++ {"path_detect_timeout", enfs_check_and_assign_int_value, ++ MIN_PATH_DETECT_TIMEOUT, MAX_PATH_DETECT_TIMEOUT}, ++ {"multipath_timeout", enfs_check_and_assign_int_value, ++ MIN_MULTIPATH_TIMEOUT, MAX_MULTIPATH_TIMEOUT}, ++ {"multipath_disable", enfs_check_and_assign_int_value, ++ MIN_MULTIPATH_STATE, MAX_MULTIPATH_STATE}, ++ {"multipath_select_policy", enfs_check_and_assign_loadbalance_mode, ++ 0, 0}, ++}; ++ ++static int enfs_read_config_file(char *buffer, char *file_path) ++{ ++ int ret; ++ struct file *filp = NULL; ++ loff_t f_pos = 0; ++ mm_segment_t fs; ++ ++ ++ filp = filp_open(file_path, O_RDONLY, 0); ++ ++ if (IS_ERR(filp)) { ++ enfs_log_error("Failed to open file %s\n", CONFIG_FILE_PATH); ++ ret = -ENOENT; ++ return ret; ++ } ++ ++ fs = get_fs(); ++ set_fs(get_ds()); ++ kernel_read(filp, buffer, MAX_FILE_SIZE, &f_pos); ++ set_fs(fs); ++ ++ ret = filp_close(filp, NULL); ++ if (ret) { ++ enfs_log_error("Close File:%s failed:%d.\n", ++ CONFIG_FILE_PATH, ret); ++ return -EINVAL; ++ } ++ return ENFS_RET_OK; ++} ++ ++static int enfs_deal_with_comment_line(char *buffer) ++{ ++ int ret; ++ char *pos = strchr(buffer, '\n'); ++ ++ if (pos != NULL) ++ ret = strlen(buffer) - strlen(pos); ++ else ++ ret = strlen(buffer); ++ ++ return ret; ++} ++ ++static int enfs_parse_key_value_from_config(char *buffer, char *key, ++ char *value, int keyLen, ++ int valueLen) ++{ ++ char *line; ++ char *tokenPtr; ++ int len; ++ char *tem; ++ char *pos = strchr(buffer, '\n'); ++ ++ if (pos != NULL) ++ len = strlen(buffer) - strlen(pos); ++ else ++ len = strlen(buffer); ++ ++ line = kmalloc(len + 1, GFP_KERNEL); ++ if (!line) { ++ enfs_log_error("Failed to allocate memory.\n"); ++ return -ENOMEM; ++ } ++ line[len] = '\0'; ++ strncpy(line, buffer, len); ++ ++ tem = line; ++ tokenPtr = strsep(&tem, "="); ++ if (tokenPtr == NULL || tem == NULL) { ++ kfree(line); ++ return len; ++ } ++ strncpy(key, strim(tokenPtr), keyLen); ++ strncpy(value, strim(tem), valueLen); ++ ++ kfree(line); ++ return len; ++} ++ ++static int enfs_get_value_from_config_file(char *buffer, char *field_name, ++ char *value, int valueLen) ++{ ++ int ret; ++ char key[STRING_BUF_SIZE + 1] = {0}; ++ char val[STRING_BUF_SIZE + 1] = {0}; ++ ++ while (buffer[0] != '\0') { ++ if (buffer[0] == '\n') { ++ buffer++; ++ } else if (buffer[0] == '#') { ++ ret = enfs_deal_with_comment_line(buffer); ++ if (ret > 0) ++ buffer += ret; ++ } else { ++ ret = enfs_parse_key_value_from_config(buffer, key, val, ++ STRING_BUF_SIZE, ++ STRING_BUF_SIZE); ++ if (ret < 0) { ++ enfs_log_error("failed parse key value, %d\n" ++ , ret); ++ return ret; ++ } ++ key[STRING_BUF_SIZE] = '\0'; ++ val[STRING_BUF_SIZE] = '\0'; ++ ++ buffer += ret; ++ ++ if (strcmp(field_name, key) == 0) { ++ strncpy(value, val, valueLen); ++ return ENFS_RET_OK; ++ } ++ } ++ } ++ enfs_log_error("can not find value which matched field_name: %s.\n", ++ field_name); ++ return -EINVAL; ++} ++ ++int enfs_config_load(void) ++{ ++ char value[STRING_BUF_SIZE + 1]; ++ int ret; ++ int table_len; ++ int min; ++ int max; ++ int i; ++ char *buffer; ++ ++ buffer = kmalloc(MAX_FILE_SIZE, GFP_KERNEL); ++ if (!buffer) { ++ enfs_log_error("Failed to allocate memory.\n"); ++ return -ENOMEM; ++ } ++ memset(buffer, 0, MAX_FILE_SIZE); ++ ++ g_enfs_config_info.path_detect_interval = DEFAULT_PATH_DETECT_INTERVAL; ++ g_enfs_config_info.path_detect_timeout = DEFAULT_PATH_DETECT_TIMEOUT; ++ g_enfs_config_info.multipath_timeout = DEFAULT_MULTIPATH_TIMEOUT; ++ g_enfs_config_info.multipath_state = DEFAULT_MULTIPATH_STATE; ++ g_enfs_config_info.loadbalance_mode = DEFAULT_LOADBALANCE_MODE; ++ ++ table_len = sizeof(g_check_and_assign_value) / ++ sizeof(g_check_and_assign_value[0]); ++ ++ ret = enfs_read_config_file(buffer, CONFIG_FILE_PATH); ++ if (ret != 0) { ++ kfree(buffer); ++ return ret; ++ } ++ ++ for (i = 0; i < table_len; i++) { ++ ret = enfs_get_value_from_config_file(buffer, ++ g_check_and_assign_value[i].field_name, ++ value, STRING_BUF_SIZE); ++ if (ret < 0) ++ continue; ++ ++ value[STRING_BUF_SIZE] = '\0'; ++ min = g_check_and_assign_value[i].min_value; ++ max = g_check_and_assign_value[i].max_value; ++ if (g_check_and_assign_value[i].func != NULL) ++ (*g_check_and_assign_value[i].func)( ++ g_check_and_assign_value[i].field_name, ++ value, min, max); ++ } ++ ++ kfree(buffer); ++ return ENFS_RET_OK; ++} ++ ++int enfs_get_config_path_detect_interval(void) ++{ ++ return g_enfs_config_info.path_detect_interval; ++} ++ ++int enfs_get_config_path_detect_timeout(void) ++{ ++ return g_enfs_config_info.path_detect_timeout; ++} ++ ++int enfs_get_config_multipath_timeout(void) ++{ ++ return g_enfs_config_info.multipath_timeout; ++} ++ ++int enfs_get_config_multipath_state(void) ++{ ++ return g_enfs_config_info.multipath_state; ++} ++ ++int enfs_get_config_loadbalance_mode(void) ++{ ++ return g_enfs_config_info.loadbalance_mode; ++} ++ ++static bool enfs_file_changed(const char *filename) ++{ ++ int err; ++ struct kstat file_stat; ++ ++ err = vfs_stat(filename, &file_stat); ++ if (err) { ++ pr_err("failed to open file:%s err:%d\n", filename, err); ++ return false; ++ } ++ ++ if (timespec64_compare(&modify_time, &file_stat.mtime) == -1) { ++ modify_time = file_stat.mtime; ++ pr_info("file change: %lld %lld\n", modify_time.tv_sec, ++ file_stat.mtime.tv_sec); ++ return true; ++ } ++ ++ return false; ++} ++ ++static int enfs_thread_func(void *data) ++{ ++ while (!kthread_should_stop()) { ++ if (enfs_file_changed(CONFIG_FILE_PATH)) ++ enfs_config_load(); ++ ++ msleep(ENFS_NOTIFY_FILE_PERIOD); ++ } ++ return 0; ++} ++ ++int enfs_config_timer_init(void) ++{ ++ thread = kthread_run(enfs_thread_func, NULL, "enfs_notiy_file_thread"); ++ if (IS_ERR(thread)) { ++ pr_err("Failed to create kernel thread\n"); ++ return PTR_ERR(thread); ++ } ++ return 0; ++} ++ ++void enfs_config_timer_exit(void) ++{ ++ pr_info("enfs_notify_file_exit\n"); ++ if (thread) ++ kthread_stop(thread); ++} +diff --git a/fs/nfs/enfs/enfs_config.h b/fs/nfs/enfs/enfs_config.h +new file mode 100644 +index 000000000..4eff0ccc3 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_config.h +@@ -0,0 +1,31 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: nfs configuration ++ * Create: 2023-07-27 ++ */ ++ ++#ifndef ENFS_CONFIG_H ++#define ENFS_CONFIG_H ++ ++#include <linux/types.h> ++ ++enum enfs_multipath_state { ++ ENFS_MULTIPATH_ENABLE = 0, ++ ENFS_MULTIPATH_DISABLE = 1, ++}; ++ ++enum enfs_loadbalance_mode { ++ ENFS_LOADBALANCE_RR, ++}; ++ ++ ++int enfs_get_config_path_detect_interval(void); ++int enfs_get_config_path_detect_timeout(void); ++int enfs_get_config_multipath_timeout(void); ++int enfs_get_config_multipath_state(void); ++int enfs_get_config_loadbalance_mode(void); ++int enfs_config_load(void); ++int enfs_config_timer_init(void); ++void enfs_config_timer_exit(void); ++#endif // ENFS_CONFIG_H +diff --git a/fs/nfs/enfs/enfs_errcode.h b/fs/nfs/enfs/enfs_errcode.h +new file mode 100644 +index 000000000..ffa089088 +--- /dev/null ++++ b/fs/nfs/enfs/enfs_errcode.h +@@ -0,0 +1,16 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: nfs errocode ++ * Create: 2023-07-31 ++ */ ++ ++#ifndef ENFS_ERRCODE_H ++#define ENFS_ERRCODE_H ++ ++enum { ++ ENFS_RET_OK = 0, ++ ENFS_RET_FAIL ++}; ++ ++#endif // ENFS_ERRCODE_H +diff --git a/fs/nfs/enfs/enfs_log.h b/fs/nfs/enfs/enfs_log.h +new file mode 100644 +index 000000000..e12f8f3ae +--- /dev/null ++++ b/fs/nfs/enfs/enfs_log.h +@@ -0,0 +1,24 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: enfs log ++ * Create: 2023-07-31 ++ */ ++#ifndef ENFS_LOG_H ++#define ENFS_LOG_H ++ ++#include <linux/printk.h> ++ ++#define enfs_log_info(fmt, ...) \ ++ pr_info("enfs:[%s]" pr_fmt(fmt), \ ++ __func__, ##__VA_ARGS__) ++ ++#define enfs_log_error(fmt, ...) \ ++ pr_err("enfs:[%s]" pr_fmt(fmt), \ ++ __func__, ##__VA_ARGS__) ++ ++#define enfs_log_debug(fmt, ...) \ ++ pr_debug("enfs:[%s]" pr_fmt(fmt), \ ++ __func__, ##__VA_ARGS__) ++ ++#endif // ENFS_ERRCODE_H +diff --git a/fs/nfs/enfs/failover_com.h b/fs/nfs/enfs/failover_com.h +new file mode 100644 +index 000000000..c52940da2 +--- /dev/null ++++ b/fs/nfs/enfs/failover_com.h +@@ -0,0 +1,23 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: failover time commont header file ++ * Create: 2023-08-02 ++ */ ++#ifndef FAILOVER_COMMON_H ++#define FAILOVER_COMMON_H ++ ++static inline bool failover_is_enfs_clnt(struct rpc_clnt *clnt) ++{ ++ struct rpc_clnt *next = clnt->cl_parent; ++ ++ while (next) { ++ if (next == next->cl_parent) ++ break; ++ next = next->cl_parent; ++ } ++ ++ return next != NULL ? next->cl_enfs : clnt->cl_enfs; ++} ++ ++#endif // FAILOVER_COMMON_H +diff --git a/fs/nfs/enfs/failover_path.c b/fs/nfs/enfs/failover_path.c +new file mode 100644 +index 000000000..2f5387216 +--- /dev/null ++++ b/fs/nfs/enfs/failover_path.c +@@ -0,0 +1,212 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: nfs path failover file ++ * Create: 2023-08-02 ++ */ ++ ++#include "failover_path.h" ++#include <linux/nfs.h> ++#include <linux/nfs3.h> ++#include <linux/nfs4.h> ++#include <linux/sunrpc/clnt.h> ++#include <linux/sunrpc/sched.h> ++#include <linux/sunrpc/xprt.h> ++#include "enfs_config.h" ++#include "enfs_log.h" ++#include "failover_com.h" ++#include "pm_state.h" ++#include "pm_ping.h" ++ ++enum failover_policy_t { ++ FAILOVER_NOACTION = 1, ++ FAILOVER_RETRY, ++ FAILOVER_RETRY_DELAY, ++}; ++ ++static void failover_retry_path(struct rpc_task *task) ++{ ++ int ret; ++ ++ ret = rpc_restart_call(task); ++ ++ if (ret == 1) { ++ xprt_release(task); ++ rpc_init_task_retry_counters(task); ++ rpc_task_release_transport(task); ++ task->tk_xprt = rpc_task_get_next_xprt(task->tk_client); ++ } ++} ++ ++static void failover_retry_path_delay(struct rpc_task *task, int32_t delay) ++{ ++ failover_retry_path(task); ++ rpc_delay(task, delay); ++} ++ ++static void failover_retry_path_by_policy(struct rpc_task *task, ++ enum failover_policy_t policy) ++{ ++ if (policy == FAILOVER_RETRY) ++ failover_retry_path(task); ++ else if (policy == FAILOVER_RETRY_DELAY) ++ failover_retry_path_delay(task, 3 * HZ); // delay 3s ++} ++ ++static ++enum failover_policy_t failover_get_nfs3_retry_policy(struct rpc_task *task) ++{ ++ enum failover_policy_t policy = FAILOVER_NOACTION; ++ const struct rpc_procinfo *procinfo = task->tk_msg.rpc_proc; ++ u32 proc; ++ ++ if (unlikely(procinfo == NULL)) { ++ enfs_log_error("the task contains no valid proc.\n"); ++ return FAILOVER_NOACTION; ++ } ++ ++ proc = procinfo->p_proc; ++ ++ switch (proc) { ++ case NFS3PROC_CREATE: ++ case NFS3PROC_MKDIR: ++ case NFS3PROC_REMOVE: ++ case NFS3PROC_RMDIR: ++ case NFS3PROC_SYMLINK: ++ case NFS3PROC_LINK: ++ case NFS3PROC_SETATTR: ++ case NFS3PROC_WRITE: ++ policy = FAILOVER_RETRY_DELAY; ++ default: ++ policy = FAILOVER_RETRY; ++ } ++ return policy; ++} ++ ++static ++enum failover_policy_t failover_get_nfs4_retry_policy(struct rpc_task *task) ++{ ++ enum failover_policy_t policy = FAILOVER_NOACTION; ++ const struct rpc_procinfo *procinfo = task->tk_msg.rpc_proc; ++ u32 proc_idx; ++ ++ if (unlikely(procinfo == NULL)) { ++ enfs_log_error("the task contains no valid proc.\n"); ++ return FAILOVER_NOACTION; ++ } ++ ++ proc_idx = procinfo->p_statidx; ++ ++ switch (proc_idx) { ++ case NFSPROC4_CLNT_CREATE: ++ case NFSPROC4_CLNT_REMOVE: ++ case NFSPROC4_CLNT_LINK: ++ case NFSPROC4_CLNT_SYMLINK: ++ case NFSPROC4_CLNT_SETATTR: ++ case NFSPROC4_CLNT_WRITE: ++ case NFSPROC4_CLNT_RENAME: ++ case NFSPROC4_CLNT_SETACL: ++ policy = FAILOVER_RETRY_DELAY; ++ default: ++ policy = FAILOVER_RETRY; ++ } ++ return policy; ++} ++ ++static enum failover_policy_t failover_get_retry_policy(struct rpc_task *task) ++{ ++ struct rpc_clnt *clnt = task->tk_client; ++ u32 version = clnt->cl_vers; ++ enum failover_policy_t policy = FAILOVER_NOACTION; ++ ++ // 1. if the task meant to send to certain xprt, take no action ++ if (task->tk_flags & RPC_TASK_FIXED) ++ return FAILOVER_NOACTION; ++ ++ // 2. get policy by different version of nfs protocal ++ if (version == 3) // nfs v3 ++ policy = failover_get_nfs3_retry_policy(task); ++ else if (version == 4) // nfs v4 ++ policy = failover_get_nfs4_retry_policy(task); ++ else ++ return FAILOVER_NOACTION; ++ ++ // 3. if the task is not send to target, retry immediately ++ if (!RPC_WAS_SENT(task)) ++ policy = FAILOVER_RETRY; ++ ++ return policy; ++} ++ ++static int failover_check_task(struct rpc_task *task) ++{ ++ struct rpc_clnt *clnt = NULL; ++ int disable_mpath = enfs_get_config_multipath_state(); ++ ++ if (disable_mpath != ENFS_MULTIPATH_ENABLE) { ++ enfs_log_debug("Multipath is not enabled.\n"); ++ return -EINVAL; ++ } ++ ++ if (unlikely((task == NULL) || (task->tk_client == NULL))) { ++ enfs_log_error("The task is not valid.\n"); ++ return -EINVAL; ++ } ++ ++ clnt = task->tk_client; ++ ++ if (clnt->cl_prog != NFS_PROGRAM) { ++ enfs_log_debug("The clnt is not prog{%u} type.\n", ++ clnt->cl_prog); ++ return -EINVAL; ++ } ++ ++ if (!failover_is_enfs_clnt(clnt)) { ++ enfs_log_debug("The clnt is not a enfs-managed type.\n"); ++ return -EINVAL; ++ } ++ return 0; ++} ++ ++void failover_handle(struct rpc_task *task) ++{ ++ enum failover_policy_t policy; ++ int ret; ++ ++ ret = failover_check_task(task); ++ if (ret != 0) ++ return; ++ ++ pm_set_path_state(task->tk_xprt, PM_STATE_FAULT); ++ ++ policy = failover_get_retry_policy(task); ++ ++ failover_retry_path_by_policy(task, policy); ++} ++ ++bool failover_task_need_call_start_again(struct rpc_task *task) ++{ ++ int ret; ++ ++ ret = failover_check_task(task); ++ if (ret != 0) ++ return false; ++ ++ return true; ++} ++ ++bool failover_prepare_transmit(struct rpc_task *task) ++{ ++ if (task->tk_flags & RPC_TASK_FIXED) ++ return true; ++ ++ if (pm_ping_is_test_xprt_task(task)) ++ return true; ++ ++ if (pm_get_path_state(task->tk_xprt) == PM_STATE_FAULT) { ++ task->tk_status = -ETIMEDOUT; ++ return false; ++ } ++ ++ return true; ++} +diff --git a/fs/nfs/enfs/failover_path.h b/fs/nfs/enfs/failover_path.h +new file mode 100644 +index 000000000..9159ada07 +--- /dev/null ++++ b/fs/nfs/enfs/failover_path.h +@@ -0,0 +1,16 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: nfs path failover header file ++ * Create: 2023-08-02 ++ */ ++ ++#ifndef FAILOVER_PATH_H ++#define FAILOVER_PATH_H ++ ++#include <linux/sunrpc/sched.h> ++ ++void failover_handle(struct rpc_task *task); ++bool failover_prepare_transmit(struct rpc_task *task); ++bool failover_task_need_call_start_again(struct rpc_task *task); ++#endif // FAILOVER_PATH_H +diff --git a/fs/nfs/enfs/failover_time.c b/fs/nfs/enfs/failover_time.c +new file mode 100644 +index 000000000..866ea82d1 +--- /dev/null ++++ b/fs/nfs/enfs/failover_time.c +@@ -0,0 +1,99 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: failover time file ++ * Create: 2023-08-02 ++ */ ++ ++#include "failover_time.h" ++#include <linux/jiffies.h> ++#include <linux/sunrpc/clnt.h> ++#include "enfs_config.h" ++#include "enfs_log.h" ++#include "failover_com.h" ++#include "pm_ping.h" ++ ++static unsigned long failover_get_mulitipath_timeout(struct rpc_clnt *clnt) ++{ ++ unsigned long config_tmo = enfs_get_config_multipath_timeout() * HZ; ++ unsigned long clnt_tmo = clnt->cl_timeout->to_initval; ++ ++ if (config_tmo == 0) ++ return clnt_tmo; ++ ++ return config_tmo > clnt_tmo ? clnt_tmo : config_tmo; ++} ++ ++void failover_adjust_task_timeout(struct rpc_task *task, void *condition) ++{ ++ struct rpc_clnt *clnt = NULL; ++ unsigned long tmo; ++ int disable_mpath = enfs_get_config_multipath_state(); ++ ++ if (disable_mpath != ENFS_MULTIPATH_ENABLE) { ++ enfs_log_debug("Multipath is not enabled.\n"); ++ return; ++ } ++ ++ clnt = task->tk_client; ++ if (unlikely(clnt == NULL)) { ++ enfs_log_error("task associate client is NULL.\n"); ++ return; ++ } ++ ++ if (!failover_is_enfs_clnt(clnt)) { ++ enfs_log_debug("The clnt is not a enfs-managed type.\n"); ++ return; ++ } ++ ++ tmo = failover_get_mulitipath_timeout(clnt); ++ if (tmo == 0) { ++ enfs_log_debug("Multipath is not enabled.\n"); ++ return; ++ } ++ ++ if (task->tk_timeout != 0) ++ task->tk_timeout = ++ task->tk_timeout < tmo ? task->tk_timeout : tmo; ++ else ++ task->tk_timeout = tmo; ++} ++ ++void failover_init_task_req(struct rpc_task *task, struct rpc_rqst *req) ++{ ++ struct rpc_clnt *clnt = NULL; ++ int disable_mpath = enfs_get_config_multipath_state(); ++ ++ if (disable_mpath != ENFS_MULTIPATH_ENABLE) { ++ enfs_log_debug("Multipath is not enabled.\n"); ++ return; ++ } ++ ++ clnt = task->tk_client; ++ if (unlikely(clnt == NULL)) { ++ enfs_log_error("task associate client is NULL.\n"); ++ return; ++ } ++ ++ if (!failover_is_enfs_clnt(clnt)) { ++ enfs_log_debug("The clnt is not a enfs-managed type.\n"); ++ return; ++ } ++ ++ if (!pm_ping_is_test_xprt_task(task)) ++ req->rq_timeout = failover_get_mulitipath_timeout(clnt); ++ else { ++ req->rq_timeout = enfs_get_config_path_detect_timeout() * HZ; ++ req->rq_majortimeo = req->rq_timeout + jiffies; ++ } ++ ++ /* ++ * when task is retried, the req is new, we lost major-timeout times, ++ * so we have to restore req major ++ * timeouts from the task, if it is stored. ++ */ ++ if (task->tk_major_timeo != 0) ++ req->rq_majortimeo = task->tk_major_timeo; ++ else ++ task->tk_major_timeo = req->rq_majortimeo; ++} +diff --git a/fs/nfs/enfs/failover_time.h b/fs/nfs/enfs/failover_time.h +new file mode 100644 +index 000000000..ede25b577 +--- /dev/null ++++ b/fs/nfs/enfs/failover_time.h +@@ -0,0 +1,16 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: failover time header file ++ * Create: 2023-08-02 ++ */ ++ ++#ifndef FAILOVER_TIME_H ++#define FAILOVER_TIME_H ++ ++#include <linux/sunrpc/sched.h> ++ ++void failover_adjust_task_timeout(struct rpc_task *task, void *condition); ++void failover_init_task_req(struct rpc_task *task, struct rpc_rqst *req); ++ ++#endif // FAILOVER_TIME_H +diff --git a/fs/nfs/enfs/init.h b/fs/nfs/enfs/init.h +new file mode 100644 +index 000000000..d81af9b02 +--- /dev/null ++++ b/fs/nfs/enfs/init.h +@@ -0,0 +1,16 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: nfs client init ++ * Create: 2023-07-31 ++ */ ++ ++#ifndef ENFS_INIT_H ++#define ENFS_INIT_H ++ ++#include <linux/types.h> ++ ++int32_t enfs_init(void); ++void enfs_fini(void); ++ ++#endif +diff --git a/fs/nfs/enfs/mgmt_init.c b/fs/nfs/enfs/mgmt_init.c +new file mode 100644 +index 000000000..122790775 +--- /dev/null ++++ b/fs/nfs/enfs/mgmt_init.c +@@ -0,0 +1,21 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: mgmt component init ++ * Create: 2023-07-31 ++ */ ++ ++#include "mgmt_init.h" ++#include <linux/printk.h> ++#include "enfs_errcode.h" ++#include "enfs_config.h" ++ ++int32_t mgmt_init(void) ++{ ++ return enfs_config_timer_init(); ++} ++ ++void mgmt_fini(void) ++{ ++ enfs_config_timer_exit(); ++} +diff --git a/fs/nfs/enfs/mgmt_init.h b/fs/nfs/enfs/mgmt_init.h +new file mode 100644 +index 000000000..779732740 +--- /dev/null ++++ b/fs/nfs/enfs/mgmt_init.h +@@ -0,0 +1,17 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: mgmt component init ++ * Create: 2023-07-31 ++ */ ++ ++#ifndef MGMT_INIT_H ++#define MGMT_INIT_H ++ ++#include <linux/types.h> ++ ++int32_t mgmt_init(void); ++void mgmt_fini(void); ++ ++ ++#endif // MGMT_INIT_H +diff --git a/fs/nfs/enfs/pm_ping.c b/fs/nfs/enfs/pm_ping.c +new file mode 100644 +index 000000000..9ecbe9a77 +--- /dev/null ++++ b/fs/nfs/enfs/pm_ping.c +@@ -0,0 +1,420 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: path state header file ++ * Create: 2023-08-21 ++ */ ++ ++#include "pm_ping.h" ++#include <linux/err.h> ++#include <linux/spinlock.h> ++#include <linux/slab.h> ++#include <linux/module.h> ++#include <linux/printk.h> ++#include <linux/kthread.h> ++#include <linux/nfs.h> ++#include <linux/errno.h> ++#include <linux/rcupdate.h> ++#include <linux/workqueue.h> ++#include <net/netns/generic.h> ++#include <linux/atomic.h> ++#include <linux/sunrpc/clnt.h> ++ ++#include "../../../net/sunrpc/netns.h" ++#include "pm_state.h" ++#include "enfs.h" ++#include "enfs_log.h" ++#include "enfs_config.h" ++ ++#define SLEEP_INTERVAL 2 ++extern unsigned int sunrpc_net_id; ++ ++static struct task_struct *pm_ping_timer_thread; ++//protect pint_execute_workq ++static spinlock_t ping_execute_workq_lock; ++// timer for test xprt workqueue ++static struct workqueue_struct *ping_execute_workq; ++// count the ping xprt work on flight ++static atomic_t check_xprt_count; ++ ++struct ping_xprt_work { ++ struct rpc_xprt *xprt; // use this specific xprt ++ struct rpc_clnt *clnt; // use this specific rpc_client ++ struct work_struct ping_work; ++}; ++ ++struct pm_ping_async_callback { ++ void *data; ++ void (*func)(void *data); ++}; ++ ++// set xprt's enum pm_check_state ++void pm_ping_set_path_check_state(struct rpc_xprt *xprt, ++ enum pm_check_state state) ++{ ++ struct enfs_xprt_context *ctx = NULL; ++ ++ if (IS_ERR(xprt)) { ++ enfs_log_error("The xprt ptr is not exist.\n"); ++ return; ++ } ++ ++ if (xprt == NULL) { ++ enfs_log_error("The xprt is not valid.\n"); ++ return; ++ } ++ ++ xprt_get(xprt); ++ ++ ctx = (struct enfs_xprt_context *)xprt->multipath_context; ++ if (ctx == NULL) { ++ enfs_log_error("The xprt multipath ctx is not valid.\n"); ++ xprt_put(xprt); ++ return; ++ } ++ ++ atomic_set(&ctx->path_check_state, state); ++ xprt_put(xprt); ++} ++ ++// get xprt's enum pm_check_state ++static enum pm_check_state pm_ping_get_path_check_state(struct rpc_xprt *xprt) ++{ ++ struct enfs_xprt_context *ctx = NULL; ++ enum pm_check_state state; ++ ++ if (xprt == NULL) { ++ enfs_log_error("The xprt is not valid.\n"); ++ return PM_CHECK_UNDEFINE; ++ } ++ ++ ctx = (struct enfs_xprt_context *)xprt->multipath_context; ++ if (ctx == NULL) { ++ enfs_log_error("The xprt multipath ctx is not valid.\n"); ++ return PM_CHECK_UNDEFINE; ++ } ++ ++ state = atomic_read(&ctx->path_check_state); ++ ++ return state; ++} ++ ++static void pm_ping_call_done_callback(void *data) ++{ ++ struct pm_ping_async_callback *callback_data = ++ (struct pm_ping_async_callback *)data; ++ ++ if (callback_data == NULL) ++ return; ++ ++ callback_data->func(callback_data->data); ++ ++ kfree(callback_data); ++} ++ ++// Default callback for async RPC calls ++static void pm_ping_call_done(struct rpc_task *task, void *data) ++{ ++ struct rpc_xprt *xprt = task->tk_xprt; ++ ++ atomic_dec(&check_xprt_count); ++ if (task->tk_status >= 0) ++ pm_set_path_state(xprt, PM_STATE_NORMAL); ++ else ++ pm_set_path_state(xprt, PM_STATE_FAULT); ++ ++ pm_ping_set_path_check_state(xprt, PM_CHECK_FINISH); ++ ++ pm_ping_call_done_callback(data); ++} ++ ++// register func to rpc_call_done ++static const struct rpc_call_ops pm_ping_set_status_ops = { ++ .rpc_call_done = pm_ping_call_done, ++}; ++ ++// execute work which in work_queue ++static void pm_ping_execute_work(struct work_struct *work) ++{ ++ int ret = 0; ++ ++ // get the work information ++ struct ping_xprt_work *work_info = ++ container_of(work, struct ping_xprt_work, ping_work); ++ ++ // if check state is pending ++ if (pm_ping_get_path_check_state(work_info->xprt) == PM_CHECK_WAITING) { ++ ++ pm_ping_set_path_check_state(work_info->xprt, ++ PM_CHECK_CHECKING); ++ ++ ret = rpc_clnt_test_xprt(work_info->clnt, ++ work_info->xprt, ++ &pm_ping_set_status_ops, ++ NULL, ++ RPC_TASK_ASYNC | RPC_TASK_FIXED); ++ ++ if (ret < 0) { ++ enfs_log_debug("ping xprt execute failed ,ret %d", ret); ++ ++ pm_ping_set_path_check_state(work_info->xprt, ++ PM_CHECK_FINISH); ++ ++ } else ++ atomic_inc(&check_xprt_count); ++ ++ } ++ ++ atomic_dec(&work_info->clnt->cl_count); ++ xprt_put(work_info->xprt); ++ kfree(work_info); ++ work_info = NULL; ++} ++ ++static bool pm_ping_workqueue_queue_work(struct work_struct *work) ++{ ++ bool ret = false; ++ ++ spin_lock(&ping_execute_workq_lock); ++ ++ if (ping_execute_workq != NULL) ++ ret = queue_work(ping_execute_workq, work); ++ ++ spin_unlock(&ping_execute_workq_lock); ++ return ret; ++} ++ ++// init test work and add this work to workqueue ++static int pm_ping_add_work(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt, void *data) ++{ ++ struct ping_xprt_work *work_info; ++ bool ret = false; ++ ++ if (IS_ERR(xprt) || xprt == NULL) { ++ enfs_log_error("The xprt ptr is not exist.\n"); ++ return -EINVAL; ++ } ++ ++ if (IS_ERR(clnt) || clnt == NULL) { ++ enfs_log_error("The clnt ptr is not exist.\n"); ++ return -EINVAL; ++ } ++ ++ if (!xprt->multipath_context) { ++ enfs_log_error("multipath_context is null.\n"); ++ return -EINVAL; ++ } ++ ++ // check xprt pending status, if pending status equals Finish ++ // means this xprt can inster to work queue ++ if (pm_ping_get_path_check_state(xprt) == ++ PM_CHECK_FINISH || ++ pm_ping_get_path_check_state(xprt) == ++ PM_CHECK_INIT) { ++ ++ enfs_log_debug("find xprt pointer. %p\n", xprt); ++ work_info = kzalloc(sizeof(struct ping_xprt_work), GFP_ATOMIC); ++ if (work_info == NULL) ++ return -ENOMEM; ++ work_info->clnt = clnt; ++ atomic_inc(&clnt->cl_count); ++ work_info->xprt = xprt; ++ xprt_get(xprt); ++ INIT_WORK(&work_info->ping_work, pm_ping_execute_work); ++ pm_ping_set_path_check_state(xprt, PM_CHECK_WAITING); ++ ++ ret = pm_ping_workqueue_queue_work(&work_info->ping_work); ++ if (!ret) { ++ atomic_dec(&work_info->clnt->cl_count); ++ xprt_put(work_info->xprt); ++ kfree(work_info); ++ return -EINVAL; ++ } ++ } ++ return 0; ++} ++ ++// encapsulate pm_ping_add_work() ++static int pm_ping_execute_xprt_test(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt, void *data) ++{ ++ pm_ping_add_work(clnt, xprt, NULL); ++ // return 0 for rpc_clnt_iterate_for_each_xprt(); ++ // because negative value will stop iterate all xprt ++ // and we need return negative value for debug ++ // Therefore, we need this function to iterate all xprt ++ return 0; ++} ++ ++// export to other module add ping work to workqueue ++int pm_ping_rpc_test_xprt(struct rpc_clnt *clnt, struct rpc_xprt *xprt) ++{ ++ int ret; ++ ++ ret = pm_ping_add_work(clnt, xprt, NULL); ++ return ret; ++} ++ ++// iterate xprt in the client ++static void pm_ping_loop_rpclnt(struct sunrpc_net *sn) ++{ ++ struct rpc_clnt *clnt; ++ ++ spin_lock(&sn->rpc_client_lock); ++ list_for_each_entry_rcu(clnt, &sn->all_clients, cl_clients) { ++ if (clnt->cl_enfs) { ++ enfs_log_debug("find rpc_clnt. %p\n", clnt); ++ rpc_clnt_iterate_for_each_xprt(clnt, ++ pm_ping_execute_xprt_test, NULL); ++ } ++ } ++ spin_unlock(&sn->rpc_client_lock); ++} ++ ++// iterate each clnt in the sunrpc_net ++static void pm_ping_loop_sunrpc_net(void) ++{ ++ struct net *net; ++ struct sunrpc_net *sn; ++ ++ rcu_read_lock(); ++ for_each_net_rcu(net) { ++ sn = net_generic(net, sunrpc_net_id); ++ if (sn == NULL) ++ continue; ++ pm_ping_loop_rpclnt(sn); ++ } ++ rcu_read_unlock(); ++} ++ ++static int pm_ping_routine(void *data) ++{ ++ while (!kthread_should_stop()) { ++ // equale 0 means open multipath ++ if (enfs_get_config_multipath_state() == ++ ENFS_MULTIPATH_ENABLE) ++ pm_ping_loop_sunrpc_net(); ++ ++ msleep( ++ enfs_get_config_path_detect_interval() * 1000); ++ } ++ return 0; ++} ++ ++// start thread to cycly ping ++static int pm_ping_start(void) ++{ ++ pm_ping_timer_thread = ++ kthread_run(pm_ping_routine, NULL, "pm_ping_routine"); ++ if (IS_ERR(pm_ping_timer_thread)) { ++ enfs_log_error("Failed to create kernel thread\n"); ++ return PTR_ERR(pm_ping_timer_thread); ++ } ++ return 0; ++} ++ ++// initialize workqueue ++static int pm_ping_workqueue_init(void) ++{ ++ struct workqueue_struct *queue = NULL; ++ ++ queue = create_workqueue("pm_ping_workqueue"); ++ ++ if (queue == NULL) { ++ enfs_log_error("create workqueue failed.\n"); ++ return -ENOMEM; ++ } ++ ++ spin_lock(&ping_execute_workq_lock); ++ ping_execute_workq = queue; ++ spin_unlock(&ping_execute_workq_lock); ++ enfs_log_info("create workqueue succeeeded.\n"); ++ return 0; ++} ++ ++static void pm_ping_workqueue_fini(void) ++{ ++ struct workqueue_struct *queue = NULL; ++ ++ spin_lock(&ping_execute_workq_lock); ++ queue = ping_execute_workq; ++ ping_execute_workq = NULL; ++ spin_unlock(&ping_execute_workq_lock); ++ ++ enfs_log_info("delete work queue\n"); ++ ++ if (queue != NULL) { ++ flush_workqueue(queue); ++ destroy_workqueue(queue); ++ } ++} ++ ++// module exit func ++void pm_ping_fini(void) ++{ ++ if (pm_ping_timer_thread) ++ kthread_stop(pm_ping_timer_thread); ++ ++ pm_ping_workqueue_fini(); ++ ++ while (atomic_read(&check_xprt_count) != 0) ++ msleep(SLEEP_INTERVAL); ++} ++ ++// module init func ++int pm_ping_init(void) ++{ ++ int ret; ++ ++ atomic_set(&check_xprt_count, 0); ++ ret = pm_ping_workqueue_init(); ++ if (ret != 0) { ++ enfs_log_error("PM_PING Module loading failed.\n"); ++ return ret; ++ } ++ ret = pm_ping_start(); ++ if (ret != 0) { ++ enfs_log_error("PM_PING Module loading failed.\n"); ++ pm_ping_workqueue_fini(); ++ return ret; ++ } ++ ++ return ret; ++} ++ ++bool pm_ping_is_test_xprt_task(struct rpc_task *task) ++{ ++ return task->tk_ops == &pm_ping_set_status_ops ? true : false; ++} ++ ++int pm_ping_rpc_test_xprt_with_callback(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt, ++ void (*func)(void *data), ++ void *data) ++{ ++ int ret; ++ ++ struct pm_ping_async_callback *callback_data = ++ kzalloc(sizeof(struct pm_ping_async_callback), GFP_KERNEL); ++ ++ if (callback_data == NULL) { ++ enfs_log_error("failed to mzalloc mem\n"); ++ return -ENOMEM; ++ } ++ ++ callback_data->data = data; ++ callback_data->func = func; ++ atomic_inc(&check_xprt_count); ++ ret = rpc_clnt_test_xprt(clnt, xprt, ++ &pm_ping_set_status_ops, ++ callback_data, ++ RPC_TASK_ASYNC | RPC_TASK_FIXED); ++ ++ if (ret < 0) { ++ enfs_log_debug("ping xprt execute failed ,ret %d", ret); ++ atomic_dec(&check_xprt_count); ++ } ++ ++ return ret; ++} +diff --git a/fs/nfs/enfs/pm_ping.h b/fs/nfs/enfs/pm_ping.h +new file mode 100644 +index 000000000..8b159b286 +--- /dev/null ++++ b/fs/nfs/enfs/pm_ping.h +@@ -0,0 +1,32 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: nfs configuration ++ * Create: 2023-07-27 ++ */ ++ ++#ifndef PM_PING_H ++#define PM_PING_H ++ ++#include <linux/sunrpc/clnt.h> ++ ++enum pm_check_state { ++ PM_CHECK_INIT, // this xprt never been queued ++ PM_CHECK_WAITING, // this xprt waiting in the queue ++ PM_CHECK_CHECKING, // this xprt is testing ++ PM_CHECK_FINISH, // this xprt has been finished ++ PM_CHECK_UNDEFINE, // undefine multipath struct ++}; ++ ++int pm_ping_init(void); ++void pm_ping_fini(void); ++int pm_ping_rpc_test_xprt(struct rpc_clnt *clnt, struct rpc_xprt *xprt); ++void pm_ping_set_path_check_state(struct rpc_xprt *xprt, ++ enum pm_check_state state); ++bool pm_ping_is_test_xprt_task(struct rpc_task *task); ++int pm_ping_rpc_test_xprt_with_callback(struct rpc_clnt *clnt, ++ struct rpc_xprt *xprt, ++ void (*func)(void *data), ++ void *data); ++ ++#endif // PM_PING_H +diff --git a/fs/nfs/enfs/pm_state.c b/fs/nfs/enfs/pm_state.c +new file mode 100644 +index 000000000..279028dc0 +--- /dev/null ++++ b/fs/nfs/enfs/pm_state.c +@@ -0,0 +1,156 @@ ++// SPDX-License-Identifier: GPL-2.0 ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: path state file ++ * Create: 2023-08-12 ++ */ ++#include "pm_state.h" ++#include <linux/sunrpc/xprt.h> ++ ++#include "enfs.h" ++#include "enfs_log.h" ++ ++enum pm_path_state pm_get_path_state(struct rpc_xprt *xprt) ++{ ++ struct enfs_xprt_context *ctx = NULL; ++ enum pm_path_state state; ++ ++ if (xprt == NULL) { ++ enfs_log_error("The xprt is not valid.\n"); ++ return PM_STATE_UNDEFINED; ++ } ++ ++ xprt_get(xprt); ++ ++ ctx = (struct enfs_xprt_context *)xprt->multipath_context; ++ if (ctx == NULL) { ++ enfs_log_error("The xprt multipath ctx is not valid.\n"); ++ xprt_put(xprt); ++ return PM_STATE_UNDEFINED; ++ } ++ ++ state = atomic_read(&ctx->path_state); ++ ++ xprt_put(xprt); ++ ++ return state; ++} ++ ++void pm_set_path_state(struct rpc_xprt *xprt, enum pm_path_state state) ++{ ++ struct enfs_xprt_context *ctx = NULL; ++ enum pm_path_state cur_state; ++ ++ if (xprt == NULL) { ++ enfs_log_error("The xprt is not valid.\n"); ++ return; ++ } ++ ++ xprt_get(xprt); ++ ++ ctx = (struct enfs_xprt_context *)xprt->multipath_context; ++ if (ctx == NULL) { ++ enfs_log_error("The xprt multipath ctx is not valid.\n"); ++ xprt_put(xprt); ++ return; ++ } ++ ++ cur_state = atomic_read(&ctx->path_state); ++ if (cur_state == state) { ++ xprt_put(xprt); ++ return; ++ } ++ ++ atomic_set(&ctx->path_state, state); ++ enfs_log_info("The xprt {%p} path state change from {%d} to {%d}.\n", ++ xprt, cur_state, state); ++ ++ xprt_put(xprt); ++} ++ ++void pm_get_path_state_desc(struct rpc_xprt *xprt, char *buf, int len) ++{ ++ enum pm_path_state state; ++ ++ if (xprt == NULL) { ++ enfs_log_error("The xprt is not valid.\n"); ++ return; ++ } ++ ++ if ((buf == NULL) || (len <= 0)) { ++ enfs_log_error("Buffer is not valid, len=%d.\n", len); ++ return; ++ } ++ ++ state = pm_get_path_state(xprt); ++ ++ switch (state) { ++ case PM_STATE_INIT: ++ (void)snprintf(buf, len, "Init"); ++ break; ++ case PM_STATE_NORMAL: ++ (void)snprintf(buf, len, "Normal"); ++ break; ++ case PM_STATE_FAULT: ++ (void)snprintf(buf, len, "Fault"); ++ break; ++ default: ++ (void)snprintf(buf, len, "Unknown"); ++ break; ++ } ++} ++ ++void pm_get_xprt_state_desc(struct rpc_xprt *xprt, char *buf, int len) ++{ ++ int i; ++ unsigned long state; ++ static unsigned long xprt_mask[] = { ++ XPRT_LOCKED, XPRT_CONNECTED, ++ XPRT_CONNECTING, XPRT_CLOSE_WAIT, ++ XPRT_BOUND, XPRT_BINDING, XPRT_CLOSING, ++ XPRT_CONGESTED}; ++ ++ static const char *const xprt_state_desc[] = { ++ "LOCKED", "CONNECTED", "CONNECTING", ++ "CLOSE_WAIT", "BOUND", "BINDING", ++ "CLOSING", "CONGESTED"}; ++ int pos = 0; ++ int ret = 0; ++ ++ if (xprt == NULL) { ++ enfs_log_error("The xprt is not valid.\n"); ++ return; ++ } ++ ++ if ((buf == NULL) || (len <= 0)) { ++ enfs_log_error( ++ "Xprt state buffer is not valid, len=%d.\n", ++ len); ++ return; ++ } ++ ++ xprt_get(xprt); ++ state = READ_ONCE(xprt->state); ++ xprt_put(xprt); ++ ++ for (i = 0; i < ARRAY_SIZE(xprt_mask); ++i) { ++ if (pos >= len) ++ break; ++ ++ if (!test_bit(xprt_mask[i], &state)) ++ continue; ++ ++ if (pos == 0) ++ ret = snprintf(buf, len, "%s", xprt_state_desc[i]); ++ else ++ ret = snprintf(buf + pos, len - pos, "|%s", ++ xprt_state_desc[i]); ++ ++ if (ret < 0) { ++ enfs_log_error("format state failed, ret %d.\n", ret); ++ break; ++ } ++ ++ pos += ret; ++ } ++} +diff --git a/fs/nfs/enfs/pm_state.h b/fs/nfs/enfs/pm_state.h +new file mode 100644 +index 000000000..469af998d +--- /dev/null ++++ b/fs/nfs/enfs/pm_state.h +@@ -0,0 +1,27 @@ ++/* SPDX-License-Identifier: GPL-2.0 */ ++/* ++ * Copyright (c) Huawei Technologies Co., Ltd. 2023-2023. All rights reserved. ++ * Description: path state header file ++ * Create: 2023-08-12 ++ */ ++ ++#ifndef PM_STATE_H ++#define PM_STATE_H ++ ++#include <linux/types.h> ++#include <linux/sunrpc/xprt.h> ++ ++enum pm_path_state { ++ PM_STATE_INIT, ++ PM_STATE_NORMAL, ++ PM_STATE_FAULT, ++ PM_STATE_UNDEFINED // xprt is not multipath xprt ++}; ++ ++void pm_set_path_state(struct rpc_xprt *xprt, enum pm_path_state state); ++enum pm_path_state pm_get_path_state(struct rpc_xprt *xprt); ++ ++void pm_get_path_state_desc(struct rpc_xprt *xprt, char *buf, int len); ++void pm_get_xprt_state_desc(struct rpc_xprt *xprt, char *buf, int len); ++ ++#endif // PM_STATE_H diff --git a/0006-add_enfs_compile_option.patch b/0006-add_enfs_compile_option.patch new file mode 100644 index 0000000..71140c1 --- /dev/null +++ b/0006-add_enfs_compile_option.patch @@ -0,0 +1,70 @@ +diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig +index b04256636d4b..ae53510c0627 100644 +--- a/arch/arm64/configs/openeuler_defconfig ++++ b/arch/arm64/configs/openeuler_defconfig +@@ -5344,6 +5344,7 @@ CONFIG_LOCKD=m + CONFIG_LOCKD_V4=y + CONFIG_NFS_ACL_SUPPORT=m + CONFIG_NFS_COMMON=y ++# CONFIG_ENFS is not set + CONFIG_SUNRPC=m + CONFIG_SUNRPC_GSS=m + CONFIG_SUNRPC_BACKCHANNEL=y +diff --git a/arch/x86/configs/openeuler_defconfig b/arch/x86/configs/openeuler_defconfig +index 59baeb2973af..ccc317f7fdb2 100644 +--- a/arch/x86/configs/openeuler_defconfig ++++ b/arch/x86/configs/openeuler_defconfig +@@ -6825,6 +6825,7 @@ CONFIG_LOCKD=m + CONFIG_LOCKD_V4=y + CONFIG_NFS_ACL_SUPPORT=m + CONFIG_NFS_COMMON=y ++# CONFIG_ENFS is not set + CONFIG_SUNRPC=m + CONFIG_SUNRPC_GSS=m + CONFIG_SUNRPC_BACKCHANNEL=y +diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig +index e55f86713948..872c9b7671b1 100644 +--- a/fs/nfs/Kconfig ++++ b/fs/nfs/Kconfig +@@ -196,3 +196,14 @@ config NFS_DEBUG + depends on NFS_FS && SUNRPC_DEBUG + select CRC32 + default y ++ ++config ENFS ++ tristate "NFS client support for ENFS" ++ depends on NFS_FS ++ default n ++ help ++ This option enables support multipath of the NFS protocol ++ in the kernel's NFS client. ++ This feature will improve performance and reliability. ++ ++ If sure, say Y. +diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile +index c587e3c4c6a6..19d0ac2ba3b8 100644 +--- a/fs/nfs/Makefile ++++ b/fs/nfs/Makefile +@@ -12,6 +12,7 @@ nfs-y := client.o dir.o file.o getroot.o inode.o super.o \ + nfs-$(CONFIG_ROOT_NFS) += nfsroot.o + nfs-$(CONFIG_SYSCTL) += sysctl.o + nfs-$(CONFIG_NFS_FSCACHE) += fscache.o fscache-index.o ++nfs-$(CONFIG_ENFS) += enfs_adapter.o + + obj-$(CONFIG_NFS_V2) += nfsv2.o + nfsv2-y := nfs2super.o proc.o nfs2xdr.o +@@ -34,3 +35,5 @@ nfsv4-$(CONFIG_NFS_V4_2) += nfs42proc.o + obj-$(CONFIG_PNFS_FILE_LAYOUT) += filelayout/ + obj-$(CONFIG_PNFS_BLOCK) += blocklayout/ + obj-$(CONFIG_PNFS_FLEXFILE_LAYOUT) += flexfilelayout/ ++ ++obj-$(CONFIG_ENFS) += enfs/ +diff --git a/net/sunrpc/Makefile b/net/sunrpc/Makefile +index 090658c3da12..fe4e3b28c5d1 100644 +--- a/net/sunrpc/Makefile ++++ b/net/sunrpc/Makefile +@@ -19,3 +19,4 @@ sunrpc-$(CONFIG_SUNRPC_DEBUG) += debugfs.o + sunrpc-$(CONFIG_SUNRPC_BACKCHANNEL) += backchannel_rqst.o + sunrpc-$(CONFIG_PROC_FS) += stats.o + sunrpc-$(CONFIG_SYSCTL) += sysctl.o ++sunrpc-$(CONFIG_ENFS) += sunrpc_enfs_adapter.o -- 2.23.0

