- Kernel - mailweb.openeuler.org

[PATCH] Disable cache readunique on 1620 which can cause performance decrease
by shenkai (D) 28 Jun '20

28 Jun '20

From 85f6f7c3bd13b3d6fd064899e2818ee70bdf29e2 Mon Sep 17 00:00:00 2001 euleros inclusion commit 85f6f7c3bd13b3d6fd064899e2818ee70bdf29e2 category: bugfix Random performance decreases appear on cases of Hackbench which test pipe or socket communication among multi-threads on Hisi 1620 SoC. Cache sharing which caused by the change of the data layout and the cache readunique 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 can trackle this problem. Test cases are like: ./hackbench -pipe 20 thread 1000 On 128 cores 1620 machine, the time cost of above test cases is basicly 0.3s when RU is off while when RU is on the time cost can be over 1s. What's more, we disble readunique only in el2 for in el1 disabling readunique may cause panic due to lack of related priority which often be set in BIOS. Signed-off-by: Kai Shen <shenkai8(a)huawei.com> Reviewed-by: Wenliang He <hewenliang4(a)huawei.com> Reviewed-by: Jinxian He <hejingxian(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Reviewed-by: Wei Li <liwei391(a)huawei.com> --- arch/arm64/Kconfig | 9 +++++++ arch/arm64/configs/euleros_defconfig | 1 + arch/arm64/include/asm/cpucaps.h | 3 ++- arch/arm64/kernel/cpu_errata.c | 37 ++++++++++++++++++++++++++++ 4 files changed, 49 insertions(+), 1 deletion(-) diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index b818273ef..33e185af5 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -642,6 +642,15 @@ config QCOM_FALKOR_ERRATUM_E1041 If unsure, say Y. +config HISILICON_ERRATUM_1620_RU + bool "Hisi 1620 cache readunique might compromise performance" + default y + help + The HiSilicon 1620 cache readunique might compromise performance, + use cmdline to enable or disable RU. + + If unsure, say Y. + endmenu diff --git a/arch/arm64/configs/euleros_defconfig b/arch/arm64/configs/euleros_defconfig index 360e69291..29403bfe4 100644 --- a/arch/arm64/configs/euleros_defconfig +++ b/arch/arm64/configs/euleros_defconfig @@ -5650,3 +5650,4 @@ CONFIG_IO_STRICT_DEVMEM=y # CONFIG_DEBUG_EFI is not set # CONFIG_ARM64_RELOC_TEST is not set # CONFIG_CORESIGHT is not set +CONFIG_HISILICON_ERRATUM_1620=y diff --git a/arch/arm64/include/asm/cpucaps.h b/arch/arm64/include/asm/cpucaps.h index d56d815b9..112c866c1 100644 --- a/arch/arm64/include/asm/cpucaps.h +++ b/arch/arm64/include/asm/cpucaps.h @@ -58,7 +58,8 @@ #define ARM64_SSBS 37 #define ARM64_CLEARPAGE_STNP 38 #define ARM64_WORKAROUND_1542419 39 +#define ARM64_HISI_1620_RU 40 -#define ARM64_NCAPS 40 +#define ARM64_NCAPS 41 #endif /* __ASM_CPUCAPS_H */ diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c index 7522163c1..fa751abed 100644 --- a/arch/arm64/kernel/cpu_errata.c +++ b/arch/arm64/kernel/cpu_errata.c @@ -25,6 +25,11 @@ #include <asm/cpufeature.h> #include <asm/smp_plat.h> +#ifdef CONFIG_HISILICON_ERRATUM_1620_RU +#include <asm/ptrace.h> +#include <asm/sysreg.h> +#endif + static bool __maybe_unused is_affected_midr_range(const struct arm64_cpu_capabilities *entry, int scope) { @@ -491,6 +496,30 @@ cpu_enable_cache_maint_trap(const struct arm64_cpu_capabilities *__unused) sysreg_clear_set(sctlr_el1, SCTLR_EL1_UCI, 0); } +#ifdef CONFIG_HISILICON_ERRATUM_1620_RU +static bool readunique_enabled = false; + +static int __init enable_readunique(char *data) +{ + readunique_enabled = true; + return 0; +} +__setup("readunique_enable", enable_readunique); + +#define CTLR_HISI_1620_RU (1L << 40) +static void __maybe_unused +hisi_1620_ru_disable(const struct arm64_cpu_capabilities *__unused) +{ + u64 el; + if (readunique_enabled) + return; + + el = read_sysreg(CurrentEL); + if (el == CurrentEL_EL2) + sysreg_clear_set(S3_1_c15_c6_4, 0, CTLR_HISI_1620_RU); +} +#endif + /* known invulnerable cores */ static const struct midr_range arm64_ssb_cpus[] = { MIDR_ALL_VERSIONS(MIDR_CORTEX_A35), @@ -884,6 +913,14 @@ const struct arm64_cpu_capabilities arm64_errata[] = { .matches = has_neoverse_n1_erratum_1542419, .cpu_enable = cpu_enable_trap_ctr_access, }, +#endif +#ifdef CONFIG_HISILICON_ERRATUM_1620_RU + { + .desc = "Hisi 1620 Cache Readunique Disable", + .capability = ARM64_HISI_1620_RU, + ERRATA_MIDR_ALL_VERSIONS(MIDR_HISI_TSV110), + .cpu_enable = hisi_1620_ru_disable, + }, #endif { } -- 2.19.1

1 0

[hulk-4.19 v3 0/6] Adding Huawei BMA drivers
by yunaixin03610＠163.com 28 Jun '20

28 Jun '20

From: yunaixin <yunaixin03610(a)163.com> This patch set contains 5 communication drivers for Huawei BMA software. The BMA software is a system management software. It supports the status monitoring, performance monitoring, and event monitoring of various components, including server CPUs, memory, hard disks, NICs, IB cards, PCIe cards, RAID controller cards, and optical modules. These 5 drivers are used to send/receive message through PCIe channel in different ways by BMA software. yunaixin (6): Huawei BMA: Adding Huawei BMA driver: host_edma_drv Huawei BMA: Adding Huawei BMA driver: host_cdev_drv Huawei BMA: Adding Huawei BMA driver: host_veth_drv Huawei BMA: Adding Huawei BMA driver: cdev_veth_drv Huawei BMA: Adding Huawei BMA driver: host_kbox_drv Huawei BMA: Add BMA driver config options arch/arm64/configs/euleros_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + drivers/net/ethernet/huawei/Kconfig | 1 + drivers/net/ethernet/huawei/Makefile | 3 +- drivers/net/ethernet/huawei/bma/Kconfig | 5 + drivers/net/ethernet/huawei/bma/Makefile | 9 + .../net/ethernet/huawei/bma/cdev_drv/Kconfig | 11 + .../net/ethernet/huawei/bma/cdev_drv/Makefile | 2 + .../ethernet/huawei/bma/cdev_drv/bma_cdev.c | 369 +++ .../ethernet/huawei/bma/cdev_veth_drv/Kconfig | 11 + .../huawei/bma/cdev_veth_drv/Makefile | 2 + .../bma/cdev_veth_drv/virtual_cdev_eth_net.c | 1839 ++++++++++++ .../bma/cdev_veth_drv/virtual_cdev_eth_net.h | 300 ++ .../net/ethernet/huawei/bma/edma_drv/Kconfig | 11 + .../net/ethernet/huawei/bma/edma_drv/Makefile | 2 + .../huawei/bma/edma_drv/bma_devintf.c | 597 ++++ .../huawei/bma/edma_drv/bma_devintf.h | 40 + .../huawei/bma/edma_drv/bma_include.h | 116 + .../ethernet/huawei/bma/edma_drv/bma_pci.c | 533 ++++ .../ethernet/huawei/bma/edma_drv/bma_pci.h | 94 + .../ethernet/huawei/bma/edma_drv/edma_host.c | 1462 ++++++++++ .../ethernet/huawei/bma/edma_drv/edma_host.h | 351 +++ .../huawei/bma/include/bma_ker_intf.h | 94 + .../net/ethernet/huawei/bma/kbox_drv/Kconfig | 11 + .../net/ethernet/huawei/bma/kbox_drv/Makefile | 2 + .../ethernet/huawei/bma/kbox_drv/kbox_dump.c | 121 + .../ethernet/huawei/bma/kbox_drv/kbox_dump.h | 33 + .../ethernet/huawei/bma/kbox_drv/kbox_hook.c | 101 + .../ethernet/huawei/bma/kbox_drv/kbox_hook.h | 33 + .../huawei/bma/kbox_drv/kbox_include.h | 40 + .../ethernet/huawei/bma/kbox_drv/kbox_main.c | 168 ++ .../ethernet/huawei/bma/kbox_drv/kbox_main.h | 23 + .../ethernet/huawei/bma/kbox_drv/kbox_mce.c | 264 ++ .../ethernet/huawei/bma/kbox_drv/kbox_mce.h | 23 + .../ethernet/huawei/bma/kbox_drv/kbox_panic.c | 187 ++ .../ethernet/huawei/bma/kbox_drv/kbox_panic.h | 25 + .../huawei/bma/kbox_drv/kbox_printk.c | 363 +++ .../huawei/bma/kbox_drv/kbox_printk.h | 33 + .../huawei/bma/kbox_drv/kbox_ram_drive.c | 188 ++ .../huawei/bma/kbox_drv/kbox_ram_drive.h | 31 + .../huawei/bma/kbox_drv/kbox_ram_image.c | 135 + .../huawei/bma/kbox_drv/kbox_ram_image.h | 84 + .../huawei/bma/kbox_drv/kbox_ram_op.c | 986 +++++++ .../huawei/bma/kbox_drv/kbox_ram_op.h | 77 + .../net/ethernet/huawei/bma/veth_drv/Kconfig | 11 + .../net/ethernet/huawei/bma/veth_drv/Makefile | 2 + .../ethernet/huawei/bma/veth_drv/veth_hb.c | 2502 +++++++++++++++++ .../ethernet/huawei/bma/veth_drv/veth_hb.h | 440 +++ 48 files changed, 11736 insertions(+), 1 deletion(-) create mode 100644 drivers/net/ethernet/huawei/bma/Kconfig create mode 100644 drivers/net/ethernet/huawei/bma/Makefile create mode 100644 drivers/net/ethernet/huawei/bma/cdev_drv/Kconfig create mode 100644 drivers/net/ethernet/huawei/bma/cdev_drv/Makefile create mode 100644 drivers/net/ethernet/huawei/bma/cdev_drv/bma_cdev.c create mode 100644 drivers/net/ethernet/huawei/bma/cdev_veth_drv/Kconfig create mode 100644 drivers/net/ethernet/huawei/bma/cdev_veth_drv/Makefile create mode 100644 drivers/net/ethernet/huawei/bma/cdev_veth_drv/virtual_cdev_eth_net.c create mode 100644 drivers/net/ethernet/huawei/bma/cdev_veth_drv/virtual_cdev_eth_net.h create mode 100644 drivers/net/ethernet/huawei/bma/edma_drv/Kconfig create mode 100644 drivers/net/ethernet/huawei/bma/edma_drv/Makefile create mode 100644 drivers/net/ethernet/huawei/bma/edma_drv/bma_devintf.c create mode 100644 drivers/net/ethernet/huawei/bma/edma_drv/bma_devintf.h create mode 100644 drivers/net/ethernet/huawei/bma/edma_drv/bma_include.h create mode 100644 drivers/net/ethernet/huawei/bma/edma_drv/bma_pci.c create mode 100644 drivers/net/ethernet/huawei/bma/edma_drv/bma_pci.h create mode 100644 drivers/net/ethernet/huawei/bma/edma_drv/edma_host.c create mode 100644 drivers/net/ethernet/huawei/bma/edma_drv/edma_host.h create mode 100644 drivers/net/ethernet/huawei/bma/include/bma_ker_intf.h create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/Kconfig create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/Makefile create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_dump.c create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_dump.h create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_hook.c create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_hook.h create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_include.h create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_main.c create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_main.h create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_mce.c create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_mce.h create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_panic.c create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_panic.h create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_printk.c create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_printk.h create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_ram_drive.c create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_ram_drive.h create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_ram_image.c create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_ram_image.h create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_ram_op.c create mode 100644 drivers/net/ethernet/huawei/bma/kbox_drv/kbox_ram_op.h create mode 100644 drivers/net/ethernet/huawei/bma/veth_drv/Kconfig create mode 100644 drivers/net/ethernet/huawei/bma/veth_drv/Makefile create mode 100644 drivers/net/ethernet/huawei/bma/veth_drv/veth_hb.c create mode 100644 drivers/net/ethernet/huawei/bma/veth_drv/veth_hb.h -- 2.26.2.windows.1

2 7

[PATCH 1/8] ovl: fix lookup failure on multi lower squashfs
by Yang Yingliang 24 Jun '20

24 Jun '20

From: Amir Goldstein <amir73il(a)gmail.com> mainline inclusion from mainline-5.5-rc2 commit 7e63c87fc2dcf3be9d3aab82d4a0ea085880bdca category: bugfix bugzilla: 37632 CVE: NA --------------------------- In the past, overlayfs required that lower fs have non null uuid in order to support nfs export and decode copy up origin file handles. Commit 9df085f3c9a2 ("ovl: relax requirement for non null uuid of lower fs") relaxed this requirement for nfs export support, as long as uuid (even if null) is unique among all lower fs. However, said commit unintentionally also relaxed the non null uuid requirement for decoding copy up origin file handles, regardless of the unique uuid requirement. Amend this mistake by disabling decoding of copy up origin file handle from lower fs with a conflicting uuid. We still encode copy up origin file handles from those fs, because file handles like those already exist in the wild and because they might provide useful information in the future. There is an unhandled corner case described by Miklos this way: - two filesystems, A and B, both have null uuid - upper layer is on A - lower layer 1 is also on A - lower layer 2 is on B In this case bad_uuid won't be set for B, because the check only involves the list of lower fs. Hence we'll try to decode a layer 2 origin on layer 1 and fail. We will deal with this corner case later. Reported-by: Colin Ian King <colin.king(a)canonical.com> Tested-by: Colin Ian King <colin.king(a)canonical.com> Link: https://lore.kernel.org/lkml/20191106234301.283006-1-colin.king@canonical.c… Fixes: 9df085f3c9a2 ("ovl: relax requirement for non null uuid ...") Cc: stable(a)vger.kernel.org # v4.20+ Signed-off-by: Amir Goldstein <amir73il(a)gmail.com> Signed-off-by: Miklos Szeredi <mszeredi(a)redhat.com> Signed-off-by: Zhihao Cheng <chengzhihao1(a)huawei.com> Reviewed-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/overlayfs/namei.c | 8 ++++++++ fs/overlayfs/ovl_entry.h | 2 ++ fs/overlayfs/super.c | 24 +++++++++++++++++------- 3 files changed, 27 insertions(+), 7 deletions(-) diff --git a/fs/overlayfs/namei.c b/fs/overlayfs/namei.c index badf039267a2..62515ec25481 100644 --- a/fs/overlayfs/namei.c +++ b/fs/overlayfs/namei.c @@ -328,6 +328,14 @@ int ovl_check_origin_fh(struct ovl_fs *ofs, struct ovl_fh *fh, bool connected, int i; for (i = 0; i < ofs->numlower; i++) { + /* + * If lower fs uuid is not unique among lower fs we cannot match + * fh->uuid to layer. + */ + if (ofs->lower_layers[i].fsid && + ofs->lower_layers[i].fs->bad_uuid) + continue; + origin = ovl_decode_real_fh(fh, ofs->lower_layers[i].mnt, connected); if (origin) diff --git a/fs/overlayfs/ovl_entry.h b/fs/overlayfs/ovl_entry.h index 1a1adc697c55..787fe4be835e 100644 --- a/fs/overlayfs/ovl_entry.h +++ b/fs/overlayfs/ovl_entry.h @@ -25,6 +25,8 @@ struct ovl_config { struct ovl_sb { struct super_block *sb; dev_t pseudo_dev; + /* Unusable (conflicting) uuid */ + bool bad_uuid; }; struct ovl_layer { diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c index ddcf8d3a83bf..3881f07a2434 100644 --- a/fs/overlayfs/super.c +++ b/fs/overlayfs/super.c @@ -1259,7 +1259,7 @@ static bool ovl_lower_uuid_ok(struct ovl_fs *ofs, const uuid_t *uuid) { unsigned int i; - if (!ofs->config.nfs_export && !(ofs->config.index && ofs->upper_mnt)) + if (!ofs->config.nfs_export && !ofs->upper_mnt) return true; for (i = 0; i < ofs->numlowerfs; i++) { @@ -1267,9 +1267,13 @@ static bool ovl_lower_uuid_ok(struct ovl_fs *ofs, const uuid_t *uuid) * We use uuid to associate an overlay lower file handle with a * lower layer, so we can accept lower fs with null uuid as long * as all lower layers with null uuid are on the same fs. + * if we detect multiple lower fs with the same uuid, we + * disable lower file handle decoding on all of them. */ - if (uuid_equal(&ofs->lower_fs[i].sb->s_uuid, uuid)) + if (uuid_equal(&ofs->lower_fs[i].sb->s_uuid, uuid)) { + ofs->lower_fs[i].bad_uuid = true; return false; + } } return true; } @@ -1281,6 +1285,7 @@ static int ovl_get_fsid(struct ovl_fs *ofs, const struct path *path) unsigned int i; dev_t dev; int err; + bool bad_uuid = false; /* fsid 0 is reserved for upper fs even with non upper overlay */ if (ofs->upper_mnt && ofs->upper_mnt->mnt_sb == sb) @@ -1292,11 +1297,15 @@ static int ovl_get_fsid(struct ovl_fs *ofs, const struct path *path) } if (!ovl_lower_uuid_ok(ofs, &sb->s_uuid)) { - ofs->config.index = false; - ofs->config.nfs_export = false; - pr_warn("overlayfs: %s uuid detected in lower fs '%pd2', falling back to index=off,nfs_export=off.\n", - uuid_is_null(&sb->s_uuid) ? "null" : "conflicting", - path->dentry); + bad_uuid = true; + if (ofs->config.index || ofs->config.nfs_export) { + ofs->config.index = false; + ofs->config.nfs_export = false; + pr_warn("overlayfs: %s uuid detected in lower fs '%pd2', falling back to index=off,nfs_export=off.\n", + uuid_is_null(&sb->s_uuid) ? "null" : + "conflicting", + path->dentry); + } } err = get_anon_bdev(&dev); @@ -1307,6 +1316,7 @@ static int ovl_get_fsid(struct ovl_fs *ofs, const struct path *path) ofs->lower_fs[ofs->numlowerfs].sb = sb; ofs->lower_fs[ofs->numlowerfs].pseudo_dev = dev; + ofs->lower_fs[ofs->numlowerfs].bad_uuid = bad_uuid; ofs->numlowerfs++; return ofs->numlowerfs; -- 2.25.1

1 7

[PATCH 1/6] perf top: Fix wrong hottest instruction highlighted
by Yang Yingliang 21 Jun '20

21 Jun '20

From: He Kuang <hekuang(a)huawei.com> mainline inclusion from mainline-5.0-rc5 commit da06d5683868 category: bugfix bugzilla: 38073 CVE: NA ------------------------------------------------- The annotation line percentage is compared and inserted into the rbtree, but the percent field of 'struct annotation_data' is an array, the comparison result between them is the address difference. This patch compares the right slot of percent array according to opts->percent_type and makes things right. The problem can be reproduced by pressing 'H' in perf top annotation view. It should highlight the instruction line which has the highest sampling percentage. Signed-off-by: He Kuang <hekuang(a)huawei.com> Reviewed-by: Jiri Olsa <jolsa(a)kernel.org> Cc: Alexander Shishkin <alexander.shishkin(a)linux.intel.com> Cc: Namhyung Kim <namhyung(a)kernel.org> Cc: Peter Zijlstra <peterz(a)infradead.org> Link: http://lkml.kernel.org/r/20190120160523.4391-1-hekuang@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme(a)redhat.com> Signed-off-by: Wei Li <liwei391(a)huawei.com> Reviewed-by: Jian Cheng <cj.chengjian(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- tools/perf/ui/browsers/annotate.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/tools/perf/ui/browsers/annotate.c b/tools/perf/ui/browsers/annotate.c index a3c255228d62..131f51131b52 100644 --- a/tools/perf/ui/browsers/annotate.c +++ b/tools/perf/ui/browsers/annotate.c @@ -225,20 +225,24 @@ static unsigned int annotate_browser__refresh(struct ui_browser *browser) return ret; } -static int disasm__cmp(struct annotation_line *a, struct annotation_line *b) +static double disasm__cmp(struct annotation_line *a, struct annotation_line *b, + int percent_type) { int i; for (i = 0; i < a->data_nr; i++) { - if (a->data[i].percent == b->data[i].percent) + if (a->data[i].percent[percent_type] == b->data[i].percent[percent_type]) continue; - return a->data[i].percent < b->data[i].percent; + return a->data[i].percent[percent_type] - + b->data[i].percent[percent_type]; } return 0; } -static void disasm_rb_tree__insert(struct rb_root *root, struct annotation_line *al) +static void disasm_rb_tree__insert(struct annotate_browser *browser, + struct annotation_line *al) { + struct rb_root *root = &browser->entries; struct rb_node **p = &root->rb_node; struct rb_node *parent = NULL; struct annotation_line *l; @@ -247,7 +251,7 @@ static void disasm_rb_tree__insert(struct rb_root *root, struct annotation_line parent = *p; l = rb_entry(parent, struct annotation_line, rb_node); - if (disasm__cmp(al, l)) + if (disasm__cmp(al, l, browser->opts->percent_type) < 0) p = &(*p)->rb_left; else p = &(*p)->rb_right; @@ -330,7 +334,7 @@ static void annotate_browser__calc_percent(struct annotate_browser *browser, RB_CLEAR_NODE(&pos->al.rb_node); continue; } - disasm_rb_tree__insert(&browser->entries, &pos->al); + disasm_rb_tree__insert(browser, &pos->al); } pthread_mutex_unlock(&notes->lock); -- 2.25.1

