- Kernel - mailweb.openeuler.org

[PATCH OLK-6.6 0/2] arm64: errata: add option to disable cache readunique prefetch on HIP08
by Yang Yingliang 07 Dec '23

07 Dec '23

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'. Kai Shen (1): arm64: errata: add option to disable cache readunique prefetch on HIP08 Xie XiuQi (1): arm64: errata: enable HISILICON_ERRATUM_HIP08_RU_PREFETCH arch/arm64/Kconfig | 18 +++++++++ arch/arm64/configs/openeuler_defconfig | 2 + arch/arm64/kernel/cpu_errata.c | 56 ++++++++++++++++++++++++++ arch/arm64/tools/cpucaps | 1 + 4 files changed, 77 insertions(+) -- 2.25.1

2 3

[PATCH openEuler-1.0-LTS] x86/cpu: Fix AMD erratum #1485 on Zen4-based CPUs
by liwei 07 Dec '23

07 Dec '23

From: "Borislav Petkov (AMD)" <bp(a)alien8.de> stable inclusion from stable-v4.19.297 commit 05868a16aab502d48a05a9528f23002fbf830335 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8MCB5 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=… --------------------------- commit f454b18e07f518bcd0c05af17a2239138bff52de upstream. Fix erratum #1485 on Zen4 parts where running with STIBP disabled can cause an #UD exception. The performance impact of the fix is negligible. Reported-by: René Rebe <rene(a)exactcode.de> Signed-off-by: Borislav Petkov (AMD) <bp(a)alien8.de> Tested-by: René Rebe <rene(a)exactcode.de> Cc: <stable(a)kernel.org> Link: https://lore.kernel.org/r/D99589F4-BC5D-430B-87B2-72C20370CF57@exactcode.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: arch/x86/kernel/cpu/amd.c Signed-off-by: liwei <liwei728(a)huawei.com> --- arch/x86/include/asm/msr-index.h | 4 ++++ arch/x86/kernel/cpu/amd.c | 9 +++++++++ 2 files changed, 13 insertions(+) diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 7a115a34f9b2..3948ba5010a2 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -442,6 +442,10 @@ #define MSR_AMD64_VIRT_SPEC_CTRL 0xc001011f +/* Zen4 */ +#define MSR_ZEN4_BP_CFG 0xc001102e +#define MSR_ZEN4_BP_CFG_SHARED_BTB_FIX_BIT 5 + /* Fam 17h MSRs */ #define MSR_F17H_IRPERF 0xc00000e9 diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 002aa4bd35cf..9a2257a64562 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -27,6 +27,7 @@ static const int amd_erratum_383[]; static const int amd_erratum_400[]; static const int amd_zenbleed[]; static const int amd_div0[]; +static const int amd_erratum_1485[]; static bool cpu_has_amd_erratum(struct cpuinfo_x86 *cpu, const int *erratum); /* @@ -1044,6 +1045,10 @@ static void init_amd(struct cpuinfo_x86 *c) pr_notice_once("AMD Zen1 DIV0 bug detected. Disable SMT for full protection.\n"); setup_force_cpu_bug(X86_BUG_DIV0); } + + if (!cpu_has(c, X86_FEATURE_HYPERVISOR) && + cpu_has_amd_erratum(c, amd_erratum_1485)) + msr_set_bit(MSR_ZEN4_BP_CFG, MSR_ZEN4_BP_CFG_SHARED_BTB_FIX_BIT); } #ifdef CONFIG_X86_32 @@ -1177,6 +1182,10 @@ static const int amd_zenbleed[] = AMD_MODEL_RANGE(0x17, 0x90, 0x0, 0x91, 0xf), AMD_MODEL_RANGE(0x17, 0xa0, 0x0, 0xaf, 0xf)); +static const int amd_erratum_1485[] = + AMD_LEGACY_ERRATUM(AMD_MODEL_RANGE(0x19, 0x10, 0x0, 0x1f, 0xf), + AMD_MODEL_RANGE(0x19, 0x60, 0x0, 0xaf, 0xf)); + static const int amd_div0[] = AMD_LEGACY_ERRATUM(AMD_MODEL_RANGE(0x17, 0x00, 0x0, 0x2f, 0xf), AMD_MODEL_RANGE(0x17, 0x50, 0x0, 0x5f, 0xf)); -- 2.25.1

2 1

[PATCH OLK-5.10 0/6] Make the rcache depot scale better
by Zhang Zekun 07 Dec '23