2 1

[PATCH openEuler-1.0-LTS] sched: smart grid: check is active in affinity timer
by Yipeng Zou 01 Dec '23

01 Dec '23

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8KZ3H CVE: NA ---------------------------------------- In the test case like below, there is change going to hang system in stop affinity timer function. for i in `seq 1 100000` do echo $1 > /sys/fs/cgroup/cpu/cpu.dynamic_affinity_mode 2> &1 done [ 259.061097] CPU: 0 PID: 10223 Comm: sh Kdump: loaded Not tainted 5.10.0OLK-5.10-SP3_SG+ #102 [ 259.061098] Hardware name: Huawei TaiShan 200 (Model 5280)/BC82AMDD, BIOS 1.08 12/14/2019 [ 259.061098] pstate: 60400089 (nZCv daIf +PAN -UAO -TCO BTYPE=--) [ 259.061099] pc : hrtimer_active+0x2c/0x7c [ 259.061099] lr : hrtimer_cancel+0x3c/0x60 [ 259.061100] sp : ffff800167f33c10 [ 259.061100] x29: ffff800167f33c10 x28: ffff003006bde040 [ 259.061101] x27: 0000000000000000 x26: 0000000000000000 [ 259.061102] x25: 0000000000000000 x24: ffff0020b2cdea20 [ 259.061103] x23: ffff800167f33d48 x22: ffff800011cc6100 [ 259.061105] x21: ffff80001170a120 x20: ffff800011728d88 [ 259.061106] x19: ffff00208e53c418 x18: 0000000000000000 [ 259.061107] x17: 0000000000000000 x16: 0000000000000000 [ 259.061108] x15: 0000aaab02369e80 x14: 0000000000000000 [ 259.061109] x13: 0000000000000000 x12: 0000000000000000 [ 259.061110] x11: 0000000000000000 x10: 0000000000000008 [ 259.061111] x9 : ffff80001017c6bc x8 : 00000000ffffffc9 [ 259.061112] x7 : ffff0020cb082581 x6 : 0000000000000000 [ 259.061114] x5 : ffff0020cb082582 x4 : 0000000000000000 [ 259.061115] x3 : ffff202fffb9adc0 x2 : ffff202fffb9ae00 [ 259.061116] x1 : 0000000000003b82 x0 : ffff00208e53c418 [ 259.061117] Call trace: [ 259.061117] hrtimer_active+0x2c/0x7c [ 259.061118] stop_auto_affinity+0xf0/0x170 [ 259.061118] cpu_affinity_mode_write_u64+0x2c/0x60 [ 259.061119] cgroup_file_write+0x128/0x1b0 [ 259.061119] kernfs_fop_write_iter+0x130/0x1c0 [ 259.061120] new_sync_write+0xec/0x18c [ 259.061121] vfs_write+0x214/0x2ac [ 259.061121] ksys_write+0x70/0xfc [ 259.061122] __arm64_sys_write+0x24/0x30 [ 259.061122] invoke_syscall+0x50/0x11c [ 259.061123] el0_svc_common.constprop.0+0x158/0x164 [ 259.061123] do_el0_svc+0x2c/0xac [ 259.061123] el0_svc+0x20/0x30 [ 259.061124] el0_sync_handler+0xb0/0xb4 [ 259.061124] el0_sync+0x160/0x180 The root cause was that hrtimer_cancel are holding spin_lock, which the timer are running. Thread: hrtimer: acqurie lock acqurie lock cancel hrtimer sync ... // deadlock ... release lock release lock Thread will wait hrtimer exit forever, and the hrtimer are waiting spin_ lock forever. Fixes: 16980bd872f6 ("sched: Introduce smart grid scheduling strategy for cfs") Signed-off-by: Yipeng Zou <zouyipeng(a)huawei.com> --- kernel/sched/fair.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index ccd2a060c2df..ce6bb4304949 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5403,6 +5403,12 @@ static enum hrtimer_restart sched_auto_affi_period_timer(struct hrtimer *timer) } raw_spin_lock_irqsave(&auto_affi->lock, flags); + /* May be re-entrant by stop_auto_affinity, So check again. */ + if (auto_affi->period_active == 0) { + raw_spin_unlock_irqrestore(&auto_affi->lock, flags); + return HRTIMER_NORESTART; + } + if (util_avg_sum * 100 >= tg_capacity * sysctl_sched_util_low_pct) { affinity_domain_up(tg); } else if (util_avg_sum * 100 < tg_capacity * @@ -5496,8 +5502,6 @@ void stop_auto_affinity(struct auto_affinity *auto_affi) mutex_unlock(&smart_grid_used_mutex); return; } - - hrtimer_cancel(&auto_affi->period_timer); auto_affi->period_active = 0; auto_affi->mode = 0; ad->curr_level = ad->dcount > 0 ? ad->dcount - 1 : 0; -- 2.34.1

2 1

[PATCH OLK-5.10] mm/readahead: stop readahead loop if memcg charge fails
by Jinjiang Tu 01 Dec '23

01 Dec '23

From: Liu Shixin <liushixin2(a)huawei.com> When a task in memcg readaheads file pages, page_cache_ra_unbounded() will try to readahead nr_to_read pages. Even if the new allocated page fails to charge, page_cache_ra_unbounded() still tries to readahead next page. This leads to too much memory reclaim. Stop readahead if mem_cgroup_charge() fails, i.e. add_to_page_cache_lru() returns -ENOMEM. Signed-off-by: Liu Shixin <liushixin2(a)huawei.com> Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> --- mm/readahead.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/mm/readahead.c b/mm/readahead.c index ed23d5dec123..22dd9c8fe808 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -220,11 +220,18 @@ void page_cache_ra_unbounded(struct readahead_control *ractl, if (mapping->a_ops->readpages) { page->index = index + i; list_add(&page->lru, &page_pool); - } else if (add_to_page_cache_lru(page, mapping, index + i, - gfp_mask) < 0) { - put_page(page); - read_pages(ractl, &page_pool, true); - continue; + } else { + int ret; + + ret = add_to_page_cache_lru(page, mapping, index + i, + gfp_mask); + if (ret < 0) { + put_page(page); + read_pages(ractl, &page_pool, true); + if (ret == -ENOMEM) + break; + continue; + } } if (i == nr_to_read - lookahead_size) SetPageReadahead(page); -- 2.25.1

2 1

[openEuler-1.0-LTS] sched: smart grid: check is active in affinity timer
by Yipeng Zou 01 Dec '23

01 Dec '23

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8KZ3H CVE: NA ---------------------------------------- In the test case like below, there is change going to hang system in stop affinity timer function. for i in `seq 1 100000` do echo $1 > /sys/fs/cgroup/cpu/cpu.dynamic_affinity_mode 2> &1 done [ 259.061097] CPU: 0 PID: 10223 Comm: sh Kdump: loaded Not tainted 5.10.0OLK-5.10-SP3_SG+ #102 [ 259.061098] Hardware name: Huawei TaiShan 200 (Model 5280)/BC82AMDD, BIOS 1.08 12/14/2019 [ 259.061098] pstate: 60400089 (nZCv daIf +PAN -UAO -TCO BTYPE=--) [ 259.061099] pc : hrtimer_active+0x2c/0x7c [ 259.061099] lr : hrtimer_cancel+0x3c/0x60 [ 259.061100] sp : ffff800167f33c10 [ 259.061100] x29: ffff800167f33c10 x28: ffff003006bde040 [ 259.061101] x27: 0000000000000000 x26: 0000000000000000 [ 259.061102] x25: 0000000000000000 x24: ffff0020b2cdea20 [ 259.061103] x23: ffff800167f33d48 x22: ffff800011cc6100 [ 259.061105] x21: ffff80001170a120 x20: ffff800011728d88 [ 259.061106] x19: ffff00208e53c418 x18: 0000000000000000 [ 259.061107] x17: 0000000000000000 x16: 0000000000000000 [ 259.061108] x15: 0000aaab02369e80 x14: 0000000000000000 [ 259.061109] x13: 0000000000000000 x12: 0000000000000000 [ 259.061110] x11: 0000000000000000 x10: 0000000000000008 [ 259.061111] x9 : ffff80001017c6bc x8 : 00000000ffffffc9 [ 259.061112] x7 : ffff0020cb082581 x6 : 0000000000000000 [ 259.061114] x5 : ffff0020cb082582 x4 : 0000000000000000 [ 259.061115] x3 : ffff202fffb9adc0 x2 : ffff202fffb9ae00 [ 259.061116] x1 : 0000000000003b82 x0 : ffff00208e53c418 [ 259.061117] Call trace: [ 259.061117] hrtimer_active+0x2c/0x7c [ 259.061118] stop_auto_affinity+0xf0/0x170 [ 259.061118] cpu_affinity_mode_write_u64+0x2c/0x60 [ 259.061119] cgroup_file_write+0x128/0x1b0 [ 259.061119] kernfs_fop_write_iter+0x130/0x1c0 [ 259.061120] new_sync_write+0xec/0x18c [ 259.061121] vfs_write+0x214/0x2ac [ 259.061121] ksys_write+0x70/0xfc [ 259.061122] __arm64_sys_write+0x24/0x30 [ 259.061122] invoke_syscall+0x50/0x11c [ 259.061123] el0_svc_common.constprop.0+0x158/0x164 [ 259.061123] do_el0_svc+0x2c/0xac [ 259.061123] el0_svc+0x20/0x30 [ 259.061124] el0_sync_handler+0xb0/0xb4 [ 259.061124] el0_sync+0x160/0x180 The root cause was that hrtimer_cancel are holding spin_lock, which the timer are running. Thread: hrtimer: acqurie lock acqurie lock cancel hrtimer sync ... // deadlock ... release lock release lock Thread will wait hrtimer exit forever, and the hrtimer are waiting spin_ lock forever. Fixes: 16980bd872f6 ("sched: Introduce smart grid scheduling strategy for cfs") Signed-off-by: Yipeng Zou <zouyipeng(a)huawei.com> --- kernel/sched/fair.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index ccd2a060c2df..ce6bb4304949 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5403,6 +5403,12 @@ static enum hrtimer_restart sched_auto_affi_period_timer(struct hrtimer *timer) } raw_spin_lock_irqsave(&auto_affi->lock, flags); + /* May be re-entrant by stop_auto_affinity, So check again. */ + if (auto_affi->period_active == 0) { + raw_spin_unlock_irqrestore(&auto_affi->lock, flags); + return HRTIMER_NORESTART; + } + if (util_avg_sum * 100 >= tg_capacity * sysctl_sched_util_low_pct) { affinity_domain_up(tg); } else if (util_avg_sum * 100 < tg_capacity * @@ -5496,8 +5502,6 @@ void stop_auto_affinity(struct auto_affinity *auto_affi) mutex_unlock(&smart_grid_used_mutex); return; } - - hrtimer_cancel(&auto_affi->period_timer); auto_affi->period_active = 0; auto_affi->mode = 0; ad->curr_level = ad->dcount > 0 ? ad->dcount - 1 : 0; -- 2.34.1

1 0

[PATCH openEuler-1.0-LTS] lib/notifier-error-inject: fix error when writing -errno to debugfs file
by Zhao Wenhui 01 Dec '23

01 Dec '23

From: Akinobu Mita <akinobu.mita(a)gmail.com> mainline inclusion from mainline-v6.2-rc1 commit f883c3edd2c432a2931ec8773c70a570115a50fe category: bugfix bugzilla: 189394,https://gitee.com/openeuler/kernel/issues/I8KYFI CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The simple attribute files do not accept a negative value since the commit 488dac0c9237 ("libfs: fix error cast of negative value in simple_attr_write()"). This restores the previous behaviour by using newly introduced DEFINE_SIMPLE_ATTRIBUTE_SIGNED instead of DEFINE_SIMPLE_ATTRIBUTE. Link: https://lkml.kernel.org/r/20220919172418.45257-3-akinobu.mita@gmail.com Fixes: 488dac0c9237 ("libfs: fix error cast of negative value in simple_attr_write()") Signed-off-by: Akinobu Mita <akinobu.mita(a)gmail.com> Reported-by: Zhao Gongyi <zhaogongyi(a)huawei.com> Reviewed-by: David Hildenbrand <david(a)redhat.com> Reviewed-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Cc: Alexander Viro <viro(a)zeniv.linux.org.uk> Cc: Jonathan Corbet <corbet(a)lwn.net> Cc: Oscar Salvador <osalvador(a)suse.de> Cc: Rafael J. Wysocki <rafael(a)kernel.org> Cc: Shuah Khan <shuah(a)kernel.org> Cc: Wei Yongjun <weiyongjun1(a)huawei.com> Cc: Yicong Yang <yangyicong(a)hisilicon.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Zhao Wenhui <zhaowenhui8(a)huawei.com> --- lib/notifier-error-inject.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/notifier-error-inject.c b/lib/notifier-error-inject.c index eb4a04afea80..125ea8ce23a4 100644 --- a/lib/notifier-error-inject.c +++ b/lib/notifier-error-inject.c @@ -14,7 +14,7 @@ static int debugfs_errno_get(void *data, u64 *val) return 0; } -DEFINE_SIMPLE_ATTRIBUTE(fops_errno, debugfs_errno_get, debugfs_errno_set, +DEFINE_SIMPLE_ATTRIBUTE_SIGNED(fops_errno, debugfs_errno_get, debugfs_errno_set, "%lld\n"); static struct dentry *debugfs_create_errno(const char *name, umode_t mode, -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] iio: industrialio-core: Fix debugfs read
by Zhao Wenhui 01 Dec '23