1 5

[PATCH 1/2] xfs: Use scnprintf() for avoiding potential buffer overflow
by Yang Yingliang 17 Jun '20

17 Jun '20

From: Takashi Iwai <tiwai(a)suse.de> mainline inclusion from mainline-v5.7-rc1 commit 17bb60b74124e9491d593e2601e3afe14daa2f57 category: bugfix bugzilla: 33322 CVE: NA --------------------------- Since snprintf() returns the would-be-output size instead of the actual output size, the succeeding calls may go beyond the given buffer limit. Fix it by replacing with scnprintf(). Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Reviewed-by: Darrick J. Wong <darrick.wong(a)oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong(a)oracle.com> Signed-off-by: yu kuai <yukuai3(a)huawei.com> Reviewed-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/xfs/xfs_stats.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/fs/xfs/xfs_stats.c b/fs/xfs/xfs_stats.c index 740ac9674848..0373d9ea7334 100644 --- a/fs/xfs/xfs_stats.c +++ b/fs/xfs/xfs_stats.c @@ -58,13 +58,13 @@ int xfs_stats_format(struct xfsstats __percpu *stats, char *buf) /* Loop over all stats groups */ for (i = j = 0; i < ARRAY_SIZE(xstats); i++) { - len += snprintf(buf + len, PATH_MAX - len, "%s", + len += scnprintf(buf + len, PATH_MAX - len, "%s", xstats[i].desc); /* inner loop does each group */ for (; j < xstats[i].endpoint; j++) - len += snprintf(buf + len, PATH_MAX - len, " %u", + len += scnprintf(buf + len, PATH_MAX - len, " %u", counter_val(stats, j)); - len += snprintf(buf + len, PATH_MAX - len, "\n"); + len += scnprintf(buf + len, PATH_MAX - len, "\n"); } /* extra precision counters */ for_each_possible_cpu(i) { @@ -73,9 +73,9 @@ int xfs_stats_format(struct xfsstats __percpu *stats, char *buf) xs_read_bytes += per_cpu_ptr(stats, i)->s.xs_read_bytes; } - len += snprintf(buf + len, PATH_MAX-len, "xpc %Lu %Lu %Lu\n", + len += scnprintf(buf + len, PATH_MAX-len, "xpc %Lu %Lu %Lu\n", xs_xstrat_bytes, xs_write_bytes, xs_read_bytes); - len += snprintf(buf + len, PATH_MAX-len, "debug %u\n", + len += scnprintf(buf + len, PATH_MAX-len, "debug %u\n", #if defined(DEBUG) 1); #else -- 2.25.1

1 1

[PATCH] nfs: set invalid blocks after NFSv4 writes
by Yang Yingliang 16 Jun '20

16 Jun '20

From: Zheng Bin <zhengbin13(a)huawei.com> mainline inclusion from mainline-bc5603c314bd commit bc5603c314bd1d2a931ad3088fcea52f20cfbac5 category: bugfix bugzilla: 34591 CVE: NA --------------------------- Use the following command to test nfsv4(size of file1M is 1MB): mount -t nfs -o vers=4.0,actimeo=60 127.0.0.1/dir1 /mnt cp file1M /mnt du -h /mnt/file1M -->0 within 60s, then 1M When write is done(cp file1M /mnt), will call this: nfs_writeback_done nfs4_write_done nfs4_write_done_cb nfs_writeback_update_inode nfs_post_op_update_inode_force_wcc_locked(change, ctime, mtime nfs_post_op_update_inode_force_wcc_locked nfs_set_cache_invalid nfs_refresh_inode_locked nfs_update_inode nfsd write response contains change, ctime, mtime, the flag will be clear after nfs_update_inode. Howerver, write response does not contain space_used, previous open response contains space_used whose value is 0, so inode->i_blocks is still 0. nfs_getattr -->called by "du -h" do_update |= force_sync || nfs_attribute_cache_expired -->false in 60s cache_validity = READ_ONCE(NFS_I(inode)->cache_validity) do_update |= cache_validity & (NFS_INO_INVALID_ATTR -->false if (do_update) { __nfs_revalidate_inode } Within 60s, does not send getattr request to nfsd, thus "du -h /mnt/file1M" is 0. Add a NFS_INO_INVALID_BLOCKS flag, set it when nfsv4 write is done. [zb: although commit 1c341b777501 ("NFS: Add deferred cache invalidation for close-to-open consistency violations") not merged, still define NFS_INO_INVALID_BLOCKS BIT(14)] Fixes: 16e143751727 ("NFS: More fine grained attribute tracking") Signed-off-by: Zheng Bin <zhengbin13(a)huawei.com> Signed-off-by: Anna Schumaker <Anna.Schumaker(a)Netapp.com> Signed-off-by: Zheng Bin <zhengbin13(a)huawei.com> Reviewed-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/nfs/inode.c | 14 +++++++++++--- include/linux/nfs_fs.h | 1 + 2 files changed, 12 insertions(+), 3 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index e4cd3a2fe698..5d8e0eb0d663 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -822,6 +822,8 @@ int nfs_getattr(const struct path *path, struct kstat *stat, do_update |= cache_validity & NFS_INO_INVALID_ATIME; if (request_mask & (STATX_CTIME|STATX_MTIME)) do_update |= cache_validity & NFS_INO_REVAL_PAGECACHE; + if (request_mask & STATX_BLOCKS) + do_update |= cache_validity & NFS_INO_INVALID_BLOCKS; if (do_update) { /* Update the attribute cache */ if (!(server->flags & NFS_MOUNT_NOAC)) @@ -1741,7 +1743,8 @@ int nfs_post_op_update_inode_force_wcc_locked(struct inode *inode, struct nfs_fa status = nfs_post_op_update_inode_locked(inode, fattr, NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_CTIME - | NFS_INO_INVALID_MTIME); + | NFS_INO_INVALID_MTIME + | NFS_INO_INVALID_BLOCKS); return status; } @@ -1851,7 +1854,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ATIME | NFS_INO_REVAL_FORCED - | NFS_INO_REVAL_PAGECACHE); + | NFS_INO_REVAL_PAGECACHE + | NFS_INO_INVALID_BLOCKS); /* Do atomic weak cache consistency updates */ nfs_wcc_update_inode(inode, fattr); @@ -2012,8 +2016,12 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used); } else if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED) inode->i_blocks = fattr->du.nfs2.blocks; - else + else { + nfsi->cache_validity |= save_cache_validity & + (NFS_INO_INVALID_BLOCKS + | NFS_INO_REVAL_FORCED); cache_revalidated = false; + } /* Update attrtimeo value if we're out of the unstable period */ if (attr_changed) { diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index a0831e9d19c9..620306b1604d 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -221,6 +221,7 @@ struct nfs4_copy_state { #define NFS_INO_INVALID_MTIME BIT(10) /* cached mtime is invalid */ #define NFS_INO_INVALID_SIZE BIT(11) /* cached size is invalid */ #define NFS_INO_INVALID_OTHER BIT(12) /* other attrs are invalid */ +#define NFS_INO_INVALID_BLOCKS BIT(14) /* cached blocks are invalid */ #define NFS_INO_INVALID_ATTR (NFS_INO_INVALID_CHANGE \ | NFS_INO_INVALID_CTIME \ -- 2.25.1

1 0

[PATCH 1/4] xfs: fix inode fork extent count overflow
by Yang Yingliang 16 Jun '20

16 Jun '20

From: Dave Chinner <dchinner(a)redhat.com> mainline inclusion from mainline-5.5-rc1 commit 3f8a4f1d876d3e3e49e50b0396eaffcc4ba71b08 category: bugfix bugzilla: 26723 CVE: NA --------------------------- [commit message is verbose for discussion purposes - will trim it down later. Some questions about implementation details at the end.] Zorro Lang recently ran a new test to stress single inode extent counts now that they are no longer limited by memory allocation. The test was simply: This test uncovered a problem where the hole punching operation appeared to finish with no error, but apparently only created 268M extents instead of the 10 billion it was supposed to. Further, trying to punch out extents that should have been present resulted in success, but no change in the extent count. It looked like a silent failure. While running the test and observing the behaviour in real time, I observed the extent coutn growing at ~2M extents/minute, and saw this after about an hour: > sleep 60 ; \ > xfs_io -f -c "stat" /mnt/scratch/big-file |grep next fsxattr.nextents = 127657993 fsxattr.nextents = 129683339 And a few minutes later this: fsxattr.nextents = 4177861124 Ah, what? Where did that 4 billion extra extents suddenly come from? Stop the workload, unmount, mount: fsxattr.nextents = 166044375 And it's back at the expected number. i.e. the extent count is correct on disk, but it's screwed up in memory. I loaded up the extent list, and immediately: fsxattr.nextents = 4192576215 It's bad again. So, where does that number come from? xfs_fill_fsxattr(): if (ip->i_df.if_flags & XFS_IFEXTENTS) fa->fsx_nextents = xfs_iext_count(&ip->i_df); else fa->fsx_nextents = ip->i_d.di_nextents; And that's the behaviour I just saw in a nutshell. The on disk count is correct, but once the tree is loaded into memory, it goes whacky. Clearly there's something wrong with xfs_iext_count(): inline xfs_extnum_t xfs_iext_count(struct xfs_ifork *ifp) { return ifp->if_bytes / sizeof(struct xfs_iext_rec); } Simple enough, but 134M extents is 2**27, and that's right about where things went wrong. A struct xfs_iext_rec is 16 bytes in size, which means 2**27 * 2**4 = 2**31 and we're right on target for an integer overflow. And, sure enough: struct xfs_ifork { int if_bytes; /* bytes in if_u1 */ .... Once we get 2**27 extents in a file, we overflow if_bytes and the in-core extent count goes wrong. And when we reach 2**28 extents, if_bytes wraps back to zero and things really start to go wrong there. This is where the silent failure comes from - only the first 2**28 extents can be looked up directly due to the overflow, all the extents above this index wrap back to somewhere in the first 2**28 extents. Hence with a regular pattern, trying to punch a hole in the range that didn't have holes mapped to a hole in the first 2**28 extents and so "succeeded" without changing anything. Hence "silent failure"... Fix this by converting if_bytes to a int64_t and converting all the index variables and size calculations to use int64_t types to avoid overflows in future. Signed integers are still used to enable easy detection of extent count underflows. This enables scalability of extent counts to the limits of the on-disk format - MAXEXTNUM (2**31) extents. Current testing is at over 500M extents and still going: fsxattr.nextents = 517310478 Reported-by: Zorro Lang <zlang(a)redhat.com> Signed-off-by: Dave Chinner <dchinner(a)redhat.com> Reviewed-by: Darrick J. Wong <darrick.wong(a)oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong(a)oracle.com> Conflict: fs/xfs/libxfs/xfs_inode_fork.h Signed-off-by: yu kuai <yukuai3(a)huawei.com> Reviewed-by: Hou Tao <houtao1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/xfs/libxfs/xfs_attr_leaf.c | 18 ++++++++++-------- fs/xfs/libxfs/xfs_dir2_sf.c | 2 +- fs/xfs/libxfs/xfs_iext_tree.c | 2 +- fs/xfs/libxfs/xfs_inode_fork.c | 8 ++++---- fs/xfs/libxfs/xfs_inode_fork.h | 14 ++++++++------ 5 files changed, 24 insertions(+), 20 deletions(-) diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index 2652d00842d6..ceb9b0045725 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -421,13 +421,15 @@ xfs_attr_namesp_match(int arg_flags, int ondisk_flags) * special case for dev/uuid inodes, they have fixed size data forks. */ int -xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes) +xfs_attr_shortform_bytesfit( + struct xfs_inode *dp, + int bytes) { - int offset; - int minforkoff; /* lower limit on valid forkoff locations */ - int maxforkoff; /* upper limit on valid forkoff locations */ - int dsize; - xfs_mount_t *mp = dp->i_mount; + struct xfs_mount *mp = dp->i_mount; + int64_t dsize; + int minforkoff; + int maxforkoff; + int offset; /* rounded down */ offset = (XFS_LITINO(mp, dp->i_d.di_version) - bytes) >> 3; @@ -493,7 +495,7 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes) * A data fork btree root must have space for at least * MINDBTPTRS key/ptr pairs if the data fork is small or empty. */ - minforkoff = max(dsize, XFS_BMDR_SPACE_CALC(MINDBTPTRS)); + minforkoff = max_t(int64_t, dsize, XFS_BMDR_SPACE_CALC(MINDBTPTRS)); minforkoff = roundup(minforkoff, 8) >> 3; /* attr fork btree root can have at least this many key/ptr pairs */ @@ -913,7 +915,7 @@ xfs_attr_shortform_verify( char *endp; struct xfs_ifork *ifp; int i; - int size; + int64_t size; ASSERT(ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL); ifp = XFS_IFORK_PTR(ip, XFS_ATTR_FORK); diff --git a/fs/xfs/libxfs/xfs_dir2_sf.c b/fs/xfs/libxfs/xfs_dir2_sf.c index 716c8c50f8bc..e126446a9484 100644 --- a/fs/xfs/libxfs/xfs_dir2_sf.c +++ b/fs/xfs/libxfs/xfs_dir2_sf.c @@ -653,7 +653,7 @@ xfs_dir2_sf_verify( int i; int i8count; int offset; - int size; + int64_t size; int error; uint8_t filetype; diff --git a/fs/xfs/libxfs/xfs_iext_tree.c b/fs/xfs/libxfs/xfs_iext_tree.c index 771dd072015d..22beaab8f9b4 100644 --- a/fs/xfs/libxfs/xfs_iext_tree.c +++ b/fs/xfs/libxfs/xfs_iext_tree.c @@ -600,7 +600,7 @@ xfs_iext_realloc_root( struct xfs_ifork *ifp, struct xfs_iext_cursor *cur) { - size_t new_size = ifp->if_bytes + sizeof(struct xfs_iext_rec); + int64_t new_size = ifp->if_bytes + sizeof(struct xfs_iext_rec); void *new; /* account for the prev/next pointers */ diff --git a/fs/xfs/libxfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c index f9acf1d436f6..80aa0f1a04dd 100644 --- a/fs/xfs/libxfs/xfs_inode_fork.c +++ b/fs/xfs/libxfs/xfs_inode_fork.c @@ -131,7 +131,7 @@ xfs_init_local_fork( struct xfs_inode *ip, int whichfork, const void *data, - int size) + int64_t size) { struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork); int mem_size = size, real_size = 0; @@ -469,11 +469,11 @@ xfs_iroot_realloc( void xfs_idata_realloc( struct xfs_inode *ip, - int byte_diff, + int64_t byte_diff, int whichfork) { struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork); - int new_size = (int)ifp->if_bytes + byte_diff; + int64_t new_size = ifp->if_bytes + byte_diff; ASSERT(new_size >= 0); ASSERT(new_size <= XFS_IFORK_SIZE(ip, whichfork)); @@ -554,7 +554,7 @@ xfs_iextents_copy( struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork); struct xfs_iext_cursor icur; struct xfs_bmbt_irec rec; - int copied = 0; + int64_t copied = 0; ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL | XFS_ILOCK_SHARED)); ASSERT(ifp->if_bytes > 0); diff --git a/fs/xfs/libxfs/xfs_inode_fork.h b/fs/xfs/libxfs/xfs_inode_fork.h index 60361d2d74a1..09e3548afdde 100644 --- a/fs/xfs/libxfs/xfs_inode_fork.h +++ b/fs/xfs/libxfs/xfs_inode_fork.h @@ -13,16 +13,16 @@ struct xfs_dinode; * File incore extent information, present for each of data & attr forks. */ struct xfs_ifork { - int if_bytes; /* bytes in if_u1 */ - unsigned int if_seq; /* cow fork mod counter */ + int64_t if_bytes; /* bytes in if_u1 */ struct xfs_btree_block *if_broot; /* file's incore btree root */ - short if_broot_bytes; /* bytes allocated for root */ - unsigned char if_flags; /* per-fork flags */ + unsigned int if_seq; /* cow fork mod counter */ int if_height; /* height of the extent tree */ union { void *if_root; /* extent tree root */ char *if_data; /* inline file data */ } if_u1; + short if_broot_bytes; /* bytes allocated for root */ + unsigned char if_flags; /* per-fork flags */ }; /* @@ -93,12 +93,14 @@ int xfs_iformat_fork(struct xfs_inode *, struct xfs_dinode *); void xfs_iflush_fork(struct xfs_inode *, struct xfs_dinode *, struct xfs_inode_log_item *, int); void xfs_idestroy_fork(struct xfs_inode *, int); -void xfs_idata_realloc(struct xfs_inode *, int, int); +void xfs_idata_realloc(struct xfs_inode *ip, int64_t byte_diff, + int whichfork); void xfs_iroot_realloc(struct xfs_inode *, int, int); int xfs_iread_extents(struct xfs_trans *, struct xfs_inode *, int); int xfs_iextents_copy(struct xfs_inode *, struct xfs_bmbt_rec *, int); -void xfs_init_local_fork(struct xfs_inode *, int, const void *, int); +void xfs_init_local_fork(struct xfs_inode *ip, int whichfork, + const void *data, int64_t size); xfs_extnum_t xfs_iext_count(struct xfs_ifork *ifp); void xfs_iext_insert(struct xfs_inode *, struct xfs_iext_cursor *cur, -- 2.25.1

1 3

[PATCH] signal: Export tracepoint symbol signal_generate
by Yang Yingliang 15 Jun '20

15 Jun '20

From: Tang Yizhou <tangyizhou(a)huawei.com> hulk inclusion category: feature bugzilla: 37631 CVE: NA ------------------------------------------------- To support signal monitoring, we need to export the defined tracepoint signal_generate so that it can be used in kernel modules. Signed-off-by: Tang Yizhou <tangyizhou(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- kernel/signal.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/kernel/signal.c b/kernel/signal.c index 718f22da0305..091a3a313410 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -52,6 +52,8 @@ #include <asm/cacheflush.h> #include "audit.h" /* audit_signal_info() */ +EXPORT_TRACEPOINT_SYMBOL(signal_generate); + /* * SLAB caches for signal bits. */ -- 2.25.1

1 0

[PATCH 1/2] perf/x86: Make perf callchains work without CONFIG_FRAME_POINTER
by Yang Yingliang 15 Jun '20