07 Dec '23

Recently we have meet a problem when running FIO test. In our kunpeng server with 320 cores, about 80% cpus comes to- 100% usage, and soft lockup appears in the kernel message, which shows it stuck in __alloc_and_insert_iova_range(). Both the call trace and high cpu occupy rate implys that the iova_rcache behaves poorly to alloc iovas. A similar problem has been addressed in early months this year, the solution is just to enlarge the MAX_GLOBAL_MAGS to 128, but it requires to growth with the cpu cores, and it is hard to set a accurate val for a specific machine. So, it is better to use solutions from the linux community, use a list to replace the array of iova_rcache->depot. It will push iova_magazine to depot list when the local cpu_rcache is full, and use schedul_delayed_work() to free it after 100ms. The minimum length of the depot list is the num of online_cpus(). The performance of this patch shows good. This patch set use a lot of timers to record the iova_magazine, and the delay of timers in system may increase when undering heavy work load, but it should not cause problem as timers itself allows a inaccurate delay for a range. We need to merge "iommu/iova: change IOVA_MAG_SIZE to 127 to save memory" first to resolve a compile error: error: static assertion failed: "!(sizeof(struct iova_magazine) & (sizeof(struct iova_magazine) - 1))" Feng Tang (1): iommu/iova: change IOVA_MAG_SIZE to 127 to save memory Zhang Zekun (5): Revert "iommu/iova: move IOVA_MAX_GLOBAL_MAGS outside of IOMMU_SUPPORT" Revert "config: enable set the max iova mag size to 128" Revert "iommu/iova: increase the iova_rcache depot max size to 128" iommu/iova: Make the rcache depot scale better iommu/iova: Manage the depot list size arch/arm64/configs/openeuler_defconfig | 1 - drivers/iommu/Kconfig | 10 ---- drivers/iommu/iova.c | 79 +++++++++++++++++++------- include/linux/iova.h | 7 ++- 4 files changed, 63 insertions(+), 34 deletions(-) -- 2.17.1

2 7

[PATCH openEuler-1.0-LTS 00/23] linux-4.19.y inclusion
by Zhang Changzhong 07 Dec '23

07 Dec '23

Chen Yu (1): genirq/matrix: Exclude managed interrupts in irq_matrix_allocated() Christophe JAILLET (1): ACPI: sysfs: Fix create_pnp_modalias() and create_of_modalias() Eric Dumazet (5): xfrm: fix a data-race in xfrm_gen_index() tcp: tsq: relax tcp_small_queue_check() when rtx queue contains a single skb tcp_metrics: add missing barriers on delete tcp_metrics: properly set tp->snd_ssthresh in tcp_init_metrics() tcp_metrics: do not create an entry from tcp_init_metrics() Gou Hao (1): ext4: move 'ix' sanity check to corrent position Jan Kara (1): quota: Fix slow quotaoff Jeff Layton (1): overlayfs: set ctime when setting mtime and atime Jorge Sanjuan Garcia (1): mcb: remove is_added flag from mcb_device struct Kuniyuki Iwashima (1): dccp/tcp: Call security_inet_conn_request() after setting IPv6 addresses. Marc Kleine-Budde (3): can: dev: move driver related infrastructure into separate subdir can: dev: can_restart(): don't crash kernel if carrier is OK can: dev: can_restart(): fix race condition between controller restart and netif_carrier_on() Michal Koutný (1): cgroup: Remove duplicates in cgroup v1 tasks file Neal Cardwell (1): tcp: fix excessive TLP and RACK timeouts from HZ rounding Peter Zijlstra (1): sched,idle,rcu: Push rcu_idle deeper into the idle path Reuben Hawkins (1): vfs: fix readahead(2) on block devices Sunil V L (1): ACPI: irq: Fix incorrect return value in acpi_register_gsi() Thomas Gleixner (1): hrtimers: Push pending hrtimers away from outgoing CPU earlier Yan Zhai (1): ipv6: avoid atomic fragment on GSO packets Yu Liao (1): cpu/hotplug: fix kabi breakage in enum cpuhp_state -- 2.34.1

1 23

[PATCH OLK-6.6 0/2] Open CONFIG_AARCH32_EL0 and close CONFIG_ARM64_ILP32
by Jinjie Ruan 07 Dec '23

07 Dec '23

Update script for check & update openeuler_defconfig. And open CONFIG_AARCH32_EL0 and close CONFIG_ARM64_ILP32. Jinjie Ruan (2): kconfig: Update script for check & update openeuler_defconfig config: Open CONFIG_AARCH32_EL0 and close CONFIG_ARM64_ILP32 arch/arm64/configs/openeuler_defconfig | 14 +++++++++++--- scripts/kconfig/makeconfig.sh | 3 +++ 2 files changed, 14 insertions(+), 3 deletions(-) -- 2.34.1

2 4

[PATCH OLK-6.6 0/2] Open CONFIG_AARCH32_EL0 and close CONFIG_ARM64_ILP32
by Jinjie Ruan 07 Dec '23