01 Dec '23

From: Alexandru Tachici <alexandru.tachici(a)analog.com> mainline inclusion from mainline-v5.7-rc1 commit 2ddc982a6c172c141183ac954d44f01bc4f2fe72 category: bugfix bugzilla: 94099,https://gitee.com/openeuler/kernel/issues/I8KXZX CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Currently iio_debugfs_read_reg calls debugfs_reg_access every time it is ran. Reading the same hardware register multiple times during the same reading of a debugfs file can cause unintended effects. For example for each: cat iio:device0/direct_reg_access the file_operations.read function will be called at least twice. First will return the full length of the string in bytes and the second will return 0. This patch makes iio_debugfs_read_reg to call debugfs_reg_access only when the user's buffer position (*ppos) is 0. (meaning it is the beginning of a new reading of the debugfs file). Fixes: e553f182d55b ("staging: iio: core: Introduce debugfs support, add support for direct register access") Signed-off-by: Alexandru Ardelean <alexandru.ardelean(a)analog.com> Signed-off-by: Alexandru Tachici <alexandru.tachici(a)analog.com> Signed-off-by: Jonathan Cameron <Jonathan.Cameron(a)huawei.com> Signed-off-by: Zhao Wenhui <zhaowenhui8(a)huawei.com> --- drivers/iio/industrialio-core.c | 15 +++++++++++---- include/linux/iio/iio.h | 2 ++ 2 files changed, 13 insertions(+), 4 deletions(-) diff --git a/drivers/iio/industrialio-core.c b/drivers/iio/industrialio-core.c index 49d4b4f1a457..421dcbae80f1 100644 --- a/drivers/iio/industrialio-core.c +++ b/drivers/iio/industrialio-core.c @@ -297,11 +297,14 @@ static ssize_t iio_debugfs_read_reg(struct file *file, char __user *userbuf, size_t count, loff_t *ppos) { struct iio_dev *indio_dev = file->private_data; - char buf[20]; unsigned val = 0; - ssize_t len; int ret; + if (*ppos > 0) + return simple_read_from_buffer(userbuf, count, ppos, + indio_dev->read_buf, + indio_dev->read_buf_len); + ret = indio_dev->info->debugfs_reg_access(indio_dev, indio_dev->cached_reg_addr, 0, &val); @@ -310,9 +313,13 @@ static ssize_t iio_debugfs_read_reg(struct file *file, char __user *userbuf, return ret; } - len = snprintf(buf, sizeof(buf), "0x%X\n", val); + indio_dev->read_buf_len = snprintf(indio_dev->read_buf, + sizeof(indio_dev->read_buf), + "0x%X\n", val); - return simple_read_from_buffer(userbuf, count, ppos, buf, len); + return simple_read_from_buffer(userbuf, count, ppos, + indio_dev->read_buf, + indio_dev->read_buf_len); } static ssize_t iio_debugfs_write_reg(struct file *file, diff --git a/include/linux/iio/iio.h b/include/linux/iio/iio.h index a74cb177dc6f..b715afa06aef 100644 --- a/include/linux/iio/iio.h +++ b/include/linux/iio/iio.h @@ -569,6 +569,8 @@ struct iio_dev { #if defined(CONFIG_DEBUG_FS) struct dentry *debugfs_dentry; unsigned cached_reg_addr; + char read_buf[20]; + unsigned int read_buf_len; #endif }; -- 2.34.1

2 1

[src-openEuler/rdma-core] libhns: Bugfix for wrong timing of modifying ibv_qp state to err
by Chengchang Tang 01 Dec '23

01 Dec '23

From: Yangyang Li <liyangyang20(a)huawei.com> driver inclusion category: bugfix bugzilla: https://gitee.com/src-openeuler/rdma-core/issues/I8KG9M -------------------------------------------------------------------------- Currently the QPC state in HW is modified inside the critical section of spinlock but the ibv_qp state is modified outside. There will be a short period when QPC state has been modified to err with ibv_qp state still remaining RTS. WQEs during this period will still be post-send by RTS-state ibv_qp but then dropped by err-state HW with no flush CQEs generated. To fix this problem, the QPC state in HW and ibv_qp state should be both modified to err inside the critical section of spinlock. Fixes: f1a80cc3dfe2 ("libhns: Bugfix for flush cqe in case multi-process") Signed-off-by: Yangyang Li <liyangyang20(a)huawei.com> --- providers/hns/hns_roce_u_hw_v2.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/providers/hns/hns_roce_u_hw_v2.c b/providers/hns/hns_roce_u_hw_v2.c index 2fb738d..68d7110 100644 --- a/providers/hns/hns_roce_u_hw_v2.c +++ b/providers/hns/hns_roce_u_hw_v2.c @@ -1936,6 +1936,8 @@ static int hns_roce_u_v2_modify_qp(struct ibv_qp *qp, struct ibv_qp_attr *attr, sizeof(resp_ex)); if (flag) { + if (!ret) + qp->state = IBV_QPS_ERR; hns_roce_spin_unlock(&hr_qp->sq.hr_lock); hns_roce_spin_unlock(&hr_qp->rq.hr_lock); } -- 2.30.0

1 0

[PATCH OLK-6.6 0/2] openeuler_defconfig: enable some mm new
by Kefeng Wang 01 Dec '23

01 Dec '23

Enable some new mm feature for openEuler OLK-6.6. Kefeng Wang (2): x86: openeuler_defconfig: update from new feature arm64: openeuler_defconfig: update for new feature arch/arm64/configs/openeuler_defconfig | 52 +++++++++++++++++++++----- arch/x86/configs/openeuler_defconfig | 35 ++++++++++++++--- 2 files changed, 72 insertions(+), 15 deletions(-) -- 2.27.0

2 3

[PATCH v2 OLK-5.10] spdxcheck.py: Fix a type error
by Jialin Zhang 01 Dec '23

01 Dec '23

From: zhujun2 <zhujun2_yewu(a)cmss.chinamobile.com> mainline inclusion from mainline-v5.18-rc1 commit 28c9f3f9a01d954d8357bdb68fbe36255997bf14 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7E2XC Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… ----------------------------- remove unused variable "col", otherwise there will be a type error as below: typeerror: not all arguments converted during string formatting Signed-off-by: Ding Xiang <dingxiang(a)cmss.chinamobile.com> Link: https://lore.kernel.org/r/20220114024058.74536-1-dingxiang@cmss.chinamobile… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: zhujun2 <zhujun2_yewu(a)cmss.chinamobile.com> Signed-off-by: Jialin Zhang <zhangjialin11(a)huawei.com> --- scripts/spdxcheck.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/spdxcheck.py b/scripts/spdxcheck.py index bc87200f9c7c..2288192967fa 100755 --- a/scripts/spdxcheck.py +++ b/scripts/spdxcheck.py @@ -200,7 +200,7 @@ class id_parser(object): tok = pe.tok.value sys.stdout.write('%s: %d:%d %s: %s\n' %(fname, self.curline, col, pe.txt, tok)) else: - sys.stdout.write('%s: %d:0 %s\n' %(fname, self.curline, col, pe.txt)) + sys.stdout.write('%s: %d:0 %s\n' %(fname, self.curline, pe.txt)) self.spdx_errors += 1 def scan_git_tree(tree): -- 2.25.1

2 1

[PATCH OLK-5.10 0/2] arm64/mpam: Fix static analysis warning
by Jialin Zhang 30 Nov '23

30 Nov '23

Wang ShaoBo (2): arm64/mpam: Fix uninitialized symbol ret arm64/mpam: Fix buffer overflow in mpam_handle_error_irq() arch/arm64/kernel/mpam/mpam_ctrlmon.c | 2 +- arch/arm64/kernel/mpam/mpam_device.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) -- 2.25.1

2 3

[PATCH 0/5 v2] crypto: hisilicon - fix the process to obtain capability register value
by Zhiqi Song 30 Nov '23

30 Nov '23

This seires patch modify the process to obtain the value of capablity registers. Pre-store the valid values of them. Changes in v2: - Modify sec_get_alg_bitmap sequence. Wenkai Lin (1): crypto: hisilicon/qm - add a function to set qm algs Zhiqi Song (4): crypto: hisilicon/qm - save capability registers in qm init process crypto: hisilicon/hpre - save capability registers in probe process crypto: hisilicon/sec2 - save capability registers in probe process crypto: hisilicon/zip - save capability registers in probe process drivers/crypto/hisilicon/hpre/hpre_main.c | 122 +++++++++++---------- drivers/crypto/hisilicon/qm.c | 98 +++++++++++++++-- drivers/crypto/hisilicon/sec2/sec.h | 7 ++ drivers/crypto/hisilicon/sec2/sec_crypto.c | 10 +- drivers/crypto/hisilicon/sec2/sec_main.c | 78 ++++++------- drivers/crypto/hisilicon/zip/zip_main.c | 120 +++++++++++--------- include/linux/hisi_acc_qm.h | 20 +++- 7 files changed, 297 insertions(+), 158 deletions(-) -- 2.30.0

1 5

[PATCH OLK-5.10] spdxcheck.py: Fix a type error
by Jialin Zhang 30 Nov '23

30 Nov '23

From: zhujun2 <zhujun2_yewu(a)cmss.chinamobile.com> mainline inclusion from mainline-v5.18-rc1 commit 28c9f3f9a01d954d8357bdb68fbe36255997bf14 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7E2XC CVE:NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… ----------------------------- remove unused variable "col", otherwise there will be a type error as below: typeerror: not all arguments converted during string formatting Signed-off-by: Ding Xiang <dingxiang(a)cmss.chinamobile.com> Link: https://lore.kernel.org/r/20220114024058.74536-1-dingxiang@cmss.chinamobile… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: zhujun2 <zhujun2_yewu(a)cmss.chinamobile.com> Signed-off-by: Jialin Zhang <zhangjialin11(a)huawei.com> --- scripts/spdxcheck.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/spdxcheck.py b/scripts/spdxcheck.py index bc87200f9c7c..2288192967fa 100755 --- a/scripts/spdxcheck.py +++ b/scripts/spdxcheck.py @@ -200,7 +200,7 @@ class id_parser(object): tok = pe.tok.value sys.stdout.write('%s: %d:%d %s: %s\n' %(fname, self.curline, col, pe.txt, tok)) else: - sys.stdout.write('%s: %d:0 %s\n' %(fname, self.curline, col, pe.txt)) + sys.stdout.write('%s: %d:0 %s\n' %(fname, self.curline, pe.txt)) self.spdx_errors += 1 def scan_git_tree(tree): -- 2.25.1

2 1

[PATCH OLK-6.6 00/27] ARM: support kaslr feature in arm32 platform
by felix 30 Nov '23

30 Nov '23

From: Felix Fu <fuzhen5(a)huawei.com> support kaslr feature in arm32 platform, using CONFIG_RANDOMIZE_BASE=y to enable this feature. Ard Biesheuvel (15): asm-generic: add .data.rel.ro sections to __ro_after_init arm-soc: exynos: replace open coded VA->PA conversions arm-soc: mvebu: replace open coded VA->PA conversion arm-soc: various: replace open coded VA->PA calculation of pen_release ARM: kernel: switch to relative exception tables ARM: kernel: make vmlinux buildable as a PIE executable ARM: kernel: use PC-relative symbol references in MMU switch code ARM: kernel: use PC relative symbol references in suspend/resume code ARM: mm: export default vmalloc base address ARM: kernel: refer to swapper_pg_dir via its symbol arm: vectors: use local symbol names for vector entry points ARM: kernel: implement randomization of the kernel load address ARM: decompressor: explicitly map decompressor binary cacheable ARM: decompressor: add KASLR support No idea why this broke ... Cui GaoSheng (6): arm32: kaslr: Add missing sections about relocatable arm32: kaslr: Fix the bug of module install failure arm32: kaslr: Fix the bug of hidden symbols when decompressing code is compiled arm32: kaslr: Adapt dts files of multiple memory nodes arm32: kaslr: Fix the bug of symbols relocation arm32: kaslr: Print the real kaslr offset when kernel panic Ye Bin (6): arm32: kaslr: Add missing sections about relocatable arm: kaslr: Fix memtop calculate, when there is no memory top info, we can't use zero instead it. arm32: kaslr: When boot with vxboot, we must adjust dtb address before kaslr_early_init, and store dtb address after init. arm32: kaslr: pop visibility when compile decompress boot code as we need relocate BSS by GOT. arm32: kaslr: print kaslr offset when kernel panic arm32: kaslr: Fix clock_gettime and gettimeofday performance degradation when configure CONFIG_RANDOMIZE_BASE arch/arm/Kconfig | 17 + arch/arm/Makefile | 6 + arch/arm/boot/compressed/Makefile | 17 +- arch/arm/boot/compressed/head.S | 139 +++++++- arch/arm/boot/compressed/kaslr.c | 461 ++++++++++++++++++++++++++ arch/arm/boot/compressed/misc.c | 1 - arch/arm/include/asm/Kbuild | 1 - arch/arm/include/asm/assembler.h | 18 +- arch/arm/include/asm/extable.h | 47 +++ arch/arm/include/asm/futex.h | 6 +- arch/arm/include/asm/memory.h | 14 + arch/arm/include/asm/pgtable.h | 1 + arch/arm/include/asm/uaccess.h | 17 +- arch/arm/include/asm/vmlinux.lds.h | 11 +- arch/arm/include/asm/word-at-a-time.h | 6 +- arch/arm/kernel/entry-armv.S | 40 +-- arch/arm/kernel/head-common.S | 60 +--- arch/arm/kernel/head.S | 107 +++++- arch/arm/kernel/setup.c | 31 ++ arch/arm/kernel/sleep.S | 7 +- arch/arm/kernel/swp_emulate.c | 8 +- arch/arm/kernel/vmlinux.lds.S | 19 ++ arch/arm/lib/backtrace.S | 13 +- arch/arm/lib/getuser.S | 24 +- arch/arm/lib/putuser.S | 15 +- arch/arm/mach-exynos/headsmp.S | 9 +- arch/arm/mach-exynos/sleep.S | 26 +- arch/arm/mach-mvebu/coherency_ll.S | 8 +- arch/arm/mach-spear/headsmp.S | 11 +- arch/arm/mach-versatile/headsmp.S | 9 +- arch/arm/mm/alignment.c | 24 +- arch/arm/mm/extable.c | 2 +- arch/arm/nwfpe/entry.S | 6 +- arch/arm/vdso/vgettimeofday.c | 5 + include/asm-generic/vmlinux.lds.h | 2 +- scripts/module.lds.S | 1 + scripts/sorttable.c | 2 +- 37 files changed, 956 insertions(+), 235 deletions(-) create mode 100644 arch/arm/boot/compressed/kaslr.c create mode 100644 arch/arm/include/asm/extable.h -- 2.34.1

1 27

[PATCH openEuler-1.0-LTS] net: ipv6: unexport __init-annotated seg6_hmac_net_init()
by Liu Jian 30 Nov '23

30 Nov '23

From: YueHaibing <yuehaibing(a)huawei.com> stable inclusion from stable-v4.19.251 commit f6d9b011d8019bb59be4b71dd3d7de7c5e769d3e category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8KNM7 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… --------------------------- commit 53ad46169fe2996fe1b623ba6c9c4fa33847876f upstream. As of commit 5801f064e351 ("net: ipv6: unexport __init-annotated seg6_hmac_init()"), EXPORT_SYMBOL and __init is a bad combination because the .init.text section is freed up after the initialization. Hence, modules cannot use symbols annotated __init. The access to a freed symbol may end up with kernel panic. This remove the EXPORT_SYMBOL to fix modpost warning: WARNING: modpost: vmlinux.o(___ksymtab+seg6_hmac_net_init+0x0): Section mismatch in reference from the variable __ksymtab_seg6_hmac_net_init to the function .init.text:seg6_hmac_net_init() The symbol seg6_hmac_net_init is exported and annotated __init Fix this by removing the __init annotation of seg6_hmac_net_init or drop the export. Fixes: bf355b8d2c30 ("ipv6: sr: add core files for SR HMAC support") Reported-by: Hulk Robot <hulkci(a)huawei.com> Signed-off-by: YueHaibing <yuehaibing(a)huawei.com> Link: https://lore.kernel.org/r/20220628033134.21088-1-yuehaibing@huawei.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> (cherry picked from commit f6d9b011d8019bb59be4b71dd3d7de7c5e769d3e) Signed-off-by: Liu Jian <liujian56(a)huawei.com> --- net/ipv6/seg6_hmac.c | 1 - 1 file changed, 1 deletion(-) diff --git a/net/ipv6/seg6_hmac.c b/net/ipv6/seg6_hmac.c index 8546f94f30d47..013573f61f045 100644 --- a/net/ipv6/seg6_hmac.c +++ b/net/ipv6/seg6_hmac.c @@ -416,7 +416,6 @@ int __net_init seg6_hmac_net_init(struct net *net) return 0; } -EXPORT_SYMBOL(seg6_hmac_net_init); void seg6_hmac_exit(void) { -- 2.34.1

2 1

[PATCH OLK-6.6 v3 00/31] block: Make blkdev_get_by_*() return
by Li Lingfeng 30 Nov '23

30 Nov '23

Create struct bdev_handle that contains all parameters that need to be passed to blkdev_put() and provide bdev_open_* functions that return this structure instead of plain bdev pointer. v1->v2: Add new bugfix 8bd90b6ae785 ("erofs: fix NULL dereference of dif->bdev_handle in fscache mode"). Fix the comment of cd34758c5238 ("block: Remove blkdev_get_by_() functions"). v2->v3: Fix the comment of acb083b55597 ("block: Use bdev_open_by_dev() in disk_scan_partitions() and blkdev_bszset()"). Jan Kara (30): block: Provide bdev_open_* functions block: Use bdev_open_by_dev() in blkdev_open() block: Use bdev_open_by_dev() in disk_scan_partitions() and blkdev_bszset() drdb: Convert to use bdev_open_by_path() pktcdvd: Convert to bdev_open_by_dev() rnbd-srv: Convert to use bdev_open_by_path() xen/blkback: Convert to bdev_open_by_dev() zram: Convert to use bdev_open_by_dev() bcache: Convert to bdev_open_by_path() dm: Convert to bdev_open_by_dev() md: Convert to bdev_open_by_dev() mtd: block2mtd: Convert to bdev_open_by_dev/path() nvmet: Convert to bdev_open_by_path() s390/dasd: Convert to bdev_open_by_path() scsi: target: Convert to bdev_open_by_path() PM: hibernate: Convert to bdev_open_by_dev() PM: hibernate: Drop unused snapshot_test argument mm/swap: Convert to use bdev_open_by_dev() fs: Convert to bdev_open_by_dev() btrfs: Convert to bdev_open_by_path() erofs: Convert to use bdev_open_by_path() ext4: Convert to bdev_open_by_dev() f2fs: Convert to bdev_open_by_dev/path() jfs: Convert to bdev_open_by_dev() nfs/blocklayout: Convert to use bdev_open_by_dev/path() ocfs2: Convert to use bdev_open_by_dev() reiserfs: Convert to bdev_open_by_dev/path() xfs: Convert to bdev_open_by_path() bcache: Fixup error handling in register_cache() block: Remove blkdev_get_by_*() functions Jingbo Xu (1): erofs: fix NULL dereference of dif->bdev_handle in fscache mode block/bdev.c | 65 ++++++++++------- block/fops.c | 44 +++++++----- block/genhd.c | 12 ++-- block/ioctl.c | 6 +- drivers/block/drbd/drbd_int.h | 2 + drivers/block/drbd/drbd_nl.c | 65 +++++++++-------- drivers/block/pktcdvd.c | 76 +++++++++++--------- drivers/block/rnbd/rnbd-srv.c | 27 +++---- drivers/block/rnbd/rnbd-srv.h | 2 +- drivers/block/xen-blkback/blkback.c | 4 +- drivers/block/xen-blkback/common.h | 4 +- drivers/block/xen-blkback/xenbus.c | 40 ++++++----- drivers/block/zram/zram_drv.c | 31 ++++---- drivers/block/zram/zram_drv.h | 2 +- drivers/md/bcache/bcache.h | 2 + drivers/md/bcache/super.c | 95 ++++++++++++------------ drivers/md/dm.c | 20 +++--- drivers/md/md.c | 23 +++--- drivers/md/md.h | 4 +- drivers/mtd/devices/block2mtd.c | 51 +++++++------ drivers/nvme/target/io-cmd-bdev.c | 20 +++--- drivers/nvme/target/nvmet.h | 1 + drivers/s390/block/dasd.c | 12 ++-- drivers/s390/block/dasd_genhd.c | 45 ++++++------ drivers/s390/block/dasd_int.h | 2 +- drivers/s390/block/dasd_ioctl.c | 2 +- drivers/target/target_core_iblock.c | 19 ++--- drivers/target/target_core_iblock.h | 1 + drivers/target/target_core_pscsi.c | 26 +++---- drivers/target/target_core_pscsi.h | 2 +- fs/btrfs/dev-replace.c | 14 ++-- fs/btrfs/ioctl.c | 18 +++-- fs/btrfs/volumes.c | 107 ++++++++++++++-------------- fs/btrfs/volumes.h | 6 +- fs/cramfs/inode.c | 2 +- fs/erofs/data.c | 5 +- fs/erofs/internal.h | 2 +- fs/erofs/super.c | 20 +++--- fs/ext4/ext4.h | 2 +- fs/ext4/fsmap.c | 9 +-- fs/ext4/super.c | 52 +++++++------- fs/f2fs/f2fs.h | 1 + fs/f2fs/super.c | 13 ++-- fs/jfs/jfs_logmgr.c | 29 ++++---- fs/jfs/jfs_logmgr.h | 2 +- fs/jfs/jfs_mount.c | 3 +- fs/nfs/blocklayout/blocklayout.h | 2 +- fs/nfs/blocklayout/dev.c | 76 ++++++++++---------- fs/ocfs2/cluster/heartbeat.c | 81 +++++++++++---------- fs/reiserfs/journal.c | 56 +++++++-------- fs/reiserfs/procfs.c | 2 +- fs/reiserfs/reiserfs.h | 11 +-- fs/romfs/super.c | 2 +- fs/super.c | 15 ++-- fs/xfs/xfs_buf.c | 22 +++--- fs/xfs/xfs_buf.h | 3 +- fs/xfs/xfs_super.c | 42 ++++++----- include/linux/blkdev.h | 12 +++- include/linux/device-mapper.h | 1 + include/linux/fs.h | 1 + include/linux/pktcdvd.h | 4 +- include/linux/swap.h | 1 + kernel/power/hibernate.c | 14 ++-- kernel/power/power.h | 2 +- kernel/power/swap.c | 37 +++++----- mm/swapfile.c | 23 +++--- 66 files changed, 740 insertions(+), 657 deletions(-) -- 2.31.1

2 32

[PATCH OLK-6.6 v2 00/31] block: Make blkdev_get_by_*() return
by Li Lingfeng 30 Nov '23

30 Nov '23

Create struct bdev_handle that contains all parameters that need to be passed to blkdev_put() and provide bdev_open_* functions that return this structure instead of plain bdev pointer. v1->v2: Add new bugfix 8bd90b6ae785 ("erofs: fix NULL dereference of dif->bdev_handle in fscache mode"). Fix the comment of fd70b3b44 ("block: Remove blkdev_get_by_() functions"). Jan Kara (30): block: Provide bdev_open_* functions block: Use bdev_open_by_dev() in blkdev_open() block: Use bdev_open_by_dev() in disk_scan_partitions() and blkdev_bszset() drdb: Convert to use bdev_open_by_path() pktcdvd: Convert to bdev_open_by_dev() rnbd-srv: Convert to use bdev_open_by_path() xen/blkback: Convert to bdev_open_by_dev() zram: Convert to use bdev_open_by_dev() bcache: Convert to bdev_open_by_path() dm: Convert to bdev_open_by_dev() md: Convert to bdev_open_by_dev() mtd: block2mtd: Convert to bdev_open_by_dev/path() nvmet: Convert to bdev_open_by_path() s390/dasd: Convert to bdev_open_by_path() scsi: target: Convert to bdev_open_by_path() PM: hibernate: Convert to bdev_open_by_dev() PM: hibernate: Drop unused snapshot_test argument mm/swap: Convert to use bdev_open_by_dev() fs: Convert to bdev_open_by_dev() btrfs: Convert to bdev_open_by_path() erofs: Convert to use bdev_open_by_path() ext4: Convert to bdev_open_by_dev() f2fs: Convert to bdev_open_by_dev/path() jfs: Convert to bdev_open_by_dev() nfs/blocklayout: Convert to use bdev_open_by_dev/path() ocfs2: Convert to use bdev_open_by_dev() reiserfs: Convert to bdev_open_by_dev/path() xfs: Convert to bdev_open_by_path() bcache: Fixup error handling in register_cache() block: Remove blkdev_get_by_*() functions Jingbo Xu (1): erofs: fix NULL dereference of dif->bdev_handle in fscache mode block/bdev.c | 65 ++++++++++------- block/fops.c | 44 +++++++----- block/genhd.c | 12 ++-- block/ioctl.c | 6 +- drivers/block/drbd/drbd_int.h | 2 + drivers/block/drbd/drbd_nl.c | 65 +++++++++-------- drivers/block/pktcdvd.c | 76 +++++++++++--------- drivers/block/rnbd/rnbd-srv.c | 27 +++---- drivers/block/rnbd/rnbd-srv.h | 2 +- drivers/block/xen-blkback/blkback.c | 4 +- drivers/block/xen-blkback/common.h | 4 +- drivers/block/xen-blkback/xenbus.c | 40 ++++++----- drivers/block/zram/zram_drv.c | 31 ++++---- drivers/block/zram/zram_drv.h | 2 +- drivers/md/bcache/bcache.h | 2 + drivers/md/bcache/super.c | 95 ++++++++++++------------ drivers/md/dm.c | 20 +++--- drivers/md/md.c | 23 +++--- drivers/md/md.h | 4 +- drivers/mtd/devices/block2mtd.c | 51 +++++++------ drivers/nvme/target/io-cmd-bdev.c | 20 +++--- drivers/nvme/target/nvmet.h | 1 + drivers/s390/block/dasd.c | 12 ++-- drivers/s390/block/dasd_genhd.c | 45 ++++++------ drivers/s390/block/dasd_int.h | 2 +- drivers/s390/block/dasd_ioctl.c | 2 +- drivers/target/target_core_iblock.c | 19 ++--- drivers/target/target_core_iblock.h | 1 + drivers/target/target_core_pscsi.c | 26 +++---- drivers/target/target_core_pscsi.h | 2 +- fs/btrfs/dev-replace.c | 14 ++-- fs/btrfs/ioctl.c | 18 +++-- fs/btrfs/volumes.c | 107 ++++++++++++++-------------- fs/btrfs/volumes.h | 6 +- fs/cramfs/inode.c | 2 +- fs/erofs/data.c | 5 +- fs/erofs/internal.h | 2 +- fs/erofs/super.c | 20 +++--- fs/ext4/ext4.h | 2 +- fs/ext4/fsmap.c | 9 +-- fs/ext4/super.c | 52 +++++++------- fs/f2fs/f2fs.h | 1 + fs/f2fs/super.c | 13 ++-- fs/jfs/jfs_logmgr.c | 29 ++++---- fs/jfs/jfs_logmgr.h | 2 +- fs/jfs/jfs_mount.c | 3 +- fs/nfs/blocklayout/blocklayout.h | 2 +- fs/nfs/blocklayout/dev.c | 76 ++++++++++---------- fs/ocfs2/cluster/heartbeat.c | 81 +++++++++++---------- fs/reiserfs/journal.c | 56 +++++++-------- fs/reiserfs/procfs.c | 2 +- fs/reiserfs/reiserfs.h | 11 +-- fs/romfs/super.c | 2 +- fs/super.c | 15 ++-- fs/xfs/xfs_buf.c | 22 +++--- fs/xfs/xfs_buf.h | 3 +- fs/xfs/xfs_super.c | 42 ++++++----- include/linux/blkdev.h | 12 +++- include/linux/device-mapper.h | 1 + include/linux/fs.h | 1 + include/linux/pktcdvd.h | 4 +- include/linux/swap.h | 1 + kernel/power/hibernate.c | 14 ++-- kernel/power/power.h | 2 +- kernel/power/swap.c | 37 +++++----- mm/swapfile.c | 23 +++--- 66 files changed, 740 insertions(+), 657 deletions(-) -- 2.31.1

2 32

[PATCH OLK-6.6 00/30] block: Make blkdev_get_by_*() return
by Li Lingfeng 30 Nov '23

30 Nov '23

Create struct bdev_handle that contains all parameters that need to be passed to blkdev_put() and provide bdev_open_* functions that return this structure instead of plain bdev pointer. Jan Kara (30): block: Provide bdev_open_* functions block: Use bdev_open_by_dev() in blkdev_open() block: Use bdev_open_by_dev() in disk_scan_partitions() and blkdev_bszset() drdb: Convert to use bdev_open_by_path() pktcdvd: Convert to bdev_open_by_dev() rnbd-srv: Convert to use bdev_open_by_path() xen/blkback: Convert to bdev_open_by_dev() zram: Convert to use bdev_open_by_dev() bcache: Convert to bdev_open_by_path() dm: Convert to bdev_open_by_dev() md: Convert to bdev_open_by_dev() mtd: block2mtd: Convert to bdev_open_by_dev/path() nvmet: Convert to bdev_open_by_path() s390/dasd: Convert to bdev_open_by_path() scsi: target: Convert to bdev_open_by_path() PM: hibernate: Convert to bdev_open_by_dev() PM: hibernate: Drop unused snapshot_test argument mm/swap: Convert to use bdev_open_by_dev() fs: Convert to bdev_open_by_dev() btrfs: Convert to bdev_open_by_path() erofs: Convert to use bdev_open_by_path() ext4: Convert to bdev_open_by_dev() f2fs: Convert to bdev_open_by_dev/path() jfs: Convert to bdev_open_by_dev() nfs/blocklayout: Convert to use bdev_open_by_dev/path() ocfs2: Convert to use bdev_open_by_dev() reiserfs: Convert to bdev_open_by_dev/path() xfs: Convert to bdev_open_by_path() bcache: Fixup error handling in register_cache() block: Remove blkdev_get_by_*() functions block/bdev.c | 65 ++++++++++------- block/fops.c | 44 +++++++----- block/genhd.c | 12 ++-- block/ioctl.c | 6 +- drivers/block/drbd/drbd_int.h | 2 + drivers/block/drbd/drbd_nl.c | 65 +++++++++-------- drivers/block/pktcdvd.c | 76 +++++++++++--------- drivers/block/rnbd/rnbd-srv.c | 27 +++---- drivers/block/rnbd/rnbd-srv.h | 2 +- drivers/block/xen-blkback/blkback.c | 4 +- drivers/block/xen-blkback/common.h | 4 +- drivers/block/xen-blkback/xenbus.c | 40 ++++++----- drivers/block/zram/zram_drv.c | 31 ++++---- drivers/block/zram/zram_drv.h | 2 +- drivers/md/bcache/bcache.h | 2 + drivers/md/bcache/super.c | 95 ++++++++++++------------ drivers/md/dm.c | 20 +++--- drivers/md/md.c | 23 +++--- drivers/md/md.h | 4 +- drivers/mtd/devices/block2mtd.c | 51 +++++++------ drivers/nvme/target/io-cmd-bdev.c | 20 +++--- drivers/nvme/target/nvmet.h | 1 + drivers/s390/block/dasd.c | 12 ++-- drivers/s390/block/dasd_genhd.c | 45 ++++++------ drivers/s390/block/dasd_int.h | 2 +- drivers/s390/block/dasd_ioctl.c | 2 +- drivers/target/target_core_iblock.c | 19 ++--- drivers/target/target_core_iblock.h | 1 + drivers/target/target_core_pscsi.c | 26 +++---- drivers/target/target_core_pscsi.h | 2 +- fs/btrfs/dev-replace.c | 14 ++-- fs/btrfs/ioctl.c | 18 +++-- fs/btrfs/volumes.c | 107 ++++++++++++++-------------- fs/btrfs/volumes.h | 6 +- fs/cramfs/inode.c | 2 +- fs/erofs/data.c | 4 +- fs/erofs/internal.h | 2 +- fs/erofs/super.c | 20 +++--- fs/ext4/ext4.h | 2 +- fs/ext4/fsmap.c | 9 +-- fs/ext4/super.c | 52 +++++++------- fs/f2fs/f2fs.h | 1 + fs/f2fs/super.c | 13 ++-- fs/jfs/jfs_logmgr.c | 29 ++++---- fs/jfs/jfs_logmgr.h | 2 +- fs/jfs/jfs_mount.c | 3 +- fs/nfs/blocklayout/blocklayout.h | 2 +- fs/nfs/blocklayout/dev.c | 76 ++++++++++---------- fs/ocfs2/cluster/heartbeat.c | 81 +++++++++++---------- fs/reiserfs/journal.c | 56 +++++++-------- fs/reiserfs/procfs.c | 2 +- fs/reiserfs/reiserfs.h | 11 +-- fs/romfs/super.c | 2 +- fs/super.c | 15 ++-- fs/xfs/xfs_buf.c | 22 +++--- fs/xfs/xfs_buf.h | 3 +- fs/xfs/xfs_super.c | 42 ++++++----- include/linux/blkdev.h | 12 +++- include/linux/device-mapper.h | 1 + include/linux/fs.h | 1 + include/linux/pktcdvd.h | 4 +- include/linux/swap.h | 1 + kernel/power/hibernate.c | 14 ++-- kernel/power/power.h | 2 +- kernel/power/swap.c | 37 +++++----- mm/swapfile.c | 23 +++--- 66 files changed, 739 insertions(+), 657 deletions(-) -- 2.31.1