15 Jun '20

From: Kairui Song <kasong(a)redhat.com> mainline inclusion from mainline-5.2-rc1 commit d15d356887e7 category: bugfix bugzilla: 35738 CVE: NA ------------------------------------------------- Currently perf callchain doesn't work well with ORC unwinder when sampling from trace point. We'll get useless in kernel callchain like this: perf 6429 [000] 22.498450: kmem:mm_page_alloc: page=0x176a17 pfn=1534487 order=0 migratetype=0 gfp_flags=GFP_KERNEL ffffffffbe23e32e __alloc_pages_nodemask+0x22e (/lib/modules/5.1.0-rc3+/build/vmlinux) 7efdf7f7d3e8 __poll+0x18 (/usr/lib64/libc-2.28.so) 5651468729c1 [unknown] (/usr/bin/perf) 5651467ee82a main+0x69a (/usr/bin/perf) 7efdf7eaf413 __libc_start_main+0xf3 (/usr/lib64/libc-2.28.so) 5541f689495641d7 [unknown] ([unknown]) The root cause is that, for trace point events, it doesn't provide a real snapshot of the hardware registers. Instead perf tries to get required caller's registers and compose a fake register snapshot which suppose to contain enough information for start a unwinding. However without CONFIG_FRAME_POINTER, if failed to get caller's BP as the frame pointer, so current frame pointer is returned instead. We get a invalid register combination which confuse the unwinder, and end the stacktrace early. So in such case just don't try dump BP, and let the unwinder start directly when the register is not a real snapshot. Use SP as the skip mark, unwinder will skip all the frames until it meet the frame of the trace point caller. Tested with frame pointer unwinder and ORC unwinder, this makes perf callchain get the full kernel space stacktrace again like this: perf 6503 [000] 1567.570191: kmem:mm_page_alloc: page=0x16c904 pfn=1493252 order=0 migratetype=0 gfp_flags=GFP_KERNEL ffffffffb523e2ae __alloc_pages_nodemask+0x22e (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb52383bd __get_free_pages+0xd (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb52fd28a __pollwait+0x8a (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb521426f perf_poll+0x2f (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb52fe3e2 do_sys_poll+0x252 (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb52ff027 __x64_sys_poll+0x37 (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb500418b do_syscall_64+0x5b (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb5a0008c entry_SYSCALL_64_after_hwframe+0x44 (/lib/modules/5.1.0-rc3+/build/vmlinux) 7f71e92d03e8 __poll+0x18 (/usr/lib64/libc-2.28.so) 55a22960d9c1 [unknown] (/usr/bin/perf) 55a22958982a main+0x69a (/usr/bin/perf) 7f71e9202413 __libc_start_main+0xf3 (/usr/lib64/libc-2.28.so) 5541f689495641d7 [unknown] ([unknown]) Co-developed-by: Josh Poimboeuf <jpoimboe(a)redhat.com> Signed-off-by: Kairui Song <kasong(a)redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Cc: Alexander Shishkin <alexander.shishkin(a)linux.intel.com> Cc: Alexei Starovoitov <alexei.starovoitov(a)gmail.com> Cc: Arnaldo Carvalho de Melo <acme(a)kernel.org> Cc: Borislav Petkov <bp(a)alien8.de> Cc: Dave Young <dyoung(a)redhat.com> Cc: Jiri Olsa <jolsa(a)redhat.com> Cc: Linus Torvalds <torvalds(a)linux-foundation.org> Cc: Namhyung Kim <namhyung(a)kernel.org> Cc: Peter Zijlstra <peterz(a)infradead.org> Cc: Thomas Gleixner <tglx(a)linutronix.de> Link: https://lkml.kernel.org/r/20190422162652.15483-1-kasong@redhat.com Signed-off-by: Ingo Molnar <mingo(a)kernel.org> Signed-off-by: Wei Li <liwei391(a)huawei.com> Reviewed-by: Jian Cheng <cj.chengjian(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/x86/events/core.c | 21 +++++++++++++++++---- arch/x86/include/asm/perf_event.h | 7 +------ arch/x86/include/asm/stacktrace.h | 13 ------------- include/linux/perf_event.h | 14 ++++++++++---- 4 files changed, 28 insertions(+), 27 deletions(-) diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c index c89e6c6e26f6..0a6b48afeb05 100644 --- a/arch/x86/events/core.c +++ b/arch/x86/events/core.c @@ -2332,6 +2332,15 @@ void arch_perf_update_userpage(struct perf_event *event, cyc2ns_read_end(); } +/* + * Determine whether the regs were taken from an irq/exception handler rather + * than from perf_arch_fetch_caller_regs(). + */ +static bool perf_hw_regs(struct pt_regs *regs) +{ + return regs->flags & X86_EFLAGS_FIXED; +} + void perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *regs) { @@ -2343,11 +2352,15 @@ perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *re return; } - if (perf_callchain_store(entry, regs->ip)) - return; + if (perf_hw_regs(regs)) { + if (perf_callchain_store(entry, regs->ip)) + return; + unwind_start(&state, current, regs, NULL); + } else { + unwind_start(&state, current, NULL, (void *)regs->sp); + } - for (unwind_start(&state, current, regs, NULL); !unwind_done(&state); - unwind_next_frame(&state)) { + for (; !unwind_done(&state); unwind_next_frame(&state)) { addr = unwind_get_return_address(&state); if (!addr || perf_callchain_store(entry, addr)) return; diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h index f6c4915a863e..620c3e3564be 100644 --- a/arch/x86/include/asm/perf_event.h +++ b/arch/x86/include/asm/perf_event.h @@ -264,14 +264,9 @@ extern unsigned long perf_misc_flags(struct pt_regs *regs); */ #define perf_arch_fetch_caller_regs(regs, __ip) { \ (regs)->ip = (__ip); \ - (regs)->bp = caller_frame_pointer(); \ + (regs)->sp = (unsigned long)__builtin_frame_address(0); \ (regs)->cs = __KERNEL_CS; \ regs->flags = 0; \ - asm volatile( \ - _ASM_MOV "%%"_ASM_SP ", %0\n" \ - : "=m" ((regs)->sp) \ - :: "memory" \ - ); \ } struct perf_guest_switch_msr { diff --git a/arch/x86/include/asm/stacktrace.h b/arch/x86/include/asm/stacktrace.h index f335aad404a4..beef7ad9e43a 100644 --- a/arch/x86/include/asm/stacktrace.h +++ b/arch/x86/include/asm/stacktrace.h @@ -98,19 +98,6 @@ struct stack_frame_ia32 { u32 return_address; }; -static inline unsigned long caller_frame_pointer(void) -{ - struct stack_frame *frame; - - frame = __builtin_frame_address(0); - -#ifdef CONFIG_FRAME_POINTER - frame = frame->next_frame; -#endif - - return (unsigned long)frame; -} - void show_opcodes(struct pt_regs *regs, const char *loglvl); void show_ip(struct pt_regs *regs, const char *loglvl); #endif /* _ASM_X86_STACKTRACE_H */ diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index d8b4d31acd18..cf3da70056f7 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1050,12 +1050,18 @@ static inline void perf_arch_fetch_caller_regs(struct pt_regs *regs, unsigned lo #endif /* - * Take a snapshot of the regs. Skip ip and frame pointer to - * the nth caller. We only need a few of the regs: + * When generating a perf sample in-line, instead of from an interrupt / + * exception, we lack a pt_regs. This is typically used from software events + * like: SW_CONTEXT_SWITCHES, SW_MIGRATIONS and the tie-in with tracepoints. + * + * We typically don't need a full set, but (for x86) do require: * - ip for PERF_SAMPLE_IP * - cs for user_mode() tests - * - bp for callchains - * - eflags, for future purposes, just in case + * - sp for PERF_SAMPLE_CALLCHAIN + * - eflags for MISC bits and CALLCHAIN (see: perf_hw_regs()) + * + * NOTE: assumes @regs is otherwise already 0 filled; this is important for + * things like PERF_SAMPLE_REGS_INTR. */ static inline void perf_fetch_caller_regs(struct pt_regs *regs) { -- 2.25.1

1 1

[PATCH] ext4: stop overwrite the errcode in ext4_setup_super
by Yang Yingliang 15 Jun '20

15 Jun '20

From: yangerkun <yangerkun(a)huawei.com> hulk inclusion category: bugfix bugzilla: 35487 CVE: NA ----------------------------------------------- Now the errcode from ext4_commit_super will overwrite EROFS exists in ext4_setup_super. Actually, no need to call ext4_commit_super since we will return EROFS. Fix it by goto done directly. Fixes: c89128a00838 ("ext4: handle errors on ext4_commit_super") Signed-off-by: yangerkun <yangerkun(a)huawei.com> Reviewed-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Ye Bin <yebin10(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/ext4/super.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index cc057001961e..b10f96860dd8 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -2305,6 +2305,7 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, ext4_msg(sb, KERN_ERR, "revision level too high, " "forcing read-only mode"); err = -EROFS; + goto done; } if (read_only) goto done; -- 2.25.1

1 0

[PATCH] panic/printk: fix zap_lock
by Yang Yingliang 15 Jun '20

15 Jun '20

From: Cheng Jian <cj.chengjian(a)huawei.com> hulk inclusion category: bugfix bugzilla: 34546 CVE: NA ---------------------------------------- There are two problems with the implementation and use of zap_lock(). Firstly, This console_sem does not require reinit in zap_lock(), this is because: 1). printk() itself does try_lock() and skips console handling when the semaphore is not available. 2). panic() tries to push the messages later in console_flush_on_panic(). It ignores the semaphore. Also most console drivers ignore their internal locks because oops_in_progress is set by bust_spinlocks(). Secondly, The situation is more complicated when NMI is not used. 1). Non-stopped CPUs are in unknown state, most likely in a busy loop. Nobody knows whether printk() is repeatedly called in the loop. When it was called, re-initializing any lock would cause double unlock and deadlock. 2). It would be possible to add some more hacks. One problem is that there are two groups of users. One prefer to risk a deadlock and have a chance to see the messages. Others prefer to always reach emergency_restart() and reboot the machine. Fixes: b15f5676d00e ("printk/panic: Avoid deadlock in printk()") Signed-off-by: Cheng Jian <cj.chengjian(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- kernel/panic.c | 25 +++++++++++++++++++++++++ kernel/printk/printk.c | 3 --- 2 files changed, 25 insertions(+), 3 deletions(-) diff --git a/kernel/panic.c b/kernel/panic.c index 9434f6c2da9e..ebdb58e761c0 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -238,7 +238,32 @@ void panic(const char *fmt, ...) crash_smp_send_stop(); } + /* + * ZAP console related locks when nmi broadcast. If a crash is occurring, + * make sure we can't deadlock. And make sure that we print immediately. + * + * A deadlock caused by logbuf_lock can be occured when panic: + * a) Panic CPU is running in non-NMI context; + * b) Panic CPU sends out shutdown IPI via NMI vector; + * c) One of the CPUs that we bring down via NMI vector holded logbuf_lock; + * d) Panic CPU try to hold logbuf_lock, then deadlock occurs. + * + * At present, only try to solve this problem for the ARCH with NMI, + * by reinit lock, this situation is more complicated when NMI is not + * used. + * 1). Non-stopped CPUs are in unknown state, most likely in a busy loop. + * Nobody knows whether printk() is repeatedly called in the loop. + * When it was called, re-initializing any lock would cause double + * unlock and deadlock. + * + * 2). It would be possible to add some more hacks. One problem is that + * there are two groups of users. One prefer to risk a deadlock and + * have a chance to see the messages. Others prefer to always + * reach emergency_restart() and reboot the machine. + */ +#ifdef CONFIG_X86 zap_locks(); +#endif /* * Run any panic handlers, including those that might need to diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index 7225f5ef0f54..f67b6a1c8f74 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -1592,9 +1592,6 @@ void zap_locks(void) if (raw_spin_is_locked(&logbuf_lock)) { debug_locks_off(); raw_spin_lock_init(&logbuf_lock); - - console_suspended = 1; - sema_init(&console_sem, 1); } if (raw_spin_is_locked(&console_owner_lock)) { -- 2.25.1

1 0

[PATCH 1/3] net: hns3: add device name valid check
by Yang Yingliang 15 Jun '20

15 Jun '20

From: Junxin Chen <chenjunxin1(a)huawei.com> driver inclusion category: bugfix bugzilla: NA CVE: NA ---------------------------------- Nowadays, we have checked device name at user tool, and have limited length of device name at kernel driver. But when get_netdev_by_ifname return NULL, it would cause print device name without check terminator. This patch fixes this bug, when device name's last character isn't a '\0', kernel driver would return -EINVAL to user. Signed-off-by: Junxin Chen <chenjunxin1(a)huawei.com> Reviewed-by: Weiwei Deng <dengweiwei(a)huawei.com> Reviewed-by: Shengzui You <youshengzui(a)huawei.com> Reviewed-by: Zhaohui Zhong <zhongzhaohui(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hns3_cae/hns3_cae_init.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_cae/hns3_cae_init.c b/drivers/net/ethernet/hisilicon/hns3/hns3_cae/hns3_cae_init.c index 7c084a6fbe18..1e012917e010 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_cae/hns3_cae_init.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_cae/hns3_cae_init.c @@ -335,6 +335,11 @@ static long hns3_cae_k_unlocked_ioctl(struct file *pfile, unsigned int cmd, * code yet, so we don't need lock. */ rtnl_lock(); + if (nt_msg.device_name[IFNAMSIZ - 1] != '\0') { + pr_err("the device name is invalid.\n"); + ret = -EINVAL; + goto out_invalid; + } ret = hns3_cae_k_get_netdev_by_ifname(nt_msg.device_name, &nic_dev); if (ret) { pr_err("can not get the netdevice correctly\n"); -- 2.25.1

1 2

[PATCH] netdevsim: Fix error handling in nsim_fib_init and nsim_fib_exit
by liaichun 15 Jun '20

15 Jun '20

From: YueHaibing <yuehaibing(a)huawei.com> mainline inclusion from mainline-5.4 commit 33902b4a4227877896dd9368ac10f4ca0d100de5 category: bugfix issue: I1JZHT bugzilla: NA ------------------------------ In nsim_fib_init(), if register_fib_notifier failed, nsim_fib_net_ops should be unregistered before return. In nsim_fib_exit(), unregister_fib_notifier should be called before nsim_fib_net_ops be unregistered, otherwise may cause use-after-free: BUG: KASAN: use-after-free in nsim_fib_event_nb+0x342/0x570 [netdevsim] Read of size 8 at addr ffff8881daaf4388 by task kworker/0:3/3499 CPU: 0 PID: 3499 Comm: kworker/0:3 Not tainted 5.3.0-rc7+ #30 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1ubuntu1 04/01/2014 Workqueue: ipv6_addrconf addrconf_dad_work [ipv6] Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0xa9/0x10e lib/dump_stack.c:113 print_address_description+0x65/0x380 mm/kasan/report.c:351 __kasan_report+0x149/0x18d mm/kasan/report.c:482 kasan_report+0xe/0x20 mm/kasan/common.c:618 nsim_fib_event_nb+0x342/0x570 [netdevsim] notifier_call_chain+0x52/0xf0 kernel/notifier.c:95 __atomic_notifier_call_chain+0x78/0x140 kernel/notifier.c:185 call_fib_notifiers+0x30/0x60 net/core/fib_notifier.c:30 call_fib6_entry_notifiers+0xc1/0x100 [ipv6] fib6_add+0x92e/0x1b10 [ipv6] __ip6_ins_rt+0x40/0x60 [ipv6] ip6_ins_rt+0x84/0xb0 [ipv6] __ipv6_ifa_notify+0x4b6/0x550 [ipv6] ipv6_ifa_notify+0xa5/0x180 [ipv6] addrconf_dad_completed+0xca/0x640 [ipv6] addrconf_dad_work+0x296/0x960 [ipv6] process_one_work+0x5c0/0xc00 kernel/workqueue.c:2269 worker_thread+0x5c/0x670 kernel/workqueue.c:2415 kthread+0x1d7/0x200 kernel/kthread.c:255 ret_from_fork+0x3a/0x50 arch/x86/entry/entry_64.S:352 Allocated by task 3388: save_stack+0x19/0x80 mm/kasan/common.c:69 set_track mm/kasan/common.c:77 [inline] __kasan_kmalloc.constprop.3+0xa0/0xd0 mm/kasan/common.c:493 kmalloc include/linux/slab.h:557 [inline] kzalloc include/linux/slab.h:748 [inline] ops_init+0xa9/0x220 net/core/net_namespace.c:127 __register_pernet_operations net/core/net_namespace.c:1135 [inline] register_pernet_operations+0x1d4/0x420 net/core/net_namespace.c:1212 register_pernet_subsys+0x24/0x40 net/core/net_namespace.c:1253 nsim_fib_init+0x12/0x70 [netdevsim] veth_get_link_ksettings+0x2b/0x50 [veth] do_one_initcall+0xd4/0x454 init/main.c:939 do_init_module+0xe0/0x330 kernel/module.c:3490 load_module+0x3c2f/0x4620 kernel/module.c:3841 __do_sys_finit_module+0x163/0x190 kernel/module.c:3931 do_syscall_64+0x72/0x2e0 arch/x86/entry/common.c:296 entry_SYSCALL_64_after_hwframe+0x49/0xbe Freed by task 3534: save_stack+0x19/0x80 mm/kasan/common.c:69 set_track mm/kasan/common.c:77 [inline] __kasan_slab_free+0x130/0x180 mm/kasan/common.c:455 slab_free_hook mm/slub.c:1423 [inline] slab_free_freelist_hook mm/slub.c:1474 [inline] slab_free mm/slub.c:3016 [inline] kfree+0xe9/0x2d0 mm/slub.c:3957 ops_free net/core/net_namespace.c:151 [inline] ops_free_list.part.7+0x156/0x220 net/core/net_namespace.c:184 ops_free_list net/core/net_namespace.c:182 [inline] __unregister_pernet_operations net/core/net_namespace.c:1165 [inline] unregister_pernet_operations+0x221/0x2a0 net/core/net_namespace.c:1224 unregister_pernet_subsys+0x1d/0x30 net/core/net_namespace.c:1271 nsim_fib_exit+0x11/0x20 [netdevsim] nsim_module_exit+0x16/0x21 [netdevsim] __do_sys_delete_module kernel/module.c:1015 [inline] __se_sys_delete_module kernel/module.c:958 [inline] __x64_sys_delete_module+0x244/0x330 kernel/module.c:958 do_syscall_64+0x72/0x2e0 arch/x86/entry/common.c:296 entry_SYSCALL_64_after_hwframe+0x49/0xbe Reported-by: Hulk Robot <hulkci(a)huawei.com> Fixes: 59c84b9fcf42 ("netdevsim: Restore per-network namespace accounting for fib entries") Signed-off-by: YueHaibing <yuehaibing(a)huawei.com> Acked-by: Jakub Kicinski <jakub.kicinski(a)netronome.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: LiAichun <liaichun(a)huawei.com> --- drivers/net/netdevsim/fib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094746c0..1a251f76d09b 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static struct pernet_operations nsim_fib_net_ops = { void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -258,6 +258,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { pr_err("Failed to register fib notifier\n"); + unregister_pernet_subsys(&nsim_fib_net_ops); goto err_out; } -- 2.20.1

4 3

[PATCH] scsi: Fix kabi change due to add offline_already member in struct scsi_device
by Yang Yingliang 12 Jun '20

12 Jun '20

From: Ye Bin <yebin10(a)huawei.com> hulk inclusion category: bugfix bugzilla: 34604 CVE: NA ----------------------------------------------- Signed-off-by: Ye Bin <yebin10(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Reviewed-by: Hou Tao <houtao1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/scsi/scsi_lib.c | 4 ++-- include/scsi/scsi_device.h | 7 ++++--- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index a8043039f53e..bdb90f5c9eeb 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1438,7 +1438,7 @@ scsi_prep_state_check(struct scsi_device *sdev, struct request *req) * before trying any recovery commands. */ if (!sdev->offline_already) { - sdev->offline_already = true; + sdev->offline_already = 1; sdev_printk(KERN_ERR, sdev, "rejecting I/O to offline device\n"); } @@ -2859,7 +2859,7 @@ scsi_device_set_state(struct scsi_device *sdev, enum scsi_device_state state) break; } - sdev->offline_already = false; + sdev->offline_already = 0; sdev->sdev_state = state; return 0; diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h index 52b255e868a9..550739a5ea96 100644 --- a/include/scsi/scsi_device.h +++ b/include/scsi/scsi_device.h @@ -201,8 +201,6 @@ struct scsi_device { unsigned lun_in_cdb:1; /* Store LUN bits in CDB[1] */ unsigned unmap_limit_for_ws:1; /* Use the UNMAP limit for WRITE SAME */ - bool offline_already; /* Device offline message logged */ - atomic_t disk_events_disable_depth; /* disable depth for disk events */ DECLARE_BITMAP(supported_events, SDEV_EVT_MAXBITS); /* supported events */ @@ -230,8 +228,11 @@ struct scsi_device { struct mutex state_mutex; enum scsi_device_state sdev_state; struct task_struct *quiesced_by; - +#ifndef __GENKSYMS__ + unsigned long offline_already; /* Device offline message logged */ +#else KABI_RESERVE(1) +#endif KABI_RESERVE(2) KABI_RESERVE(3) KABI_RESERVE(4) -- 2.25.1

1 0

[PATCH] ALSA: proc: Avoid possible leaks of snd_info_entry objects
by Yang Yingliang 12 Jun '20

12 Jun '20

From: Takashi Iwai <tiwai(a)suse.de> mainline inclusion from mainline-v5.1-rc1 commit 3a55437141a1d287dead685b37fe240185144f15 category: bugfix bugzilla: 34614 CVE: NA ----------------------------------------------- This patch changes the parent pointer assignment of snd_info_entry object to be always non-NULL. More specifically,check the parent argument in snd_info_create_module_entry() & co, and assign snd_proc_root if NULL is passed there. This assures that the proc object is always freed when the root is freed, so avoid possible memory leaks. For example, some error paths (e.g. snd_info_register() error at snd_minor_info_init()) may leave snd_info_entry object although the proc file itself is freed. Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Signed-off-by: Ye Bin <yebin10(a)huawei.com> Reviewed-by: Yang Yingliang <yangyingliang(a)huawei.com> Signed-off-by: Ye Bin <yebin10(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- sound/core/info.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/sound/core/info.c b/sound/core/info.c index 679136fba730..3411bea54d66 100644 --- a/sound/core/info.c +++ b/sound/core/info.c @@ -744,7 +744,11 @@ struct snd_info_entry *snd_info_create_module_entry(struct module * module, const char *name, struct snd_info_entry *parent) { - struct snd_info_entry *entry = snd_info_create_entry(name, parent); + struct snd_info_entry *entry; + + if (!parent) + parent = snd_proc_root; + entry = snd_info_create_entry(name, parent); if (entry) entry->module = module; return entry; @@ -765,7 +769,11 @@ struct snd_info_entry *snd_info_create_card_entry(struct snd_card *card, const char *name, struct snd_info_entry * parent) { - struct snd_info_entry *entry = snd_info_create_entry(name, parent); + struct snd_info_entry *entry; + + if (!parent) + parent = card->proc_root; + entry = snd_info_create_entry(name, parent); if (entry) { entry->module = card->module; entry->card = card; -- 2.25.1

1 0

[PATCH 01/12] jbd2: clean __jbd2_journal_abort_hard() and __journal_abort_soft()
by Yang Yingliang 12 Jun '20