07 Dec '23

Update script for check & update openeuler_defconfig. And open CONFIG_AARCH32_EL0 and close CONFIG_ARM64_ILP32. Jinjie Ruan (2): kconfig: Update script for check & update openeuler_defconfig config: Open CONFIG_AARCH32_EL0 and close CONFIG_ARM64_ILP32 arch/arm64/configs/openeuler_defconfig | 14 +++++++++++--- scripts/kconfig/makeconfig.sh | 3 +++ 2 files changed, 14 insertions(+), 3 deletions(-) -- 2.34.1

2 3

[PATCH OLK-6.6 0/2] support clear free list
by Yu Liao 07 Dec '23

07 Dec '23

support clear free list Yu Liao (2): mm: Add sysctl to clear free list pages config: make CONFIG_CLEAR_FREELIST_PAGE not set by default .../admin-guide/kernel-parameters.txt | 3 + Documentation/admin-guide/sysctl/vm.rst | 13 ++ arch/arm64/configs/openeuler_defconfig | 1 + mm/Kconfig | 13 ++ mm/Makefile | 2 + mm/clear_freelist_page.c | 187 ++++++++++++++++++ 6 files changed, 219 insertions(+) create mode 100644 mm/clear_freelist_page.c -- 2.33.0

2 3

[PATCH OLK-5.10 0/1] iommu/arm-smmu-v3: disable stall for quiet_cd
by Chen Jun 07 Dec '23

07 Dec '23

*** BLURB HERE *** Wenkai Lin (1): iommu/arm-smmu-v3: disable stall for quiet_cd drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 2 ++ 1 file changed, 2 insertions(+) -- 2.17.1

2 3

答复: openEuler Kernel SIG 双周例会提醒//openEuler Kernel SIG双周例会
by liaotao (C) 07 Dec '23

07 Dec '23

主题：openEuler Kernel SIG双周例会时间：2023年12月8日星期五 14:00-15:45 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：https://etherpad.openeuler.org/p/Kernel-meetings） -----原始约会----- 发件人: openEuler conference <public(a)openeuler.org> 发送时间: 2023年12月4日 16:00 收件人: dev@openeuler.org,kernel-discuss@openeuler.org,kernel@openeuler.org 主题: openEuler Kernel SIG双周例会时间: 2023年12月8日星期五 14:00-15:45(UTC+08:00) 北京，重庆，香港特别行政区，乌鲁木齐。地点: 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

[PATCH openEuler-1.0-LTS] workqueue: Override implicit ordered attribute in workqueue_apply_unbound_cpumask()
by Yu Liao 07 Dec '23

07 Dec '23

From: Waiman Long <longman(a)redhat.com> stable inclusion from stable-v4.19.299 commit 55d1e6231a0b98ffcc5eacdae995fa7bd74bf6a0 bugzilla: https://gitee.com/openeuler/kernel/issues/I8MCB5 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit ca10d851b9ad0338c19e8e3089e24d565ebfffd7 ] Commit 5c0338c68706 ("workqueue: restore WQ_UNBOUND/max_active==1 to be ordered") enabled implicit ordered attribute to be added to WQ_UNBOUND workqueues with max_active of 1. This prevented the changing of attributes to these workqueues leading to fix commit 0a94efb5acbb ("workqueue: implicit ordered attribute should be overridable"). However, workqueue_apply_unbound_cpumask() was not updated at that time. So sysfs changes to wq_unbound_cpumask has no effect on WQ_UNBOUND workqueues with implicit ordered attribute. Since not all WQ_UNBOUND workqueues are visible on sysfs, we are not able to make all the necessary cpumask changes even if we iterates all the workqueue cpumasks in sysfs and changing them one by one. Fix this problem by applying the corresponding change made to apply_workqueue_attrs_locked() in the fix commit to workqueue_apply_unbound_cpumask(). Fixes: 5c0338c68706 ("workqueue: restore WQ_UNBOUND/max_active==1 to be ordered") Signed-off-by: Waiman Long <longman(a)redhat.com> Signed-off-by: Tejun Heo <tj(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: kernel/workqueue.c Signed-off-by: Yu Liao <liaoyu15(a)huawei.com> --- kernel/workqueue.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 3ce68188ad0b..0b2eba803e30 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -5293,8 +5293,11 @@ static int workqueue_apply_unbound_cpumask(void) if (!(wq->flags & WQ_UNBOUND)) continue; /* creating multiple pwqs breaks ordering guarantee */ - if ((wq->flags & __WQ_ORDERED) && !(wq->flags & __WQ_DYNAMIC)) - continue; + if (!list_empty(&wq->pwqs)) { + if ((wq->flags & __WQ_ORDERED_EXPLICIT) && !(wq->flags & __WQ_DYNAMIC)) + continue; + wq->flags &= ~__WQ_ORDERED; + } ctx = apply_wqattrs_prepare(wq, wq->unbound_attrs); if (!ctx) { -- 2.25.1

2 1