2 31

[PATCH OLK-6.6 0/6] ipvlan: Introduce l2e mode
by Zhengchao Shao 30 Nov '23

30 Nov '23

Inherited features of olk5.10 version. Keefe LIU (4): ipvlan: Introduce l2e mode ipvlan: check the value of module parameter ipvlan_default_mode ipvlan: Introduce local xmit queue for l2e mode ipvlan: limit loop_qlen ranges Keefe Liu (1): ipvlan: limit loop_delay ranges Lu Wei (1): ipvlan: Modify the value of ipvlan modes drivers/net/ipvlan/ipvlan.h | 7 + drivers/net/ipvlan/ipvlan_core.c | 222 +++++++++++++++++++++++++++++++ drivers/net/ipvlan/ipvlan_main.c | 85 +++++++++++- include/uapi/linux/if_link.h | 1 + 4 files changed, 314 insertions(+), 1 deletion(-) -- 2.34.1

2 7

[PATCH OLK-5.10 00/20] KVM: arm64: Add vtimer interrupt bypass
by Dongxu Sun 30 Nov '23

30 Nov '23

Haibin Wang (2): KVM: arm64: arch_timer: Probe vtimer irqbypass capability KVM: arm64: arch_timer: Rework vcpu init/reset logic Kunkun Jiang (6): mbigen: vtimer mbigen driver support irqchip/gic-v4.1: Detect ITS vtimer interrupt bypass capability irqchip/gic-v4.1: Extend VSGI command to support the new vPPI mbigen: vtimer: add support for MBIX1_CPPI_NEGEDGE_CLR_EN_SETR(CLRR) KVM: arm64: vgic-v3: Clearing pending status of vtimer on guest reset mbigen: Sets the regs related to vtimer irqbypass Zenghui Yu (5): KVM: arm64: GICv4.1: Add direct injection capability to PPI registers KVM: arm64: arch_timer: Extract vtimer hyp_init logic KVM: arm64: GICv4.1: Allow non-trapping WFI when using direct vtimer interrupt KVM: arm64: vtimer: Expose HW-based vtimer interrupt in debugfs KVM: arm64: arch_timer: Make vtimer_irqbypass a Distributor attr wanghaibin (7): irqchip/gic-v4.1: Rework get/set_irqchip_state callbacks of GICv4.1-sgi chip irqchip/gic-v4.1: Rework its_alloc_vcpu_sgis() to support vPPI allocation irqchip/gic-v4.1: Probe vtimer irqbypass capability at RD level KVM: arm64: GICv4.1: Inform the HiSilicon vtimer irqbypass capability KVM: arm64: vgic: Add helper for vtimer vppi info register KVM: arm64: GICv4.1: Enable vtimer vPPI irqbypass config KVM: arm64: GICv4.1: Add support for MBIGEN save/restore arch/arm64/include/asm/kvm_emulate.h | 1 + arch/arm64/kvm/arch_timer.c | 273 ++++++++++-- arch/arm64/kvm/arm.c | 4 + arch/arm64/kvm/vgic/vgic-debug.c | 2 +- arch/arm64/kvm/vgic/vgic-init.c | 3 + arch/arm64/kvm/vgic/vgic-mmio-v3.c | 15 +- arch/arm64/kvm/vgic/vgic-mmio.c | 24 +- arch/arm64/kvm/vgic/vgic-v3.c | 8 + arch/arm64/kvm/vgic/vgic-v4.c | 57 +++ arch/arm64/kvm/vgic/vgic.c | 24 + arch/arm64/kvm/vgic/vgic.h | 11 + drivers/irqchip/irq-gic-phytium-2500-its.c | 2 +- drivers/irqchip/irq-gic-v3-its.c | 94 +++- drivers/irqchip/irq-gic-v3.c | 37 +- drivers/irqchip/irq-gic-v4.c | 23 +- drivers/irqchip/irq-mbigen.c | 487 ++++++++++++++++++++- include/clocksource/arm_arch_timer.h | 18 + include/kvm/arm_arch_timer.h | 13 + include/kvm/arm_vgic.h | 34 ++ include/linux/irqchip/arm-gic-common.h | 3 + include/linux/irqchip/arm-gic-v3.h | 49 ++- include/linux/irqchip/arm-gic-v4.h | 6 +- 22 files changed, 1099 insertions(+), 89 deletions(-) -- 2.33.0

2 21

[PATCH openEuler-1.0-LTS] net: ipv6: unexport __init-annotated seg6_hmac_net_init()
by Liu Jian 30 Nov '23

30 Nov '23

From: YueHaibing <yuehaibing(a)huawei.com> stable inclusion from stable-v4.19.251 commit f6d9b011d8019bb59be4b71dd3d7de7c5e769d3e category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8KNM7 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… --------------------------- commit 53ad46169fe2996fe1b623ba6c9c4fa33847876f upstream. As of commit 5801f064e351 ("net: ipv6: unexport __init-annotated seg6_hmac_init()"), EXPORT_SYMBOL and __init is a bad combination because the .init.text section is freed up after the initialization. Hence, modules cannot use symbols annotated __init. The access to a freed symbol may end up with kernel panic. This remove the EXPORT_SYMBOL to fix modpost warning: WARNING: modpost: vmlinux.o(___ksymtab+seg6_hmac_net_init+0x0): Section mismatch in reference from the variable __ksymtab_seg6_hmac_net_init to the function .init.text:seg6_hmac_net_init() The symbol seg6_hmac_net_init is exported and annotated __init Fix this by removing the __init annotation of seg6_hmac_net_init or drop the export. Fixes: bf355b8d2c30 ("ipv6: sr: add core files for SR HMAC support") Reported-by: Hulk Robot <hulkci(a)huawei.com> Signed-off-by: YueHaibing <yuehaibing(a)huawei.com> Link: https://lore.kernel.org/r/20220628033134.21088-1-yuehaibing@huawei.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> (cherry picked from commit f6d9b011d8019bb59be4b71dd3d7de7c5e769d3e) Signed-off-by: Liu Jian <liujian56(a)huawei.com> --- net/ipv6/seg6_hmac.c | 1 - 1 file changed, 1 deletion(-) diff --git a/net/ipv6/seg6_hmac.c b/net/ipv6/seg6_hmac.c index 8546f94f30d47..013573f61f045 100644 --- a/net/ipv6/seg6_hmac.c +++ b/net/ipv6/seg6_hmac.c @@ -416,7 +416,6 @@ int __net_init seg6_hmac_net_init(struct net *net) return 0; } -EXPORT_SYMBOL(seg6_hmac_net_init); void seg6_hmac_exit(void) { -- 2.34.1

1 0

[PATCH OLK-6.6] hugetlbfs: avoid overflow in hugetlbfs_fallocate
by Zizhi Wo 30 Nov '23

30 Nov '23

From: yangerkun <yangerkun(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8JDQ7?from=project-issue --------------------------- luojiajun report a problem[1] two years ago which seems still exists in mainline. vfs_fallocate can avoid 'offset + len' trigger overflow, but 'offset + len + hpage_size - 1' may overflow too and will lead to a wrong 'end'. luojiajun give a solution which can fix the wrong 'end' but leave the overflow still happened. Fix it with DIV_ROUND_UP_ULL. [1] https://patchwork.kernel.org/project/linux-mm/patch/1554775226-67213-1-git-… Signed-off-by: yangerkun <yangerkun(a)huawei.com> Signed-off-by: Zizhi Wo <wozizhi(a)huawei.com> --- fs/hugetlbfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 316c4cebd3f3..cb6a5ea74cda 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -836,7 +836,7 @@ static long hugetlbfs_fallocate(struct file *file, int mode, loff_t offset, * as well as being converted to page offsets. */ start = offset >> hpage_shift; - end = (offset + len + hpage_size - 1) >> hpage_shift; + end = DIV_ROUND_UP_ULL(offset + len, hpage_size); inode_lock(inode); -- 2.39.2

2 1

[PATCH OLK-6.6] fs/dcache.c: avoid softlock since too many negative dentry
by Zizhi Wo 30 Nov '23

30 Nov '23

From: yangerkun <yangerkun(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8JH5M?from=project-issue --------------------------- Parallel thread to add negative dentry under root dir. Sometimes later, 'systemctl daemon-reload' will report softlockup since __fsnotify_update_child_dentry_flags need update all child under root dentry without distinguish does it active or not. It will waste so long time with catching d_lock of root dentry. And other thread try to spin_lock d_lock will run overtime. Limit negative dentry under dir can avoid this. Signed-off-by: yangerkun <yangerkun(a)huawei.com> Signed-off-by: Zizhi Wo <wozizhi(a)huawei.com> --- fs/dcache.c | 43 ++++++++++++++++++++++++++++++++++++++++-- include/linux/dcache.h | 4 ++++ 2 files changed, 45 insertions(+), 2 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index 7cdd8bbedd1c..c27aa8ff78f4 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -341,10 +341,18 @@ static inline void __d_set_inode_and_type(struct dentry *dentry, unsigned type_flags) { unsigned flags; + struct dentry *parent; + + parent = dentry->d_parent; + if ((dentry->d_flags & DCACHE_NEGATIVE_ACCOUNT) && parent) { + WARN_ON(!inode); + atomic_dec(&parent->d_neg_dnum); + } dentry->d_inode = inode; flags = READ_ONCE(dentry->d_flags); - flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU); + flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU | + DCACHE_NEGATIVE_ACCOUNT); flags |= type_flags; smp_store_release(&dentry->d_flags, flags); } @@ -363,6 +371,7 @@ static inline void __d_clear_type_and_inode(struct dentry *dentry) static void dentry_free(struct dentry *dentry) { WARN_ON(!hlist_unhashed(&dentry->d_u.d_alias)); + WARN_ON(dentry->d_flags & DCACHE_NEGATIVE_ACCOUNT); if (unlikely(dname_external(dentry))) { struct external_name *p = external_name(dentry); if (likely(atomic_dec_and_test(&p->u.count))) { @@ -601,8 +610,14 @@ static void __dentry_kill(struct dentry *dentry) /* if it was on the hash then remove it */ __d_drop(dentry); dentry_unlist(dentry, parent); - if (parent) + if (parent) { + if (dentry->d_flags & DCACHE_NEGATIVE_ACCOUNT) { + atomic_dec(&parent->d_neg_dnum); + dentry->d_flags &= ~DCACHE_NEGATIVE_ACCOUNT; + } + spin_unlock(&parent->d_lock); + } if (dentry->d_inode) dentry_unlink_inode(dentry); else @@ -662,6 +677,8 @@ static inline struct dentry *lock_parent(struct dentry *dentry) static inline bool retain_dentry(struct dentry *dentry) { + struct dentry *parent; + WARN_ON(d_in_lookup(dentry)); /* Unreachable? Get rid of it */ @@ -679,6 +696,27 @@ static inline bool retain_dentry(struct dentry *dentry) if (unlikely(dentry->d_flags & DCACHE_DONTCACHE)) return false; + if (unlikely(!dentry->d_parent)) + goto noparent; + + parent = dentry->d_parent; + /* Return false if it's negative */ + WARN_ON((atomic_read(&parent->d_neg_dnum) < 0)); + if (!dentry->d_inode) { + if (!(dentry->d_flags & DCACHE_NEGATIVE_ACCOUNT)) { + unsigned int flags = READ_ONCE(dentry->d_flags); + + flags |= DCACHE_NEGATIVE_ACCOUNT; + WRITE_ONCE(dentry->d_flags, flags); + atomic_inc(&parent->d_neg_dnum); + } + } + + if (!dentry->d_inode && + atomic_read(&parent->d_neg_dnum) >= NEG_DENTRY_LIMIT) + return false; + +noparent: /* retain; LRU fodder */ dentry->d_lockref.count--; if (unlikely(!(dentry->d_flags & DCACHE_LRU_LIST))) @@ -1809,6 +1847,7 @@ static struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) seqcount_spinlock_init(&dentry->d_seq, &dentry->d_lock); dentry->d_inode = NULL; dentry->d_parent = dentry; + atomic_set(&dentry->d_neg_dnum, 0); dentry->d_sb = sb; dentry->d_op = NULL; dentry->d_fsdata = NULL; diff --git a/include/linux/dcache.h b/include/linux/dcache.h index 6b351e009f59..1c5b08d804ce 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -77,6 +77,7 @@ extern const struct qstr dotdot_name; # endif #endif +#define NEG_DENTRY_LIMIT 16384 #define d_lock d_lockref.lock struct dentry { @@ -111,6 +112,8 @@ struct dentry { struct hlist_bl_node d_in_lookup_hash; /* only for in-lookup ones */ struct rcu_head d_rcu; } d_u; + /* negative dentry under this dentry, if it's dir */ + atomic_t d_neg_dnum; } __randomize_layout; /* @@ -212,6 +215,7 @@ struct dentry_operations { #define DCACHE_PAR_LOOKUP 0x10000000 /* being looked up (with parent locked shared) */ #define DCACHE_DENTRY_CURSOR 0x20000000 #define DCACHE_NORCU 0x40000000 /* No RCU delay for freeing */ +#define DCACHE_NEGATIVE_ACCOUNT 0x80000000 extern seqlock_t rename_lock; -- 2.39.2

2 1

[PATCH OLK-6.6] proc: fix ubsan warning in mem_lseek
by Zizhi Wo 30 Nov '23

30 Nov '23

From: yangerkun <yangerkun(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8JE4E?from=project-issue --------------------------- UBSAN has reported a overflow with mem_lseek. And it's fine with mem_open set file mode with FMODE_UNSIGNED_OFFSET(memory_lseek). However, another file use mem_lseek do lseek can have not FMODE_UNSIGNED_OFFSET(proc_kpagecount_operations/proc_pagemap_operations), fix it by checking overflow and FMODE_UNSIGNED_OFFSET. Reviewed-by: zhangyi (F) <yi.zhang(a)huawei.com> ================================================================== UBSAN: Undefined behaviour in ../fs/proc/base.c:941:15 signed integer overflow: 4611686018427387904 + 4611686018427387904 cannot be represented in type 'long long int' CPU: 4 PID: 4762 Comm: syz-executor.1 Not tainted 4.4.189 #3 Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015 Call trace: [<ffffff90080a5f28>] dump_backtrace+0x0/0x590 arch/arm64/kernel/traps.c:91 [<ffffff90080a64f0>] show_stack+0x38/0x60 arch/arm64/kernel/traps.c:234 [<ffffff9008986a34>] __dump_stack lib/dump_stack.c:15 [inline] [<ffffff9008986a34>] dump_stack+0x128/0x184 lib/dump_stack.c:51 [<ffffff9008a2d120>] ubsan_epilogue+0x34/0x9c lib/ubsan.c:166 [<ffffff9008a2d8b8>] handle_overflow+0x228/0x280 lib/ubsan.c:197 [<ffffff9008a2da2c>] __ubsan_handle_add_overflow+0x4c/0x68 lib/ubsan.c:204 [<ffffff900862b9f4>] mem_lseek+0x12c/0x130 fs/proc/base.c:941 [<ffffff90084ef78c>] vfs_llseek fs/read_write.c:260 [inline] [<ffffff90084ef78c>] SYSC_lseek fs/read_write.c:285 [inline] [<ffffff90084ef78c>] SyS_lseek+0x164/0x1f0 fs/read_write.c:276 [<ffffff9008093c80>] el0_svc_naked+0x30/0x34 ================================================================== Signed-off-by: yangerkun <yangerkun(a)huawei.com> Signed-off-by: Zizhi Wo <wozizhi(a)huawei.com> --- fs/proc/base.c | 32 ++++++++++++++++++++++++-------- fs/read_write.c | 5 ----- include/linux/fs.h | 4 ++++ 3 files changed, 28 insertions(+), 13 deletions(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index ffd54617c354..00642e7fe99f 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -903,18 +903,34 @@ static ssize_t mem_write(struct file *file, const char __user *buf, loff_t mem_lseek(struct file *file, loff_t offset, int orig) { + loff_t ret = 0; + + spin_lock(&file->f_lock); switch (orig) { - case 0: - file->f_pos = offset; - break; - case 1: - file->f_pos += offset; + case SEEK_CUR: + offset += file->f_pos; + __attribute__((__fallthrough__)); + case SEEK_SET: + /* to avoid userland mistaking f_pos=-9 as -EBADF=-9 */ + if ((unsigned long long)offset >= -MAX_ERRNO) + ret = -EOVERFLOW; break; default: - return -EINVAL; + ret = -EINVAL; + } + + if (!ret) { + if (offset < 0 && !(unsigned_offsets(file))) { + ret = -EINVAL; + } else { + file->f_pos = offset; + ret = file->f_pos; + force_successful_syscall_return(); + } } - force_successful_syscall_return(); - return file->f_pos; + + spin_unlock(&file->f_lock); + return ret; } static int mem_release(struct inode *inode, struct file *file) diff --git a/fs/read_write.c b/fs/read_write.c index 4771701c896b..2f456d5a1df5 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -34,11 +34,6 @@ const struct file_operations generic_ro_fops = { EXPORT_SYMBOL(generic_ro_fops); -static inline bool unsigned_offsets(struct file *file) -{ - return file->f_mode & FMODE_UNSIGNED_OFFSET; -} - /** * vfs_setpos - update the file offset for lseek * @file: file structure in question diff --git a/include/linux/fs.h b/include/linux/fs.h index 4f6b15e960ec..59d4edee1b96 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2932,6 +2932,10 @@ extern long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, extern void file_ra_state_init(struct file_ra_state *ra, struct address_space *mapping); extern loff_t noop_llseek(struct file *file, loff_t offset, int whence); +static inline bool unsigned_offsets(struct file *file) +{ + return file->f_mode & FMODE_UNSIGNED_OFFSET; +} #define no_llseek NULL extern loff_t vfs_setpos(struct file *file, loff_t offset, loff_t maxsize); extern loff_t generic_file_llseek(struct file *file, loff_t offset, int whence); -- 2.39.2

2 1

[PATCH OLK-6.6] fs/dcache.c: avoid panic while lockref of dentry overflow
by Zizhi Wo 30 Nov '23

30 Nov '23

From: yangerkun <yangerkun(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8JDP6?from=project-issue --------------------------- We use lockref for dentry reference without notice that so many negative dentry under one dir can lead to overflow of lockref. This can lead to system crash if we do this under root dir. Since there is not a perfect solution, we just limit max number of dentry count up to INT_MAX / 2. Also, it will cost a lot of time from INT_MAX / 2 to INT_MAX, so we no need to do this under protection of dentry lock. Also, we limit the FILES_MAX to INT_MAX / 2, since a lot open for same file can lead to overflow too. Changelog: v1->v2: add a function to do check / add a Macro to mean INT_MAX / 2 Signed-off-by: yangerkun <yangerkun(a)huawei.com> Signed-off-by: Zizhi Wo <wozizhi(a)huawei.com> --- fs/dcache.c | 32 ++++++++++++++++++++++++++++---- include/linux/fs.h | 3 +++ 2 files changed, 31 insertions(+), 4 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index 25ac74d30bff..7cdd8bbedd1c 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1834,6 +1834,18 @@ static struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) return dentry; } +static inline bool d_forbid_overflow(struct dentry *dentry) +{ + if (unlikely(d_count(dentry) >= D_COUNT_MAX)) { + shrink_dcache_parent(dentry); + + if (d_count(dentry) >= D_COUNT_MAX) + return false; + } + + return true; +} + /** * d_alloc - allocate a dcache entry * @parent: parent of entry to allocate @@ -1845,9 +1857,15 @@ static struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) */ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name) { - struct dentry *dentry = __d_alloc(parent->d_sb, name); + struct dentry *dentry = NULL; + + if (unlikely(!d_forbid_overflow(parent))) + goto out; + + dentry = __d_alloc(parent->d_sb, name); if (!dentry) - return NULL; + goto out; + spin_lock(&parent->d_lock); /* * don't need child lock because it is not subject @@ -1857,7 +1875,7 @@ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name) dentry->d_parent = parent; list_add(&dentry->d_child, &parent->d_subdirs); spin_unlock(&parent->d_lock); - +out: return dentry; } EXPORT_SYMBOL(d_alloc); @@ -1870,11 +1888,17 @@ EXPORT_SYMBOL(d_alloc_anon); struct dentry *d_alloc_cursor(struct dentry * parent) { - struct dentry *dentry = d_alloc_anon(parent->d_sb); + struct dentry *dentry = NULL; + + if (unlikely(!d_forbid_overflow(parent))) + goto out; + + dentry = d_alloc_anon(parent->d_sb); if (dentry) { dentry->d_flags |= DCACHE_DENTRY_CURSOR; dentry->d_parent = dget(parent); } +out: return dentry; } diff --git a/include/linux/fs.h b/include/linux/fs.h index 4a40823c3c67..4f6b15e960ec 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -47,6 +47,9 @@ #include <asm/byteorder.h> #include <uapi/linux/fs.h> +#define D_COUNT_MAX (INT_MAX / 2) + + struct backing_dev_info; struct bdi_writeback; struct bio; -- 2.39.2

2 1

[PATCH OLK-5.10 v5 00/21] psi irq in cgroupv1 and psi fine grained
by Lu Jialin 29 Nov '23

29 Nov '23

Chen Wandun (3): mm: disable psi cgroup v1 by default mm: add config isolation for psi under cgroup v1 psi: dont alloc memory for psi by default Chengming Zhou (3): sched/psi: Fix periodic aggregation shut off sched/psi: Optimize task switch inside shared cgroups again sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ pressure Haifeng Xu (1): sched/psi: Bail out early from irq time accounting Hao Jia (1): sched/psi: Zero the memory of struct psi_group Johannes Weiner (1): sched/psi: Remove NR_ONCPU task accounting Lu Jialin (11): psi: support irq.pressure under cgroup v1 psi: enable CONFIG_PSI_CGROUP_V1 in openeuler_defconfig psi: update psi irqtime when the irq delta is nozero memcg: Modify memcg async reclaim psi: add struct psi_group_ext PSI: Introduce fine grained stall time collect for cgroup reclaim PSI: Introduce avgs and total calculation for cgroup reclaim PSI: Introduce pressure.stat in psi PSI: add more memory fine grained stall tracking in pressure.stat add cpu fine grained stall tracking in pressure.stat PSI: enable CONFIG_PSI_FINE_GRAINED in openeuler_defconfig Suren Baghdasaryan (1): psi: Fix "defined but not used" warnings when CONFIG_PROC_FS=n Documentation/admin-guide/cgroup-v2.rst | 6 + arch/arm64/configs/openeuler_defconfig | 2 + arch/x86/configs/openeuler_defconfig | 2 + block/blk-cgroup.c | 2 +- block/blk-core.c | 2 +- include/linux/cgroup-defs.h | 6 +- include/linux/cgroup.h | 2 +- include/linux/psi.h | 9 +- include/linux/psi_types.h | 101 +++- include/linux/sched.h | 4 + init/Kconfig | 20 + kernel/cgroup/cgroup.c | 64 ++- kernel/sched/core.c | 2 + kernel/sched/cpuacct.c | 6 +- kernel/sched/fair.c | 6 - kernel/sched/psi.c | 694 ++++++++++++++++++++---- kernel/sched/stats.h | 9 + mm/compaction.c | 2 +- mm/filemap.c | 4 +- mm/memcontrol.c | 118 ++-- mm/page_alloc.c | 6 + mm/page_io.c | 3 + mm/vmscan.c | 5 +- 23 files changed, 887 insertions(+), 188 deletions(-) -- 2.34.1

2 22

[PATCH OLK-5.10] fs/dirty_pages: introduce a new config option 'CONFIG_DIRTY_PAGES'
by Zizhi Wo 29 Nov '23

29 Nov '23

From: yu kuai <yukuai3(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8KE0Z?from=project-issue --------------------------- The new config option 'CONFIG_DIRTY_PAGES' can control whether to configure the feather or not. Signed-off-by: yu kuai <yukuai3(a)huawei.com> Signed-off-by: Zizhi Wo <wozizhi(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + fs/Kconfig | 13 +++++++++++++ fs/Makefile | 2 +- 4 files changed, 16 insertions(+), 1 deletion(-) diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index a5bb7a7076a7..7150ea6672ff 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -6392,6 +6392,7 @@ CONFIG_PROC_PAGE_MONITOR=y CONFIG_PROC_CHILDREN=y CONFIG_KERNFS=y CONFIG_SYSFS=y +CONFIG_DIRTY_PAGES=y CONFIG_TMPFS=y CONFIG_TMPFS_POSIX_ACL=y CONFIG_TMPFS_XATTR=y diff --git a/arch/x86/configs/openeuler_defconfig b/arch/x86/configs/openeuler_defconfig index 3c04e75eec52..1634fd97be38 100644 --- a/arch/x86/configs/openeuler_defconfig +++ b/arch/x86/configs/openeuler_defconfig @@ -7459,6 +7459,7 @@ CONFIG_PROC_PID_ARCH_STATUS=y CONFIG_PROC_CPU_RESCTRL=y CONFIG_KERNFS=y CONFIG_SYSFS=y +CONFIG_DIRTY_PAGES=y CONFIG_TMPFS=y CONFIG_TMPFS_POSIX_ACL=y CONFIG_TMPFS_XATTR=y diff --git a/fs/Kconfig b/fs/Kconfig index 385602ba0d99..d3fff13ca13d 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -156,6 +156,19 @@ source "fs/proc/Kconfig" source "fs/kernfs/Kconfig" source "fs/sysfs/Kconfig" +config DIRTY_PAGES + bool "Dumps the number of dirty pages of each file" + depends on PROC_FS + default y + help + This config supports the rendering of dirty page data to the user, + which may be useful to analyze the IO performance when using buffer + IO. + + It create 3 interfaces by using procfs. /proc/dirty/buffer_size for + buffer allocation and release; /proc/dirty/page_threshold to filter + result; /proc/dirty/dirty_list to get dirty pages. + config TMPFS bool "Tmpfs virtual memory file system support (former shm fs)" depends on SHMEM diff --git a/fs/Makefile b/fs/Makefile index 720fccc8bffa..2a756dda52e5 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -50,7 +50,7 @@ obj-$(CONFIG_FS_POSIX_ACL) += posix_acl.o obj-$(CONFIG_NFS_COMMON) += nfs_common/ obj-$(CONFIG_COREDUMP) += coredump.o obj-$(CONFIG_SYSCTL) += drop_caches.o -obj-$(CONFIG_SYSCTL) += dirty_pages.o +obj-$(CONFIG_DIRTY_PAGES) += dirty_pages.o obj-$(CONFIG_FHANDLE) += fhandle.o obj-$(CONFIG_CGROUP_FILES) += filescontrol.o -- 2.39.2

2 1

[PATCH OLK-5.10] fs/dirty_pages: add last read check in seq_read_dirty()
by Zizhi Wo 29 Nov '23

29 Nov '23

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8JUUC?from=project-issue -------------------------------- When we call seq_read_dirty() for the last read, m->count is the same value as m->from, and we need to return 0 instead of an error. Fixes: 5d93db93b9e7 ("fs/dirty_pages: fix some errors in seq_read_dirty()") Signed-off-by: Zizhi Wo <wozizhi(a)huawei.com> --- fs/dirty_pages.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/dirty_pages.c b/fs/dirty_pages.c index b0cf442c5730..f75a5158d94b 100644 --- a/fs/dirty_pages.c +++ b/fs/dirty_pages.c @@ -218,6 +218,9 @@ static ssize_t seq_read_dirty( } n = min(m->count - m->from, size); + /* check if this is the last read */ + if (n == 0) + goto done; n -= copy_to_user(buf, m->buf + m->from, n); if (unlikely(!n)) { err = -EFAULT; -- 2.39.2

2 1

[PATCH OLK-6.6 0/5] LoongArch: add old BPI compatibility
by Hongchen Zhang 29 Nov '23

29 Nov '23

add old BPI compatibility for LoongArch machines. Hongchen Zhang (1): LoongArch: use arch specific phys_to_dma liuyun (3): LoongArch: add kernel setvirtmap for runtime LoongArch: Old BPI compatibility LoongArch: Fix virtual machine startup error yangqiming (1): LoongArch: Fixed EIOINTC structure members arch/loongarch/Kconfig | 1 + arch/loongarch/include/asm/addrspace.h | 1 + arch/loongarch/include/asm/efi.h | 1 + arch/loongarch/include/asm/irq.h | 1 + arch/loongarch/include/asm/loongarch.h | 1 + arch/loongarch/kernel/Makefile | 1 + arch/loongarch/kernel/acpi.c | 7 +- arch/loongarch/kernel/dma.c | 27 ++ arch/loongarch/kernel/efi.c | 175 ++++++++- arch/loongarch/kernel/env.c | 6 + arch/loongarch/kernel/irq.c | 25 +- arch/loongarch/kernel/legacy_boot.c | 488 +++++++++++++++++++++++++ arch/loongarch/kernel/legacy_boot.h | 90 +++++ arch/loongarch/kernel/mem.c | 26 +- arch/loongarch/kernel/numa.c | 39 +- arch/loongarch/kernel/reset.c | 3 +- arch/loongarch/kernel/setup.c | 18 +- arch/loongarch/kernel/smp.c | 6 +- arch/loongarch/pci/acpi.c | 148 +++++++- drivers/firmware/efi/Makefile | 1 + drivers/irqchip/irq-loongarch-cpu.c | 7 +- drivers/irqchip/irq-loongson-eiointc.c | 44 ++- drivers/irqchip/irq-loongson-pch-pic.c | 5 + 23 files changed, 1081 insertions(+), 40 deletions(-) create mode 100644 arch/loongarch/kernel/legacy_boot.c create mode 100644 arch/loongarch/kernel/legacy_boot.h -- 2.33.0

1 5

[PATCH OLK-5.10 00/11] KVM/arm64: Add host/guest KVM-PTP support
by Dongxu Sun 29 Nov '23

29 Nov '23

Jianyong Wu (4): ptp: Reorganize ptp_kvm.c to make it arch-independent clocksource: Add clocksource id for arm arch counter KVM: arm64: Add support for the KVM PTP service ptp: arm/arm64: Enable ptp_kvm for arm/arm64 Jon Hunter (1): ptp: Don't print an error if ptp_kvm is not supported Kele Huang (1): ptp: fix error print of ptp_kvm on X86_64 platform Thomas Gleixner (1): time: Add mechanism to recognize clocksource in time_get_snapshot Will Deacon (2): arm/arm64: Probe for the presence of KVM hypervisor KVM: arm64: Advertise KVM UID to guests via SMCCC Yanan Wang (1): KVM: arm64: fix compile error/warning Zenghui Yu (1): KVM: arm64: Fix Function ID typo for PTP_KVM service Documentation/virt/kvm/api.rst | 9 ++ Documentation/virt/kvm/arm/index.rst | 1 + Documentation/virt/kvm/arm/ptp_kvm.rst | 25 ++++++ arch/arm/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/kvm_para.h | 2 +- arch/arm64/include/uapi/asm/Kbuild | 2 +- arch/arm64/include/uapi/asm/kvm_para.h | 7 ++ arch/arm64/kvm/arm.c | 1 + arch/arm64/kvm/hypercalls.c | 94 ++++++++++++++++---- drivers/clocksource/arm_arch_timer.c | 36 ++++++++ drivers/firmware/psci/psci.c | 2 + drivers/firmware/smccc/Makefile | 2 +- drivers/firmware/smccc/kvm_guest.c | 50 +++++++++++ drivers/firmware/smccc/smccc.c | 1 + drivers/ptp/Kconfig | 2 +- drivers/ptp/Makefile | 2 + drivers/ptp/ptp_kvm_arm.c | 28 ++++++ drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} | 85 +++++------------- drivers/ptp/ptp_kvm_x86.c | 97 +++++++++++++++++++++ include/linux/arm-smccc.h | 41 +++++++++ include/linux/clocksource.h | 6 ++ include/linux/clocksource_ids.h | 12 +++ include/linux/ptp_kvm.h | 19 ++++ include/linux/timekeeping.h | 12 +-- include/uapi/linux/kvm.h | 1 + kernel/time/clocksource.c | 2 + kernel/time/timekeeping.c | 1 + 28 files changed, 459 insertions(+), 87 deletions(-) create mode 100644 Documentation/virt/kvm/arm/ptp_kvm.rst create mode 100644 arch/arm64/include/uapi/asm/kvm_para.h create mode 100644 drivers/firmware/smccc/kvm_guest.c create mode 100644 drivers/ptp/ptp_kvm_arm.c rename drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} (60%) create mode 100644 drivers/ptp/ptp_kvm_x86.c create mode 100644 include/linux/clocksource_ids.h create mode 100644 include/linux/ptp_kvm.h -- 2.33.0