12 Jun '20

From: "zhangyi (F)" <yi.zhang(a)huawei.com> mainline inclusion from mainline-5.6-rc1 commit 7f6225e446cc8dfa4c3c7959a4de3dd03ec277bf category: bugfix bugzilla: 34619 CVE: NA --------------------------- __jbd2_journal_abort_hard() is no longer used, so now we can merge __jbd2_journal_abort_hard() and __journal_abort_soft() these two functions into jbd2_journal_abort() and remove them. Signed-off-by: zhangyi (F) <yi.zhang(a)huawei.com> Reviewed-by: Jan Kara <jack(a)suse.cz> Link: https://lore.kernel.org/r/20191204124614.45424-5-yi.zhang@huawei.com Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Reviewed-by: Zhang Xiaoxu <zhangxiaoxu5(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/jbd2/journal.c | 103 ++++++++++++++++++------------------------- include/linux/jbd2.h | 1 - 2 files changed, 42 insertions(+), 62 deletions(-) diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 1f085f0bf908..7301bb766172 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -101,7 +101,6 @@ EXPORT_SYMBOL(jbd2_journal_release_jbd_inode); EXPORT_SYMBOL(jbd2_journal_begin_ordered_truncate); EXPORT_SYMBOL(jbd2_inode_cache); -static void __journal_abort_soft (journal_t *journal, int errno); static int jbd2_journal_create_slab(size_t slab_size); #ifdef CONFIG_JBD2_DEBUG @@ -810,7 +809,7 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, "at offset %lu on %s\n", __func__, blocknr, journal->j_devname); err = -EIO; - __journal_abort_soft(journal, err); + jbd2_journal_abort(journal, err); } } else { *retp = blocknr; /* +journal->j_blk_offset */ @@ -2108,64 +2107,6 @@ int jbd2_journal_wipe(journal_t *journal, int write) return err; } -/* - * Journal abort has very specific semantics, which we describe - * for journal abort. - * - * Two internal functions, which provide abort to the jbd layer - * itself are here. - */ - -/* - * Quick version for internal journal use (doesn't lock the journal). - * Aborts hard --- we mark the abort as occurred, but do _nothing_ else, - * and don't attempt to make any other journal updates. - */ -void __jbd2_journal_abort_hard(journal_t *journal) -{ - transaction_t *transaction; - - if (journal->j_flags & JBD2_ABORT) - return; - - printk(KERN_ERR "Aborting journal on device %s.\n", - journal->j_devname); - - write_lock(&journal->j_state_lock); - journal->j_flags |= JBD2_ABORT; - transaction = journal->j_running_transaction; - if (transaction) - __jbd2_log_start_commit(journal, transaction->t_tid); - write_unlock(&journal->j_state_lock); -} - -/* Soft abort: record the abort error status in the journal superblock, - * but don't do any other IO. */ -static void __journal_abort_soft (journal_t *journal, int errno) -{ - int old_errno; - - write_lock(&journal->j_state_lock); - old_errno = journal->j_errno; - if (!journal->j_errno || errno == -ESHUTDOWN) - journal->j_errno = errno; - - if (journal->j_flags & JBD2_ABORT) { - write_unlock(&journal->j_state_lock); - if (old_errno != -ESHUTDOWN && errno == -ESHUTDOWN) - jbd2_journal_update_sb_errno(journal); - return; - } - write_unlock(&journal->j_state_lock); - - __jbd2_journal_abort_hard(journal); - - jbd2_journal_update_sb_errno(journal); - write_lock(&journal->j_state_lock); - journal->j_flags |= JBD2_REC_ERR; - write_unlock(&journal->j_state_lock); -} - /** * void jbd2_journal_abort () - Shutdown the journal immediately. * @journal: the journal to shutdown. @@ -2209,7 +2150,47 @@ static void __journal_abort_soft (journal_t *journal, int errno) void jbd2_journal_abort(journal_t *journal, int errno) { - __journal_abort_soft(journal, errno); + transaction_t *transaction; + + /* + * ESHUTDOWN always takes precedence because a file system check + * caused by any other journal abort error is not required after + * a shutdown triggered. + */ + write_lock(&journal->j_state_lock); + if (journal->j_flags & JBD2_ABORT) { + int old_errno = journal->j_errno; + + write_unlock(&journal->j_state_lock); + if (old_errno != -ESHUTDOWN && errno == -ESHUTDOWN) { + journal->j_errno = errno; + jbd2_journal_update_sb_errno(journal); + } + return; + } + + /* + * Mark the abort as occurred and start current running transaction + * to release all journaled buffer. + */ + pr_err("Aborting journal on device %s.\n", journal->j_devname); + + journal->j_flags |= JBD2_ABORT; + journal->j_errno = errno; + transaction = journal->j_running_transaction; + if (transaction) + __jbd2_log_start_commit(journal, transaction->t_tid); + write_unlock(&journal->j_state_lock); + + /* + * Record errno to the journal super block, so that fsck and jbd2 + * layer could realise that a filesystem check is needed. + */ + jbd2_journal_update_sb_errno(journal); + + write_lock(&journal->j_state_lock); + journal->j_flags |= JBD2_REC_ERR; + write_unlock(&journal->j_state_lock); } /** diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 7dd9d25a9cbb..12b935c9ec1e 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1427,7 +1427,6 @@ extern int jbd2_journal_skip_recovery (journal_t *); extern void jbd2_journal_update_sb_errno(journal_t *); extern int jbd2_journal_update_sb_log_tail (journal_t *, tid_t, unsigned long, int); -extern void __jbd2_journal_abort_hard (journal_t *); extern void jbd2_journal_abort (journal_t *, int); extern int jbd2_journal_errno (journal_t *); extern void jbd2_journal_ack_err (journal_t *); -- 2.25.1

1 11

Minutes of the meeting //: 【Meeting Notice】Kernel & testing sync between Linaro and openEuler Time: 2020-06-10 16:00-17:30
by Guohanjun (Hanjun Guo) 11 Jun '20

11 Jun '20

Attendees (I can see more people join later but can’t remember now): [cid:image001.png@01D63F6C.270B2CF0] Following action items: * Vincent to share the JIRA link for thermal pressure * Vincent to share the JIRA link for scheduler improvements * Anmar to work on the opening of Power and Performance Project * Hanjun/Jonathan to share some plan for openEuler kernel in the coming 6 or 12 months * Anmar to create the open mailing list for LKQ project, so that Fengguang and the team can participate And see attached for Linaro slide, thanks! 发件人: Meeting Book via Kernel [mailto:kernel@openeuler.org] 发送时间: 2020年6月3日 16:01 收件人: kernel(a)openeuler.org 主题: 【Meeting Notice】Kernel & testing sync between Linaro and openEuler Time: 2020-06-10 16:00-17:30 Topic Kernel & testing sync between Linaro and openEuler Time 2020-06-10 16:00-17:30((UTC+08:00)Beijing) Join Conference Join (External) >><https://welink-meeting.zoom.us/j/834726983> Meeting ID 834 726 983 Convener 郭寒军 Agenda * Linux Kernel Update by Linaro - 15 mins * Linux Kernel Functional Testing (LKFT) introduction - 15mins * Brief introduction for openEuler and the kernel - 15mins * Brief introduction for Crystal testing system - 15 mins * Open discussions - about 30 mins Tips: Dial the access number to join conference<http://app.huawei.com/eshare/voiceMeetings> 注：仅支持部分国家和地区。( Only some countries and regions are supported.) 请点击会议链接按照提示下载Zoom客户端入会，并升级Zoom最新版本。 Please click the conference link and download the Zoom to join the conference, and upgrade to the latest version of Zoom.

2 1

[PATCH] irqchip/gic-v3-its: its supoport s3/s4
by Hongbo Yao 09 Jun '20

09 Jun '20

commit dba0bc7b76dcf(irqchip/gic-v3-its: add ability to save/restore ITS state) add support to save/restore its suspend state. However, it needs to set ITS_FLAGS_SAVE_SUSPEND_STATE to 1 when we want to resend MAPC. In GIC500, GITS_typer.HCC is not zero, and it can be used to judge if we can resend MAPC if the firmware restores the state, but when GITS_typer.HCC is zero, and it will never have chance to set the flag, which will lead to a failuer to power off GIC logic in suspend. Signed-off-by: Hongbo Yao <yaohongbo(a)huawei.com> --- drivers/irqchip/irq-gic-v3-its.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c index 881466163c99..6ea2809b8680 100644 --- a/drivers/irqchip/irq-gic-v3-its.c +++ b/drivers/irqchip/irq-gic-v3-its.c @@ -3733,8 +3733,7 @@ static int __init its_probe_one(struct resource *res, ctlr |= GITS_CTLR_ImDe; writel_relaxed(ctlr, its->base + GITS_CTLR); - if (GITS_TYPER_HCC(typer)) - its->flags |= ITS_FLAGS_SAVE_SUSPEND_STATE; + its->flags |= ITS_FLAGS_SAVE_SUSPEND_STATE; err = its_init_domain(handle, its); if (err) -- 2.20.1

1 0

(no subject)
by liaichun 09 Jun '20

09 Jun '20

From: LiAichun <liaichun(a)huawei.com> euleros inclusion commit NA category:bugfix bugzilla:NA CVE:NA issue:I1JZHT ------------------------------------------------- From:YueHaibing<> Subject: [PATCH] netdevsim: Fix error handling in nsim_fib_init and neim_fib_exit Date: Mon, 8 Jun 2020 15:13:40 +0800 In nsim_fib_init(),if register_fib_notifier failed,nsim_fib_net_ops should be unregistered before return; In nsim_fib_exit(),unregister_fib_notifier should be called before nsim_fib_net_ops be unregistered, otherwise may cause use-after-free More detailed information can refer to: https://lkml.org/lkml/2019/10/11/216 Reported-by:Hulk Robot <hulkci(a)huawei.com> Fixes: 59c84b9fcf42("netdevsim: Restore per-network namespace accounting for fib entries") Signed-off-by: YueHaibing <yuehaibing(a)huawei.com> Signed-off-by: LiAichun <liaichun(a)huawei.com> --- drivers/net/netdevsim/fib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094..d5c01b8 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static int __net_init nsim_fib_netns_init(struct net *net) void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -257,6 +257,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { + unregister_pernet_subsys(&nsim_fib_net_ops); pr_err("Failed to register fib notifier\n"); goto err_out; } -- 1.8.3.1

1 0

(no subject)
by liaichun 09 Jun '20

09 Jun '20

From: LiAichun <liaichun(a)huawei.com> euleros inclusion commit NA category:bugfix bugzilla:NA CVE:NA Fix issue:gitee.com/openeuler/kernel/issues/I1JZHT ------------------------------------------------- From:YueHaibing<> Subject: [PATCH] netdevsim: Fix error handling in nsim_fib_init and neim_fib_exit Date: Mon, 8 Jun 2020 15:13:40 +0800 In nsim_fib_init(),if register_fib_notifier failed,nsim_fib_net_ops should be unregistered before return; In nsim_fib_exit(),unregister_fib_notifier should be called before nsim_fib_net_ops be unregistered, otherwise may cause use-after-free More detailed information can refer to: https://lkml.org/lkml/2019/10/11/216 Reported-by:Hulk Robot <hulkci(a)huawei.com> Fixes: 59c84b9fcf42("netdevsim: Restore per-network namespace accounting for fib entries") Signed-off-by: YueHaibing <yuehaibing(a)huawei.com> Signed-off-by: LiAichun <liaichun(a)huawei.com> --- drivers/net/netdevsim/fib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094..d5c01b8 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static int __net_init nsim_fib_netns_init(struct net *net) void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -257,6 +257,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { + unregister_pernet_subsys(&nsim_fib_net_ops); pr_err("Failed to register fib notifier\n"); goto err_out; } -- 1.8.3.1

1 0

(no subject)
by liaichun 09 Jun '20

09 Jun '20

From: LiAichun <liaichun(a)huawei.com> mainline inclusion from mainline-4.19.90-2003.4.0 commit a91b7a28e76a702bd81357bf41a45dab6486716b category:bugfix bugzilla:NA CVE:NA ------------------------------------------------- From:YueHaibing<> Subject: [PATCH] netdevsim: Fix error handling in nsim_fib_init and neim_fib_exit Date: Mon, 8 Jun 2020 15:13:40 +0800 In nsim_fib_init(),if register_fib_notifier failed,nsim_fib_net_ops should be unregistered before return; In nsim_fib_exit(),unregister_fib_notifier should be called before nsim_fib_net_ops be unregistered, otherwise may cause use-after-free More detailed information can refer to: https://lkml.org/lkml/2019/10/11/216 Reported-by:Hulk Robot <hulkci(a)huawei.com> Fixes: 59c84b9fcf42("netdevsim: Restore per-network namespace accounting for fib entries") Signed-off-by: YueHaibing <yuehaibing(a)huawei.com> Signed-off-by: LiAichun <liaichun(a)huawei.com> --- drivers/net/netdevsim/fib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094..d5c01b8 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static int __net_init nsim_fib_netns_init(struct net *net) void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -257,6 +257,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { + unregister_pernet_subsys(&nsim_fib_net_ops); pr_err("Failed to register fib notifier\n"); goto err_out; } -- 1.8.3.1

2 1

[PATCH] net-netdevsim-fix-resource-leak
by liaichun 09 Jun '20

09 Jun '20

In nsim_fib_init(),if register_fib_notifier failed,nsim_fib_net_ops should be unregistered before return; In nsim_fib_exit(),unregister_fib_notifier should be called before nsim_fib_net_ops be unregistered, otherwise may cause use-after-free More detailed information can refer to: https://lkml.org/lkml/2019/10/11/216 kernel inclusion from:kernel-4.19 commit:2dca76fa95bf03c15bdcacd7cfa8f974c1a799f6 category:bugfix bugzilla:NA CVE:NA Signed-off-by: liaichun <513565428(a)qq.com> DESC:fix netdevsim resource leak --- drivers/net/netdevsim/fib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094..d5c01b8 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static int __net_init nsim_fib_netns_init(struct net *net) void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -257,6 +257,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { + unregister_pernet_subsys(&nsim_fib_net_ops); pr_err("Failed to register fib notifier\n"); goto err_out; } -- 1.8.3.1

2 1

[PATCH] net-netdevsim-fix-resource-leak
by liaichun 08 Jun '20

08 Jun '20

In nsim_fib_init(),if register_fib_notifier failed,nsim_fib_net_ops should be unregistered before return; In nsim_fib_exit(),unregister_fib_notifier should be called before nsim_fib_net_ops be unregistered, otherwise may cause use-after-free More detailed information can refer to: https://lkml.org/lkml/2019/10/11/216 Signed-off-by: liaichun <513565428(a)qq.com> Tested-by: liaichun <513565428(a)qq.com> Type:bugfix ID:NA SUG:restart DESC:fix netdevsim resource leak --- drivers/net/netdevsim/fib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094..d5c01b8 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static int __net_init nsim_fib_netns_init(struct net *net) void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -257,6 +257,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { + unregister_pernet_subsys(&nsim_fib_net_ops); pr_err("Failed to register fib notifier\n"); goto err_out; } -- 1.8.3.1

1 0

[PATCH] net-netdevsim-fix-resource-leak
by liaichun 08 Jun '20

08 Jun '20

diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094..d5c01b8 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static int __net_init nsim_fib_netns_init(struct net *net) void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -257,6 +257,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { + unregister_pernet_subsys(&nsim_fib_net_ops); pr_err("Failed to register fib notifier\n"); goto err_out; } -- 1.8.3.1

2 1

[PATCH] net-netdevsim-fix-resource-leak
by liaichun 08 Jun '20

08 Jun '20

From: liaichun <liaichun(a)huawei.com> diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094..d5c01b8 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static int __net_init nsim_fib_netns_init(struct net *net) void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -257,6 +257,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { + unregister_pernet_subsys(&nsim_fib_net_ops); pr_err("Failed to register fib notifier\n"); goto err_out; } -- 1.8.3.1

1 0

[PATCH] net-netdevsim-fix-resource-leak
by liaichun 08 Jun '20

08 Jun '20

diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094..d5c01b8 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static int __net_init nsim_fib_netns_init(struct net *net) void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -257,6 +257,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { + unregister_pernet_subsys(&nsim_fib_net_ops); pr_err("Failed to register fib notifier\n"); goto err_out; } -- 1.8.3.1

1 0

[PATCH] net-netdevsim-fix-resource-leak
by liaichun 08 Jun '20

08 Jun '20

From: liaichun <liaichun(a)huawei.com> diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094..d5c01b8 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static int __net_init nsim_fib_netns_init(struct net *net) void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -257,6 +257,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { + unregister_pernet_subsys(&nsim_fib_net_ops); pr_err("Failed to register fib notifier\n"); goto err_out; } -- 1.8.3.1

1 0

[PATCH] net-netdevsim-fix-resource-leak
by liaichun 08 Jun '20

08 Jun '20

diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094..d5c01b8 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static int __net_init nsim_fib_netns_init(struct net *net) void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -257,6 +257,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { + unregister_pernet_subsys(&nsim_fib_net_ops); pr_err("Failed to register fib notifier\n"); goto err_out; } -- 1.8.3.1

1 0

[PATCH v2] scsi: hisi_sas: directly trigger error handle of scsi instead of waiting for timeout
by Luo Jiaxing 05 Jun '20

05 Jun '20

We used timeout mechanism of scsi layer to trigger some IO's error handle, this type of abnormal IO require driver to enter error handle to clear the residue in the hardware. But timeout mechanism caught error handle time to be longer, some threads need to wait for tens of seconds when IO failed and enter error handle, so we try to trigger error handling directly. Signed-off-by: Luo Jiaxing <luojiaxing(a)huawei.com> --- drivers/scsi/hisi_sas/hisi_sas_v1_hw.c | 4 +++- drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 4 +++- drivers/scsi/hisi_sas/hisi_sas_v3_hw.c | 4 +++- 3 files changed, 9 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c index 031091e..49ea3b7 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c +++ b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c @@ -1263,8 +1263,10 @@ static int slot_complete_v1_hw(struct hisi_hba *hisi_hba, !(cmplt_hdr_data & CMPLT_HDR_RSPNS_XFRD_MSK)) { slot_err_v1_hw(hisi_hba, task, slot); - if (unlikely(slot->abort)) + if (unlikely(slot->abort)) { + sas_task_abort(task); return ts->stat; + } goto out; } diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c index 6568e20..8ebddf5 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c +++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c @@ -2406,8 +2406,10 @@ slot_complete_v2_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot) error_info[0], error_info[1], error_info[2], error_info[3]); - if (unlikely(slot->abort)) + if (unlikely(slot->abort)) { + sas_task_abort(task); return ts->stat; + } goto out; } diff --git a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c index 5b80856..7fd067c 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c +++ b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c @@ -2408,8 +2408,10 @@ slot_complete_v3_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot) dev_info(dev, "data underflow without sense, rsp_code:0x%x, rc:%d.\n", iu->resp_data[0], rc); } - if (unlikely(slot->abort)) + if (unlikely(slot->abort)) { + sas_task_abort(task); return ts->stat; + } goto out; } -- 2.7.4

1 0

[PATCH v1 hulk-4.19-next] scsi: hisi_sas: directly trigger error handle of scsi instead of waiting for timeout
by Luo Jiaxing 05 Jun '20

05 Jun '20

driver inclusion category: bugfix bugzilla: NA DTS: NA ----------------------------------------------------------------------- We used timeout mechanism of scsi layer to trigger some IO's error handle, this type of abnormal IO require driver to enter error handle to clear the residue in the hardware. But timeout mechanism caught error handle time to be longer, some threads need to wait for tens of seconds when IO failed and enter error handle, so we try to trigger error handling directly. Signed-off-by: Luo Jiaxing <luojiaxing(a)huawei.com> --- drivers/scsi/hisi_sas/hisi_sas_v1_hw.c | 4 +++- drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 4 +++- drivers/scsi/hisi_sas/hisi_sas_v3_hw.c | 4 +++- 3 files changed, 9 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c index 031091e..49ea3b7 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c +++ b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c @@ -1263,8 +1263,10 @@ static int slot_complete_v1_hw(struct hisi_hba *hisi_hba, !(cmplt_hdr_data & CMPLT_HDR_RSPNS_XFRD_MSK)) { slot_err_v1_hw(hisi_hba, task, slot); - if (unlikely(slot->abort)) + if (unlikely(slot->abort)) { + sas_task_abort(task); return ts->stat; + } goto out; } diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c index 6568e20..8ebddf5 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c +++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c @@ -2406,8 +2406,10 @@ slot_complete_v2_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot) error_info[0], error_info[1], error_info[2], error_info[3]); - if (unlikely(slot->abort)) + if (unlikely(slot->abort)) { + sas_task_abort(task); return ts->stat; + } goto out; } diff --git a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c index 5b80856..7fd067c 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c +++ b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c @@ -2408,8 +2408,10 @@ slot_complete_v3_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot) dev_info(dev, "data underflow without sense, rsp_code:0x%x, rc:%d.\n", iu->resp_data[0], rc); } - if (unlikely(slot->abort)) + if (unlikely(slot->abort)) { + sas_task_abort(task); return ts->stat; + } goto out; } -- 2.7.4

1 0

【Meeting Notice】Kernel & testing sync between Linaro and openEuler Time: 2020-06-10 16:00-17:30
by Meeting Book 03 Jun '20

03 Jun '20

1 0

【Meeting Notice】Kernel & testing sync between Linaro and openEuler Time: 2020-06-10 16:00-17:30
by Meeting Book 03 Jun '20