1 11

[PATCH OLK-5.10 00/11] KVM/arm64: Add host/guest KVM-PTP support
by Dongxu Sun 29 Nov '23

29 Nov '23

Jianyong Wu (4): ptp: Reorganize ptp_kvm.c to make it arch-independent clocksource: Add clocksource id for arm arch counter KVM: arm64: Add support for the KVM PTP service ptp: arm/arm64: Enable ptp_kvm for arm/arm64 Jon Hunter (1): ptp: Don't print an error if ptp_kvm is not supported Kele Huang (1): ptp: fix error print of ptp_kvm on X86_64 platform Thomas Gleixner (1): time: Add mechanism to recognize clocksource in time_get_snapshot Will Deacon (2): arm/arm64: Probe for the presence of KVM hypervisor KVM: arm64: Advertise KVM UID to guests via SMCCC Yanan Wang (1): KVM: arm64: fix compile error/warning Zenghui Yu (1): KVM: arm64: Fix Function ID typo for PTP_KVM service Documentation/virt/kvm/api.rst | 9 ++ Documentation/virt/kvm/arm/index.rst | 1 + Documentation/virt/kvm/arm/ptp_kvm.rst | 25 ++++++ arch/arm/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/kvm_para.h | 2 +- arch/arm64/include/uapi/asm/Kbuild | 2 +- arch/arm64/include/uapi/asm/kvm_para.h | 7 ++ arch/arm64/kvm/arm.c | 1 + arch/arm64/kvm/hypercalls.c | 94 ++++++++++++++++---- drivers/clocksource/arm_arch_timer.c | 36 ++++++++ drivers/firmware/psci/psci.c | 2 + drivers/firmware/smccc/Makefile | 2 +- drivers/firmware/smccc/kvm_guest.c | 50 +++++++++++ drivers/firmware/smccc/smccc.c | 1 + drivers/ptp/Kconfig | 2 +- drivers/ptp/Makefile | 2 + drivers/ptp/ptp_kvm_arm.c | 28 ++++++ drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} | 85 +++++------------- drivers/ptp/ptp_kvm_x86.c | 97 +++++++++++++++++++++ include/linux/arm-smccc.h | 41 +++++++++ include/linux/clocksource.h | 6 ++ include/linux/clocksource_ids.h | 12 +++ include/linux/ptp_kvm.h | 19 ++++ include/linux/timekeeping.h | 12 +-- include/uapi/linux/kvm.h | 1 + kernel/time/clocksource.c | 2 + kernel/time/timekeeping.c | 1 + 28 files changed, 459 insertions(+), 87 deletions(-) create mode 100644 Documentation/virt/kvm/arm/ptp_kvm.rst create mode 100644 arch/arm64/include/uapi/asm/kvm_para.h create mode 100644 drivers/firmware/smccc/kvm_guest.c create mode 100644 drivers/ptp/ptp_kvm_arm.c rename drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} (60%) create mode 100644 drivers/ptp/ptp_kvm_x86.c create mode 100644 include/linux/clocksource_ids.h create mode 100644 include/linux/ptp_kvm.h -- 2.33.0

2 13

[PATCH OLK-6.6 0/5] LoongArch: add old BPI compatibility
by Hongchen Zhang 29 Nov '23

29 Nov '23

add old BPI compatibility for LoongArch machines. Hongchen Zhang (1): LoongArch: use arch specific phys_to_dma liuyun (3): LoongArch: add kernel setvirtmap for runtime LoongArch: Old BPI compatibility LoongArch: Fix virtual machine startup error yangqiming (1): LoongArch: Fixed EIOINTC structure members arch/loongarch/Kconfig | 1 + arch/loongarch/include/asm/addrspace.h | 1 + arch/loongarch/include/asm/efi.h | 1 + arch/loongarch/include/asm/irq.h | 1 + arch/loongarch/include/asm/loongarch.h | 1 + arch/loongarch/kernel/Makefile | 1 + arch/loongarch/kernel/acpi.c | 7 +- arch/loongarch/kernel/dma.c | 27 ++ arch/loongarch/kernel/efi.c | 175 ++++++++- arch/loongarch/kernel/env.c | 6 + arch/loongarch/kernel/irq.c | 25 +- arch/loongarch/kernel/legacy_boot.c | 488 +++++++++++++++++++++++++ arch/loongarch/kernel/legacy_boot.h | 90 +++++ arch/loongarch/kernel/mem.c | 26 +- arch/loongarch/kernel/numa.c | 39 +- arch/loongarch/kernel/reset.c | 3 +- arch/loongarch/kernel/setup.c | 18 +- arch/loongarch/kernel/smp.c | 6 +- arch/loongarch/pci/acpi.c | 148 +++++++- drivers/firmware/efi/Makefile | 1 + drivers/irqchip/irq-loongarch-cpu.c | 7 +- drivers/irqchip/irq-loongson-eiointc.c | 44 ++- drivers/irqchip/irq-loongson-pch-pic.c | 5 + 23 files changed, 1081 insertions(+), 40 deletions(-) create mode 100644 arch/loongarch/kernel/legacy_boot.c create mode 100644 arch/loongarch/kernel/legacy_boot.h -- 2.33.0

1 5

[PATCH OLK-5.10 00/11] KVM: arm64: Add host/guest KVM-PTP support
by Dongxu Sun 29 Nov '23

29 Nov '23

Jianyong Wu (4): ptp: Reorganize ptp_kvm.c to make it arch-independent clocksource: Add clocksource id for arm arch counter KVM: arm64: Add support for the KVM PTP service ptp: arm/arm64: Enable ptp_kvm for arm/arm64 Jon Hunter (1): ptp: Don't print an error if ptp_kvm is not supported Kele Huang (1): ptp: fix error print of ptp_kvm on X86_64 platform Thomas Gleixner (1): time: Add mechanism to recognize clocksource in time_get_snapshot Will Deacon (2): arm/arm64: Probe for the presence of KVM hypervisor KVM: arm64: Advertise KVM UID to guests via SMCCC Yanan Wang (1): KVM: arm64: fix compile error/warning Zenghui Yu (1): KVM: arm64: Fix Function ID typo for PTP_KVM service Documentation/virt/kvm/api.rst | 9 ++ Documentation/virt/kvm/arm/index.rst | 1 + Documentation/virt/kvm/arm/ptp_kvm.rst | 25 ++++++ arch/arm/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/kvm_para.h | 2 +- arch/arm64/include/uapi/asm/Kbuild | 2 +- arch/arm64/include/uapi/asm/kvm_para.h | 7 ++ arch/arm64/kvm/arm.c | 1 + arch/arm64/kvm/hypercalls.c | 94 ++++++++++++++++---- drivers/clocksource/arm_arch_timer.c | 36 ++++++++ drivers/firmware/psci/psci.c | 2 + drivers/firmware/smccc/Makefile | 2 +- drivers/firmware/smccc/kvm_guest.c | 50 +++++++++++ drivers/firmware/smccc/smccc.c | 1 + drivers/ptp/Kconfig | 2 +- drivers/ptp/Makefile | 2 + drivers/ptp/ptp_kvm_arm.c | 28 ++++++ drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} | 85 +++++------------- drivers/ptp/ptp_kvm_x86.c | 97 +++++++++++++++++++++ include/linux/arm-smccc.h | 41 +++++++++ include/linux/clocksource.h | 6 ++ include/linux/clocksource_ids.h | 12 +++ include/linux/ptp_kvm.h | 19 ++++ include/linux/timekeeping.h | 12 +-- include/uapi/linux/kvm.h | 1 + kernel/time/clocksource.c | 2 + kernel/time/timekeeping.c | 1 + 28 files changed, 459 insertions(+), 87 deletions(-) create mode 100644 Documentation/virt/kvm/arm/ptp_kvm.rst create mode 100644 arch/arm64/include/uapi/asm/kvm_para.h create mode 100644 drivers/firmware/smccc/kvm_guest.c create mode 100644 drivers/ptp/ptp_kvm_arm.c rename drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} (60%) create mode 100644 drivers/ptp/ptp_kvm_x86.c create mode 100644 include/linux/clocksource_ids.h create mode 100644 include/linux/ptp_kvm.h -- 2.33.0

1 11

[PATCH OLK-5.10] printk: ringbuffer: Fix truncating buffer size min_t cast
by Ye Weihua 29 Nov '23

29 Nov '23

From: Kees Cook <keescook(a)chromium.org> stable inclusion from stable-v5.10.192 commit 8c90c4e61929eb6af0b227497556a09e0691d2eb category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8KG4V Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 53e9e33ede37a247d926db5e4a9e56b55204e66c upstream. If an output buffer size exceeded U16_MAX, the min_t(u16, ...) cast in copy_data() was causing writes to truncate. This manifested as output bytes being skipped, seen as %NUL bytes in pstore dumps when the available record size was larger than 65536. Fix the cast to no longer truncate the calculation. Cc: Petr Mladek <pmladek(a)suse.com> Cc: Sergey Senozhatsky <senozhatsky(a)chromium.org> Cc: Steven Rostedt <rostedt(a)goodmis.org> Cc: John Ogness <john.ogness(a)linutronix.de> Reported-by: Vijay Balakrishna <vijayb(a)linux.microsoft.com> Link: https://lore.kernel.org/lkml/d8bb1ec7-a4c5-43a2-9de0-9643a70b899f@linux.mic… Fixes: b6cf8b3f3312 ("printk: add lockless ringbuffer") Cc: stable(a)vger.kernel.org Signed-off-by: Kees Cook <keescook(a)chromium.org> Tested-by: Vijay Balakrishna <vijayb(a)linux.microsoft.com> Tested-by: Guilherme G. Piccoli <gpiccoli(a)igalia.com> # Steam Deck Reviewed-by: Tyler Hicks (Microsoft) <code(a)tyhicks.com> Tested-by: Tyler Hicks (Microsoft) <code(a)tyhicks.com> Reviewed-by: John Ogness <john.ogness(a)linutronix.de> Reviewed-by: Sergey Senozhatsky <senozhatsky(a)chromium.org> Reviewed-by: Petr Mladek <pmladek(a)suse.com> Signed-off-by: Petr Mladek <pmladek(a)suse.com> Link: https://lore.kernel.org/r/20230811054528.never.165-kees@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Ye Weihua <yeweihua4(a)huawei.com> --- kernel/printk/printk_ringbuffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/printk/printk_ringbuffer.c b/kernel/printk/printk_ringbuffer.c index 617dd6358965..3e8214e9bf4a 100644 --- a/kernel/printk/printk_ringbuffer.c +++ b/kernel/printk/printk_ringbuffer.c @@ -1726,7 +1726,7 @@ static bool copy_data(struct prb_data_ring *data_ring, if (!buf || !buf_size) return true; - data_size = min_t(u16, buf_size, len); + data_size = min_t(unsigned int, buf_size, len); memcpy(&buf[0], data, data_size); /* LMM(copy_data:A) */ return true; -- 2.34.1

2 1

[PATCH OLK-5.10 00/11] KVM: arm64: Add host/guest KVM-PTP support
by Dongxu Sun 29 Nov '23

29 Nov '23

Jianyong Wu (4): ptp: Reorganize ptp_kvm.c to make it arch-independent clocksource: Add clocksource id for arm arch counter KVM: arm64: Add support for the KVM PTP service ptp: arm/arm64: Enable ptp_kvm for arm/arm64 Jon Hunter (1): ptp: Don't print an error if ptp_kvm is not supported Kele Huang (1): ptp: fix error print of ptp_kvm on X86_64 platform Thomas Gleixner (1): time: Add mechanism to recognize clocksource in time_get_snapshot Will Deacon (2): arm/arm64: Probe for the presence of KVM hypervisor KVM: arm64: Advertise KVM UID to guests via SMCCC Yanan Wang (1): KVM: arm64: fix compile error/warning Zenghui Yu (1): KVM: arm64: Fix Function ID typo for PTP_KVM service Documentation/virt/kvm/api.rst | 9 ++ Documentation/virt/kvm/arm/index.rst | 1 + Documentation/virt/kvm/arm/ptp_kvm.rst | 25 ++++++ arch/arm/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/kvm_para.h | 2 +- arch/arm64/include/uapi/asm/Kbuild | 2 +- arch/arm64/include/uapi/asm/kvm_para.h | 7 ++ arch/arm64/kvm/arm.c | 1 + arch/arm64/kvm/hypercalls.c | 94 ++++++++++++++++---- drivers/clocksource/arm_arch_timer.c | 36 ++++++++ drivers/firmware/psci/psci.c | 2 + drivers/firmware/smccc/Makefile | 2 +- drivers/firmware/smccc/kvm_guest.c | 50 +++++++++++ drivers/firmware/smccc/smccc.c | 1 + drivers/ptp/Kconfig | 2 +- drivers/ptp/Makefile | 2 + drivers/ptp/ptp_kvm_arm.c | 28 ++++++ drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} | 85 +++++------------- drivers/ptp/ptp_kvm_x86.c | 97 +++++++++++++++++++++ include/linux/arm-smccc.h | 41 +++++++++ include/linux/clocksource.h | 6 ++ include/linux/clocksource_ids.h | 12 +++ include/linux/ptp_kvm.h | 19 ++++ include/linux/timekeeping.h | 12 +-- include/uapi/linux/kvm.h | 1 + kernel/time/clocksource.c | 2 + kernel/time/timekeeping.c | 1 + 28 files changed, 459 insertions(+), 87 deletions(-) create mode 100644 Documentation/virt/kvm/arm/ptp_kvm.rst create mode 100644 arch/arm64/include/uapi/asm/kvm_para.h create mode 100644 drivers/firmware/smccc/kvm_guest.c create mode 100644 drivers/ptp/ptp_kvm_arm.c rename drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} (60%) create mode 100644 drivers/ptp/ptp_kvm_x86.c create mode 100644 include/linux/clocksource_ids.h create mode 100644 include/linux/ptp_kvm.h -- 2.33.0

1 11

[PATCH openEuler-1.0-LTS v2] perf/core: Fix perf_mmap fail when CONFIG_PERF_USE_VMALLOC enabled
by Yang Jihong 29 Nov '23

29 Nov '23

From: Zhipeng Xie <xiezhipeng1(a)huawei.com> mainline inclusion from mainline-v5.18-rc4 commit 60490e7966659b26d74bf1fa4aa8693d9a94ca88 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8KFB3 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- This problem can be reproduced with CONFIG_PERF_USE_VMALLOC enabled on both x86_64 and aarch64 arch when using sysdig -B(using ebpf)[1]. sysdig -B works fine after rebuilding the kernel with CONFIG_PERF_USE_VMALLOC disabled. I tracked it down to the if condition event->rb->nr_pages != nr_pages in perf_mmap is true when CONFIG_PERF_USE_VMALLOC is enabled where event->rb->nr_pages = 1 and nr_pages = 2048 resulting perf_mmap to return -EINVAL. This is because when CONFIG_PERF_USE_VMALLOC is enabled, rb->nr_pages is always equal to 1. Arch with CONFIG_PERF_USE_VMALLOC enabled by default: arc/arm/csky/mips/sh/sparc/xtensa Arch with CONFIG_PERF_USE_VMALLOC disabled by default: x86_64/aarch64/... Fix this problem by using data_page_nr() [1] https://github.com/draios/sysdig Fixes: 906010b2134e ("perf_event: Provide vmalloc() based mmap() backing") Signed-off-by: Zhipeng Xie <xiezhipeng1(a)huawei.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Link: https://lkml.kernel.org/r/20220209145417.6495-1-xiezhipeng1@huawei.com conflict: kernel/events/core.c kernel/events/internal.h kernel/events/ring_buffer.c Signed-off-by: Yang Jihong <yangjihong1(a)huawei.com> --- kernel/events/core.c | 2 +- kernel/events/internal.h | 5 +++++ kernel/events/ring_buffer.c | 4 ---- 3 files changed, 6 insertions(+), 5 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index a49860879777..a1f12009e5ef 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5760,7 +5760,7 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) again: mutex_lock(&event->mmap_mutex); if (event->rb) { - if (event->rb->nr_pages != nr_pages) { + if (data_page_nr(event->rb) != nr_pages) { ret = -EINVAL; goto unlock; } diff --git a/kernel/events/internal.h b/kernel/events/internal.h index 8fc0ddc38cb6..578de8c2694b 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -114,6 +114,11 @@ static inline int page_order(struct ring_buffer *rb) } #endif +static inline int data_page_nr(struct ring_buffer *rb) +{ + return rb->nr_pages << page_order(rb); +} + static inline unsigned long perf_data_size(struct ring_buffer *rb) { return rb->nr_pages << (PAGE_SHIFT + page_order(rb)); diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index 34e1667e4254..a69378d901fa 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -803,10 +803,6 @@ void rb_free(struct ring_buffer *rb) } #else -static int data_page_nr(struct ring_buffer *rb) -{ - return rb->nr_pages << page_order(rb); -} static struct page * __perf_mmap_to_page(struct ring_buffer *rb, unsigned long pgoff) -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] perf/core: Fix perf_mmap fail when CONFIG_PERF_USE_VMALLOC enabled
by Yang Jihong 29 Nov '23

29 Nov '23

From: Zhipeng Xie <xiezhipeng1(a)huawei.com> mainline inclusion from mainline-v5.18-rc4 commit 60490e7966659b26d74bf1fa4aa category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8KFB3 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- This problem can be reproduced with CONFIG_PERF_USE_VMALLOC enabled on both x86_64 and aarch64 arch when using sysdig -B(using ebpf)[1]. sysdig -B works fine after rebuilding the kernel with CONFIG_PERF_USE_VMALLOC disabled. I tracked it down to the if condition event->rb->nr_pages != nr_pages in perf_mmap is true when CONFIG_PERF_USE_VMALLOC is enabled where event->rb->nr_pages = 1 and nr_pages = 2048 resulting perf_mmap to return -EINVAL. This is because when CONFIG_PERF_USE_VMALLOC is enabled, rb->nr_pages is always equal to 1. Arch with CONFIG_PERF_USE_VMALLOC enabled by default: arc/arm/csky/mips/sh/sparc/xtensa Arch with CONFIG_PERF_USE_VMALLOC disabled by default: x86_64/aarch64/... Fix this problem by using data_page_nr() [1] https://github.com/draios/sysdig Fixes: 906010b2134e ("perf_event: Provide vmalloc() based mmap() backing") Signed-off-by: Zhipeng Xie <xiezhipeng1(a)huawei.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Link: https://lkml.kernel.org/r/20220209145417.6495-1-xiezhipeng1@huawei.com conflict: kernel/events/core.c kernel/events/internal.h kernel/events/ring_buffer.c Signed-off-by: Yang Jihong <yangjihong1(a)huawei.com> --- kernel/events/core.c | 2 +- kernel/events/internal.h | 5 +++++ kernel/events/ring_buffer.c | 4 ---- 3 files changed, 6 insertions(+), 5 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index a49860879777..a1f12009e5ef 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5760,7 +5760,7 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) again: mutex_lock(&event->mmap_mutex); if (event->rb) { - if (event->rb->nr_pages != nr_pages) { + if (data_page_nr(event->rb) != nr_pages) { ret = -EINVAL; goto unlock; } diff --git a/kernel/events/internal.h b/kernel/events/internal.h index 8fc0ddc38cb6..578de8c2694b 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -114,6 +114,11 @@ static inline int page_order(struct ring_buffer *rb) } #endif +static inline int data_page_nr(struct ring_buffer *rb) +{ + return rb->nr_pages << page_order(rb); +} + static inline unsigned long perf_data_size(struct ring_buffer *rb) { return rb->nr_pages << (PAGE_SHIFT + page_order(rb)); diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index 34e1667e4254..a69378d901fa 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -803,10 +803,6 @@ void rb_free(struct ring_buffer *rb) } #else -static int data_page_nr(struct ring_buffer *rb) -{ - return rb->nr_pages << page_order(rb); -} static struct page * __perf_mmap_to_page(struct ring_buffer *rb, unsigned long pgoff) -- 2.34.1

2 1

[PATCH v2 OLK-5.10] drivers: perf: Not enabled ARM64_BRBE by default
by Junhao He 29 Nov '23

29 Nov '23

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8EC9K CVE: NA -------------------------------- The BRBE feature depend on ARMV8 PMU hardware support for branch stack sampling. Not enabled ARM64_BRBE by default. Signed-off-by: Junhao He <hejunhao3(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 2 +- drivers/perf/Kconfig | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 2c6aa6404190..f9371ebf21e7 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -6169,7 +6169,7 @@ CONFIG_QCOM_L3_PMU=y CONFIG_THUNDERX2_PMU=m CONFIG_XGENE_PMU=y CONFIG_ARM_SPE_PMU=y -CONFIG_ARM64_BRBE=y +# CONFIG_ARM64_BRBE is not set CONFIG_HISI_PMU=m CONFIG_HISI_PCIE_PMU=m CONFIG_HNS3_PMU=m diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig index 25a577530da4..67ad53cde11f 100644 --- a/drivers/perf/Kconfig +++ b/drivers/perf/Kconfig @@ -133,7 +133,6 @@ config ARM_SPE_PMU config ARM64_BRBE bool "Enable support for Branch Record Buffer Extension (BRBE)" depends on PERF_EVENTS && ARM64 && ARM_PMU - default y help Enable perf support for Branch Record Buffer Extension (BRBE) which records all branches taken in an execution path. This supports some -- 2.33.0

2 1

[PACH OLK-5.10 00/11] KVM: arm64: Add host/guest KVM-PTP support
by Dongxu Sun 29 Nov '23

29 Nov '23

Jianyong Wu (4): ptp: Reorganize ptp_kvm.c to make it arch-independent clocksource: Add clocksource id for arm arch counter KVM: arm64: Add support for the KVM PTP service ptp: arm/arm64: Enable ptp_kvm for arm/arm64 Jon Hunter (1): ptp: Don't print an error if ptp_kvm is not supported Kele Huang (1): ptp: fix error print of ptp_kvm on X86_64 platform Thomas Gleixner (1): time: Add mechanism to recognize clocksource in time_get_snapshot Will Deacon (2): arm/arm64: Probe for the presence of KVM hypervisor KVM: arm64: Advertise KVM UID to guests via SMCCC Yanan Wang (1): KVM: arm64: fix compile error/warning Zenghui Yu (1): KVM: arm64: Fix Function ID typo for PTP_KVM service Documentation/virt/kvm/api.rst | 9 ++ Documentation/virt/kvm/arm/index.rst | 1 + Documentation/virt/kvm/arm/ptp_kvm.rst | 25 ++++++ arch/arm/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/kvm_para.h | 2 +- arch/arm64/include/uapi/asm/Kbuild | 2 +- arch/arm64/include/uapi/asm/kvm_para.h | 7 ++ arch/arm64/kvm/arm.c | 1 + arch/arm64/kvm/hypercalls.c | 94 ++++++++++++++++---- drivers/clocksource/arm_arch_timer.c | 36 ++++++++ drivers/firmware/psci/psci.c | 2 + drivers/firmware/smccc/Makefile | 2 +- drivers/firmware/smccc/kvm_guest.c | 50 +++++++++++ drivers/firmware/smccc/smccc.c | 1 + drivers/ptp/Kconfig | 2 +- drivers/ptp/Makefile | 2 + drivers/ptp/ptp_kvm_arm.c | 28 ++++++ drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} | 85 +++++------------- drivers/ptp/ptp_kvm_x86.c | 97 +++++++++++++++++++++ include/linux/arm-smccc.h | 41 +++++++++ include/linux/clocksource.h | 6 ++ include/linux/clocksource_ids.h | 12 +++ include/linux/ptp_kvm.h | 19 ++++ include/linux/timekeeping.h | 12 +-- include/uapi/linux/kvm.h | 1 + kernel/time/clocksource.c | 2 + kernel/time/timekeeping.c | 1 + 28 files changed, 459 insertions(+), 87 deletions(-) create mode 100644 Documentation/virt/kvm/arm/ptp_kvm.rst create mode 100644 arch/arm64/include/uapi/asm/kvm_para.h create mode 100644 drivers/firmware/smccc/kvm_guest.c create mode 100644 drivers/ptp/ptp_kvm_arm.c rename drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} (60%) create mode 100644 drivers/ptp/ptp_kvm_x86.c create mode 100644 include/linux/clocksource_ids.h create mode 100644 include/linux/ptp_kvm.h -- 2.33.0

1 11

[PATCH V3 OLK-5.10] vhost-vdpa: add reset state params to indicate reset level
by Jiang Dongxu 29 Nov '23

29 Nov '23

From: jiangdongxu <jiangdongxu1(a)huawei.com> virt inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I86ITO ---------------------------------------------------------------------- Add the interface parameter state to the vdpa reset interface, which respectively identifies the reset when the device is turned on/off and the virtio reset issued by the virtual machine. Signed-off-by: jiangdongxu <jiangdongxu1(a)huawei.com> --- drivers/vdpa/ifcvf/ifcvf_main.c | 2 +- drivers/vdpa/mlx5/net/mlx5_vnet.c | 2 +- drivers/vdpa/vdpa_sim/vdpa_sim.c | 2 +- drivers/vdpa/virtio_pci/vp_vdpa.c | 2 +- drivers/vhost/vdpa.c | 10 +++++----- drivers/virtio/virtio_vdpa.c | 2 +- include/linux/vdpa.h | 16 +++++++++++++--- 7 files changed, 23 insertions(+), 13 deletions(-) diff --git a/drivers/vdpa/ifcvf/ifcvf_main.c b/drivers/vdpa/ifcvf/ifcvf_main.c index 862f4591fff2..2a4991464cfd 100644 --- a/drivers/vdpa/ifcvf/ifcvf_main.c +++ b/drivers/vdpa/ifcvf/ifcvf_main.c @@ -233,7 +233,7 @@ static void ifcvf_vdpa_set_status(struct vdpa_device *vdpa_dev, u8 status) ifcvf_set_status(vf, status); } -static int ifcvf_vdpa_reset(struct vdpa_device *vdpa_dev) +static int ifcvf_vdpa_reset(struct vdpa_device *vdpa_dev, int state) { struct ifcvf_adapter *adapter; struct ifcvf_hw *vf; diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c index 76e7b4bab9d3..c676bb70a453 100644 --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c @@ -1800,7 +1800,7 @@ static void mlx5_vdpa_set_status(struct vdpa_device *vdev, u8 status) ndev->mvdev.status |= VIRTIO_CONFIG_S_FAILED; } -static int mlx5_vdpa_reset(struct vdpa_device *vdev) +static int mlx5_vdpa_reset(struct vdpa_device *vdev, int state) { struct mlx5_vdpa_dev *mvdev = to_mvdev(vdev); struct mlx5_vdpa_net *ndev = to_mlx5_vdpa_ndev(mvdev); diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c index c141f08b8274..be925462922d 100644 --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c @@ -454,7 +454,7 @@ static void vdpasim_set_status(struct vdpa_device *vdpa, u8 status) spin_unlock(&vdpasim->lock); } -static int vdpasim_reset(struct vdpa_device *vdpa) +static int vdpasim_reset(struct vdpa_device *vdpa, int state) { struct vdpasim *vdpasim = vdpa_to_sim(vdpa); diff --git a/drivers/vdpa/virtio_pci/vp_vdpa.c b/drivers/vdpa/virtio_pci/vp_vdpa.c index f4e375b1d903..bc0bdcc7bcea 100644 --- a/drivers/vdpa/virtio_pci/vp_vdpa.c +++ b/drivers/vdpa/virtio_pci/vp_vdpa.c @@ -220,7 +220,7 @@ static void vp_vdpa_set_status(struct vdpa_device *vdpa, u8 status) vp_modern_set_status(mdev, status); } -static int vp_vdpa_reset(struct vdpa_device *vdpa) +static int vp_vdpa_reset(struct vdpa_device *vdpa, int state) { struct vp_vdpa *vp_vdpa = vdpa_to_vp(vdpa); struct virtio_pci_modern_device *mdev = vp_vdpa_to_mdev(vp_vdpa); diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index eec8027dfc4f..eed51d004531 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -212,13 +212,13 @@ static void vhost_vdpa_unsetup_vq_irq(struct vhost_vdpa *v, u16 qid) irq_bypass_unregister_producer(&vq->call_ctx.producer); } -static int vhost_vdpa_reset(struct vhost_vdpa *v) +static int vhost_vdpa_reset(struct vhost_vdpa *v, int state) { struct vdpa_device *vdpa = v->vdpa; v->in_batch = 0; - return vdpa_reset(vdpa); + return vdpa_reset(vdpa, state); } static long vhost_vdpa_bind_mm(struct vhost_vdpa *v) @@ -297,7 +297,7 @@ static long vhost_vdpa_set_status(struct vhost_vdpa *v, u8 __user *statusp) vhost_vdpa_unsetup_vq_irq(v, i); if (status == 0) { - ret = vdpa_reset(vdpa); + ret = vdpa_reset(vdpa, VDPA_DEV_RESET_VIRTIO); if (ret) return ret; } else @@ -1496,7 +1496,7 @@ static int vhost_vdpa_open(struct inode *inode, struct file *filep) return r; nvqs = v->nvqs; - r = vhost_vdpa_reset(v); + r = vhost_vdpa_reset(v, VDPA_DEV_RESET_OPEN); if (r) goto err; @@ -1542,7 +1542,7 @@ static int vhost_vdpa_release(struct inode *inode, struct file *filep) mutex_lock(&d->mutex); filep->private_data = NULL; vhost_vdpa_clean_irq(v); - vhost_vdpa_reset(v); + vhost_vdpa_reset(v, VDPA_DEV_RESET_CLOSE); vhost_dev_stop(&v->vdev); vhost_vdpa_unbind_mm(v); vhost_vdpa_config_put(v); diff --git a/drivers/virtio/virtio_vdpa.c b/drivers/virtio/virtio_vdpa.c index 46c71653f508..8126a9be22cd 100644 --- a/drivers/virtio/virtio_vdpa.c +++ b/drivers/virtio/virtio_vdpa.c @@ -99,7 +99,7 @@ static void virtio_vdpa_reset(struct virtio_device *vdev) { struct vdpa_device *vdpa = vd_get_vdpa(vdev); - vdpa_reset(vdpa); + vdpa_reset(vdpa, VDPA_DEV_RESET_VIRTIO); } static bool virtio_vdpa_notify(struct virtqueue *vq) diff --git a/include/linux/vdpa.h b/include/linux/vdpa.h index 83df2380eeb9..ef53829c7a7a 100644 --- a/include/linux/vdpa.h +++ b/include/linux/vdpa.h @@ -120,6 +120,12 @@ struct vdpa_map_file { u64 offset; }; +enum vdpa_reset_state { + VDPA_DEV_RESET_VIRTIO = 0, + VDPA_DEV_RESET_OPEN = 1, + VDPA_DEV_RESET_CLOSE = 2, +}; + /** * struct vdpa_config_ops - operations for configuring a vDPA device. * Note: vDPA device drivers are required to implement all of the @@ -218,6 +224,10 @@ struct vdpa_map_file { * @status: virtio device status * @reset: Reset device * @vdev: vdpa device + * @state: state for reset + * VDPA_DEV_RESET_VIRTIO for virtio reset + * VDPA_DEV_RESET_OPEN for vhost-vdpa device open + * VDPA_DEV_RESET_CLOSE for vhost-vdpa device close * Returns integer: success (0) or error (< 0) * @suspend: Suspend the device (optional) * @vdev: vdpa device @@ -359,7 +369,7 @@ struct vdpa_config_ops { u32 (*get_vendor_id)(struct vdpa_device *vdev); u8 (*get_status)(struct vdpa_device *vdev); void (*set_status)(struct vdpa_device *vdev, u8 status); - int (*reset)(struct vdpa_device *vdev); + int (*reset)(struct vdpa_device *vdev, int state); int (*suspend)(struct vdpa_device *vdev); int (*resume)(struct vdpa_device *vdev); size_t (*get_config_size)(struct vdpa_device *vdev); @@ -482,14 +492,14 @@ static inline struct device *vdpa_get_dma_dev(struct vdpa_device *vdev) return vdev->dma_dev; } -static inline int vdpa_reset(struct vdpa_device *vdev) +static inline int vdpa_reset(struct vdpa_device *vdev, int state) { const struct vdpa_config_ops *ops = vdev->config; int ret; down_write(&vdev->cf_lock); vdev->features_valid = false; - ret = ops->reset(vdev); + ret = ops->reset(vdev, state); up_write(&vdev->cf_lock); return ret; } -- 2.27.0

2 1

[PATCH OLK-5.10 v3 00/10] rcu: Add RCU stall diagnosis information
by Zhen Lei 29 Nov '23

29 Nov '23

v2 --> v3: Add "# CONFIG_RCU_CPU_STALL_CPUTIME is not set" into openeuler_defconfig. v1 --> v2: Remove patch 1 in v1: rcu: Prevent lockdep-RCU splats on lock acquisition/release It had arleady been merged. v1: In some extreme cases, such as the I/O pressure test, the CPU usage may be 100%, causing RCU stall. In this case, the printed information about current is not useful. Displays the number and usage of hard interrupts, soft interrupts, and context switches that are generated within half of the CPU stall timeout, can help us make a general judgment. In other cases, we can preliminarily determine whether an infinite loop occurs when local_irq, local_bh or preempt is disabled. Neeraj Upadhyay (1): rcu: Check and report missed fqs timer wakeup on RCU stall Paul E. McKenney (2): rcu: For RCU grace-period kthread starvation, dump last CPU it ran on rcu: Do not NMI offline CPUs Zhen Lei (7): sched/debug: Try trigger_single_cpu_backtrace(cpu) in dump_cpu_task() sched/debug: Show the registers of 'current' in dump_cpu_task() sched: Add helper kstat_cpu_softirqs_sum() sched: Add helper nr_context_switches_cpu() rcu: Add RCU stall diagnosis information rcu: Align the output of RCU CPU stall warning messages config: update openeuler_defconfig for arm64 and x86 Documentation/RCU/stallwarn.rst | 23 ++++- .../admin-guide/kernel-parameters.txt | 6 ++ arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + include/linux/kernel_stat.h | 12 +++ kernel/rcu/Kconfig.debug | 13 +++ kernel/rcu/rcu.h | 1 + kernel/rcu/tree.c | 43 +++++++--- kernel/rcu/tree.h | 19 ++++ kernel/rcu/tree_stall.h | 86 +++++++++++++++++-- kernel/rcu/update.c | 2 + kernel/sched/core.c | 19 ++++ kernel/smp.c | 3 +- 13 files changed, 211 insertions(+), 18 deletions(-) -- 2.25.1

2 11

[PATCH v3 OLK-5.10 0/3] support page eject
by Wupeng Ma 29 Nov '23

29 Nov '23

From: Ma Wupeng <mawupeng1(a)huawei.com> Page ejection is achieved by soft_offline_page and unpoison_memory. A linked list is maintained to log the pfns which are offlined by this module. Only the pfns present in the list is allowed to go online. The follow sysfs is introduced to online/offline page via this module, since there is a internal linked list to record all offlined pages, remove_page is introduced when a user no longer wishes to online this particular page. To reduce the length of the internal list, this function removes the specified page from the list. - /sys/kernel/page_eject/online_page - /sys/kernel/page_eject/offline_page - /sys/kernel/page_eject/remove_page Changelog since v2: - bugfix for page_eject_clear_list_locked Changelog since v1: - remove mce related patches temporarily. Ma Wupeng (3): mm/hwpoison: Export symbol soft_offline_page mm: page_eject: Introuduce page ejection config: update defconfig for PAGE_EJECT arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + drivers/ras/Kconfig | 10 + drivers/ras/Makefile | 1 + drivers/ras/page_eject.c | 248 +++++++++++++++++++++++++ mm/memory-failure.c | 1 + 6 files changed, 262 insertions(+) create mode 100644 drivers/ras/page_eject.c -- 2.25.1

2 4

[PATCH OLK-6.6 0/9] mm: PCP high auto-tuning
by Ze Zuo 29 Nov '23

29 Nov '23

The page allocation performance requirements of different workloads are often different. So, we need to tune the PCP (Per-CPU Pageset) high on each CPU automatically to optimize the page allocation performance. The list of patches in series is as follows, [1/9] mm, pcp: avoid to drain PCP when process exit [2/9] cacheinfo: calculate per-CPU data cache size [3/9] mm, pcp: reduce lock contention for draining high-order pages [4/9] mm: restrict the pcp batch scale factor to avoid too long latency [5/9] mm, page_alloc: scale the number of pages that are batch allocated [6/9] mm: add framework for PCP high auto-tuning [7/9] mm: tune PCP high automatically [8/9] mm, pcp: decrease PCP high if free pages < high watermark [9/9] mm, pcp: reduce detecting time of consecutive high order page freeing Patch [1/9], [2/9], [3/9] optimize the PCP draining for consecutive high-order pages freeing. Patch [4/9], [5/9] optimize batch freeing and allocating. Patch [6/9], [7/9], [8/9] implement and optimize a PCP high auto-tuning method. Patch [9/9] optimize the PCP draining for consecutive high order page freeing based on PCP high auto-tuning. The test results for patches with performance impact are as follows, kbuild ====== On a 2-socket Intel server with 224 logical CPU, we run 8 kbuild instances in parallel (each with `make -j 28`) in 8 cgroup. This simulates the kbuild server that is used by 0-Day kbuild service. build time lock contend% free_high alloc_zone ---------- ---------- --------- ---------- base 100.0 14.0 100.0 100.0 patch1 99.5 12.8 19.5 95.6 patch3 99.4 12.6 7.1 95.6 patch5 98.6 11.0 8.1 97.1 patch7 95.1 0.5 2.8 15.6 patch9 95.0 1.0 8.8 20.0 The PCP draining optimization (patch [1/9], [3/9]) and PCP batch allocation optimization (patch [5/9]) reduces zone lock contention a little. The PCP high auto-tuning (patch [7/9], [9/9]) reduces build time visibly. Where the tuning target: the number of pages allocated from zone reduces greatly. So, the zone contention cycles% reduces greatly. With PCP tuning patches (patch [7/9], [9/9]), the average used memory during test increases up to 18.4% because more pages are cached in PCP. But at the end of the test, the number of the used memory decreases to the same level as that of the base patch. That is, the pages cached in PCP will be released to zone after not being used actively. netperf SCTP_STREAM_MANY ======================== On a 2-socket Intel server with 128 logical CPU, we tested SCTP_STREAM_MANY test case of netperf test suite with 64-pair processes. score lock contend% free_high alloc_zone cache miss rate% ----- ---------- --------- ---------- ---------------- base 100.0 2.1 100.0 100.0 1.3 patch1 99.4 2.1 99.4 99.4 1.3 patch3 106.4 1.3 13.3 106.3 1.3 patch5 106.0 1.2 13.2 105.9 1.3 patch7 103.4 1.9 6.7 90.3 7.6 patch9 108.6 1.3 13.7 108.6 1.3 The PCP draining optimization (patch [1/9]+[3/9]) improves performance. The PCP high auto-tuning (patch [7/9]) reduces performance a little because PCP draining cannot be triggered in time sometimes. So, the cache miss rate% increases. The further PCP draining optimization (patch [9/9]) based on PCP tuning restore the performance. lmbench3 UNIX (AF_UNIX) ======================= On a 2-socket Intel server with 128 logical CPU, we tested UNIX (AF_UNIX socket) test case of lmbench3 test suite with 16-pair processes. score lock contend% free_high alloc_zone cache miss rate% ----- ---------- --------- ---------- ---------------- base 100.0 51.4 100.0 100.0 0.2 patch1 116.8 46.1 69.5 104.3 0.2 patch3 199.1 21.3 7.0 104.9 0.2 patch5 200.0 20.8 7.1 106.9 0.3 patch7 191.6 19.9 6.8 103.8 2.8 patch9 193.4 21.7 7.0 104.7 2.1 The PCP draining optimization (patch [1/9], [3/9]) improves performance much. The PCP tuning (patch [7/9]) reduces performance a little because PCP draining cannot be triggered in time sometimes. The further PCP draining optimization (patch [9/9]) based on PCP tuning restores the performance partly. The patchset adds several fields in struct per_cpu_pages. The struct layout before/after the patchset is as follows, base ==== struct per_cpu_pages { spinlock_t lock; /* 0 4 */ int count; /* 4 4 */ int high; /* 8 4 */ int batch; /* 12 4 */ short int free_factor; /* 16 2 */ short int expire; /* 18 2 */ /* XXX 4 bytes hole, try to pack */ struct list_head lists[13]; /* 24 208 */ /* size: 256, cachelines: 4, members: 7 */ /* sum members: 228, holes: 1, sum holes: 4 */ /* padding: 24 */ } __attribute__((__aligned__(64))); patched ======= struct per_cpu_pages { spinlock_t lock; /* 0 4 */ int count; /* 4 4 */ int high; /* 8 4 */ int high_min; /* 12 4 */ int high_max; /* 16 4 */ int batch; /* 20 4 */ u8 flags; /* 24 1 */ u8 alloc_factor; /* 25 1 */ u8 expire; /* 26 1 */ /* XXX 1 byte hole, try to pack */ short int free_count; /* 28 2 */ /* XXX 2 bytes hole, try to pack */ struct list_head lists[13]; /* 32 208 */ /* size: 256, cachelines: 4, members: 11 */ /* sum members: 237, holes: 2, sum holes: 3 */ /* padding: 16 */ } __attribute__((__aligned__(64))); The size of the struct doesn't changed with the patchset. Huang Ying (9): mm, pcp: avoid to drain PCP when process exit cacheinfo: calculate size of per-CPU data cache slice mm, pcp: reduce lock contention for draining high-order pages mm: restrict the pcp batch scale factor to avoid too long latency mm, page_alloc: scale the number of pages that are batch allocated mm: add framework for PCP high auto-tuning mm: tune PCP high automatically mm, pcp: decrease PCP high if free pages < high watermark mm, pcp: reduce detecting time of consecutive high order page freeing drivers/base/cacheinfo.c | 51 ++++++- include/linux/cacheinfo.h | 1 + include/linux/gfp.h | 2 + include/linux/mmzone.h | 27 +++- mm/Kconfig | 11 ++ mm/page_alloc.c | 310 +++++++++++++++++++++++++++++++------- mm/vmstat.c | 8 +- 7 files changed, 345 insertions(+), 65 deletions(-) -- 2.25.1

2 10

[PATCH OLK-6.6 0/9] mm: PCP high auto-tuning
by Ze Zuo 29 Nov '23

29 Nov '23

The page allocation performance requirements of different workloads are often different. So, we need to tune the PCP (Per-CPU Pageset) high on each CPU automatically to optimize the page allocation performance. The list of patches in series is as follows, [1/9] mm, pcp: avoid to drain PCP when process exit [2/9] cacheinfo: calculate per-CPU data cache size [3/9] mm, pcp: reduce lock contention for draining high-order pages [4/9] mm: restrict the pcp batch scale factor to avoid too long latency [5/9] mm, page_alloc: scale the number of pages that are batch allocated [6/9] mm: add framework for PCP high auto-tuning [7/9] mm: tune PCP high automatically [8/9] mm, pcp: decrease PCP high if free pages < high watermark [9/9] mm, pcp: reduce detecting time of consecutive high order page freeing Patch [1/9], [2/9], [3/9] optimize the PCP draining for consecutive high-order pages freeing. Patch [4/9], [5/9] optimize batch freeing and allocating. Patch [6/9], [7/9], [8/9] implement and optimize a PCP high auto-tuning method. Patch [9/9] optimize the PCP draining for consecutive high order page freeing based on PCP high auto-tuning. The test results for patches with performance impact are as follows, kbuild ====== On a 2-socket Intel server with 224 logical CPU, we run 8 kbuild instances in parallel (each with `make -j 28`) in 8 cgroup. This simulates the kbuild server that is used by 0-Day kbuild service. build time lock contend% free_high alloc_zone ---------- ---------- --------- ---------- base 100.0 14.0 100.0 100.0 patch1 99.5 12.8 19.5 95.6 patch3 99.4 12.6 7.1 95.6 patch5 98.6 11.0 8.1 97.1 patch7 95.1 0.5 2.8 15.6 patch9 95.0 1.0 8.8 20.0 The PCP draining optimization (patch [1/9], [3/9]) and PCP batch allocation optimization (patch [5/9]) reduces zone lock contention a little. The PCP high auto-tuning (patch [7/9], [9/9]) reduces build time visibly. Where the tuning target: the number of pages allocated from zone reduces greatly. So, the zone contention cycles% reduces greatly. With PCP tuning patches (patch [7/9], [9/9]), the average used memory during test increases up to 18.4% because more pages are cached in PCP. But at the end of the test, the number of the used memory decreases to the same level as that of the base patch. That is, the pages cached in PCP will be released to zone after not being used actively. netperf SCTP_STREAM_MANY ======================== On a 2-socket Intel server with 128 logical CPU, we tested SCTP_STREAM_MANY test case of netperf test suite with 64-pair processes. score lock contend% free_high alloc_zone cache miss rate% ----- ---------- --------- ---------- ---------------- base 100.0 2.1 100.0 100.0 1.3 patch1 99.4 2.1 99.4 99.4 1.3 patch3 106.4 1.3 13.3 106.3 1.3 patch5 106.0 1.2 13.2 105.9 1.3 patch7 103.4 1.9 6.7 90.3 7.6 patch9 108.6 1.3 13.7 108.6 1.3 The PCP draining optimization (patch [1/9]+[3/9]) improves performance. The PCP high auto-tuning (patch [7/9]) reduces performance a little because PCP draining cannot be triggered in time sometimes. So, the cache miss rate% increases. The further PCP draining optimization (patch [9/9]) based on PCP tuning restore the performance. lmbench3 UNIX (AF_UNIX) ======================= On a 2-socket Intel server with 128 logical CPU, we tested UNIX (AF_UNIX socket) test case of lmbench3 test suite with 16-pair processes. score lock contend% free_high alloc_zone cache miss rate% ----- ---------- --------- ---------- ---------------- base 100.0 51.4 100.0 100.0 0.2 patch1 116.8 46.1 69.5 104.3 0.2 patch3 199.1 21.3 7.0 104.9 0.2 patch5 200.0 20.8 7.1 106.9 0.3 patch7 191.6 19.9 6.8 103.8 2.8 patch9 193.4 21.7 7.0 104.7 2.1 The PCP draining optimization (patch [1/9], [3/9]) improves performance much. The PCP tuning (patch [7/9]) reduces performance a little because PCP draining cannot be triggered in time sometimes. The further PCP draining optimization (patch [9/9]) based on PCP tuning restores the performance partly. The patchset adds several fields in struct per_cpu_pages. The struct layout before/after the patchset is as follows, base ==== struct per_cpu_pages { spinlock_t lock; /* 0 4 */ int count; /* 4 4 */ int high; /* 8 4 */ int batch; /* 12 4 */ short int free_factor; /* 16 2 */ short int expire; /* 18 2 */ /* XXX 4 bytes hole, try to pack */ struct list_head lists[13]; /* 24 208 */ /* size: 256, cachelines: 4, members: 7 */ /* sum members: 228, holes: 1, sum holes: 4 */ /* padding: 24 */ } __attribute__((__aligned__(64))); patched ======= struct per_cpu_pages { spinlock_t lock; /* 0 4 */ int count; /* 4 4 */ int high; /* 8 4 */ int high_min; /* 12 4 */ int high_max; /* 16 4 */ int batch; /* 20 4 */ u8 flags; /* 24 1 */ u8 alloc_factor; /* 25 1 */ u8 expire; /* 26 1 */ /* XXX 1 byte hole, try to pack */ short int free_count; /* 28 2 */ /* XXX 2 bytes hole, try to pack */ struct list_head lists[13]; /* 32 208 */ /* size: 256, cachelines: 4, members: 11 */ /* sum members: 237, holes: 2, sum holes: 3 */ /* padding: 16 */ } __attribute__((__aligned__(64))); The size of the struct doesn't changed with the patchset. Huang Ying (9): mm, pcp: avoid to drain PCP when process exit cacheinfo: calculate size of per-CPU data cache slice mm, pcp: reduce lock contention for draining high-order pages mm: restrict the pcp batch scale factor to avoid too long latency mm, page_alloc: scale the number of pages that are batch allocated mm: add framework for PCP high auto-tuning mm: tune PCP high automatically mm, pcp: decrease PCP high if free pages < high watermark mm, pcp: reduce detecting time of consecutive high order page freeing drivers/base/cacheinfo.c | 51 ++++++- include/linux/cacheinfo.h | 1 + include/linux/gfp.h | 2 + include/linux/mmzone.h | 27 +++- mm/Kconfig | 11 ++ mm/page_alloc.c | 310 +++++++++++++++++++++++++++++++------- mm/vmstat.c | 8 +- 7 files changed, 345 insertions(+), 65 deletions(-) -- 2.25.1

1 9

[PACH OLK-5.10 00/11] KVM: arm64: Add host/guest KVM-PTP support
by Dongxu Sun 29 Nov '23

29 Nov '23

Jianyong Wu (4): ptp: Reorganize ptp_kvm.c to make it arch-independent clocksource: Add clocksource id for arm arch counter KVM: arm64: Add support for the KVM PTP service ptp: arm/arm64: Enable ptp_kvm for arm/arm64 Jon Hunter (1): ptp: Don't print an error if ptp_kvm is not supported Kele Huang (1): ptp: fix error print of ptp_kvm on X86_64 platform Thomas Gleixner (1): time: Add mechanism to recognize clocksource in time_get_snapshot Will Deacon (2): arm/arm64: Probe for the presence of KVM hypervisor KVM: arm64: Advertise KVM UID to guests via SMCCC Yanan Wang (1): KVM: arm64: fix compile error/warning Zenghui Yu (1): KVM: arm64: Fix Function ID typo for PTP_KVM service Documentation/virt/kvm/api.rst | 9 ++ Documentation/virt/kvm/arm/index.rst | 1 + Documentation/virt/kvm/arm/ptp_kvm.rst | 25 ++++++ arch/arm/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/kvm_para.h | 2 +- arch/arm64/include/uapi/asm/Kbuild | 2 +- arch/arm64/include/uapi/asm/kvm_para.h | 7 ++ arch/arm64/kvm/arm.c | 1 + arch/arm64/kvm/hypercalls.c | 94 ++++++++++++++++---- drivers/clocksource/arm_arch_timer.c | 36 ++++++++ drivers/firmware/psci/psci.c | 2 + drivers/firmware/smccc/Makefile | 2 +- drivers/firmware/smccc/kvm_guest.c | 50 +++++++++++ drivers/firmware/smccc/smccc.c | 1 + drivers/ptp/Kconfig | 2 +- drivers/ptp/Makefile | 2 + drivers/ptp/ptp_kvm_arm.c | 28 ++++++ drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} | 85 +++++------------- drivers/ptp/ptp_kvm_x86.c | 97 +++++++++++++++++++++ include/linux/arm-smccc.h | 41 +++++++++ include/linux/clocksource.h | 6 ++ include/linux/clocksource_ids.h | 12 +++ include/linux/ptp_kvm.h | 19 ++++ include/linux/timekeeping.h | 12 +-- include/uapi/linux/kvm.h | 1 + kernel/time/clocksource.c | 2 + kernel/time/timekeeping.c | 1 + 28 files changed, 459 insertions(+), 87 deletions(-) create mode 100644 Documentation/virt/kvm/arm/ptp_kvm.rst create mode 100644 arch/arm64/include/uapi/asm/kvm_para.h create mode 100644 drivers/firmware/smccc/kvm_guest.c create mode 100644 drivers/ptp/ptp_kvm_arm.c rename drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} (60%) create mode 100644 drivers/ptp/ptp_kvm_x86.c create mode 100644 include/linux/clocksource_ids.h create mode 100644 include/linux/ptp_kvm.h -- 2.33.0

1 11

[PACH OLK-5.10 00/11] KVM: arm64: Add host/guest KVM-PTP support
by Dongxu Sun 29 Nov '23

29 Nov '23

Jianyong Wu (4): ptp: Reorganize ptp_kvm.c to make it arch-independent clocksource: Add clocksource id for arm arch counter KVM: arm64: Add support for the KVM PTP service ptp: arm/arm64: Enable ptp_kvm for arm/arm64 Jon Hunter (1): ptp: Don't print an error if ptp_kvm is not supported Kele Huang (1): ptp: fix error print of ptp_kvm on X86_64 platform Thomas Gleixner (1): time: Add mechanism to recognize clocksource in time_get_snapshot Will Deacon (2): arm/arm64: Probe for the presence of KVM hypervisor KVM: arm64: Advertise KVM UID to guests via SMCCC Yanan Wang (1): KVM: arm64: fix compile error/warning Zenghui Yu (1): KVM: arm64: Fix Function ID typo for PTP_KVM service Documentation/virt/kvm/api.rst | 9 ++ Documentation/virt/kvm/arm/index.rst | 1 + Documentation/virt/kvm/arm/ptp_kvm.rst | 25 ++++++ arch/arm/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/hypervisor.h | 3 + arch/arm64/include/asm/kvm_para.h | 2 +- arch/arm64/include/uapi/asm/Kbuild | 2 +- arch/arm64/include/uapi/asm/kvm_para.h | 7 ++ arch/arm64/kvm/arm.c | 1 + arch/arm64/kvm/hypercalls.c | 94 ++++++++++++++++---- drivers/clocksource/arm_arch_timer.c | 36 ++++++++ drivers/firmware/psci/psci.c | 2 + drivers/firmware/smccc/Makefile | 2 +- drivers/firmware/smccc/kvm_guest.c | 50 +++++++++++ drivers/firmware/smccc/smccc.c | 1 + drivers/ptp/Kconfig | 2 +- drivers/ptp/Makefile | 2 + drivers/ptp/ptp_kvm_arm.c | 28 ++++++ drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} | 85 +++++------------- drivers/ptp/ptp_kvm_x86.c | 97 +++++++++++++++++++++ include/linux/arm-smccc.h | 41 +++++++++ include/linux/clocksource.h | 6 ++ include/linux/clocksource_ids.h | 12 +++ include/linux/ptp_kvm.h | 19 ++++ include/linux/timekeeping.h | 12 +-- include/uapi/linux/kvm.h | 1 + kernel/time/clocksource.c | 2 + kernel/time/timekeeping.c | 1 + 28 files changed, 459 insertions(+), 87 deletions(-) create mode 100644 Documentation/virt/kvm/arm/ptp_kvm.rst create mode 100644 arch/arm64/include/uapi/asm/kvm_para.h create mode 100644 drivers/firmware/smccc/kvm_guest.c create mode 100644 drivers/ptp/ptp_kvm_arm.c rename drivers/ptp/{ptp_kvm.c => ptp_kvm_common.c} (60%) create mode 100644 drivers/ptp/ptp_kvm_x86.c create mode 100644 include/linux/clocksource_ids.h create mode 100644 include/linux/ptp_kvm.h -- 2.33.0

1 4

[PATCH OLK-5.10] drivers: perf: Add feature flag check in armpmu_add/del
by Junhao He 29 Nov '23

29 Nov '23

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8EC9K CVE: NA -------------------------------- Function armpmu_add will call armv8pmu_branch_reset() to reset BRBE. The BRBE feature depend on ARMV8 PMU hardware support for branch stack sampling. On older platforms the kernel panics. Internal error: Oops - Undefined instruction: 0000000002000000 [#1] SMP Call trace: armv8pmu_branch_reset+0xc/0x20 event_sched_in+0xc8/0x1a0 merge_sched_in+0x16c/0x41c Fixes: 2690e65ac9f9 ("drivers: perf: arm_pmuv3: Enable branch stack sampling via FEAT_BRBE") Signed-off-by: Junhao He <hejunhao3(a)huawei.com> --- drivers/perf/arm_pmu.c | 34 +++++++++++++++++++--------------- 1 file changed, 19 insertions(+), 15 deletions(-) diff --git a/drivers/perf/arm_pmu.c b/drivers/perf/arm_pmu.c index d02a85c0bedc..4ef8aee84eea 100644 --- a/drivers/perf/arm_pmu.c +++ b/drivers/perf/arm_pmu.c @@ -313,10 +313,12 @@ armpmu_del(struct perf_event *event, int flags) struct hw_perf_event *hwc = &event->hw; int idx = hwc->idx; - WARN_ON_ONCE(!hw_events->brbe_users); - hw_events->brbe_users--; - if (!hw_events->brbe_users) - hw_events->brbe_context = NULL; + if (has_branch_stack(event)) { + WARN_ON_ONCE(!hw_events->brbe_users); + hw_events->brbe_users--; + if (!hw_events->brbe_users) + hw_events->brbe_context = NULL; + } armpmu_stop(event, PERF_EF_UPDATE); hw_events->events[idx] = NULL; @@ -334,18 +336,20 @@ armpmu_add(struct perf_event *event, int flags) struct hw_perf_event *hwc = &event->hw; int idx; - /* - * Reset branch records buffer if a new task event gets - * scheduled on a PMU which might have existing records. - * Otherwise older branch records present in the buffer - * might leak into the new task event. - */ - if (event->ctx->task && hw_events->brbe_context != event->ctx) { - hw_events->brbe_context = event->ctx; - if (armpmu->branch_reset) - armpmu->branch_reset(); + if (has_branch_stack(event)) { + /* + * Reset branch records buffer if a new task event gets + * scheduled on a PMU which might have existing records. + * Otherwise older branch records present in the buffer + * might leak into the new task event. + */ + if (event->ctx->task && hw_events->brbe_context != event->ctx) { + hw_events->brbe_context = event->ctx; + if (armpmu->branch_reset) + armpmu->branch_reset(); + } + hw_events->brbe_users++; } - hw_events->brbe_users++; /* An event following a process won't be stopped earlier */ if (!cpumask_test_cpu(smp_processor_id(), &armpmu->supported_cpus)) -- 2.33.0

2 1

[PATCH OLK-6.6 v2 0/6] Handle more faults under the VMA lock
by Tong Tiangen 29 Nov '23

29 Nov '23

At this point, we're handling the majority of file-backed page faults under the VMA lock, using the ->map_pages entry point. This patch set attempts to expand that for the following siutations: - We have to do a read. This could be because we've hit the point in the readahead window where we need to kick off the next readahead, or because the page is simply not present in cache. - We're handling a write fault. Most applications don't do I/O by writes to shared mmaps for very good reasons, but some do, and it'd be nice to not make that slow unnecessarily. - We're doing a COW of a private mapping (both PTE already present and PTE not-present). These are two different codepaths and I handle both of them in this patch set. There is no support in this patch set for drivers to mark themselves as being VMA lock friendly; they could implement the ->map_pages vm_operation, but if they do, they would be the first. This is probably something we want to change at some point in the future, and I've marked where to make that change in the code. There is very little performance change in the benchmarks we've run; mostly because the vast majority of page faults are handled through the other paths. I still think this patch series is useful for workloads that may take these paths more often, and just for cleaning up the fault path in general (it's now clearer why we have to retry in these cases). Matthew Wilcox (Oracle) (6): mm: make lock_folio_maybe_drop_mmap() VMA lock aware mm: call wp_page_copy() under the VMA lock mm: handle shared faults under the VMA lock mm: handle COW faults under the VMA lock mm: handle read faults under the VMA lock mm: handle write faults to RO pages under the VMA lock mm/filemap.c | 13 ++++---- mm/memory.c | 93 ++++++++++++++++++++++++++++++++-------------------- 2 files changed, 65 insertions(+), 41 deletions(-) -- 2.25.1

2 7

[PATCH OLK-6.6] tcp/dccp: Add another way to allocate local ports in connect()
by Zhengchao Shao 29 Nov '23

29 Nov '23

From: Lu Wei <luwei32(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8KBYH -------------------------------- Commit 07f4c90062f8 ("tcp/dccp: try to not exhaust ip_local_port_range in connect()") allocates even ports for connect() first while leaving odd ports for bind() and this works well in busy servers. But this strategy causes severe performance degradation in busy clients. when a client has used more than half of the local ports setted in proc/sys/net/ipv4/ip_local_port_range, if this client trys to connect to a server again, the connect time increases rapidly since it will traverse all the even ports though they are exhausted. So this path provides another strategy by introducing a system option: local_port_allocation. If it is a busy client, users should set it to 1 to use sequential allocation while it should be set to 0 in other situations. Its default value is 0. Signed-off-by: Lu Wei <luwei32(a)huawei.com> Signed-off-by: Liu Jian <liujian56(a)huawei.com> Conflicts: include/net/tcp.h net/ipv4/inet_hashtables.c net/ipv4/sysctl_net_ipv4.c Signed-off-by: Zhengchao Shao <shaozhengchao(a)huawei.com> --- include/net/tcp.h | 1 + net/ipv4/inet_hashtables.c | 11 ++++++++--- net/ipv4/sysctl_net_ipv4.c | 8 ++++++++ 3 files changed, 17 insertions(+), 3 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index 0239e815edf7..e9d387fffe22 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -258,6 +258,7 @@ DECLARE_PER_CPU(int, tcp_memory_per_cpu_fw_alloc); extern struct percpu_counter tcp_sockets_allocated; extern unsigned long tcp_memory_pressure; +extern int sysctl_local_port_allocation; /* optimized version of sk_under_memory_pressure() for TCP sockets */ static inline bool tcp_under_memory_pressure(const struct sock *sk) diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 598c1b114d2c..919f0f869118 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -1011,7 +1011,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, struct inet_bind_bucket *tb; bool tb_created = false; u32 remaining, offset; - int ret, i, low, high; + int ret, i, low, high, span_size; int l3mdev; u32 index; @@ -1021,6 +1021,11 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, local_bh_enable(); return ret; } + /* local_port_allocation 0 means even and odd port allocation strategy + * will be applied, so span size is 2; otherwise sequential allocation + * will be used and span size is 1. Default value is 0. + */ + span_size = sysctl_local_port_allocation ? 1 : 2; l3mdev = inet_sk_bound_l3mdev(sk); @@ -1043,7 +1048,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, offset &= ~1U; other_parity_scan: port = low + offset; - for (i = 0; i < remaining; i += 2, port += 2) { + for (i = 0; i < remaining; i += span_size, port += span_size) { if (unlikely(port >= high)) port -= remaining; if (inet_is_local_reserved_port(net, port)) @@ -1084,7 +1089,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, } offset++; - if ((offset & 1) && remaining > 1) + if ((offset & 1) && remaining > 1 && span_size == 2) goto other_parity_scan; return -EADDRNOTAVAIL; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 6ac890b4073f..b17eb28a9690 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -39,6 +39,7 @@ static unsigned long ip_ping_group_range_min[] = { 0, 0 }; static unsigned long ip_ping_group_range_max[] = { GID_T_MAX, GID_T_MAX }; static u32 u32_max_div_HZ = UINT_MAX / HZ; static int one_day_secs = 24 * 3600; +int sysctl_local_port_allocation; static u32 fib_multipath_hash_fields_all_mask __maybe_unused = FIB_MULTIPATH_HASH_FIELD_ALL_MASK; static unsigned int tcp_child_ehash_entries_max = 16 * 1024 * 1024; @@ -579,6 +580,13 @@ static struct ctl_table ipv4_table[] = { .extra1 = &sysctl_fib_sync_mem_min, .extra2 = &sysctl_fib_sync_mem_max, }, + { + .procname = "local_port_allocation", + .data = &sysctl_local_port_allocation, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { } }; -- 2.34.1

2 1

[PACH OLK-5.10 v5 00/20] psi irq in cgroupv1 and psi fine grained
by Lu Jialin 29 Nov '23

29 Nov '23

Chen Wandun (3): mm: disable psi cgroup v1 by default mm: add config isolation for psi under cgroup v1 psi: dont alloc memory for psi by default Chengming Zhou (3): sched/psi: Fix periodic aggregation shut off sched/psi: Optimize task switch inside shared cgroups again sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ pressure Hao Jia (1): sched/psi: Zero the memory of struct psi_group Johannes Weiner (1): sched/psi: Remove NR_ONCPU task accounting Lu Jialin (11): psi: support irq.pressure under cgroup v1 psi: enable CONFIG_PSI_CGROUP_V1 in openeuler_defconfig psi: update psi irqtime when the irq delta is nozero memcg: split async memcg reclaim from reclaim_high psi: add struct psi_group_ext PSI: Introduce fine grained stall time collect for cgroup reclaim PSI: Introduce avgs and total calculation for cgroup reclaim PSI: Introduce pressure.stat in psi PSI: add more memory fine grained stall tracking in pressure.stat add cpu fine grained stall tracking in pressure.stat PSI: enable CONFIG_PSI_FINE_GRAINED in openeuler_defconfig Suren Baghdasaryan (1): psi: Fix "defined but not used" warnings when CONFIG_PROC_FS=n Documentation/admin-guide/cgroup-v2.rst | 6 + arch/arm64/configs/openeuler_defconfig | 2 + arch/x86/configs/openeuler_defconfig | 2 + block/blk-cgroup.c | 2 +- block/blk-core.c | 2 +- include/linux/cgroup-defs.h | 6 +- include/linux/cgroup.h | 2 +- include/linux/psi.h | 9 +- include/linux/psi_types.h | 99 +++- include/linux/sched.h | 4 + init/Kconfig | 20 + kernel/cgroup/cgroup.c | 64 ++- kernel/sched/core.c | 2 + kernel/sched/cpuacct.c | 6 +- kernel/sched/fair.c | 6 - kernel/sched/psi.c | 691 ++++++++++++++++++++---- kernel/sched/stats.h | 9 + mm/compaction.c | 2 +- mm/filemap.c | 4 +- mm/memcontrol.c | 92 ++-- mm/page_alloc.c | 6 + mm/page_io.c | 3 + mm/vmscan.c | 5 +- 23 files changed, 873 insertions(+), 171 deletions(-) -- 2.34.1