03 Jun '20

1 0

[PATCH 01/76] net: dsa: Do not make user port errors fatal
by Yang Yingliang 01 Jun '20

01 Jun '20

From: Florian Fainelli <f.fainelli(a)gmail.com> commit 86f8b1c01a0a537a73d2996615133be63cdf75db upstream. Prior to 1d27732f411d ("net: dsa: setup and teardown ports"), we would not treat failures to set-up an user port as fatal, but after this commit we would, which is a regression for some systems where interfaces may be declared in the Device Tree, but the underlying hardware may not be present (pluggable daughter cards for instance). Fixes: 1d27732f411d ("net: dsa: setup and teardown ports") Signed-off-by: Florian Fainelli <f.fainelli(a)gmail.com> Reviewed-by: Andrew Lunn <andrew(a)lunn.ch> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- net/dsa/dsa2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 46ae4de..b036c55 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -412,7 +412,7 @@ static int dsa_tree_setup_switches(struct dsa_switch_tree *dst) err = dsa_switch_setup(ds); if (err) - return err; + continue; for (port = 0; port < ds->num_ports; port++) { dp = &ds->ports[port]; -- 1.8.3

1 75

[PATCH 1/9] net/hinic: Fix the firmware compatibility bug in the MAC reuse scenario
by Yang Yingliang 01 Jun '20

01 Jun '20

From: Chiqijun <chiqijun(a)huawei.com> driver inclusion category: bugfix bugzilla: 4472 ----------------------------------------------------------------------- 2.3.0.1 and later drivers and firmware add MAC reuse feature, when the driver is compatible with firmware version 2.3.0.0 and before, if the ip link command is used to configure the MAC and VLAN for the VF, it will cause the VF network anomaly. To solve this problem, the driver obtains the function capability from the firmware when loading, and determines whether to support MAC reuse for different processing. Signed-off-by: Chiqijun <chiqijun(a)huawei.com> Reviewed-by: Zengweiliang <zengweiliang.zengweiliang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/net/ethernet/huawei/hinic/hinic_hwdev.h | 1 + .../ethernet/huawei/hinic/hinic_mgmt_interface.h | 11 ++++ drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c | 45 +++++++++++++++-- drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.h | 4 ++ drivers/net/ethernet/huawei/hinic/hinic_nic_io.c | 10 +++- drivers/net/ethernet/huawei/hinic/hinic_port_cmd.h | 2 + drivers/net/ethernet/huawei/hinic/hinic_sriov.c | 59 ++++++++++++++-------- 7 files changed, 106 insertions(+), 26 deletions(-) diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hwdev.h b/drivers/net/ethernet/huawei/hinic/hinic_hwdev.h index a096f91..2a81867 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hwdev.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hwdev.h @@ -319,6 +319,7 @@ struct hinic_hwdev { struct hinic_board_info board_info; #define MGMT_VERSION_MAX_LEN 32 u8 mgmt_ver[MGMT_VERSION_MAX_LEN]; + u64 fw_support_func_flag; }; int hinic_init_comm_ch(struct hinic_hwdev *hwdev); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_mgmt_interface.h b/drivers/net/ethernet/huawei/hinic/hinic_mgmt_interface.h index 3c3be6e..8673cf5 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_mgmt_interface.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_mgmt_interface.h @@ -690,6 +690,15 @@ struct hinic_port_rt_cmd { u8 rsvd1[6]; }; +struct fw_support_func { + u8 status; + u8 version; + u8 rsvd0[6]; + + u64 flag; + u64 rsvd; +}; + struct hinic_vf_dcb_state { u8 status; u8 version; @@ -741,6 +750,8 @@ struct hinic_set_link_follow { int hinic_get_base_qpn(void *hwdev, u16 *global_qpn); +int hinic_get_fw_support_func(void *hwdev); + int hinic_vf_func_init(struct hinic_hwdev *hwdev); void hinic_vf_func_free(struct hinic_hwdev *hwdev); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c b/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c index f9e16ec..7dfedbd 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c @@ -201,6 +201,34 @@ int hinic_get_base_qpn(void *hwdev, u16 *global_qpn) return 0; } +int hinic_get_fw_support_func(void *hwdev) +{ + struct fw_support_func support_flag = {0}; + struct hinic_hwdev *dev = hwdev; + u16 out_size = sizeof(support_flag); + int err; + + if (!hwdev) + return -EINVAL; + + err = hinic_msg_to_mgmt_sync(hwdev, HINIC_MOD_L2NIC, + HINIC_PORT_CMD_GET_FW_SUPPORT_FLAG, + &support_flag, sizeof(support_flag), + &support_flag, &out_size, 0); + if (support_flag.status == HINIC_MGMT_CMD_UNSUPPORTED) { + nic_info(dev->dev_hdl, "Current firmware doesn't support to get function capability\n"); + support_flag.flag = 0; + } else if (support_flag.status || err || !out_size) { + nic_err(dev->dev_hdl, "Failed to get function capability, err: %d, status: 0x%x, out size: 0x%x\n", + err, support_flag.status, out_size); + return -EFAULT; + } + + dev->fw_support_func_flag = support_flag.flag; + + return 0; +} + #define HINIC_ADD_VLAN_IN_MAC 0x8000 #define HINIC_VLAN_ID_MASK 0x7FFF @@ -352,6 +380,11 @@ int hinic_update_mac_vlan(void *hwdev, u16 old_vlan, u16 new_vlan, int vf_id) if (!vf_info->pf_set_mac) return 0; + if (!FW_SUPPORT_MAC_REUSE_FUNC(dev)) { + nic_info(dev->dev_hdl, "Current firmware doesn't support mac reuse\n"); + return 0; + } + func_id = hinic_glb_pf_vf_offset(dev) + (u16)vf_id; vlan_id = old_vlan; if (vlan_id) @@ -2562,9 +2595,14 @@ static int hinic_init_vf_config(struct hinic_hwdev *hwdev, u16 vf_id) vf_info = hwdev->nic_io->vf_infos + HW_VF_ID_TO_OS(vf_id); if (vf_info->pf_set_mac) { func_id = hinic_glb_pf_vf_offset(hwdev) + vf_id; - vlan_id = vf_info->pf_vlan; - if (vlan_id) - vlan_id |= HINIC_ADD_VLAN_IN_MAC; + if (FW_SUPPORT_MAC_REUSE_FUNC(hwdev)) { + vlan_id = vf_info->pf_vlan; + if (vlan_id) + vlan_id |= HINIC_ADD_VLAN_IN_MAC; + } else { + vlan_id = 0; + } + err = hinic_set_mac(hwdev, vf_info->vf_mac_addr, vlan_id, func_id); if (err) { @@ -4065,3 +4103,4 @@ int hinic_get_sfp_type(void *hwdev, u8 *data0, u8 *data1) return 0; } + diff --git a/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.h b/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.h index 427d469..e0979df 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.h @@ -19,6 +19,10 @@ #define OS_VF_ID_TO_HW(os_vf_id) ((os_vf_id) + 1) #define HW_VF_ID_TO_OS(hw_vf_id) ((hw_vf_id) - 1) +#define FW_SUPPORT_MAC_REUSE 0x1 +#define FW_SUPPORT_MAC_REUSE_FUNC(hwdev) \ + ((hwdev)->fw_support_func_flag & FW_SUPPORT_MAC_REUSE) + #define HINIC_VLAN_PRIORITY_SHIFT 13 #define HINIC_RSS_INDIR_SIZE 256 diff --git a/drivers/net/ethernet/huawei/hinic/hinic_nic_io.c b/drivers/net/ethernet/huawei/hinic/hinic_nic_io.c index 1cb4e26..f2fb0bc 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_nic_io.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_nic_io.c @@ -831,9 +831,17 @@ int hinic_init_nic_hwdev(void *hwdev, u16 rx_buff_len) } /* VFs don't set port routine command report */ - if (hinic_func_type(dev) != TYPE_VF) + if (hinic_func_type(dev) != TYPE_VF) { + /* Get the fw support mac reuse flag */ + err = hinic_get_fw_support_func(hwdev); + if (err) { + nic_err(dev->dev_hdl, "Failed to get function capability\n"); + return err; + } + /* Inform mgmt to send sfp's information to driver */ err = hinic_set_port_routine_cmd_report(hwdev, true); + } return err; } diff --git a/drivers/net/ethernet/huawei/hinic/hinic_port_cmd.h b/drivers/net/ethernet/huawei/hinic/hinic_port_cmd.h index 46464b1f..d428f48 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_port_cmd.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_port_cmd.h @@ -146,6 +146,8 @@ enum hinic_port_cmd { HINIC_PORT_CMD_SET_PORT_LINK_STATUS = 0x76, HINIC_PORT_CMD_SET_CGE_PAUSE_TIME_CFG = 0x77, + HINIC_PORT_CMD_GET_FW_SUPPORT_FLAG = 0x79, + HINIC_PORT_CMD_SET_PORT_REPORT = 0x7B, HINIC_PORT_CMD_LINK_STATUS_REPORT = 0xa0, diff --git a/drivers/net/ethernet/huawei/hinic/hinic_sriov.c b/drivers/net/ethernet/huawei/hinic/hinic_sriov.c index 37d0116..73ab2d7 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_sriov.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_sriov.c @@ -266,6 +266,42 @@ int hinic_ndo_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) /*lint -save -e574 -e734*/ #ifdef IFLA_VF_MAX +static int set_hw_vf_vlan(struct hinic_sriov_info *sriov_info, + u16 cur_vlanprio, int vf, u16 vlan, u8 qos) +{ + int err = 0; + u16 old_vlan = cur_vlanprio & VLAN_VID_MASK; + + if (vlan || qos) { + if (cur_vlanprio) { + err = hinic_kill_vf_vlan(sriov_info->hwdev, + OS_VF_ID_TO_HW(vf)); + if (err) { + nic_err(&sriov_info->pdev->dev, "Failed to delete vf %d old vlan %d\n", + vf, old_vlan); + return err; + } + } + err = hinic_add_vf_vlan(sriov_info->hwdev, + OS_VF_ID_TO_HW(vf), vlan, qos); + if (err) { + nic_err(&sriov_info->pdev->dev, "Failed to add vf %d new vlan %d\n", + vf, vlan); + return err; + } + } else { + err = hinic_kill_vf_vlan(sriov_info->hwdev, OS_VF_ID_TO_HW(vf)); + if (err) { + nic_err(&sriov_info->pdev->dev, "Failed to delete vf %d vlan %d\n", + vf, old_vlan); + return err; + } + } + + return hinic_update_mac_vlan(sriov_info->hwdev, old_vlan, vlan, + OS_VF_ID_TO_HW(vf)); +} + #ifdef IFLA_VF_VLAN_INFO_MAX int hinic_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos, __be16 vlan_proto) @@ -276,7 +312,6 @@ int hinic_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos) struct hinic_nic_dev *adapter = netdev_priv(netdev); struct hinic_sriov_info *sriov_info; u16 vlanprio, cur_vlanprio; - int err = 0; if (!FUNC_SUPPORT_SET_VF_MAC_VLAN(adapter->hwdev)) { nicif_err(adapter, drv, netdev, @@ -298,27 +333,7 @@ int hinic_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos) if (vlanprio == cur_vlanprio) return 0; - if (vlan || qos) { - if (cur_vlanprio) - err = hinic_kill_vf_vlan(sriov_info->hwdev, - OS_VF_ID_TO_HW(vf)); - if (err) - goto out; - err = hinic_add_vf_vlan(sriov_info->hwdev, OS_VF_ID_TO_HW(vf), - vlan, qos); - } else { - err = hinic_kill_vf_vlan(sriov_info->hwdev, OS_VF_ID_TO_HW(vf)); - } - - if (err) - return err; - - err = hinic_update_mac_vlan(sriov_info->hwdev, - cur_vlanprio & VLAN_VID_MASK, vlan, - OS_VF_ID_TO_HW(vf)); - -out: - return err; + return set_hw_vf_vlan(sriov_info, cur_vlanprio, vf, vlan, qos); } #endif -- 1.8.3

1 8

[PATCH] irqchip/gic-v3-its: Probe ITS page size for all GITS_BASERn registers
by Yang Yingliang 01 Jun '20

01 Jun '20

From: Marc Zyngier <maz(a)kernel.org> mainline inclusion from mainline-v5.6-rc4 commit d5df9dc96eb7423d3f742b13d5e1e479ff795eaa category: bugfix bugzilla: NA CVE: NA --------------------------- The GICv3 ITS driver assumes that once it has latched on a page size for a given BASER register, it can use the same page size as the maximum page size for all subsequent BASER registers. Although it worked so far, nothing in the architecture guarantees this, and Nianyao Tang hit this problem on some undisclosed implementation. Let's bite the bullet and probe the the supported page size on all BASER registers before starting to populate the tables. This simplifies the setup a bit, at the expense of a few additional MMIO accesses. Signed-off-by: Marc Zyngier <maz(a)kernel.org> Reported-by: Nianyao Tang <tangnianyao(a)huawei.com> Tested-by: Nianyao Tang <tangnianyao(a)huawei.com> Link: https://lore.kernel.org/r/1584089195-63897-1-git-send-email-zhangshaokun@hi… Signed-off-by: Hongbo Yao <yaohongbo(a)huawei.com> conflicts: drivers/irqchip/irq-gic-v3-its.c Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/irqchip/irq-gic-v3-its.c | 101 ++++++++++++++++++++++++++------------- 1 file changed, 67 insertions(+), 34 deletions(-) diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c index 59c6c0d..bfedab2 100644 --- a/drivers/irqchip/irq-gic-v3-its.c +++ b/drivers/irqchip/irq-gic-v3-its.c @@ -17,6 +17,7 @@ #include <linux/acpi.h> #include <linux/acpi_iort.h> +#include <linux/bitfield.h> #include <linux/bitmap.h> #include <linux/cpu.h> #include <linux/crash_dump.h> @@ -1768,17 +1769,16 @@ static void its_write_baser(struct its_node *its, struct its_baser *baser, } static int its_setup_baser(struct its_node *its, struct its_baser *baser, - u64 cache, u64 shr, u32 psz, u32 order, - bool indirect) + u64 cache, u64 shr, u32 order, bool indirect) { u64 val = its_read_baser(its, baser); u64 esz = GITS_BASER_ENTRY_SIZE(val); u64 type = GITS_BASER_TYPE(val); u64 baser_phys, tmp; - u32 alloc_pages; + u32 alloc_pages, psz; void *base; -retry_alloc_baser: + psz = baser->psz; alloc_pages = (PAGE_ORDER_TO_SIZE(order) / psz); if (alloc_pages > GITS_BASER_PAGES_MAX) { pr_warn("ITS@%pa: %s too large, reduce ITS pages %u->%u\n", @@ -1851,25 +1851,6 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser, goto retry_baser; } - if ((val ^ tmp) & GITS_BASER_PAGE_SIZE_MASK) { - /* - * Page size didn't stick. Let's try a smaller - * size and retry. If we reach 4K, then - * something is horribly wrong... - */ - free_pages((unsigned long)base, order); - baser->base = NULL; - - switch (psz) { - case SZ_16K: - psz = SZ_4K; - goto retry_alloc_baser; - case SZ_64K: - psz = SZ_16K; - goto retry_alloc_baser; - } - } - if (val != tmp) { pr_err("ITS@%pa: %s doesn't stick: %llx %llx\n", &its->phys_base, its_base_type_string[type], @@ -1895,13 +1876,14 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser, static bool its_parse_indirect_baser(struct its_node *its, struct its_baser *baser, - u32 psz, u32 *order, u32 ids) + u32 *order, u32 ids) { u64 tmp = its_read_baser(its, baser); u64 type = GITS_BASER_TYPE(tmp); u64 esz = GITS_BASER_ENTRY_SIZE(tmp); u64 val = GITS_BASER_InnerShareable | GITS_BASER_RaWaWb; u32 new_order = *order; + u32 psz = baser->psz; bool indirect = false; /* No need to enable Indirection if memory requirement < (psz*2)bytes */ @@ -1960,11 +1942,58 @@ static void its_free_tables(struct its_node *its) } } +static int its_probe_baser_psz(struct its_node *its, struct its_baser *baser) +{ + u64 psz = SZ_64K; + + while (psz) { + u64 val, gpsz; + + val = its_read_baser(its, baser); + val &= ~GITS_BASER_PAGE_SIZE_MASK; + + switch (psz) { + case SZ_64K: + gpsz = GITS_BASER_PAGE_SIZE_64K; + break; + case SZ_16K: + gpsz = GITS_BASER_PAGE_SIZE_16K; + break; + case SZ_4K: + default: + gpsz = GITS_BASER_PAGE_SIZE_4K; + break; + } + + gpsz >>= GITS_BASER_PAGE_SIZE_SHIFT; + + val |= FIELD_PREP(GITS_BASER_PAGE_SIZE_MASK, gpsz); + its_write_baser(its, baser, val); + + if (FIELD_GET(GITS_BASER_PAGE_SIZE_MASK, baser->val) == gpsz) + break; + + switch (psz) { + case SZ_64K: + psz = SZ_16K; + break; + case SZ_16K: + psz = SZ_4K; + break; + case SZ_4K: + default: + return -1; + } + } + + baser->psz = psz; + return 0; +} + static int its_alloc_tables(struct its_node *its) { u64 shr = GITS_BASER_InnerShareable; u64 cache = GITS_BASER_RaWaWb; - u32 psz = SZ_64K; int err, i; if (its->flags & ITS_FLAGS_WORKAROUND_CAVIUM_22375) @@ -1975,34 +2004,38 @@ static int its_alloc_tables(struct its_node *its) struct its_baser *baser = its->tables + i; u64 val = its_read_baser(its, baser); u64 type = GITS_BASER_TYPE(val); - u32 order = get_order(psz); bool indirect = false; + u32 order; - switch (type) { - case GITS_BASER_TYPE_NONE: + if (type == GITS_BASER_TYPE_NONE) continue; + if (its_probe_baser_psz(its, baser)) { + its_free_tables(its); + return -ENXIO; + } + + order = get_order(baser->psz); + + switch (type) { case GITS_BASER_TYPE_DEVICE: - indirect = its_parse_indirect_baser(its, baser, - psz, &order, + indirect = its_parse_indirect_baser(its, baser, &order, its->device_ids); break; case GITS_BASER_TYPE_VCPU: - indirect = its_parse_indirect_baser(its, baser, - psz, &order, + indirect = its_parse_indirect_baser(its, baser, &order, ITS_MAX_VPEID_BITS); break; } - err = its_setup_baser(its, baser, cache, shr, psz, order, indirect); + err = its_setup_baser(its, baser, cache, shr, order, indirect); if (err < 0) { its_free_tables(its); return err; } /* Update settings which will be used for next BASERn */ - psz = baser->psz; cache = baser->val & GITS_BASER_CACHEABILITY_MASK; shr = baser->val & GITS_BASER_SHAREABILITY_MASK; } -- 1.8.3

1 0

[PATCH 1/4] vfio/mdev: Improve the create/remove sequence
by Yang Yingliang 01 Jun '20