1 20

[PATCH OLK-5.10] livepatch/powerpc: Fix issue that miss one layer on stack checking
by Zheng Yejian 29 Nov '23

29 Nov '23

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7BMC9 CVE: NA -------------------------------- Suppose task 'T' is running in function 'B', and user is doing livepatch for function 'A', and function 'A' is calling 'B'. Then task 'T' will be interrupted by stop_machine, and suppose that 'B' has not saved the link register (which point to address in 'A') into stack memory, and data in link register position may be invalid, then 'A' would be missed on stack checking. To find a solution, we list following cases (suppose calling chain: P -> A -> B, A runs `bl B`, 'nip' means instruction pointer found in interrupt frame, 'lr' means link register found in interrupt frame, 'frame->pc' means link register saved in first stack frame): 1. 'nip' is in 'A', `bl B` is not executed, then: 'lr' is in 'P', 'frame->pc' should be ignored; 2. 'nip' is in some ppc64_stub or ppc32_plt which also not save link register, then: 'lr' is in 'A', 'frame->pc' should be ignored; 3. 'nip' is in 'B', but 'B' has not saved link register, then: 'lr' is in 'A', 'frame->pc' should be ignored; 4. 'nip' is in 'B', 'B' has saved link register, then: 'lr' is in 'A', 'frame->pc' is also in 'A'; 5. 'nip' is in 'A', 'B' has returned but its stack pointer is not moved, then 'lr' is in 'A', 'frame->pc' is also in 'A'; 6. 'nip' is in 'A', 'B' has returned and its stack pointer is moved, then 'lr' is in 'A', 'frame->pc' is in 'P'. As a conclusion, we need to: 1. check 'nip' or 'lr' or both if they are not in same function; 2. ignore 'frame->pc' if 'nip' and 'lr' are not in same function. Signed-off-by: Zheng Yejian <zhengyejian1(a)huawei.com> --- arch/powerpc/include/asm/livepatch.h | 11 ++++++++++ arch/powerpc/kernel/livepatch.c | 28 +++++++++++++++++++++++-- arch/powerpc/kernel/livepatch_32.c | 31 +++++++++++++++++++--------- arch/powerpc/kernel/livepatch_64.c | 31 +++++++++++++++++++--------- 4 files changed, 79 insertions(+), 22 deletions(-) diff --git a/arch/powerpc/include/asm/livepatch.h b/arch/powerpc/include/asm/livepatch.h index ae674ea59ab3..285602e637f1 100644 --- a/arch/powerpc/include/asm/livepatch.h +++ b/arch/powerpc/include/asm/livepatch.h @@ -121,9 +121,20 @@ struct arch_klp_data { #define KLP_MAX_REPLACE_SIZE sizeof_field(struct arch_klp_data, old_insns) struct stackframe { + /* stack frame to be unwinded */ unsigned long sp; + /* link register saved in last stack frame */ unsigned long pc; + /* instruction register saved in pt_regs */ unsigned long nip; + /* link register saved in pt_regs */ + unsigned long link; + /* stack frame pointer (r1 register) saved in pt_regs */ + unsigned long sfp; + /* check if nip and link are in same function */ + unsigned int nip_link_in_same_func; + /* check if it is top frame before interrupt */ + unsigned int is_top_frame; }; #ifdef PPC64_ELF_ABI_v1 diff --git a/arch/powerpc/kernel/livepatch.c b/arch/powerpc/kernel/livepatch.c index d568e8c8b16b..5ba38c2c7c5c 100644 --- a/arch/powerpc/kernel/livepatch.c +++ b/arch/powerpc/kernel/livepatch.c @@ -23,6 +23,7 @@ #include <linux/module.h> #include <linux/ftrace.h> #include <linux/livepatch.h> +#include <linux/kallsyms.h> #include <asm/probes.h> #include <asm/livepatch.h> #include <asm/code-patching.h> @@ -67,6 +68,22 @@ int klp_brk_handler(struct pt_regs *regs) return 1; } +static int check_addr_in_same_func(unsigned long addr1, unsigned long addr2) +{ + unsigned long size = 0; + unsigned long offset = 0; + unsigned long start; + + if (addr1 == 0 || addr2 == 0) + return 0; + if (addr1 == addr2) + return 1; + if (!kallsyms_lookup_size_offset(addr1, &size, &offset)) + return 0; + start = addr1 - offset; + return (addr2 >= start) && (addr2 - start < size); +} + int klp_unwind_frame(struct task_struct *tsk, struct stackframe *frame) { unsigned long *stack; @@ -79,7 +96,10 @@ int klp_unwind_frame(struct task_struct *tsk, struct stackframe *frame) if (frame->nip != 0) frame->nip = 0; + if (frame->link != 0) + frame->link = 0; + frame->is_top_frame = (frame->sfp == frame->sp); stack = (unsigned long *)frame->sp; /* @@ -94,10 +114,14 @@ int klp_unwind_frame(struct task_struct *tsk, struct stackframe *frame) struct pt_regs *regs = (struct pt_regs *) (frame->sp + STACK_FRAME_OVERHEAD); frame->nip = regs->nip; - pr_debug("--- interrupt: task = %d/%s, trap %lx at NIP=x%lx/%pS, LR=0x%lx/%pS\n", + frame->link = regs->link; + frame->sfp = regs->gpr[PT_R1]; + frame->nip_link_in_same_func = check_addr_in_same_func(frame->nip, frame->link); + pr_debug("--- interrupt: task = %d/%s, trap %lx at NIP=0x%lx/%pS, LR=0x%lx/%pS, SFP=0x%lx, nip_link_in_same_func=%u\n", tsk->pid, tsk->comm, regs->trap, regs->nip, (void *)regs->nip, - regs->link, (void *)regs->link); + regs->link, (void *)regs->link, + frame->sfp, frame->nip_link_in_same_func); } frame->sp = stack[0]; diff --git a/arch/powerpc/kernel/livepatch_32.c b/arch/powerpc/kernel/livepatch_32.c index 3fe4f3c5790b..134b08e12e74 100644 --- a/arch/powerpc/kernel/livepatch_32.c +++ b/arch/powerpc/kernel/livepatch_32.c @@ -272,13 +272,21 @@ static int klp_check_jump_func(struct stackframe *frame, void *data) struct walk_stackframe_args *args = data; struct klp_func_list *check_funcs = args->check_funcs; - /* check the PC first */ - if (!check_func_list(check_funcs, &args->ret, frame->pc)) - return args->ret; - /* check NIP when the exception stack switching */ if (frame->nip && !check_func_list(check_funcs, &args->ret, frame->nip)) return args->ret; + if (frame->link && !frame->nip_link_in_same_func && + !check_func_list(check_funcs, &args->ret, frame->link)) + return args->ret; + /* + * There are two cases that frame->pc is reliable: + * 1. frame->pc is not in top frame before interrupt; + * 2. nip and link are in same function; + */ + if (!frame->is_top_frame || frame->nip_link_in_same_func) { + if (!check_func_list(check_funcs, &args->ret, frame->pc)) + return args->ret; + } return 0; } @@ -298,10 +306,11 @@ static int do_check_calltrace(struct walk_stackframe_args *args, int (*fn)(struct stackframe *, void *)) { struct task_struct *g, *t; - struct stackframe frame; unsigned long *stack; for_each_process_thread(g, t) { + struct stackframe frame = { 0 }; + if (t == current) { /* * Handle the current carefully on each CPUs, we shouldn't @@ -329,7 +338,6 @@ static int do_check_calltrace(struct walk_stackframe_args *args, frame.sp = (unsigned long)stack; frame.pc = stack[STACK_FRAME_LR_SAVE]; - frame.nip = 0; klp_walk_stackframe(&frame, fn, t, args); if (args->ret) { pr_info("PID: %d Comm: %.20s\n", t->pid, t->comm); @@ -368,13 +376,16 @@ static int check_module_calltrace(struct stackframe *frame, void *data) { struct walk_stackframe_args *args = data; - /* check the PC first */ - if (within_module_core(frame->pc, args->mod)) - goto err_out; - /* check NIP when the exception stack switching */ if (frame->nip && within_module_core(frame->nip, args->mod)) goto err_out; + if (frame->link && !frame->nip_link_in_same_func && + within_module_core(frame->link, args->mod)) + goto err_out; + if (!frame->is_top_frame || frame->nip_link_in_same_func) { + if (within_module_core(frame->pc, args->mod)) + goto err_out; + } return 0; diff --git a/arch/powerpc/kernel/livepatch_64.c b/arch/powerpc/kernel/livepatch_64.c index a2ec7c8c1bad..b33839b5916a 100644 --- a/arch/powerpc/kernel/livepatch_64.c +++ b/arch/powerpc/kernel/livepatch_64.c @@ -273,13 +273,21 @@ static int klp_check_jump_func(struct stackframe *frame, void *data) struct walk_stackframe_args *args = data; struct klp_func_list *check_funcs = args->check_funcs; - /* check the PC first */ - if (!check_func_list(check_funcs, &args->ret, frame->pc)) - return args->ret; - /* check NIP when the exception stack switching */ if (frame->nip && !check_func_list(check_funcs, &args->ret, frame->nip)) return args->ret; + if (frame->link && !frame->nip_link_in_same_func && + !check_func_list(check_funcs, &args->ret, frame->link)) + return args->ret; + /* + * There are two cases that frame->pc is reliable: + * 1. frame->pc is not in top frame before interrupt; + * 2. nip and link are in same function; + */ + if (!frame->is_top_frame || frame->nip_link_in_same_func) { + if (!check_func_list(check_funcs, &args->ret, frame->pc)) + return args->ret; + } return 0; } @@ -299,10 +307,11 @@ static int do_check_calltrace(struct walk_stackframe_args *args, int (*fn)(struct stackframe *, void *)) { struct task_struct *g, *t; - struct stackframe frame; unsigned long *stack; for_each_process_thread(g, t) { + struct stackframe frame = { 0 }; + if (t == current) { /* * Handle the current carefully on each CPUs, @@ -332,7 +341,6 @@ static int do_check_calltrace(struct walk_stackframe_args *args, frame.sp = (unsigned long)stack; frame.pc = stack[STACK_FRAME_LR_SAVE]; - frame.nip = 0; klp_walk_stackframe(&frame, fn, t, args); if (args->ret) { pr_debug("%s FAILED when %s\n", __func__, @@ -373,13 +381,16 @@ static int check_module_calltrace(struct stackframe *frame, void *data) { struct walk_stackframe_args *args = data; - /* check the PC first */ - if (within_module_core(frame->pc, args->mod)) - goto err_out; - /* check NIP when the exception stack switching */ if (frame->nip && within_module_core(frame->nip, args->mod)) goto err_out; + if (frame->link && !frame->nip_link_in_same_func && + within_module_core(frame->link, args->mod)) + goto err_out; + if (!frame->is_top_frame || frame->nip_link_in_same_func) { + if (within_module_core(frame->pc, args->mod)) + goto err_out; + } return 0; -- 2.25.1

2 1

[PACH OLK-5.10 v5 00/20] psi irq in cgroupv1 and psi fine grained
by Lu Jialin 29 Nov '23

29 Nov '23

Chen Wandun (3): mm: disable psi cgroup v1 by default mm: add config isolation for psi under cgroup v1 psi: dont alloc memory for psi by default Chengming Zhou (3): sched/psi: Fix periodic aggregation shut off sched/psi: Optimize task switch inside shared cgroups again sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ pressure Hao Jia (1): sched/psi: Zero the memory of struct psi_group Johannes Weiner (1): sched/psi: Remove NR_ONCPU task accounting Lu Jialin (11): psi: support irq.pressure under cgroup v1 psi: enable CONFIG_PSI_CGROUP_V1 in openeuler_defconfig psi: update psi irqtime when the irq delta is nozero memcg: split async memcg reclaim from reclaim_high psi: add struct psi_group_ext PSI: Introduce fine grained stall time collect for cgroup reclaim PSI: Introduce avgs and total calculation for cgroup reclaim PSI: Introduce pressure.stat in psi PSI: add more memory fine grained stall tracking in pressure.stat add cpu fine grained stall tracking in pressure.stat PSI: enable CONFIG_PSI_FINE_GRAINED in openeuler_defconfig Suren Baghdasaryan (1): psi: Fix "defined but not used" warnings when CONFIG_PROC_FS=n Documentation/admin-guide/cgroup-v2.rst | 6 + arch/arm64/configs/openeuler_defconfig | 2 + arch/x86/configs/openeuler_defconfig | 2 + block/blk-cgroup.c | 2 +- block/blk-core.c | 2 +- include/linux/cgroup-defs.h | 6 +- include/linux/cgroup.h | 2 +- include/linux/psi.h | 9 +- include/linux/psi_types.h | 99 +++- include/linux/sched.h | 4 + init/Kconfig | 20 + kernel/cgroup/cgroup.c | 64 ++- kernel/sched/core.c | 2 + kernel/sched/cpuacct.c | 6 +- kernel/sched/fair.c | 6 - kernel/sched/psi.c | 691 ++++++++++++++++++++---- kernel/sched/stats.h | 9 + mm/compaction.c | 2 +- mm/filemap.c | 4 +- mm/memcontrol.c | 92 ++-- mm/page_alloc.c | 6 + mm/page_io.c | 3 + mm/vmscan.c | 5 +- 23 files changed, 873 insertions(+), 171 deletions(-) -- 2.34.1

1 20

[PATCH v2 OLK-5.10 0/3] support page eject
by Wupeng Ma 29 Nov '23

29 Nov '23

From: Ma Wupeng <mawupeng1(a)huawei.com> Page ejection is achieved by soft_offline_page and unpoison_memory. A linked list is maintained to log the pfns which are offlined by this module. Only the pfns present in the list is allowed to go online. The follow sysfs is introduced to online/offline page via this module, since there is a internal linked list to record all offlined pages, remove_page is introduce to remove this page from list. - /sys/kernel/page_eject/online_page - /sys/kernel/page_eject/offline_page - /sys/kernel/page_eject/remove_page Changelog since v1: - remove mce related patches temporarily. Ma Wupeng (3): mm/hwpoison: Export symbol soft_offline_page mm: page_eject: Introuduce page ejection config: update defconfig for PAGE_EJECT arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + drivers/ras/Kconfig | 10 + drivers/ras/Makefile | 1 + drivers/ras/page_eject.c | 248 +++++++++++++++++++++++++ mm/memory-failure.c | 1 + 6 files changed, 262 insertions(+) create mode 100644 drivers/ras/page_eject.c -- 2.25.1

2 4

[PATCH v2 OLK-5.10 0/3] support page eject
by Wupeng Ma 29 Nov '23

29 Nov '23

From: Ma Wupeng <mawupeng1(a)huawei.com> Page ejection is achieved by soft_offline_page and unpoison_memory. A linked list is maintained to log the pfns which are offlined by this module. Only the pfns present in the list is allowed to go online. The follow sysfs is introduced to online/offline page via this module, since there is a internal linked list to record all offlined pages, remove_page is introduce to remove this page from list. - /sys/kernel/page_eject/online_page - /sys/kernel/page_eject/offline_page - /sys/kernel/page_eject/remove_page Besides pag eject, machine check safe is enhanced as follow: During soft offline page, page is copied to new page in kernel. If the origin page has UCE, there will lead to kernel panic. In order to solve this problem, use machine check safe to catch this error which can be achieved by using copy_mc_to_kernel to replace copy_page. Signal SIGBUS will be send to user task if this UCE is consumed by this situation to avoid kernel panic. Ma Wupeng (3): mm/hwpoison: Export symbol soft_offline_page mm: page_eject: Introuduce page ejection config: update defconfig for PAGE_EJECT arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + drivers/ras/Kconfig | 10 + drivers/ras/Makefile | 1 + drivers/ras/page_eject.c | 248 +++++++++++++++++++++++++ mm/memory-failure.c | 1 + 6 files changed, 262 insertions(+) create mode 100644 drivers/ras/page_eject.c -- 2.25.1

2 4

[PATCH OLK-5.10] driver/perf: Not enabled ARM64_BRBE by default
by Junhao He 29 Nov '23

29 Nov '23

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8EC9K CVE: NA -------------------------------- The BRBE feature depend on ARMV8 PMU hardware support for branch stack sampling. Not enabled ARM64_BRBE by default. Signed-off-by: Junhao He <hejunhao3(a)huawei.com> --- drivers/perf/Kconfig | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig index 25a577530da4..67ad53cde11f 100644 --- a/drivers/perf/Kconfig +++ b/drivers/perf/Kconfig @@ -133,7 +133,6 @@ config ARM_SPE_PMU config ARM64_BRBE bool "Enable support for Branch Record Buffer Extension (BRBE)" depends on PERF_EVENTS && ARM64 && ARM_PMU - default y help Enable perf support for Branch Record Buffer Extension (BRBE) which records all branches taken in an execution path. This supports some -- 2.33.0

2 1

[PACH OLK-5.10 v5 00/20] psi irq in cgroupv1 and psi fine grained
by Lu Jialin 29 Nov '23

29 Nov '23

Chen Wandun (3): mm: disable psi cgroup v1 by default mm: add config isolation for psi under cgroup v1 psi: dont alloc memory for psi by default Chengming Zhou (3): sched/psi: Fix periodic aggregation shut off sched/psi: Optimize task switch inside shared cgroups again sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ pressure Hao Jia (1): sched/psi: Zero the memory of struct psi_group Johannes Weiner (1): sched/psi: Remove NR_ONCPU task accounting Lu Jialin (11): psi: support irq.pressure under cgroup v1 psi: enable CONFIG_PSI_CGROUP_V1 in openeuler_defconfig psi: update psi irqtime when the irq delta is nozero memcg: split async memcg reclaim from reclaim_high psi: add struct psi_group_ext PSI: Introduce fine grained stall time collect for cgroup reclaim PSI: Introduce avgs and total calculation for cgroup reclaim PSI: Introduce pressure.stat in psi PSI: add more memory fine grained stall tracking in pressure.stat add cpu fine grained stall tracking in pressure.stat PSI: enable CONFIG_PSI_FINE_GRAINED in openeuler_defconfig Suren Baghdasaryan (1): psi: Fix "defined but not used" warnings when CONFIG_PROC_FS=n Documentation/admin-guide/cgroup-v2.rst | 6 + arch/arm64/configs/openeuler_defconfig | 2 + arch/x86/configs/openeuler_defconfig | 2 + block/blk-cgroup.c | 2 +- block/blk-core.c | 2 +- include/linux/cgroup-defs.h | 6 +- include/linux/cgroup.h | 2 +- include/linux/psi.h | 9 +- include/linux/psi_types.h | 99 +++- include/linux/sched.h | 4 + init/Kconfig | 20 + kernel/cgroup/cgroup.c | 64 ++- kernel/sched/core.c | 2 + kernel/sched/cpuacct.c | 6 +- kernel/sched/fair.c | 6 - kernel/sched/psi.c | 691 ++++++++++++++++++++---- kernel/sched/stats.h | 9 + mm/compaction.c | 2 +- mm/filemap.c | 4 +- mm/memcontrol.c | 92 ++-- mm/page_alloc.c | 6 + mm/page_io.c | 3 + mm/vmscan.c | 5 +- 23 files changed, 873 insertions(+), 171 deletions(-) -- 2.34.1

1 20

[PATCH OLK-6.6 v2 00/31] Backport ARM64-ILP32 patches
by Jinjie Ruan 29 Nov '23

29 Nov '23

Backport ILP32 for ARM64 patches to OLK6.6 form openEuler-23.09. Andrew Pinski (3): arm64: rename COMPAT to AARCH32_EL0 arm64: uapi: set __BITS_PER_LONG correctly for ILP32 and LP64 arm64:ilp32: add ARM64_ILP32 to Kconfig Chen Jiahao (4): arm64: fix AUDIT_ARCH_AARCH64ILP32 bug on audit subsystem arm64: fix address limit problem with TASK_SIZE_MAX arm64: set 32-bit compatible TASK_SIZE_MAX to fix U32 libc_write_01 error arm64: fix image size inflation with CONFIG_COMPAT_TASK_SIZE Dave Martin (1): arm64: signal: Make parse_user_sigframe() independent of rt_sigframe layout James Morse (1): ptrace: Add compat PTRACE_{G,S}ETSIGMASK handlers Philipp Tomsich (1): arm64:ilp32: add vdso-ilp32 and use for signal return Xiongfeng Wang (6): arm64: rename functions that reference compat term arm64: secomp: fix the secure computing mode 1 syscall check for ilp32 ilp32: avoid clearing upper 32 bits of syscall return value for ilp32 ilp32: skip ARM erratum 1418040 for ilp32 application arm64: fix abi change caused by ILP32 ilp32: fix compile problem when ARM64_ILP32 and UBSAN are both enabled Yury Norov (14): thread: move thread bits accessors to separated file arm64: ilp32: add documentation on the ILP32 ABI for ARM64 arm64: introduce is_a32_compat_{task,thread} for AArch32 compat arm64: ilp32: add is_ilp32_compat_{task,thread} and TIF_32BIT_AARCH64 arm64: introduce AUDIT_ARCH_AARCH64ILP32 for ilp32 arm64: introduce binfmt_elf32.c arm64: change compat_elf_hwcap and compat_elf_hwcap2 prefix to a32 arm64: ilp32: introduce binfmt_ilp32.c arm64: ilp32: share aarch32 syscall handlers arm64: ilp32: introduce syscall table for ILP32 arm64: signal: share lp64 signal structures and routines to ilp32 arm64: signal32: move ilp32 and aarch32 common code to separated file arm64: ilp32: introduce ilp32-specific sigframe and ucontext arm64: ptrace: handle ptrace_request differently for aarch32 and ilp32 Zhen Lei (1): arm64: replace is_compat_task() with is_ilp32_compat_task() in TASK_SIZE_MAX Documentation/arm64/ilp32.txt | 52 +++ arch/arm64/Kconfig | 26 +- arch/arm64/Makefile | 3 + arch/arm64/include/asm/arch_timer.h | 4 +- arch/arm64/include/asm/compat.h | 18 +- arch/arm64/include/asm/elf.h | 32 +- arch/arm64/include/asm/fpsimd.h | 2 +- arch/arm64/include/asm/ftrace.h | 2 +- arch/arm64/include/asm/hwcap.h | 8 +- arch/arm64/include/asm/is_compat.h | 83 ++++ arch/arm64/include/asm/processor.h | 24 +- arch/arm64/include/asm/ptrace.h | 16 +- arch/arm64/include/asm/seccomp.h | 32 +- arch/arm64/include/asm/signal32.h | 31 +- arch/arm64/include/asm/signal32_common.h | 13 + arch/arm64/include/asm/signal_common.h | 391 +++++++++++++++++ arch/arm64/include/asm/signal_ilp32.h | 23 + arch/arm64/include/asm/syscall.h | 14 +- arch/arm64/include/asm/thread_info.h | 4 +- arch/arm64/include/asm/unistd.h | 5 + arch/arm64/include/asm/vdso.h | 9 + arch/arm64/include/uapi/asm/bitsperlong.h | 9 +- arch/arm64/include/uapi/asm/unistd.h | 15 +- arch/arm64/kernel/Makefile | 8 +- arch/arm64/kernel/armv8_deprecated.c | 14 +- arch/arm64/kernel/asm-offsets.c | 13 +- arch/arm64/kernel/binfmt_elf32.c | 27 ++ arch/arm64/kernel/binfmt_ilp32.c | 89 ++++ arch/arm64/kernel/compat_alignment.c | 2 +- arch/arm64/kernel/cpufeature.c | 36 +- arch/arm64/kernel/cpuinfo.c | 18 +- arch/arm64/kernel/debug-monitors.c | 4 +- arch/arm64/kernel/entry-common.c | 6 +- arch/arm64/kernel/hw_breakpoint.c | 8 +- arch/arm64/kernel/perf_callchain.c | 28 +- arch/arm64/kernel/perf_regs.c | 4 +- arch/arm64/kernel/process.c | 15 +- arch/arm64/kernel/proton-pack.c | 2 +- arch/arm64/kernel/ptrace.c | 38 +- arch/arm64/kernel/signal.c | 397 +++--------------- arch/arm64/kernel/signal32.c | 97 ++--- arch/arm64/kernel/signal32_common.c | 37 ++ arch/arm64/kernel/signal_ilp32.c | 67 +++ arch/arm64/kernel/sys32.c | 104 +---- arch/arm64/kernel/sys32_common.c | 106 +++++ arch/arm64/kernel/sys_compat.c | 12 +- arch/arm64/kernel/sys_ilp32.c | 82 ++++ arch/arm64/kernel/syscall.c | 37 +- arch/arm64/kernel/traps.c | 5 +- arch/arm64/kernel/vdso-ilp32/.gitignore | 2 + arch/arm64/kernel/vdso-ilp32/Makefile | 111 +++++ arch/arm64/kernel/vdso-ilp32/vdso-ilp32.S | 22 + arch/arm64/kernel/vdso-ilp32/vdso-ilp32.lds.S | 88 ++++ arch/arm64/kernel/vdso.c | 46 +- arch/arm64/mm/fault.c | 2 +- include/linux/sched.h | 1 + include/linux/thread_bits.h | 87 ++++ include/linux/thread_info.h | 73 +--- kernel/ptrace.c | 52 ++- 59 files changed, 1790 insertions(+), 766 deletions(-) create mode 100644 Documentation/arm64/ilp32.txt create mode 100644 arch/arm64/include/asm/is_compat.h create mode 100644 arch/arm64/include/asm/signal32_common.h create mode 100644 arch/arm64/include/asm/signal_common.h create mode 100644 arch/arm64/include/asm/signal_ilp32.h create mode 100644 arch/arm64/kernel/binfmt_elf32.c create mode 100644 arch/arm64/kernel/binfmt_ilp32.c create mode 100644 arch/arm64/kernel/signal32_common.c create mode 100644 arch/arm64/kernel/signal_ilp32.c create mode 100644 arch/arm64/kernel/sys32_common.c create mode 100644 arch/arm64/kernel/sys_ilp32.c create mode 100644 arch/arm64/kernel/vdso-ilp32/.gitignore create mode 100644 arch/arm64/kernel/vdso-ilp32/Makefile create mode 100644 arch/arm64/kernel/vdso-ilp32/vdso-ilp32.S create mode 100644 arch/arm64/kernel/vdso-ilp32/vdso-ilp32.lds.S create mode 100644 include/linux/thread_bits.h -- 2.34.1

1 31

[openEuler-20.03-LTS-SP4 PATCH] arm64: cpufeature: Extract capped perfmon fields
by Kunkun Jiang 29 Nov '23