01 Jun '20

From: Parav Pandit <parav(a)mellanox.com> mainline inclusion from mainline-5.2 commit 522ecce08ab20b57342d65b05601818e0f95fb2c category: bugfix bugzilla: NA CVE: NA ------------------------------------------------- This patch addresses below two issues and prepares the code to address 3rd issue listed below. 1. mdev device is placed on the mdev bus before it is created in the vendor driver. Once a device is placed on the mdev bus without creating its supporting underlying vendor device, mdev driver's probe() gets triggered. However there isn't a stable mdev available to work on. create_store() mdev_create_device() device_register() ... vfio_mdev_probe() [...] parent->ops->create() vfio_ap_mdev_create() mdev_set_drvdata(mdev, matrix_mdev); /* Valid pointer set above */ Due to this way of initialization, mdev driver who wants to use the mdev, doesn't have a valid mdev to work on. 2. Current creation sequence is, parent->ops_create() groups_register() Remove sequence is, parent->ops->remove() groups_unregister() However, remove sequence should be exact mirror of creation sequence. Once this is achieved, all users of the mdev will be terminated first before removing underlying vendor device. (Follow standard linux driver model). At that point vendor's remove() ops shouldn't fail because taking the device off the bus should terminate any usage. 3. When remove operation fails, mdev sysfs removal attempts to add the file back on already removed device. Following call trace [1] is observed. [1] call trace: kernel: WARNING: CPU: 2 PID: 9348 at fs/sysfs/file.c:327 sysfs_create_file_ns+0x7f/0x90 kernel: CPU: 2 PID: 9348 Comm: bash Kdump: loaded Not tainted 5.1.0-rc6-vdevbus+ #6 kernel: Hardware name: Supermicro SYS-6028U-TR4+/X10DRU-i+, BIOS 2.0b 08/09/2016 kernel: RIP: 0010:sysfs_create_file_ns+0x7f/0x90 kernel: Call Trace: kernel: remove_store+0xdc/0x100 [mdev] kernel: kernfs_fop_write+0x113/0x1a0 kernel: vfs_write+0xad/0x1b0 kernel: ksys_write+0x5a/0xe0 kernel: do_syscall_64+0x5a/0x210 kernel: entry_SYSCALL_64_after_hwframe+0x49/0xbe Therefore, mdev core is improved in following ways. 1. Split the device registration/deregistration sequence so that some things can be done between initialization of the device and hooking it up to the bus respectively after deregistering it from the bus but before giving up our final reference. In particular, this means invoking the ->create() and ->remove() callbacks in those new windows. This gives the vendor driver an initialized mdev device to work with during creation. At the same time, a bus driver who wish to bind to mdev driver also gets initialized mdev device. This follows standard Linux kernel bus and device model. 2. During remove flow, first remove the device from the bus. This ensures that any bus specific devices are removed. Once device is taken off the mdev bus, invoke remove() of mdev from the vendor driver. 3. The driver core device model provides way to register and auto unregister the device sysfs attribute groups at dev->groups. Make use of dev->groups to let core create the groups and eliminate code to avoid explicit groups creation and removal. To ensure, that new sequence is solid, a below stack dump of a process is taken who attempts to remove the device while device is in use by vfio driver and user application. This stack dump validates that vfio driver guards against such device removal when device is in use. cat /proc/21962/stack [<0>] vfio_del_group_dev+0x216/0x3c0 [vfio] [<0>] mdev_remove+0x21/0x40 [mdev] [<0>] device_release_driver_internal+0xe8/0x1b0 [<0>] bus_remove_device+0xf9/0x170 [<0>] device_del+0x168/0x350 [<0>] mdev_device_remove_common+0x1d/0x50 [mdev] [<0>] mdev_device_remove+0x8c/0xd0 [mdev] [<0>] remove_store+0x71/0x90 [mdev] [<0>] kernfs_fop_write+0x113/0x1a0 [<0>] vfs_write+0xad/0x1b0 [<0>] ksys_write+0x5a/0xe0 [<0>] do_syscall_64+0x5a/0x210 [<0>] entry_SYSCALL_64_after_hwframe+0x49/0xbe [<0>] 0xffffffffffffffff This prepares the code to eliminate calling device_create_file() in subsequent patch. Reviewed-by: Cornelia Huck <cohuck(a)redhat.com> Signed-off-by: Parav Pandit <parav(a)mellanox.com> Signed-off-by: Alex Williamson <alex.williamson(a)redhat.com> Conflicts: drivers/vfio/mdev/mdev_core.c -mdev_device_remove_ops() and -mdev_device_remove_cb() are removed anyway, although they are not identical compared within the original patch. The type of uuid is still uuid_le, and it has not been changed to guid_t. drivers/vfio/mdev/mdev_private.h The type of uuid is still uuid_le, and it has not been changed to guid_t. Signed-off-by: Xiaoyang Xu <xuxiaoyang2(a)huawei.com> Reviewed-by: Yingtai Xie <xieyingtai(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/vfio/mdev/mdev_core.c | 94 +++++++++++----------------------------- drivers/vfio/mdev/mdev_private.h | 2 +- drivers/vfio/mdev/mdev_sysfs.c | 2 +- 3 files changed, 27 insertions(+), 71 deletions(-) diff --git a/drivers/vfio/mdev/mdev_core.c b/drivers/vfio/mdev/mdev_core.c index e052f62..9705a4a 100644 --- a/drivers/vfio/mdev/mdev_core.c +++ b/drivers/vfio/mdev/mdev_core.c @@ -103,55 +103,10 @@ static inline void mdev_put_parent(struct mdev_parent *parent) kref_put(&parent->ref, mdev_release_parent); } -static int mdev_device_create_ops(struct kobject *kobj, - struct mdev_device *mdev) -{ - struct mdev_parent *parent = mdev->parent; - int ret; - - ret = parent->ops->create(kobj, mdev); - if (ret) - return ret; - - ret = sysfs_create_groups(&mdev->dev.kobj, - parent->ops->mdev_attr_groups); - if (ret) - parent->ops->remove(mdev); - - return ret; -} - -/* - * mdev_device_remove_ops gets called from sysfs's 'remove' and when parent - * device is being unregistered from mdev device framework. - * - 'force_remove' is set to 'false' when called from sysfs's 'remove' which - * indicates that if the mdev device is active, used by VMM or userspace - * application, vendor driver could return error then don't remove the device. - * - 'force_remove' is set to 'true' when called from mdev_unregister_device() - * which indicate that parent device is being removed from mdev device - * framework so remove mdev device forcefully. - */ -static int mdev_device_remove_ops(struct mdev_device *mdev, bool force_remove) -{ - struct mdev_parent *parent = mdev->parent; - int ret; - - /* - * Vendor driver can return error if VMM or userspace application is - * using this mdev device. - */ - ret = parent->ops->remove(mdev); - if (ret && !force_remove) - return -EBUSY; - - sysfs_remove_groups(&mdev->dev.kobj, parent->ops->mdev_attr_groups); - return 0; -} - static int mdev_device_remove_cb(struct device *dev, void *data) { if (dev_is_mdev(dev)) - mdev_device_remove(dev, true); + mdev_device_remove(dev); return 0; } @@ -311,41 +266,43 @@ int mdev_device_create(struct kobject *kobj, struct device *dev, uuid_le uuid) mdev->parent = parent; kref_init(&mdev->ref); + device_initialize(&mdev->dev); mdev->dev.parent = dev; mdev->dev.bus = &mdev_bus_type; mdev->dev.release = mdev_device_release; dev_set_name(&mdev->dev, "%pUl", uuid.b); + mdev->dev.groups = parent->ops->mdev_attr_groups; + mdev->type_kobj = kobj; - ret = device_register(&mdev->dev); - if (ret) { - put_device(&mdev->dev); - goto mdev_fail; - } + ret = parent->ops->create(kobj, mdev); + if (ret) + goto ops_create_fail; - ret = mdev_device_create_ops(kobj, mdev); + ret = device_add(&mdev->dev); if (ret) - goto create_fail; + goto add_fail; ret = mdev_create_sysfs_files(&mdev->dev, type); - if (ret) { - mdev_device_remove_ops(mdev, true); - goto create_fail; - } + if (ret) + goto sysfs_fail; - mdev->type_kobj = kobj; mdev->active = true; dev_dbg(&mdev->dev, "MDEV: created\n"); return 0; -create_fail: - device_unregister(&mdev->dev); +sysfs_fail: + device_del(&mdev->dev); +add_fail: + parent->ops->remove(mdev); +ops_create_fail: + put_device(&mdev->dev); mdev_fail: mdev_put_parent(parent); return ret; } -int mdev_device_remove(struct device *dev, bool force_remove) +int mdev_device_remove(struct device *dev) { struct mdev_device *mdev, *tmp; struct mdev_parent *parent; @@ -374,16 +331,15 @@ int mdev_device_remove(struct device *dev, bool force_remove) mutex_unlock(&mdev_list_lock); type = to_mdev_type(mdev->type_kobj); + mdev_remove_sysfs_files(dev, type); + device_del(&mdev->dev); parent = mdev->parent; + ret = parent->ops->remove(mdev); + if (ret) + dev_err(&mdev->dev, "Remove failed: err=%d\n", ret); - ret = mdev_device_remove_ops(mdev, force_remove); - if (ret) { - mdev->active = true; - return ret; - } - - mdev_remove_sysfs_files(dev, type); - device_unregister(dev); + /* Balances with device_initialize() */ + put_device(&mdev->dev); mdev_put_parent(parent); return 0; diff --git a/drivers/vfio/mdev/mdev_private.h b/drivers/vfio/mdev/mdev_private.h index b5819b7..9c65d8e 100644 --- a/drivers/vfio/mdev/mdev_private.h +++ b/drivers/vfio/mdev/mdev_private.h @@ -59,6 +59,6 @@ struct mdev_type { void mdev_remove_sysfs_files(struct device *dev, struct mdev_type *type); int mdev_device_create(struct kobject *kobj, struct device *dev, uuid_le uuid); -int mdev_device_remove(struct device *dev, bool force_remove); +int mdev_device_remove(struct device *dev); #endif /* MDEV_PRIVATE_H */ diff --git a/drivers/vfio/mdev/mdev_sysfs.c b/drivers/vfio/mdev/mdev_sysfs.c index e7770b5..1286366 100644 --- a/drivers/vfio/mdev/mdev_sysfs.c +++ b/drivers/vfio/mdev/mdev_sysfs.c @@ -236,7 +236,7 @@ static ssize_t remove_store(struct device *dev, struct device_attribute *attr, if (val && device_remove_file_self(dev, attr)) { int ret; - ret = mdev_device_remove(dev, false); + ret = mdev_device_remove(dev); if (ret) { device_create_file(dev, attr); return ret; -- 1.8.3

1 3

[PATCH 1/2] arm64/mpam: Fix unreset resources when mkdir ctrl group or umount resctrl
by Yang Yingliang 01 Jun '20

01 Jun '20

From: Wang ShaoBo <bobo.shaobowang(a)huawei.com> hulk inclusion category: bugfix bugzilla: 34588 CVE: NA ----------------------------------------------- There are two problems related to schemata: 1) When rmdir a group and then mkdir a new group under resctrl root directory, the new group still inherits the schemata configuration from old. e.g. > mount -t resctrl resctrl /sys/fs/resctrl > cd /sys/fs/resctrl > mkdir p1 && cd p1 > echo 'L3:0=7f' > schemata > cd .. && rmdir p1 && mkdir p1 && cd p1 > cat schemata L3:0=7f;1=7fff;2=7fff;3=7fff MB:0=100;1=100;2=100;3=100 2) It still exists when umount /sys/fs/resctrl and remount. e.g. > mount -t resctrl resctrl /sys/fs/resctrl > cd /sys/fs/resctrl > echo 'L3:0=7f' > schemata > umount /sys/fs/resctrl > mount -t resctrl resctrl /sys/fs/resctrl > cat schemata L3:0=7f;1=7fff;2=7fff;3=7fff MB:0=100;1=100;2=100;3=100 Firstly we make each resctrl resource obtains their corresponding default configuration. NOTE we use zero to initialize L3 default value instead of max cpbm bits, as zero configurarion equals to maximum configuration for L3 MSCs. And we use max-percentage masks of max bandwidth to generate maximum configuration for MB. Then we reset resources' configuration settings to default value and back MSCs to default state, when mkdir or umount happended. Fixes: caf75b6b2540 ("resctrlfs: mpam: init struct for mpam") Fixes: 916dd9321e3c ("resctrlfs: init support resctrlfs") Signed-off-by: Wang ShaoBo <bobo.shaobowang(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/include/asm/mpam_resource.h | 4 ++++ arch/arm64/include/asm/resctrl.h | 2 ++ arch/arm64/kernel/mpam.c | 17 +++++++++++++++++ arch/arm64/kernel/mpam_ctrlmon.c | 28 ++++++++++++++++++++++++++++ fs/resctrlfs.c | 5 +++++ 5 files changed, 56 insertions(+) diff --git a/arch/arm64/include/asm/mpam_resource.h b/arch/arm64/include/asm/mpam_resource.h index ab90596..1a6904c 100644 --- a/arch/arm64/include/asm/mpam_resource.h +++ b/arch/arm64/include/asm/mpam_resource.h @@ -89,6 +89,10 @@ /* [FIXME] hard code for hardlim */ #define MBW_MAX_SET(v) (MBW_MAX_HARDLIM|((v) << (16 - BWA_WD))) #define MBW_MAX_GET(v) (((v) & MBW_MAX_MASK) >> (16 - BWA_WD)) + +/* hard code for mbw_max max-percentage's cresponding masks */ +#define MBA_MAX_WD 63u + /* * emulate the mpam nodes * These should be reported by ACPI MPAM Table. diff --git a/arch/arm64/include/asm/resctrl.h b/arch/arm64/include/asm/resctrl.h index 0a0a12b..fb5fa6c 100644 --- a/arch/arm64/include/asm/resctrl.h +++ b/arch/arm64/include/asm/resctrl.h @@ -65,4 +65,6 @@ int mkdir_mondata_all(struct kernfs_node *parent_kn, mongroup_create_dir(struct kernfs_node *parent_kn, struct resctrl_group *prgrp, char *name, struct kernfs_node **dest_kn); +int rdtgroup_init_alloc(struct rdtgroup *rdtgrp); + #endif /* _ASM_ARM64_RESCTRL_H */ diff --git a/arch/arm64/kernel/mpam.c b/arch/arm64/kernel/mpam.c index 202e49a..120795e 100644 --- a/arch/arm64/kernel/mpam.c +++ b/arch/arm64/kernel/mpam.c @@ -86,6 +86,9 @@ static inline void mpam_node_assign_val(struct mpam_node *n, n->addr = hwpage_address; n->component_id = component_id; n->cpus_list = "0"; + + if (n->type == MPAM_RESOURCE_MC) + n->default_ctrl = MBA_MAX_WD; } #define MPAM_NODE_NAME_SIZE (10) @@ -544,6 +547,20 @@ void post_resctrl_mount(void) static int reset_all_ctrls(struct resctrl_resource *r) { + struct raw_resctrl_resource *rr; + struct rdt_domain *d; + int partid; + + rr = (struct raw_resctrl_resource *)r->res; + for (partid = 0; partid < rr->num_partid; partid++) { + list_for_each_entry(d, &r->domains, list) { + d->new_ctrl = rr->default_ctrl; + d->ctrl_val[partid] = rr->default_ctrl; + d->have_new_ctrl = true; + rr->msr_update(d, partid); + } + } + return 0; } diff --git a/arch/arm64/kernel/mpam_ctrlmon.c b/arch/arm64/kernel/mpam_ctrlmon.c index b9f9495..2b94efc 100644 --- a/arch/arm64/kernel/mpam_ctrlmon.c +++ b/arch/arm64/kernel/mpam_ctrlmon.c @@ -585,3 +585,31 @@ int resctrl_mkdir_ctrlmon_mondata(struct kernfs_node *parent_kn, } return ret; } + +/* Initialize the RDT group's allocations. */ +int rdtgroup_init_alloc(struct rdtgroup *rdtgrp) +{ + struct resctrl_resource *r; + struct raw_resctrl_resource *rr; + struct rdt_domain *d; + int ret; + + for_each_resctrl_resource(r) { + if (!r->alloc_enabled) + continue; + + rr = (struct raw_resctrl_resource *)r->res; + list_for_each_entry(d, &r->domains, list) { + d->new_ctrl = rr->default_ctrl; + d->have_new_ctrl = true; + } + + ret = update_domains(r, rdtgrp); + if (ret < 0) { + rdt_last_cmd_puts("Failed to initialize allocations\n"); + return ret; + } + } + + return 0; +} diff --git a/fs/resctrlfs.c b/fs/resctrlfs.c index 09e56384..8d463b8 100644 --- a/fs/resctrlfs.c +++ b/fs/resctrlfs.c @@ -670,6 +670,11 @@ static int resctrl_group_mkdir_ctrl_mon(struct kernfs_node *parent_kn, ret = 0; rdtgrp->closid = closid; + + ret = rdtgroup_init_alloc(rdtgrp); + if (ret < 0) + goto out_id_free; + list_add(&rdtgrp->resctrl_group_list, &resctrl_all_groups); if (resctrl_mon_capable) { -- 1.8.3

1 1

[PATCH] ext4: report error to userspace by netlink
by Yang Yingliang 01 Jun '20

01 Jun '20

From: Zhao Minmin <zhaominmin1(a)huawei.com> hulk inclusion category: feature bugzilla: 34592 CVE: NA ------------------------------------------------- Implement the ext3/ext4 file system error report. This patch is used to implement abnormal alarm of ext3/ext4 filesystem. You can archieve this by setting "FILESYSTEM_MONITOR" or "FILESYSTEM_ALARM" on in configuration file. With this setting, alarm will be raised when ext3/ext4 file system expection occurs. Signed-off-by: Zhao Minmin <zhaominmin1(a)huawei.com> Reviewed-by: Yi Zhang <yi.zhang(a)huawei.com> Link: http://hulk.huawei.com/pipermail/kernel.openeuler/2016-March/009711.html Signed-off-by: Wang Hui <john.wanghui(a)huawei.com> Signed-off-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> [yebin: cherry-pick this patch from redhat-7.5, commit 13848b1856c1] Signed-off-by: Ye Bin <yebin10(a)huawei.com> Reviewed-by: Wenan Mao <maowenan(a)huawei.com> Reviewed-by: Yi Zhang <yi.zhang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/ext4/ext4.h | 9 ++++++++ fs/ext4/super.c | 51 ++++++++++++++++++++++++++++++++++++++++++++ include/uapi/linux/netlink.h | 1 + 3 files changed, 61 insertions(+) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index b1bcc6e..83bec07 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -54,6 +54,15 @@ #endif #endif +#define NL_EXT4_ERROR_GROUP 1 +#define EXT4_ERROR_MAGIC 0xAE32014U +struct ext4_err_msg { + int magic; + char s_id[32]; + unsigned long s_flags; + int ext4_errno; +}; + /* * The fourth extended filesystem constants/structures */ diff --git a/fs/ext4/super.c b/fs/ext4/super.c index bd6c1cf..33f1e7c 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -54,6 +54,10 @@ #include "mballoc.h" #include "fsmap.h" +#include <uapi/linux/netlink.h> +#include <net/sock.h> +#include <net/net_namespace.h> + #define CREATE_TRACE_POINTS #include <trace/events/ext4.h> @@ -84,6 +88,8 @@ static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags, static void ext4_clear_request_list(void); static struct inode *ext4_get_journal_inode(struct super_block *sb, unsigned int journal_inum); +static void ext4_netlink_send_info(struct super_block *sb, int ext4_errno); +static struct sock *ext4nl; /* * Lock ordering @@ -439,6 +445,42 @@ static bool system_going_down(void) || system_state == SYSTEM_RESTART; } +static void ext4_netlink_send_info(struct super_block *sb, int ext4_errno) +{ + int size; + sk_buff_data_t old_tail; + struct sk_buff *skb; + struct nlmsghdr *nlh; + struct ext4_err_msg *msg; + + if (ext4nl) { + size = NLMSG_SPACE(sizeof(struct ext4_err_msg)); + skb = alloc_skb(size, GFP_ATOMIC); + if (!skb) { + printk(KERN_ERR "Cannot alloc skb!"); + return; + } + old_tail = skb->tail; + nlh = nlmsg_put(skb, 0, 0, NLMSG_ERROR, size - sizeof(*nlh), 0); + if (!nlh) + goto nlmsg_failure; + msg = (struct ext4_err_msg *)NLMSG_DATA(nlh); + msg->magic = EXT4_ERROR_MAGIC; + memcpy(msg->s_id, sb->s_id, sizeof(sb->s_id)); + msg->s_flags = sb->s_flags; + msg->ext4_errno = ext4_errno; + nlh->nlmsg_len = skb->tail - old_tail; + NETLINK_CB(skb).portid = 0; + NETLINK_CB(skb).dst_group = NL_EXT4_ERROR_GROUP; + netlink_broadcast(ext4nl, skb, 0, NL_EXT4_ERROR_GROUP, + GFP_ATOMIC); + return; +nlmsg_failure: + if (skb) + kfree_skb(skb); + } +} + /* Deal with the reporting of failure conditions on a filesystem such as * inconsistencies detected or read IO failures. * @@ -469,6 +511,9 @@ static void ext4_handle_error(struct super_block *sb) if (journal) jbd2_journal_abort(journal, -EIO); } + + ext4_netlink_send_info(sb, 1); + /* * We force ERRORS_RO behavior when system is rebooting. Otherwise we * could panic during 'reboot -f' as the underlying device got already @@ -700,6 +745,7 @@ void __ext4_abort(struct super_block *sb, const char *function, if (EXT4_SB(sb)->s_journal) jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO); save_error_info(sb, function, line); + ext4_netlink_send_info(sb, 2); } if (test_opt(sb, ERRORS_PANIC) && !system_going_down()) { if (EXT4_SB(sb)->s_journal && @@ -6063,6 +6109,7 @@ static inline int ext3_feature_set_ok(struct super_block *sb) static int __init ext4_init_fs(void) { int i, err; + struct netlink_kernel_cfg cfg = {.groups = NL_EXT4_ERROR_GROUP,}; ratelimit_state_init(&ext4_mount_msg_ratelimit, 30 * HZ, 64); ext4_li_info = NULL; @@ -6102,6 +6149,9 @@ static int __init ext4_init_fs(void) if (err) goto out; + ext4nl = netlink_kernel_create(&init_net, NETLINK_FILESYSTEM, &cfg); + if (!ext4nl) + printk(KERN_ERR "EXT4-fs: Cannot create netlink socket.\n"); return 0; out: unregister_as_ext2(); @@ -6133,6 +6183,7 @@ static void __exit ext4_exit_fs(void) ext4_exit_system_zone(); ext4_exit_pageio(); ext4_exit_es(); + netlink_kernel_release(ext4nl); } MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); diff --git a/include/uapi/linux/netlink.h b/include/uapi/linux/netlink.h index 776bc92..ecaefee 100644 --- a/include/uapi/linux/netlink.h +++ b/include/uapi/linux/netlink.h @@ -29,6 +29,7 @@ #define NETLINK_RDMA 20 #define NETLINK_CRYPTO 21 /* Crypto layer */ #define NETLINK_SMC 22 /* SMC monitoring */ +#define NETLINK_FILESYSTEM 28 /* filesystem alarm*/ #define NETLINK_INET_DIAG NETLINK_SOCK_DIAG -- 1.8.3

1 0

[PATCH 1/3] Added ethtool_ops interface to query optical module information
by Yang Yingliang 01 Jun '20

01 Jun '20

From: Chiqijun <chiqijun(a)huawei.com> driver inclusion category: feature bugzilla: 4472 ----------------------------------------------------------------------- Obtaining optical module information is time-consuming. If the user frequently acquires, it will affect the firmware to process other commands. To prevent such events, the firmware will actively report the optical module information to the driver when it is idle. Obtained from the driver cache information. When the user obtains through the command, directly obtain from the information saved by the driver. Signed-off-by: Chiqijun <chiqijun(a)huawei.com> Reviewed-by: Luoshaokai <luoshaokai(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/net/ethernet/huawei/hinic/hinic_ethtool.c | 77 ++++++++++ drivers/net/ethernet/huawei/hinic/hinic_hw.h | 54 +++++++ drivers/net/ethernet/huawei/hinic/hinic_hwdev.c | 168 +++++++++++++++++++++ drivers/net/ethernet/huawei/hinic/hinic_lld.c | 2 + .../ethernet/huawei/hinic/hinic_mgmt_interface.h | 12 ++ drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c | 139 ++++++++++++++++- drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.h | 2 + drivers/net/ethernet/huawei/hinic/hinic_nic_io.c | 11 +- drivers/net/ethernet/huawei/hinic/hinic_port_cmd.h | 3 + 9 files changed, 465 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c b/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c index f321365..78d32ea 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c @@ -2039,6 +2039,74 @@ static void hinic_diag_test(struct net_device *netdev, hinic_lp_test(netdev, eth_test, data, 0); } +#ifdef ETHTOOL_GMODULEEEPROM +static int hinic_get_module_info(struct net_device *netdev, + struct ethtool_modinfo *modinfo) +{ + struct hinic_nic_dev *nic_dev = netdev_priv(netdev); + u8 sfp_type; + u8 sfp_type_ext; + int err; + + err = hinic_get_sfp_type(nic_dev->hwdev, &sfp_type, &sfp_type_ext); + if (err) + return err; + + switch (sfp_type) { + case MODULE_TYPE_SFP: + modinfo->type = ETH_MODULE_SFF_8472; + modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN; + break; + case MODULE_TYPE_QSFP: + modinfo->type = ETH_MODULE_SFF_8436; + modinfo->eeprom_len = STD_SFP_INFO_MAX_SIZE; + break; + case MODULE_TYPE_QSFP_PLUS: + if (sfp_type_ext >= 0x3) { + modinfo->type = ETH_MODULE_SFF_8636; + modinfo->eeprom_len = STD_SFP_INFO_MAX_SIZE; + + } else { + modinfo->type = ETH_MODULE_SFF_8436; + modinfo->eeprom_len = STD_SFP_INFO_MAX_SIZE; + } + break; + case MODULE_TYPE_QSFP28: + modinfo->type = ETH_MODULE_SFF_8636; + modinfo->eeprom_len = STD_SFP_INFO_MAX_SIZE; + break; + default: + nicif_warn(nic_dev, drv, netdev, + "Optical module unknown: 0x%x\n", sfp_type); + return -EINVAL; + } + + return 0; +} + +static int hinic_get_module_eeprom(struct net_device *netdev, + struct ethtool_eeprom *ee, u8 *data) +{ + struct hinic_nic_dev *nic_dev = netdev_priv(netdev); + u8 sfp_data[STD_SFP_INFO_MAX_SIZE]; + u16 len; + int err; + + if (!ee->len || ((ee->len + ee->offset) > STD_SFP_INFO_MAX_SIZE)) + return -EINVAL; + + memset(data, 0, ee->len); + + err = hinic_get_sfp_eeprom(nic_dev->hwdev, sfp_data, &len); + if (err) + return err; + + memcpy(data, sfp_data + ee->offset, ee->len); + + return 0; +} +#endif /* ETHTOOL_GMODULEEEPROM */ + static int set_l4_rss_hash_ops(struct ethtool_rxnfc *cmd, struct nic_rss_type *rss_type) { @@ -2532,6 +2600,10 @@ static int hinic_set_rxfh_indir(struct net_device *netdev, const u32 *indir) #ifndef HAVE_RHEL6_ETHTOOL_OPS_EXT_STRUCT .get_channels = hinic_get_channels, .set_channels = hinic_set_channels, +#ifdef ETHTOOL_GMODULEEEPROM + .get_module_info = hinic_get_module_info, + .get_module_eeprom = hinic_get_module_eeprom, +#endif #ifndef NOT_HAVE_GET_RXFH_INDIR_SIZE .get_rxfh_indir_size = hinic_get_rxfh_indir_size, #endif @@ -2552,6 +2624,11 @@ static int hinic_set_rxfh_indir(struct net_device *netdev, const u32 *indir) .set_phys_id = hinic_set_phys_id, .get_channels = hinic_get_channels, .set_channels = hinic_set_channels, +#ifdef ETHTOOL_GMODULEEEPROM + .get_module_info = hinic_get_module_info, + .get_module_eeprom = hinic_get_module_eeprom, +#endif + #ifndef NOT_HAVE_GET_RXFH_INDIR_SIZE .get_rxfh_indir_size = hinic_get_rxfh_indir_size, #endif diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw.h b/drivers/net/ethernet/huawei/hinic/hinic_hw.h index e489a00..d4becb6 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw.h @@ -264,6 +264,56 @@ struct hinic_init_para { #define INIT_SUCCESS 1 #define MAX_DRV_BUF_SIZE 4096 +struct hinic_cmd_get_light_module_abs { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 port_id; + u8 abs_status; /* 0:present, 1:absent */ + u8 rsv[2]; +}; + +#define MODULE_TYPE_SFP 0x3 +#define MODULE_TYPE_QSFP28 0x11 +#define MODULE_TYPE_QSFP 0x0C +#define MODULE_TYPE_QSFP_PLUS 0x0D + +#define SFP_INFO_MAX_SIZE 512 +struct hinic_cmd_get_sfp_qsfp_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 port_id; + u8 wire_type; + u16 out_len; + u8 sfp_qsfp_info[SFP_INFO_MAX_SIZE]; +}; + +#define STD_SFP_INFO_MAX_SIZE 640 +struct hinic_cmd_get_std_sfp_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 port_id; + u8 wire_type; + u16 eeprom_len; + u32 rsvd; + u8 sfp_info[STD_SFP_INFO_MAX_SIZE]; +}; + +#define HINIC_MAX_PORT_ID 4 + +struct hinic_port_routine_cmd { + int up_send_sfp_info; + int up_send_sfp_abs; + + struct hinic_cmd_get_sfp_qsfp_info sfp_info; + struct hinic_cmd_get_light_module_abs abs; +}; + struct card_node { struct list_head node; struct list_head func_list; @@ -282,6 +332,10 @@ struct card_node { bool disable_vf_load[HINIC_MAX_PF_NUM]; u32 vf_mbx_old_rand_id[MAX_FUNCTION_NUM]; u32 vf_mbx_rand_id[MAX_FUNCTION_NUM]; + struct hinic_port_routine_cmd rt_cmd[HINIC_MAX_PORT_ID]; + + /* mutex used for copy sfp info */ + struct mutex sfp_mutex; }; enum hinic_hwdev_init_state { diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hwdev.c b/drivers/net/ethernet/huawei/hinic/hinic_hwdev.c index 101b671..172be6c 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hwdev.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hwdev.c @@ -59,6 +59,8 @@ #define HINIC_OK_FLAG_FAILED 1 +#define HINIC_GET_SFP_INFO_REAL_TIME 0x1 + #define HINIC_GLB_SO_RO_CFG_SHIFT 0x0 #define HINIC_GLB_SO_RO_CFG_MASK 0x1 #define HINIC_DISABLE_ORDER 0 @@ -913,6 +915,80 @@ int hinic_pf_msg_to_mgmt_sync(void *hwdev, enum hinic_mod_type mod, u8 cmd, return err; } +static bool is_sfp_info_cmd_cached(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hinic_cmd_get_sfp_qsfp_info *sfp_info = NULL; + struct hinic_port_routine_cmd *rt_cmd = NULL; + struct card_node *chip_node = hwdev->chip_node; + + sfp_info = buf_in; + if (sfp_info->port_id >= HINIC_MAX_PORT_ID || + *out_size < sizeof(*sfp_info)) + return false; + + if (sfp_info->version == HINIC_GET_SFP_INFO_REAL_TIME) + return false; + + rt_cmd = &chip_node->rt_cmd[sfp_info->port_id]; + mutex_lock(&chip_node->sfp_mutex); + memcpy(buf_out, &rt_cmd->sfp_info, sizeof(*sfp_info)); + mutex_unlock(&chip_node->sfp_mutex); + + return true; +} + +static bool is_sfp_abs_cmd_cached(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hinic_cmd_get_light_module_abs *abs = NULL; + struct hinic_port_routine_cmd *rt_cmd = NULL; + struct card_node *chip_node = hwdev->chip_node; + + abs = buf_in; + if (abs->port_id >= HINIC_MAX_PORT_ID || + *out_size < sizeof(*abs)) + return false; + + if (abs->version == HINIC_GET_SFP_INFO_REAL_TIME) + return false; + + rt_cmd = &chip_node->rt_cmd[abs->port_id]; + mutex_lock(&chip_node->sfp_mutex); + memcpy(buf_out, &rt_cmd->abs, sizeof(*abs)); + mutex_unlock(&chip_node->sfp_mutex); + + return true; +} + +static bool driver_processed_cmd(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct card_node *chip_node = hwdev->chip_node; + + if (mod == HINIC_MOD_L2NIC) { + if (cmd == HINIC_PORT_CMD_GET_SFP_INFO && + chip_node->rt_cmd->up_send_sfp_info) { + return is_sfp_info_cmd_cached(hwdev, mod, cmd, buf_in, + in_size, buf_out, + out_size); + } else if (cmd == HINIC_PORT_CMD_GET_SFP_ABS && + chip_node->rt_cmd->up_send_sfp_abs) { + return is_sfp_abs_cmd_cached(hwdev, mod, cmd, buf_in, + in_size, buf_out, + out_size); + } + } + + return false; +} + int hinic_msg_to_mgmt_sync(void *hwdev, enum hinic_mod_type mod, u8 cmd, void *buf_in, u16 in_size, void *buf_out, u16 *out_size, u32 timeout) @@ -949,6 +1025,10 @@ int hinic_msg_to_mgmt_sync(void *hwdev, enum hinic_mod_type mod, u8 cmd, err = __func_send_mbox(hwdev, mod, cmd, buf_in, in_size, buf_out, out_size, timeout); } else { + if (driver_processed_cmd(hwdev, mod, cmd, buf_in, in_size, + buf_out, out_size)) + return 0; + do { if (!hinic_get_mgmt_channel_status(hwdev) || !hinic_get_chip_present_flag(hwdev)) @@ -3202,6 +3282,8 @@ enum hinic_event_cmd { HINIC_EVENT_MGMT_PCIE_DFX, HINIC_EVENT_MCTP_HOST_INFO, HINIC_EVENT_MGMT_HEARTBEAT_EHD, + HINIC_EVENT_SFP_INFO_REPORT, + HINIC_EVENT_SFP_ABS_REPORT, HINIC_EVENT_MAX_TYPE, }; @@ -3284,6 +3366,16 @@ struct hinic_event_convert { .cmd = HINIC_MGMT_CMD_HEARTBEAT_EVENT, .event = HINIC_EVENT_MGMT_HEARTBEAT_EHD, }, + { + .mod = HINIC_MOD_L2NIC, + .cmd = HINIC_PORT_CMD_GET_SFP_INFO, + .event = HINIC_EVENT_SFP_INFO_REPORT, + }, + { + .mod = HINIC_MOD_L2NIC, + .cmd = HINIC_PORT_CMD_GET_SFP_ABS, + .event = HINIC_EVENT_SFP_ABS_REPORT, + }, }; static enum hinic_event_cmd __get_event_type(u8 mod, u8 cmd) @@ -3587,12 +3679,22 @@ static void module_status_event(struct hinic_hwdev *hwdev, struct hinic_cable_plug_event *plug_event; struct hinic_link_err_event *link_err; struct hinic_event_info event_info = {0}; + struct hinic_port_routine_cmd *rt_cmd; + struct card_node *chip_node = hwdev->chip_node; event_info.type = HINIC_EVENT_PORT_MODULE_EVENT; if (cmd == HINIC_EVENT_CABLE_PLUG) { plug_event = buf_in; + if (plug_event->port_id < HINIC_MAX_PORT_ID) { + rt_cmd = &chip_node->rt_cmd[plug_event->port_id]; + mutex_lock(&chip_node->sfp_mutex); + rt_cmd->up_send_sfp_abs = false; + rt_cmd->up_send_sfp_info = false; + mutex_unlock(&chip_node->sfp_mutex); + } + event_info.module_event.type = plug_event->plugged ? HINIC_PORT_MODULE_CABLE_PLUGGED : HINIC_PORT_MODULE_CABLE_UNPLUGGED; @@ -3733,6 +3835,64 @@ static void mgmt_watchdog_timeout_event_handler(struct hinic_hwdev *hwdev, queue_work(hwdev->workq, &hwdev->fault_work); } +static void port_sfp_info_event(struct hinic_hwdev *hwdev, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size) +{ + struct hinic_cmd_get_sfp_qsfp_info *sfp_info = buf_in; + struct hinic_port_routine_cmd *rt_cmd; + struct card_node *chip_node = hwdev->chip_node; + + if (in_size != sizeof(*sfp_info)) { + sdk_err(hwdev->dev_hdl, "Invalid sfp info cmd, length: %d, should be %ld\n", + in_size, sizeof(*sfp_info)); + return; + } + + if (sfp_info->port_id >= HINIC_MAX_PORT_ID) { + sdk_err(hwdev->dev_hdl, "Invalid sfp port id: %d, max port is %d\n", + sfp_info->port_id, HINIC_MAX_PORT_ID - 1); + return; + } + + if (!chip_node->rt_cmd) + return; + + rt_cmd = &chip_node->rt_cmd[sfp_info->port_id]; + mutex_lock(&chip_node->sfp_mutex); + memcpy(&rt_cmd->sfp_info, sfp_info, sizeof(rt_cmd->sfp_info)); + rt_cmd->up_send_sfp_info = true; + mutex_unlock(&chip_node->sfp_mutex); +} + +static void port_sfp_abs_event(struct hinic_hwdev *hwdev, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size) +{ + struct hinic_cmd_get_light_module_abs *sfp_abs = buf_in; + struct hinic_port_routine_cmd *rt_cmd; + struct card_node *chip_node = hwdev->chip_node; + + if (in_size != sizeof(*sfp_abs)) { + sdk_err(hwdev->dev_hdl, "Invalid sfp absent cmd, length: %d, should be %ld\n", + in_size, sizeof(*sfp_abs)); + return; + } + + if (sfp_abs->port_id >= HINIC_MAX_PORT_ID) { + sdk_err(hwdev->dev_hdl, "Invalid sfp port id: %d, max port is %d\n", + sfp_abs->port_id, HINIC_MAX_PORT_ID - 1); + return; + } + + if (!chip_node->rt_cmd) + return; + + rt_cmd = &chip_node->rt_cmd[sfp_abs->port_id]; + mutex_lock(&chip_node->sfp_mutex); + memcpy(&rt_cmd->abs, sfp_abs, sizeof(rt_cmd->abs)); + rt_cmd->up_send_sfp_abs = true; + mutex_unlock(&chip_node->sfp_mutex); +} + static void mgmt_reset_event_handler(struct hinic_hwdev *hwdev) { sdk_info(hwdev->dev_hdl, "Mgmt is reset\n"); @@ -4241,6 +4401,14 @@ static void _event_handler(struct hinic_hwdev *hwdev, enum hinic_event_cmd cmd, buf_out, out_size); break; + case HINIC_EVENT_SFP_INFO_REPORT: + port_sfp_info_event(hwdev, buf_in, in_size, buf_out, out_size); + break; + + case HINIC_EVENT_SFP_ABS_REPORT: + port_sfp_abs_event(hwdev, buf_in, in_size, buf_out, out_size); + break; + default: sdk_warn(hwdev->dev_hdl, "Unsupported event %d to process\n", cmd); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_lld.c b/drivers/net/ethernet/huawei/hinic/hinic_lld.c index e8e446a..3bf47ac 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_lld.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_lld.c @@ -1990,6 +1990,8 @@ static int alloc_chip_node(struct hinic_pcidev *pci_adapter) INIT_LIST_HEAD(&chip_node->func_list); pci_adapter->chip_node = chip_node; + mutex_init(&chip_node->sfp_mutex); + return 0; alloc_dbgtool_attr_file_err: diff --git a/drivers/net/ethernet/huawei/hinic/hinic_mgmt_interface.h b/drivers/net/ethernet/huawei/hinic/hinic_mgmt_interface.h index 13bc351..3c3be6e 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_mgmt_interface.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_mgmt_interface.h @@ -680,6 +680,16 @@ struct hinic_capture_info { u32 data_vlan; }; +struct hinic_port_rt_cmd { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 pf_id; + u8 enable; + u8 rsvd1[6]; +}; + struct hinic_vf_dcb_state { u8 status; u8 version; @@ -737,6 +747,8 @@ struct hinic_set_link_follow { void hinic_unregister_vf_msg_handler(void *hwdev, u16 vf_id); +int hinic_set_port_routine_cmd_report(void *hwdev, bool enable); + int hinic_refresh_nic_cfg(void *hwdev, struct nic_port_info *port_info); int hinic_save_dcb_state(struct hinic_hwdev *hwdev, diff --git a/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c b/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c index 07858be..6ddae36 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c @@ -35,7 +35,6 @@ #include "hinic_nic.h" #include "hinic_mgmt_interface.h" #include "hinic_hwif.h" -#include "hinic_eqs.h" static unsigned char set_vf_link_state; module_param(set_vf_link_state, byte, 0444); @@ -3571,6 +3570,35 @@ int hinic_set_super_cqe_state(void *hwdev, bool enable) return 0; } +int hinic_set_port_routine_cmd_report(void *hwdev, bool enable) +{ + struct hinic_port_rt_cmd rt_cmd = { 0 }; + struct hinic_hwdev *dev = hwdev; + u16 out_size = sizeof(rt_cmd); + int err; + + if (!hwdev) + return -EINVAL; + + rt_cmd.pf_id = (u8)hinic_global_func_id(hwdev); + rt_cmd.enable = enable; + + err = l2nic_msg_to_mgmt_sync(hwdev, + HINIC_PORT_CMD_SET_PORT_REPORT, + &rt_cmd, sizeof(rt_cmd), &rt_cmd, + &out_size); + if (rt_cmd.status == HINIC_MGMT_CMD_UNSUPPORTED) { + nic_info(dev->dev_hdl, "Current firmware doesn't support to set port routine command report\n"); + } else if (rt_cmd.status || err || !out_size) { + nic_err(dev->dev_hdl, + "Failed to set port routine command report, err: %d, status: 0x%x, out size: 0x%x\n", + err, rt_cmd.status, out_size); + return -EFAULT; + } + + return 0; +} + int hinic_set_func_capture_en(void *hwdev, u16 func_id, bool cap_en) { struct hinic_hwdev *dev = hwdev; @@ -3889,3 +3917,112 @@ int hinic_disable_tx_promisc(void *hwdev) } return 0; } + +static bool hinic_if_sfp_absent(void *hwdev) +{ + struct card_node *chip_node = ((struct hinic_hwdev *)hwdev)->chip_node; + struct hinic_port_routine_cmd *rt_cmd; + struct hinic_cmd_get_light_module_abs sfp_abs = {0}; + u8 port_id = hinic_physical_port_id(hwdev); + u16 out_size = sizeof(sfp_abs); + int err; + bool sfp_abs_valid; + bool sfp_abs_status; + + rt_cmd = &chip_node->rt_cmd[port_id]; + mutex_lock(&chip_node->sfp_mutex); + sfp_abs_valid = rt_cmd->up_send_sfp_abs; + sfp_abs_status = (bool)rt_cmd->abs.abs_status; + if (sfp_abs_valid) { + mutex_unlock(&chip_node->sfp_mutex); + return sfp_abs_status; + } + mutex_unlock(&chip_node->sfp_mutex); + + sfp_abs.port_id = port_id; + err = l2nic_msg_to_mgmt_sync(hwdev, HINIC_PORT_CMD_GET_SFP_ABS, + &sfp_abs, sizeof(sfp_abs), &sfp_abs, + &out_size); + if (sfp_abs.status || err || !out_size) { + nic_err(((struct hinic_hwdev *)hwdev)->dev_hdl, + "Failed to get port%d sfp absent status, err: %d, status: 0x%x, out size: 0x%x\n", + port_id, err, sfp_abs.status, out_size); + return true; + } + + return ((sfp_abs.abs_status == 0) ? false : true); +} + +int hinic_get_sfp_eeprom(void *hwdev, u8 *data, u16 *len) +{ + struct hinic_cmd_get_std_sfp_info sfp_info = {0}; + u8 port_id; + u16 out_size = sizeof(sfp_info); + int err; + + if (!hwdev || !data || !len) + return -EINVAL; + + port_id = hinic_physical_port_id(hwdev); + if (port_id >= HINIC_MAX_PORT_ID) + return -EINVAL; + + if (hinic_if_sfp_absent(hwdev)) + return -ENXIO; + + sfp_info.port_id = port_id; + err = l2nic_msg_to_mgmt_sync(hwdev, HINIC_PORT_CMD_GET_STD_SFP_INFO, + &sfp_info, sizeof(sfp_info), &sfp_info, + &out_size); + if (sfp_info.status || err || !out_size) { + nic_err(((struct hinic_hwdev *)hwdev)->dev_hdl, + "Failed to get port%d sfp eeprom information, err: %d, status: 0x%x, out size: 0x%x\n", + port_id, err, sfp_info.status, out_size); + return -EIO; + } + + *len = min_t(u16, sfp_info.eeprom_len, STD_SFP_INFO_MAX_SIZE); + memcpy(data, sfp_info.sfp_info, STD_SFP_INFO_MAX_SIZE); + + return 0; +} + +int hinic_get_sfp_type(void *hwdev, u8 *data0, u8 *data1) +{ + struct card_node *chip_node = NULL; + struct hinic_port_routine_cmd *rt_cmd; + u8 sfp_data[STD_SFP_INFO_MAX_SIZE]; + u16 len; + u8 port_id; + int err; + + if (!hwdev || !data0 || !data1) + return -EINVAL; + + port_id = hinic_physical_port_id(hwdev); + if (port_id >= HINIC_MAX_PORT_ID) + return -EINVAL; + + if (hinic_if_sfp_absent(hwdev)) + return -ENXIO; + + chip_node = ((struct hinic_hwdev *)hwdev)->chip_node; + rt_cmd = &chip_node->rt_cmd[port_id]; + mutex_lock(&chip_node->sfp_mutex); + if (rt_cmd->up_send_sfp_info) { + *data0 = rt_cmd->sfp_info.sfp_qsfp_info[0]; + *data1 = rt_cmd->sfp_info.sfp_qsfp_info[1]; + mutex_unlock(&chip_node->sfp_mutex); + return 0; + } + mutex_unlock(&chip_node->sfp_mutex); + + err = hinic_get_sfp_eeprom(hwdev, sfp_data, &len); + if (err) + return err; + + *data0 = sfp_data[0]; + *data1 = sfp_data[1]; + + return 0; +} diff --git a/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.h b/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.h index 395f6d8..eb060fb 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.h @@ -620,5 +620,7 @@ int hinic_add_hw_rqfilter(void *hwdev, struct hinic_rq_filter_info *filter_info); int hinic_del_hw_rqfilter(void *hwdev, struct hinic_rq_filter_info *filter_info); +int hinic_get_sfp_eeprom(void *hwdev, u8 *data, u16 *len); +int hinic_get_sfp_type(void *hwdev, u8 *data0, u8 *data1); #endif diff --git a/drivers/net/ethernet/huawei/hinic/hinic_nic_io.c b/drivers/net/ethernet/huawei/hinic/hinic_nic_io.c index 8d13105..1cb4e26 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_nic_io.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_nic_io.c @@ -829,13 +829,20 @@ int hinic_init_nic_hwdev(void *hwdev, u16 rx_buff_len) } } } - return 0; + + /* VFs don't set port routine command report */ + if (hinic_func_type(dev) != TYPE_VF) + /* Inform mgmt to send sfp's information to driver */ + err = hinic_set_port_routine_cmd_report(hwdev, true); + + return err; } EXPORT_SYMBOL(hinic_init_nic_hwdev); void hinic_free_nic_hwdev(void *hwdev) { - /* nothing to do for now */ + if (hinic_func_type(hwdev) != TYPE_VF) + hinic_set_port_routine_cmd_report(hwdev, false); } EXPORT_SYMBOL(hinic_free_nic_hwdev); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_port_cmd.h b/drivers/net/ethernet/huawei/hinic/hinic_port_cmd.h index 8730565..46464b1f 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_port_cmd.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_port_cmd.h @@ -146,6 +146,8 @@ enum hinic_port_cmd { HINIC_PORT_CMD_SET_PORT_LINK_STATUS = 0x76, HINIC_PORT_CMD_SET_CGE_PAUSE_TIME_CFG = 0x77, + HINIC_PORT_CMD_SET_PORT_REPORT = 0x7B, + HINIC_PORT_CMD_LINK_STATUS_REPORT = 0xa0, HINIC_PORT_CMD_SET_LOSSLESS_ETH = 0xa3, @@ -200,6 +202,7 @@ enum hinic_port_cmd { HINIC_PORT_CMD_SET_LINK_FOLLOW = 0xF8, HINIC_PORT_CMD_SET_VF_MAX_MIN_RATE = 0xF9, HINIC_PORT_CMD_SET_RXQ_LRO_ADPT = 0xFA, + HINIC_PORT_CMD_GET_SFP_ABS = 0xFB, HINIC_PORT_CMD_Q_FILTER = 0xFC, HINIC_PORT_CMD_TCAM_FILTER = 0xFE, HINIC_PORT_CMD_SET_VLAN_FILTER = 0xFF, -- 1.8.3