29 Nov '23

From: zhangyuan <zhangyuan162(a)huawei.com> KVM: arm64: limit PMU version to PMUv3 for ARMv8.1 Backport patches from upstream to fix the issue of PMU. Signed-off-by: Yuan Zhang <zhangyuan162(a)huawei.com> --- kernel.spec | 5 +- ...eature-Extract-capped-perfmon-fields.patch | 80 +++++++++++++++++ ...mit-PMU-version-to-PMUv3-for-ARMv8.1.patch | 90 +++++++++++++++++++ series.conf | 2 + 4 files changed, 176 insertions(+), 1 deletion(-) create mode 100644 patches/0758-arm64-cpufeature-Extract-capped-perfmon-fields.patch create mode 100644 patches/0759-KVM-arm64-limit-PMU-version-to-PMUv3-for-ARMv8.1.patch diff --git a/kernel.spec b/kernel.spec index 0ddbd93..7a06c2c 100644 --- a/kernel.spec +++ b/kernel.spec @@ -32,7 +32,7 @@ Name: kernel Version: 4.19.90 -Release: %{hulkrelease}.0252 +Release: %{hulkrelease}.0253 Summary: Linux Kernel License: GPLv2 URL: http://www.kernel.org/ @@ -849,6 +849,9 @@ fi %endif %changelog +* Tue Nov 28 2023 Yuan Zhang <zhangyuan162(a)huawei.com> - 4.19.90-2311.5.0.0253 +- arm64: cpufeature: Extract capped perfmon fields +- KVM: arm64: limit PMU version to PMUv3 for ARMv8.1 * Tue Nov 28 2023 Li Xiaodong <lixiaodong67(a)huawei.com> - 4.19.90-2311.5.0.0252 - !3000 [openEuler-1.0-LTS] add Phytium drivers CONFIG diff --git a/patches/0758-arm64-cpufeature-Extract-capped-perfmon-fields.patch b/patches/0758-arm64-cpufeature-Extract-capped-perfmon-fields.patch new file mode 100644 index 0000000..de1db70 --- /dev/null +++ b/patches/0758-arm64-cpufeature-Extract-capped-perfmon-fields.patch @@ -0,0 +1,80 @@ +From a207eaeac98757a86cce704be6c4e097439503c5 Mon Sep 17 00:00:00 2001 +From: Andrew Murray <andrew.murray(a)arm.com> +Date: Tue, 28 Nov 2023 15:46:32 +0800 +Subject: [PATCH 1/2] arm64: cpufeature: Extract capped perfmon fields + +mainline inclusion +from mainline-v5.7-rc1 +commit 8e35aa642ee4dab01b16cc4b2df59d1936f3b3c2 +category: bugfix +bugzilla: https://gitee.com/openeuler/kernel/issues/I8K8XV +CVE: NA + +Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… + +----------------------------------- + +commit 8e35aa642ee4dab01b16cc4b2df59d1936f3b3c2 upstream. + +When emulating ID registers there is often a need to cap the version +bits of a feature such that the guest will not use features that the +host is not aware of. For example, when KVM mediates access to the PMU +by emulating register accesses. + +Let's add a helper that extracts a performance monitors ID field and +caps the version to a given value. + +Fields that identify the version of the Performance Monitors Extension +do not follow the standard ID scheme, and instead follow the scheme +described in ARM DDI 0487E.a page D13-2825 "Alternative ID scheme used +for the Performance Monitors Extension version". The value 0xF means an +IMPLEMENTATION DEFINED PMU is present, and values 0x0-OxE can be treated +the same as an unsigned field with 0x0 meaning no PMU is present. + +Signed-off-by: Andrew Murray <andrew.murray(a)arm.com> +Reviewed-by: Suzuki K Poulose <suzuki.poulose(a)arm.com> +[Mark: rework to handle perfmon fields] +Signed-off-by: Mark Rutland <mark.rutland(a)arm.com> +Signed-off-by: Will Deacon <will(a)kernel.org> +Signed-off-by: zhangyuan <zhangyuan162(a)huawei.com> +--- + arch/arm64/include/asm/cpufeature.h | 23 +++++++++++++++++++++++ + 1 file changed, 23 insertions(+) + +diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h +index eef5a9c9b823..e2ba573644d5 100644 +--- a/arch/arm64/include/asm/cpufeature.h ++++ b/arch/arm64/include/asm/cpufeature.h +@@ -449,6 +449,29 @@ cpuid_feature_extract_unsigned_field(u64 features, int field) + return cpuid_feature_extract_unsigned_field_width(features, field, 4); + } + ++/* ++ * Fields that identify the version of the Performance Monitors Extension do ++ * not follow the standard ID scheme. See ARM DDI 0487E.a page D13-2825, ++ * "Alternative ID scheme used for the Performance Monitors Extension version". ++ */ ++static inline u64 __attribute_const__ ++cpuid_feature_cap_perfmon_field(u64 features, int field, u64 cap) ++{ ++ u64 val = cpuid_feature_extract_unsigned_field(features, field); ++ u64 mask = GENMASK_ULL(field + 3, field); ++ ++ /* Treat IMPLEMENTATION DEFINED functionality as unimplemented */ ++ if (val == 0xf) ++ val = 0; ++ ++ if (val > cap) { ++ features &= ~mask; ++ features |= (cap << field) & mask; ++ } ++ ++ return features; ++} ++ + static inline u64 arm64_ftr_mask(const struct arm64_ftr_bits *ftrp) + { + return (u64)GENMASK(ftrp->shift + ftrp->width - 1, ftrp->shift); +-- +2.33.0 + diff --git a/patches/0759-KVM-arm64-limit-PMU-version-to-PMUv3-for-ARMv8.1.patch b/patches/0759-KVM-arm64-limit-PMU-version-to-PMUv3-for-ARMv8.1.patch new file mode 100644 index 0000000..1f72f92 --- /dev/null +++ b/patches/0759-KVM-arm64-limit-PMU-version-to-PMUv3-for-ARMv8.1.patch @@ -0,0 +1,90 @@ +From 6a416c5efa3ce99f5f0f7e9c603b42b0e047481c Mon Sep 17 00:00:00 2001 +From: Andrew Murray <andrew.murray(a)arm.com> +Date: Tue, 28 Nov 2023 15:46:33 +0800 +Subject: [PATCH 2/2] KVM: arm64: limit PMU version to PMUv3 for ARMv8.1 + +mainline inclusion +from mainline-v5.7-rc1 +commit c854188ea01062f5a5fd7f05658feb1863774eaa +category: bugfix +bugzilla: https://gitee.com/openeuler/kernel/issues/I8K8XV +CVE: NA + +Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… + +----------------------------------- + +commit c854188ea01062f5a5fd7f05658feb1863774eaa upstream. + +We currently expose the PMU version of the host to the guest via +emulation of the DFR0_EL1 and AA64DFR0_EL1 debug feature registers. +However many of the features offered beyond PMUv3 for 8.1 are not +supported in KVM. Examples of this include support for the PMMIR +registers (added in PMUv3 for ARMv8.4) and 64-bit event counters +added in (PMUv3 for ARMv8.5). + +Let's trap the Debug Feature Registers in order to limit +PMUVer/PerfMon in the Debug Feature Registers to PMUv3 for ARMv8.1 +to avoid unexpected behaviour. + +Both ID_AA64DFR0.PMUVer and ID_DFR0.PerfMon follow the "Alternative ID +scheme used for the Performance Monitors Extension version" where 0xF +means an IMPLEMENTATION DEFINED PMU is implemented, and values 0x0-0xE +are treated as with an unsigned field (with 0x0 meaning no PMU is +present). As we don't expect to expose an IMPLEMENTATION DEFINED PMU, +and our cap is below 0xF, we can treat these fields as unsigned when +applying the cap. + +Signed-off-by: Andrew Murray <andrew.murray(a)arm.com> +Reviewed-by: Suzuki K Poulose <suzuki.poulose(a)arm.com> +[Mark: make field names consistent, use perfmon cap] +Signed-off-by: Mark Rutland <mark.rutland(a)arm.com> +Signed-off-by: Will Deacon <will(a)kernel.org> +[zhangyuan162(a)huawei.com: adjust the context in read_id_reg()] +Signed-off-by: zhangyuan <zhangyuan162(a)huawei.com> +--- + arch/arm64/include/asm/sysreg.h | 6 ++++++ + arch/arm64/kvm/sys_regs.c | 10 ++++++++++ + 2 files changed, 16 insertions(+) + +diff --git a/arch/arm64/include/asm/sysreg.h b/arch/arm64/include/asm/sysreg.h +index de4c31f073bb..041100c28cab 100644 +--- a/arch/arm64/include/asm/sysreg.h ++++ b/arch/arm64/include/asm/sysreg.h +@@ -637,6 +637,12 @@ + #define ID_AA64DFR0_TRACEVER_SHIFT 4 + #define ID_AA64DFR0_DEBUGVER_SHIFT 0 + ++#define ID_AA64DFR0_PMUVER_8_1 0x4 ++ ++#define ID_DFR0_PERFMON_SHIFT 24 ++ ++#define ID_DFR0_PERFMON_8_1 0x4 ++ + #define ID_AA64DFR0_PMSVER_8_2 0x1 + #define ID_AA64DFR0_PMSVER_8_3 0x2 + +diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c +index 25caa9ec95e4..cc824a208684 100644 +--- a/arch/arm64/kvm/sys_regs.c ++++ b/arch/arm64/kvm/sys_regs.c +@@ -1082,6 +1082,16 @@ static u64 read_id_reg(struct kvm_vcpu *vcpu, + kvm_debug("LORegions unsupported for guests, suppressing\n"); + + val &= ~(0xfUL << ID_AA64MMFR1_LOR_SHIFT); ++ } else if (id == SYS_ID_AA64DFR0_EL1) { ++ /* Limit guests to PMUv3 for ARMv8.1 */ ++ val = cpuid_feature_cap_perfmon_field(val, ++ ID_AA64DFR0_PMUVER_SHIFT, ++ ID_AA64DFR0_PMUVER_8_1); ++ } else if (id == SYS_ID_DFR0_EL1) { ++ /* Limit guests to PMUv3 for ARMv8.1 */ ++ val = cpuid_feature_cap_perfmon_field(val, ++ ID_DFR0_PERFMON_SHIFT, ++ ID_DFR0_PERFMON_8_1); + } + + return val; +-- +2.33.0 + diff --git a/series.conf b/series.conf index f5f2522..67b9be0 100644 --- a/series.conf +++ b/series.conf @@ -758,3 +758,5 @@ patches/0754-scsi-hisi_sas_v3_hw-Remove-extra-function-calls-for-.patch patches/0755-config-arm64-Enable-dubugfs-config-of-hisi-sas.patch patches/0756-crypto-hisilicon-Add-value-profile-support-for-kerne.patch patches/0757-Revert-genirq-Increase-the-number-of-IRQ-descriptors.patch +patches/arm64-cpufeature-Extract-capped-perfmon-fields.patch +patches/KVM-arm64-limit-PMU-version-to-PMUv3-for-ARMv8.1.patch -- 2.33.0

1 0

[PATCH OLK-5.10 0/5] SME bugfixes
by Wang ShaoBo 28 Nov '23

28 Nov '23

SME bugfixes Catalin Marinas (1): arm64: Remove the __user annotation for the restore_za_context() argument Christophe JAILLET (1): arm64/ptrace: Clean up error handling path in sve_set_common() Mark Brown (3): arm64: cpufeature: Add missing .field_width for GIC system registers arm64/signal: Flush FPSIMD register state when disabling streaming mode arm64/fpsimd: Sync and zero pad FPSIMD state for streaming SVE arch/arm64/kernel/cpufeature.c | 1 + arch/arm64/kernel/fpsimd.c | 3 ++- arch/arm64/kernel/ptrace.c | 3 ++- arch/arm64/kernel/signal.c | 12 +++++++++++- 4 files changed, 16 insertions(+), 3 deletions(-) -- 2.25.1

2 6

[PATCH OLK-5.10 0/7] support page eject
by Wupeng Ma 28 Nov '23

28 Nov '23

From: Ma Wupeng <mawupeng1(a)huawei.com> Page ejection is achieved by soft_offline_page and unpoison_memory. A linked list is maintained to log the pfns which are offlined by this module. Only the pfns present in the list is allowed to go online. The follow sysfs is introduced to online/offline page via this module, since there is a internal linked list to record all offlined pages, remove_page is introduce to remove this page from list. - /sys/kernel/page_eject/online_page - /sys/kernel/page_eject/offline_page - /sys/kernel/page_eject/remove_page Besides pag eject, machine check safe is enhanced as follow: During soft offline page, page is copied to new page in kernel. If the origin page has UCE, there will lead to kernel panic. In order to solve this problem, use machine check safe to catch this error which can be achieved by using copy_mc_to_kernel to replace copy_page. Signal SIGBUS will be send to user task if this UCE is consumed by this situation to avoid kernel panic. Jiaqi Yan (1): mm/hwpoison: introduce copy_mc_highpage Ma Wupeng (6): mm: introduce copy_mc_highpages mm: arm64: Add migrate_page_copy_mc() mm: Update PF_COREDUMP_MCS to PF_MCS mm/hwpoison: Export symbol soft_offline_page mm: page_eject: Introuduce page ejection mm: page_eject: Add mc support during offline page drivers/ras/Kconfig | 10 ++ drivers/ras/Makefile | 1 + drivers/ras/page_eject.c | 249 +++++++++++++++++++++++++++++++++++++++ fs/coredump.c | 4 +- include/linux/highmem.h | 49 ++++++++ include/linux/sched.h | 2 +- lib/iov_iter.c | 2 +- mm/memory-failure.c | 1 + mm/migrate.c | 63 ++++++++-- 9 files changed, 365 insertions(+), 16 deletions(-) create mode 100644 drivers/ras/page_eject.c -- 2.25.1

2 8

[PATCH OLK-5.10] mpam: Fix uninitialized value
by Yu Liao 28 Nov '23

28 Nov '23

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7ZRRD -------------------------------- Fix the following smatch warning: drivers/acpi/arm64/mpam.c:219 acpi_mpam_parse_version() error: uninitialized symbol 'ret'. Fixes: b1e2f3d84839 ("[Backport] ACPI/MPAM: Adapt to Arm's MPAM ACPI table version 2") Signed-off-by: Yu Liao <liaoyu15(a)huawei.com> --- drivers/acpi/arm64/mpam.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/arm64/mpam.c b/drivers/acpi/arm64/mpam.c index f625f539d50a..9d59652e49f8 100644 --- a/drivers/acpi/arm64/mpam.c +++ b/drivers/acpi/arm64/mpam.c @@ -219,7 +219,7 @@ int __init acpi_mpam_parse_version(void) { struct acpi_table_header *mpam, *pptt; acpi_status status; - int ret; + int ret = -EINVAL; if (!cpus_have_const_cap(ARM64_HAS_MPAM)) return 0; -- 2.25.1

2 1

[PATCH OLK-5.10] Revert "arm64/mpam: Fix mpam corrupt when cpu online"
by Yu Liao 28 Nov '23

28 Nov '23

From: Wang ShaoBo <bobo.shaobowang(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7R0WK ------------------------------------------------- BUG 'sleeping function called from invalid context' reported when setup MPAM driver, it was blamed to bc9e3f9895ef2 ("arm64/mpam: Fix mpam corrupt when cpu online"), which reported a 'Bad PC' BUG, but missing the right conclusion, finally disabling irqs before calling cpuhp_setup_state() may only affect the probability of reproduction. The reason why triggerring 'Bad PC' BUG report is because mpam_enable() is __init type function, and may schedule out after calling __cpuhp_setup_state()->__might_sleep(), so the space of mpam_enable() might be freed after scheduling back. As we have changed mpam_enable() to non-init type function, we can revert commit bc9e3f9895ef2 directly, to solve these both two problems. Fixes: bc9e3f9895ef2 ("arm64/mpam: Fix mpam corrupt when cpu online") Signed-off-by: Wang ShaoBo <bobo.shaobowang(a)huawei.com> Signed-off-by: Yu Liao <liaoyu15(a)huawei.com> --- arch/arm64/kernel/mpam/mpam_device.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/arm64/kernel/mpam/mpam_device.c b/arch/arm64/kernel/mpam/mpam_device.c index 96288f66f619..b3de7f518423 100644 --- a/arch/arm64/kernel/mpam/mpam_device.c +++ b/arch/arm64/kernel/mpam/mpam_device.c @@ -621,11 +621,9 @@ static void mpam_enable(struct work_struct *work) pr_err("Failed to setup/init resctrl\n"); mutex_unlock(&mpam_devices_lock); - local_irq_disable(); mpam_cpuhp_state = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "mpam:online", mpam_cpu_online, mpam_cpu_offline); - local_irq_enable(); if (mpam_cpuhp_state <= 0) pr_err("Failed to re-register 'dyn' cpuhp callbacks"); mutex_unlock(&mpam_cpuhp_lock); -- 2.25.1

2 1

[PATCH OLK-5.10] posix-cpu-timers: Implement timer_wait_running callback
by Xiongfeng Wang 28 Nov '23

28 Nov '23

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7EQ8B -------------------------------- For posix CPU timers, the .timer_wait_running() callback is missing, which cause WARN_ON() in timer_wait_running(). Commit f7abf14f0001 ("posix-cpu-timers: Implement the missing timer_wait_running callback") solved the problem. But it introduce a kabi broken which is hard to fix. The commit also describe a livelock in preemptible kernel. To fix the livelock, we add cond_resched() in the retry loop. *** below is from origin commit merge tag The reason is that for CONFIG_POSIX_CPU_TIMERS_TASK_WORK enabled systems there is a livelock issue independent of RT. CONFIG_POSIX_CPU_TIMERS_TASK_WORK=y moves the expiry of POSIX CPU timers out from hard interrupt context to task work, which is handled before returning to user space or to a VM. The expiry mechanism moves the expired timers to a stack local list head with sighand lock held. Once sighand is dropped the task can be preempted and a task which wants to delete a timer will spin-wait until the expiry task is scheduled back in. In the worst case this will end up in a livelock when the preempting task and the expiry task are pinned on the same CPU. Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> --- kernel/time/posix-cpu-timers.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index c97bc3e3b210..578d0ebedb67 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -1446,6 +1446,9 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, spin_lock_irq(&timer.it_lock); error = posix_cpu_timer_del(&timer); spin_unlock_irq(&timer.it_lock); +#ifdef CONFIG_POSIX_CPU_TIMERS_TASK_WORK + cond_resched(); +#endif } if ((it.it_value.tv_sec | it.it_value.tv_nsec) == 0) { @@ -1546,6 +1549,18 @@ static int thread_cpu_timer_create(struct k_itimer *timer) return posix_cpu_timer_create(timer); } +static void posix_cpu_timer_wait_running(struct k_itimer *timr) +{ +#ifdef CONFIG_POSIX_CPU_TIMERS_TASK_WORK + rcu_read_unlock(); + cond_resched(); + rcu_read_lock(); +#else + cpu_relax(); +#endif +} + + const struct k_clock clock_posix_cpu = { .clock_getres = posix_cpu_clock_getres, .clock_set = posix_cpu_clock_set, @@ -1556,6 +1571,7 @@ const struct k_clock clock_posix_cpu = { .timer_del = posix_cpu_timer_del, .timer_get = posix_cpu_timer_get, .timer_rearm = posix_cpu_timer_rearm, + .timer_wait_running = posix_cpu_timer_wait_running, }; const struct k_clock clock_process = { -- 2.20.1

2 1

[PATCH OLK-6.6 0/6] Handle more faults under the VMA lock
by Tong Tiangen 28 Nov '23

28 Nov '23

At this point, we're handling the majority of file-backed page faults under the VMA lock, using the ->map_pages entry point. This patch set attempts to expand that for the following siutations: - We have to do a read. This could be because we've hit the point in the readahead window where we need to kick off the next readahead, or because the page is simply not present in cache. - We're handling a write fault. Most applications don't do I/O by writes to shared mmaps for very good reasons, but some do, and it'd be nice to not make that slow unnecessarily. - We're doing a COW of a private mapping (both PTE already present and PTE not-present). These are two different codepaths and I handle both of them in this patch set. There is no support in this patch set for drivers to mark themselves as being VMA lock friendly; they could implement the ->map_pages vm_operation, but if they do, they would be the first. This is probably something we want to change at some point in the future, and I've marked where to make that change in the code. There is very little performance change in the benchmarks we've run; mostly because the vast majority of page faults are handled through the other paths. I still think this patch series is useful for workloads that may take these paths more often, and just for cleaning up the fault path in general (it's now clearer why we have to retry in these cases). Matthew Wilcox (Oracle) (6): mm: make lock_folio_maybe_drop_mmap() VMA lock aware mm: call wp_page_copy() under the VMA lock mm: handle shared faults under the VMA lock mm: handle COW faults under the VMA lock mm: handle read faults under the VMA lock mm: handle write faults to RO pages under the VMA lock mm/filemap.c | 13 ++++---- mm/memory.c | 93 ++++++++++++++++++++++++++++++++-------------------- 2 files changed, 65 insertions(+), 41 deletions(-) -- 2.25.1

2 7

[PATCH OLK-5.10 00/80] sme backport from mainline
by Wang ShaoBo 28 Nov '23

28 Nov '23

sme backport from mainline Alexandru Elisei (1): arm64: Do not trap PMSNEVFR_EL1 Marc Zyngier (10): KVM: arm64: Let vcpu_sve_pffr() handle HYP VAs KVM: arm64: Provide KVM's own save/restore SVE primitives KVM: arm64: Use {read,write}_sysreg_el1 to access ZCR_EL1 KVM: arm64: Introduce vcpu_sve_vq() helper arm64: sve: Provide a conditional update accessor for ZCR_ELx KVM: arm64: Map SVE context at EL2 when available KVM: arm64: Rework SVE host-save/guest-restore KVM: arm64: Save guest's ZCR_EL1 before saving the FPSIMD state KVM: arm64: Save/restore SVE state for nVHE KVM: arm64: Always start with clearing SME flag on load Mark Brown (65): arm64/sve: Remove redundant system_supports_sve() tests arm64/sve: Add compile time checks for SVE hooks in generic functions arm64/sve: Rework SVE access trap to convert state in registers arm64/sve: Split _sve_flush macro into separate Z and predicate flushes arm64/sve: Skip flushing Z registers with 128 bit vectors arm64/sve: Use the sve_flush macros in sve_load_from_fpsimd_state() arm64/sve: Remove sve_load_from_fpsimd_state() arm64/sve: Better handle failure to allocate SVE register storage arm64/fp: Reindent fpsimd_save() arm64/sve: Make access to FFR optional arm64/sve: Rename find_supported_vector_length() arm64/sve: Use accessor functions for vector lengths in thread_struct arm64/sve: Put system wide vector length information into structs arm64/sve: Explicitly load vector length when restoring SVE state arm64/sve: Track vector lengths for tasks in an array arm64/sve: Make sysctl interface for SVE reusable by SME arm64/sve: Generalise vector length configuration prctl() for SME arm64/sve: Minor clarification of ABI documentation arm64: cpufeature: Always specify and use a field width for capabilities arm64/sme: Provide ABI documentation for SME arm64/sme: System register and exception syndrome definitions arm64/sme: Manually encode SME instructions arm64: Disable fine grained traps on boot arm64/sme: Early CPU setup for SME arm64/sme: Basic enumeration support arm64/sme: Identify supported SME vector lengths at boot arm64/sme: Implement sysctl to set the default vector length arm64/sme: Implement vector length configuration prctl()s arm64/sme: Implement support for TPIDR2 arm64/sme: Implement SVCR context switching arm64/sme: Implement streaming SVE context switching arm64/sme: Implement ZA context switching arm64/sme: Implement traps and syscall handling for SME arm64/sme: Disable ZA and streaming mode when handling signals arm64/sme: Implement streaming SVE signal handling arm64/sme: Implement ZA signal handling arm64/sme: Implement ptrace support for streaming mode SVE registers arm64/sme: Add ptrace support for ZA arm64/sme: Disable streaming mode and ZA when flushing CPU state arm64/sme: Save and restore streaming mode over EFI runtime calls arm64/sme: Provide Kconfig for SME arm64/sme: Add ID_AA64SMFR0_EL1 to __read_sysreg_by_encoding() arm64/sme: More sensibly define the size for the ZA register set KVM: arm64: Hide SME system registers from guests KVM: arm64: Trap SME usage in guest KVM: arm64: Handle SME host state when running guests arm64/fp: Make SVE and SME length register definition match architecture arm64/fp: Rename SVE and SME LEN field name to _WIDTH arm64/sme: Drop SYS_ from SMIDR_EL1 defines arm64/sme: Standardise bitfield names for SVCR arm64/sme: Remove _EL0 from name of SVCR - FIXME sysreg.h arm64/sme: Fix tests for 0b1111 value ID registers arm64/sme: Fix SVE/SME typo in ABI documentation arm64/sme: Fix EFI save/restore arm64/sme: Don't flush SVE register state when allocating SME storage arm64/sme: Don't flush SVE register state when handling SME traps arm64/ptrace: Don't clear calling process' TIF_SME on OOM arm64/signal: Always allocate SVE signal frames on SME only systems arm64/sme: Don't use streaming mode to probe the maximum SME VL arm64/sme: Optimise SME exit on syscall entry arm64/fpsimd: Ensure SME storage is allocated after SVE VL changes arm64/sme: Set new vector length before reallocating arm64/fpsimd: Clear SME state in the target task when setting the VL arm64/fpsimd: Sync FPSIMD state with SVE for SME only systems arm64/ptrace: Ensure that SME is set up for target when writing SSVE state Wan Jiabing (1): arm64/sme: Fix NULL check after kzalloc Will Deacon (1): arm64: sme: Use STR P to clear FFR context field in streaming SVE mode Xiaofei Tan (1): arm64: sve: Provide sve_cond_update_zcr_vq fallback when !ARM64_SVE Zenghui Yu (1): arm64: ptrace: Use ARM64_SME to guard the SME register enumerations Documentation/arm64/elf_hwcaps.rst | 34 + Documentation/arm64/index.rst | 1 + Documentation/arm64/sme.rst | 428 ++++++++++ Documentation/arm64/sve.rst | 72 +- arch/arm64/Kconfig | 11 + arch/arm64/include/asm/cpu.h | 4 + arch/arm64/include/asm/cpucaps.h | 2 + arch/arm64/include/asm/cpufeature.h | 25 + arch/arm64/include/asm/esr.h | 13 +- arch/arm64/include/asm/exception.h | 1 + arch/arm64/include/asm/fpsimd.h | 266 ++++++- arch/arm64/include/asm/fpsimdmacros.h | 108 ++- arch/arm64/include/asm/hwcap.h | 8 + arch/arm64/include/asm/kvm_arm.h | 1 + arch/arm64/include/asm/kvm_host.h | 14 +- arch/arm64/include/asm/kvm_hyp.h | 2 + arch/arm64/include/asm/processor.h | 78 +- arch/arm64/include/asm/signal_common.h | 19 +- arch/arm64/include/asm/sysreg.h | 79 +- arch/arm64/include/asm/thread_info.h | 4 +- arch/arm64/include/uapi/asm/hwcap.h | 8 + arch/arm64/include/uapi/asm/ptrace.h | 69 +- arch/arm64/include/uapi/asm/sigcontext.h | 55 +- arch/arm64/kernel/cpufeature.c | 292 +++++-- arch/arm64/kernel/cpuinfo.c | 13 + arch/arm64/kernel/entry-common.c | 11 + arch/arm64/kernel/entry-fpsimd.S | 72 +- arch/arm64/kernel/fpsimd.c | 961 ++++++++++++++++++----- arch/arm64/kernel/head.S | 87 ++ arch/arm64/kernel/process.c | 45 +- arch/arm64/kernel/ptrace.c | 380 +++++++-- arch/arm64/kernel/signal.c | 179 ++++- arch/arm64/kernel/syscall.c | 25 +- arch/arm64/kernel/traps.c | 1 + arch/arm64/kvm/fpsimd.c | 70 +- arch/arm64/kvm/guest.c | 6 +- arch/arm64/kvm/hyp/fpsimd.S | 12 + arch/arm64/kvm/hyp/include/hyp/switch.h | 75 +- arch/arm64/kvm/hyp/nvhe/switch.c | 38 +- arch/arm64/kvm/hyp/vhe/switch.c | 10 +- arch/arm64/kvm/reset.c | 14 +- arch/arm64/kvm/sys_regs.c | 8 +- include/uapi/linux/elf.h | 2 + include/uapi/linux/prctl.h | 9 + kernel/sys.c | 12 + 45 files changed, 3148 insertions(+), 476 deletions(-) create mode 100644 Documentation/arm64/sme.rst -- 2.25.1

1 80

[PATCH OLK-6.6 0/8] memcg_qos and memcg_async_reclaim
by Lu Jialin 28 Nov '23

28 Nov '23

Lu Jialin (8): cgroup: Export cgroup.kill from cgroupv2 to cgroupv1 memcg: Export memcg.{min/low} from cgroupv2 to cgroupv1 memcg: Export memcg.high from cgroupv2 to cgroupv1 memcg: Export memory.events and memory.events.local from cgroupv2 to cgroupv1 memcg: Fix inconsistent oom event behavior for OOM_MEMCG_KILL memcg: enable memcg async reclaim memcg: export high_async_ratio to userland memcg: Introduce CONFIG_MEMCG_V1_THRESHOLD_QOS arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + include/linux/memcontrol.h | 24 +- init/Kconfig | 6 + kernel/cgroup/cgroup-internal.h | 3 + kernel/cgroup/cgroup-v1.c | 5 + kernel/cgroup/cgroup.c | 4 +- mm/memcontrol.c | 378 ++++++++++++++++++------- 8 files changed, 310 insertions(+), 112 deletions(-) -- 2.34.1

2 9

[PATCH OLK-5.10 00/80] sme backport from mainline
by Wang ShaoBo 28 Nov '23

28 Nov '23

sme backport from mainline Alexandru Elisei (1): arm64: Do not trap PMSNEVFR_EL1 Marc Zyngier (10): KVM: arm64: Let vcpu_sve_pffr() handle HYP VAs KVM: arm64: Provide KVM's own save/restore SVE primitives KVM: arm64: Use {read,write}_sysreg_el1 to access ZCR_EL1 KVM: arm64: Introduce vcpu_sve_vq() helper arm64: sve: Provide a conditional update accessor for ZCR_ELx KVM: arm64: Map SVE context at EL2 when available KVM: arm64: Rework SVE host-save/guest-restore KVM: arm64: Save guest's ZCR_EL1 before saving the FPSIMD state KVM: arm64: Save/restore SVE state for nVHE KVM: arm64: Always start with clearing SME flag on load Mark Brown (65): arm64/sve: Remove redundant system_supports_sve() tests arm64/sve: Add compile time checks for SVE hooks in generic functions arm64/sve: Rework SVE access trap to convert state in registers arm64/sve: Split _sve_flush macro into separate Z and predicate flushes arm64/sve: Skip flushing Z registers with 128 bit vectors arm64/sve: Use the sve_flush macros in sve_load_from_fpsimd_state() arm64/sve: Remove sve_load_from_fpsimd_state() arm64/sve: Better handle failure to allocate SVE register storage arm64/fp: Reindent fpsimd_save() arm64/sve: Make access to FFR optional arm64/sve: Rename find_supported_vector_length() arm64/sve: Use accessor functions for vector lengths in thread_struct arm64/sve: Put system wide vector length information into structs arm64/sve: Explicitly load vector length when restoring SVE state arm64/sve: Track vector lengths for tasks in an array arm64/sve: Make sysctl interface for SVE reusable by SME arm64/sve: Generalise vector length configuration prctl() for SME arm64/sve: Minor clarification of ABI documentation arm64: cpufeature: Always specify and use a field width for capabilities arm64/sme: Provide ABI documentation for SME arm64/sme: System register and exception syndrome definitions arm64/sme: Manually encode SME instructions arm64: Disable fine grained traps on boot arm64/sme: Early CPU setup for SME arm64/sme: Basic enumeration support arm64/sme: Identify supported SME vector lengths at boot arm64/sme: Implement sysctl to set the default vector length arm64/sme: Implement vector length configuration prctl()s arm64/sme: Implement support for TPIDR2 arm64/sme: Implement SVCR context switching arm64/sme: Implement streaming SVE context switching arm64/sme: Implement ZA context switching arm64/sme: Implement traps and syscall handling for SME arm64/sme: Disable ZA and streaming mode when handling signals arm64/sme: Implement streaming SVE signal handling arm64/sme: Implement ZA signal handling arm64/sme: Implement ptrace support for streaming mode SVE registers arm64/sme: Add ptrace support for ZA arm64/sme: Disable streaming mode and ZA when flushing CPU state arm64/sme: Save and restore streaming mode over EFI runtime calls arm64/sme: Provide Kconfig for SME arm64/sme: Add ID_AA64SMFR0_EL1 to __read_sysreg_by_encoding() arm64/sme: More sensibly define the size for the ZA register set KVM: arm64: Hide SME system registers from guests KVM: arm64: Trap SME usage in guest KVM: arm64: Handle SME host state when running guests arm64/fp: Make SVE and SME length register definition match architecture arm64/fp: Rename SVE and SME LEN field name to _WIDTH arm64/sme: Drop SYS_ from SMIDR_EL1 defines arm64/sme: Standardise bitfield names for SVCR arm64/sme: Remove _EL0 from name of SVCR - FIXME sysreg.h arm64/sme: Fix tests for 0b1111 value ID registers arm64/sme: Fix SVE/SME typo in ABI documentation arm64/sme: Fix EFI save/restore arm64/sme: Don't flush SVE register state when allocating SME storage arm64/sme: Don't flush SVE register state when handling SME traps arm64/ptrace: Don't clear calling process' TIF_SME on OOM arm64/signal: Always allocate SVE signal frames on SME only systems arm64/sme: Don't use streaming mode to probe the maximum SME VL arm64/sme: Optimise SME exit on syscall entry arm64/fpsimd: Ensure SME storage is allocated after SVE VL changes arm64/sme: Set new vector length before reallocating arm64/fpsimd: Clear SME state in the target task when setting the VL arm64/fpsimd: Sync FPSIMD state with SVE for SME only systems arm64/ptrace: Ensure that SME is set up for target when writing SSVE state Wan Jiabing (1): arm64/sme: Fix NULL check after kzalloc Will Deacon (1): arm64: sme: Use STR P to clear FFR context field in streaming SVE mode Xiaofei Tan (1): arm64: sve: Provide sve_cond_update_zcr_vq fallback when !ARM64_SVE Zenghui Yu (1): arm64: ptrace: Use ARM64_SME to guard the SME register enumerations Documentation/arm64/elf_hwcaps.rst | 34 + Documentation/arm64/index.rst | 1 + Documentation/arm64/sme.rst | 428 ++++++++++ Documentation/arm64/sve.rst | 72 +- arch/arm64/Kconfig | 11 + arch/arm64/include/asm/cpu.h | 4 + arch/arm64/include/asm/cpucaps.h | 2 + arch/arm64/include/asm/cpufeature.h | 25 + arch/arm64/include/asm/esr.h | 13 +- arch/arm64/include/asm/exception.h | 1 + arch/arm64/include/asm/fpsimd.h | 266 ++++++- arch/arm64/include/asm/fpsimdmacros.h | 108 ++- arch/arm64/include/asm/hwcap.h | 8 + arch/arm64/include/asm/kvm_arm.h | 1 + arch/arm64/include/asm/kvm_host.h | 14 +- arch/arm64/include/asm/kvm_hyp.h | 2 + arch/arm64/include/asm/processor.h | 78 +- arch/arm64/include/asm/signal_common.h | 19 +- arch/arm64/include/asm/sysreg.h | 79 +- arch/arm64/include/asm/thread_info.h | 4 +- arch/arm64/include/uapi/asm/hwcap.h | 8 + arch/arm64/include/uapi/asm/ptrace.h | 69 +- arch/arm64/include/uapi/asm/sigcontext.h | 55 +- arch/arm64/kernel/cpufeature.c | 292 +++++-- arch/arm64/kernel/cpuinfo.c | 13 + arch/arm64/kernel/entry-common.c | 11 + arch/arm64/kernel/entry-fpsimd.S | 72 +- arch/arm64/kernel/fpsimd.c | 961 ++++++++++++++++++----- arch/arm64/kernel/head.S | 87 ++ arch/arm64/kernel/process.c | 45 +- arch/arm64/kernel/ptrace.c | 380 +++++++-- arch/arm64/kernel/signal.c | 179 ++++- arch/arm64/kernel/syscall.c | 25 +- arch/arm64/kernel/traps.c | 1 + arch/arm64/kvm/fpsimd.c | 70 +- arch/arm64/kvm/guest.c | 6 +- arch/arm64/kvm/hyp/fpsimd.S | 12 + arch/arm64/kvm/hyp/include/hyp/switch.h | 75 +- arch/arm64/kvm/hyp/nvhe/switch.c | 38 +- arch/arm64/kvm/hyp/vhe/switch.c | 10 +- arch/arm64/kvm/reset.c | 14 +- arch/arm64/kvm/sys_regs.c | 8 +- include/uapi/linux/elf.h | 2 + include/uapi/linux/prctl.h | 9 + kernel/sys.c | 12 + 45 files changed, 3148 insertions(+), 476 deletions(-) create mode 100644 Documentation/arm64/sme.rst -- 2.25.1

1 80