1 2

[PATCH] Revert "consolemap: Fix a memory leaking bug in drivers/tty/vt/consolemap.c"
by Yang Yingliang 01 Jun '20

01 Jun '20

From: Ben Hutchings <ben(a)decadent.org.uk> mainline inclusion from mainline-v5.3-rc1 commit 15b3cd8ef46ad1b100e0d3c7e38774f330726820 category: bugfix bugzilla: 34617 CVE: CVE-2019-12379 --------------------------- This reverts commit 84ecc2f6eb1cb12e6d44818f94fa49b50f06e6ac. con_insert_unipair() is working with a sparse 3-dimensional array: - p->uni_pgdir[] is the top layer - p1 points to a middle layer - p2 points to a bottom layer If it needs to allocate a new middle layer, and then fails to allocate a new bottom layer, it would previously free only p2, and now it frees both p1 and p2. But since the new middle layer was already registered in the top layer, it was not leaked. However, if it looks up an *existing* middle layer and then fails to allocate a bottom layer, it now frees both p1 and p2 but does *not* free any other bottom layers under p1. So it *introduces* a memory leak. The error path also cleared the wrong index in p->uni_pgdir[], introducing a use-after-free. Signed-off-by: Ben Hutchings <ben(a)decadent.org.uk> Fixes: 84ecc2f6eb1c ("consolemap: Fix a memory leaking bug in drivers/tty/vt/consolemap.c") Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Zheng Bin <zhengbin13(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/tty/vt/consolemap.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/tty/vt/consolemap.c b/drivers/tty/vt/consolemap.c index 814d1b7..7c7ada0 100644 --- a/drivers/tty/vt/consolemap.c +++ b/drivers/tty/vt/consolemap.c @@ -489,11 +489,7 @@ static int con_unify_unimap(struct vc_data *conp, struct uni_pagedir *p) p2 = p1[n = (unicode >> 6) & 0x1f]; if (!p2) { p2 = p1[n] = kmalloc_array(64, sizeof(u16), GFP_KERNEL); - if (!p2) { - kfree(p1); - p->uni_pgdir[n] = NULL; - return -ENOMEM; - } + if (!p2) return -ENOMEM; memset(p2, 0xff, 64*sizeof(u16)); /* No glyphs for the characters (yet) */ } -- 1.8.3

1 0

[PATCH 1/3] sunrpc: check that domain table is empty at module unload.
by Yang Yingliang 26 May '20

26 May '20

From: NeilBrown <neilb(a)suse.de> hulk inclusion category: bugfix bugzilla: NA CVE: CVE-2020-12656 --------------------------- The domain table should be empty at module unload. If it isn't there is a bug somewhere. So check and report. Link: https://bugzilla.kernel.org/show_bug.cgi?id=206651 Signed-off-by: NeilBrown <neilb(a)suse.de> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Wenan Mao <maowenan(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- net/sunrpc/sunrpc.h | 1 + net/sunrpc/sunrpc_syms.c | 2 ++ net/sunrpc/svcauth.c | 27 +++++++++++++++++++++++++++ 3 files changed, 30 insertions(+) diff --git a/net/sunrpc/sunrpc.h b/net/sunrpc/sunrpc.h index c9bacb3..82035fa 100644 --- a/net/sunrpc/sunrpc.h +++ b/net/sunrpc/sunrpc.h @@ -56,4 +56,5 @@ int svc_send_common(struct socket *sock, struct xdr_buf *xdr, int rpc_clients_notifier_register(void); void rpc_clients_notifier_unregister(void); +void auth_domain_cleanup(void); #endif /* _NET_SUNRPC_SUNRPC_H */ diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c index 56f9eff..7b4c6ee 100644 --- a/net/sunrpc/sunrpc_syms.c +++ b/net/sunrpc/sunrpc_syms.c @@ -22,6 +22,7 @@ #include <linux/sunrpc/rpc_pipe_fs.h> #include <linux/sunrpc/xprtsock.h> +#include "sunrpc.h" #include "netns.h" unsigned int sunrpc_net_id; @@ -130,6 +131,7 @@ static __net_exit void sunrpc_exit_net(struct net *net) unregister_rpc_pipefs(); rpc_destroy_mempool(); unregister_pernet_subsys(&sunrpc_net_ops); + auth_domain_cleanup(); #if IS_ENABLED(CONFIG_SUNRPC_DEBUG) rpc_unregister_sysctl(); #endif diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c index bb8db3c..0107350 100644 --- a/net/sunrpc/svcauth.c +++ b/net/sunrpc/svcauth.c @@ -18,6 +18,10 @@ #include <linux/err.h> #include <linux/hash.h> +#include <trace/events/sunrpc.h> + +#include "sunrpc.h" + #define RPCDBG_FACILITY RPCDBG_AUTH @@ -170,3 +174,26 @@ struct auth_domain *auth_domain_find(char *name) return auth_domain_lookup(name, NULL); } EXPORT_SYMBOL_GPL(auth_domain_find); + +/** + * auth_domain_cleanup - check that the auth_domain table is empty + * + * On module unload the auth_domain_table must be empty. To make it + * easier to catch bugs which don't clean up domains properly, we + * warn if anything remains in the table at cleanup time. + * + * Note that we cannot proactively remove the domains at this stage. + * The ->release() function might be in a module that has already been + * unloaded. + */ + +void auth_domain_cleanup(void) +{ + int h; + struct auth_domain *hp; + + for (h = 0; h < DN_HASHMAX; h++) + hlist_for_each_entry(hp, &auth_domain_table[h], hash) + pr_warn("svc: domain %s still present at module unload.\n", + hp->name); +} -- 1.8.3

1 2

[PATCH 1/7] USB: gadget: fix illegal array access in binding with UDC
by Yang Yingliang 26 May '20

26 May '20

From: Kyungtae Kim <kt0755(a)gmail.com> mainline inclusion from mainline-v5.7-rc6 commit 15753588bcd4bbffae1cca33c8ced5722477fe1f category: bugfix bugzilla: 13690 CVE: CVE-2020-13143 ------------------------------------------------- FuzzUSB (a variant of syzkaller) found an illegal array access using an incorrect index while binding a gadget with UDC. Reference: https://www.spinics.net/lists/linux-usb/msg194331.html This bug occurs when a size variable used for a buffer is misused to access its strcpy-ed buffer. Given a buffer along with its size variable (taken from user input), from which, a new buffer is created using kstrdup(). Due to the original buffer containing 0 value in the middle, the size of the kstrdup-ed buffer becomes smaller than that of the original. So accessing the kstrdup-ed buffer with the same size variable triggers memory access violation. The fix makes sure no zero value in the buffer, by comparing the strlen() of the orignal buffer with the size variable, so that the access to the kstrdup-ed buffer is safe. BUG: KASAN: slab-out-of-bounds in gadget_dev_desc_UDC_store+0x1ba/0x200 drivers/usb/gadget/configfs.c:266 Read of size 1 at addr ffff88806a55dd7e by task syz-executor.0/17208 CPU: 2 PID: 17208 Comm: syz-executor.0 Not tainted 5.6.8 #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0xce/0x128 lib/dump_stack.c:118 print_address_description.constprop.4+0x21/0x3c0 mm/kasan/report.c:374 __kasan_report+0x131/0x1b0 mm/kasan/report.c:506 kasan_report+0x12/0x20 mm/kasan/common.c:641 __asan_report_load1_noabort+0x14/0x20 mm/kasan/generic_report.c:132 gadget_dev_desc_UDC_store+0x1ba/0x200 drivers/usb/gadget/configfs.c:266 flush_write_buffer fs/configfs/file.c:251 [inline] configfs_write_file+0x2f1/0x4c0 fs/configfs/file.c:283 __vfs_write+0x85/0x110 fs/read_write.c:494 vfs_write+0x1cd/0x510 fs/read_write.c:558 ksys_write+0x18a/0x220 fs/read_write.c:611 __do_sys_write fs/read_write.c:623 [inline] __se_sys_write fs/read_write.c:620 [inline] __x64_sys_write+0x73/0xb0 fs/read_write.c:620 do_syscall_64+0x9e/0x510 arch/x86/entry/common.c:294 entry_SYSCALL_64_after_hwframe+0x49/0xbe Signed-off-by: Kyungtae Kim <kt0755(a)gmail.com> Reported-and-tested-by: Kyungtae Kim <kt0755(a)gmail.com> Cc: Felipe Balbi <balbi(a)kernel.org> Cc: stable <stable(a)vger.kernel.org> Link: https://lore.kernel.org/r/20200510054326.GA19198@pizza01 Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/usb/gadget/configfs.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/gadget/configfs.c b/drivers/usb/gadget/configfs.c index ab9ac48..a7709d1 100644 --- a/drivers/usb/gadget/configfs.c +++ b/drivers/usb/gadget/configfs.c @@ -260,6 +260,9 @@ static ssize_t gadget_dev_desc_UDC_store(struct config_item *item, char *name; int ret; + if (strlen(page) < len) + return -EOVERFLOW; + name = kstrdup(page, GFP_KERNEL); if (!name) return -ENOMEM; -- 1.8.3

1 6

[PATCH] signal: fix kabi changes in struct task_struct
by Yang Yingliang 20 May '20

20 May '20

hulk inclusion category: kabi bugzilla: NA CVE: CVE-2020-10741, CVE-2020-12826 --------------------------- Commit d1e7fd6462ca ("signal: Extend exec_id to 64bits") can fixes CVE-2020-10741 and CVE-2020-12826, but it introduces a kabi change in struct task_strcut. Fix this kabi broken by using another new 64bits variables parent_exec_id_u64 and self_exec_id_u64. Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/exec.c | 1 + include/linux/sched.h | 9 +++++++-- kernel/fork.c | 2 ++ kernel/signal.c | 2 +- 4 files changed, 11 insertions(+), 3 deletions(-) diff --git a/fs/exec.c b/fs/exec.c index 15d9974..19c0700 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1380,6 +1380,7 @@ void setup_new_exec(struct linux_binprm * bprm) /* An exec changes our domain. We are no longer part of the thread group */ WRITE_ONCE(current->self_exec_id, current->self_exec_id + 1); + WRITE_ONCE(current->self_exec_id_u64, current->self_exec_id_u64 + 1); flush_signal_handlers(current, 0); } EXPORT_SYMBOL(setup_new_exec); diff --git a/include/linux/sched.h b/include/linux/sched.h index 1d15ab4..302fa00 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -892,8 +892,8 @@ struct task_struct { struct seccomp seccomp; /* Thread group tracking: */ - u64 parent_exec_id; - u64 self_exec_id; + u32 parent_exec_id; + u32 self_exec_id; /* Protection against (de-)allocation: mm, files, fs, tty, keyrings, mems_allowed, mempolicy: */ spinlock_t alloc_lock; @@ -1212,8 +1212,13 @@ struct task_struct { */ randomized_struct_fields_end +#ifndef __GENKSYMS__ + u64 parent_exec_id_u64; + u64 self_exec_id_u64; +#else KABI_RESERVE(1) KABI_RESERVE(2) +#endif KABI_RESERVE(3) KABI_RESERVE(4) KABI_RESERVE(5) diff --git a/kernel/fork.c b/kernel/fork.c index 2839961..951aa6f 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -2055,9 +2055,11 @@ static __latent_entropy struct task_struct *copy_process( if (clone_flags & (CLONE_PARENT|CLONE_THREAD)) { p->real_parent = current->real_parent; p->parent_exec_id = current->parent_exec_id; + p->parent_exec_id_u64 = current->parent_exec_id_u64; } else { p->real_parent = current; p->parent_exec_id = current->self_exec_id; + p->parent_exec_id_u64 = current->self_exec_id_u64; } klp_copy_process(p); diff --git a/kernel/signal.c b/kernel/signal.c index 60ea2ee..a58af7d 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1827,7 +1827,7 @@ bool do_notify_parent(struct task_struct *tsk, int sig) * This is only possible if parent == real_parent. * Check if it has changed security domain. */ - if (tsk->parent_exec_id != READ_ONCE(tsk->parent->self_exec_id)) + if (tsk->parent_exec_id_u64 != READ_ONCE(tsk->parent->self_exec_id_u64)) sig = SIGCHLD; } -- 1.8.3

1 0

[PATCH 01/14] net: hns3: solve the unlock 2 times when rocee init fault
by Yang Yingliang 20 May '20

20 May '20

From: Hao Shen <shenhao21(a)huawei.com> driver inclusion category: bugfix bugzilla: NA CVE: NA -------------------------------------------------------- When rocee init fault, the rtnl_unlock execute 2 times Signed-off-by: Hao Shen <shenhao21(a)huawei.com> Reviewed-by: Zhong Zhaohui <zhongzhaohui(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 0b7e159..3e0afc4 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -3865,7 +3865,7 @@ static int hclge_resume(struct hnae3_ae_dev *ae_dev) ret = hclge_notify_roce_client(hdev, HNAE3_INIT_CLIENT); if (ret) - goto err_reset_lock; + return ret; rtnl_lock(); -- 1.8.3

1 13

[PATCH 1/2] KVM: SVM: Fix potential memory leak in svm_cpu_init()
by Yang Yingliang 13 May '20

13 May '20

From: Miaohe Lin <linmiaohe(a)huawei.com> mainline inclusion from mainline-v5.6-rc4 commit d80b64ff297e40c2b6f7d7abc1b3eba70d22a068 category: bugfix bugzilla: 13690 CVE: CVE-2020-12768 ------------------------------------------------- When kmalloc memory for sd->sev_vmcbs failed, we forget to free the page held by sd->save_area. Also get rid of the var r as '-ENOMEM' is actually the only possible outcome here. Reviewed-by: Liran Alon <liran.alon(a)oracle.com> Reviewed-by: Vitaly Kuznetsov <vkuznets(a)redhat.com> Signed-off-by: Miaohe Lin <linmiaohe(a)huawei.com> Signed-off-by: Paolo Bonzini <pbonzini(a)redhat.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/x86/kvm/svm.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index df22744..226db3dc 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -998,33 +998,32 @@ static void svm_cpu_uninit(int cpu) static int svm_cpu_init(int cpu) { struct svm_cpu_data *sd; - int r; sd = kzalloc(sizeof(struct svm_cpu_data), GFP_KERNEL); if (!sd) return -ENOMEM; sd->cpu = cpu; - r = -ENOMEM; sd->save_area = alloc_page(GFP_KERNEL); if (!sd->save_area) - goto err_1; + goto free_cpu_data; if (svm_sev_enabled()) { - r = -ENOMEM; sd->sev_vmcbs = kmalloc_array(max_sev_asid + 1, sizeof(void *), GFP_KERNEL); if (!sd->sev_vmcbs) - goto err_1; + goto free_save_area; } per_cpu(svm_data, cpu) = sd; return 0; -err_1: +free_save_area: + __free_page(sd->save_area); +free_cpu_data: kfree(sd); - return r; + return -ENOMEM; } -- 1.8.3

1 1

[PATCH] config: disable CONFIG_ARM64_ERRATUM_1542419 by default
by Yang Yingliang 13 May '20

13 May '20

hulk inclusion category: config bugzilla: NA CVE: NA ------------------------------------------------- Introduced by 05460849c3b5 ("arm64: errata: Hide CTR_EL0.DIC on systems affected by Neoverse-N1 #1542419"). Disable CONFIG_ARM64_ERRATUM_1542419 by default. Reviewed-by: Xuefeng Wang <wxf.wang(a)hisilicon.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/configs/euleros_defconfig | 1 + arch/arm64/configs/hulk_defconfig | 1 + arch/arm64/configs/openeuler_defconfig | 1 + arch/arm64/configs/storage_ci_defconfig | 1 + arch/arm64/configs/syzkaller_defconfig | 1 + 5 files changed, 5 insertions(+) diff --git a/arch/arm64/configs/euleros_defconfig b/arch/arm64/configs/euleros_defconfig index 04456f3..360e692 100644 --- a/arch/arm64/configs/euleros_defconfig +++ b/arch/arm64/configs/euleros_defconfig @@ -387,6 +387,7 @@ CONFIG_ARM64_ERRATUM_845719=y CONFIG_ARM64_ERRATUM_843419=y CONFIG_ARM64_ERRATUM_1024718=y # CONFIG_ARM64_ERRATUM_1463225 is not set +# CONFIG_ARM64_ERRATUM_1542419 is not set CONFIG_CAVIUM_ERRATUM_22375=y CONFIG_CAVIUM_ERRATUM_23144=y CONFIG_CAVIUM_ERRATUM_23154=y diff --git a/arch/arm64/configs/hulk_defconfig b/arch/arm64/configs/hulk_defconfig index 6ee2472..bb020a1 100644 --- a/arch/arm64/configs/hulk_defconfig +++ b/arch/arm64/configs/hulk_defconfig @@ -386,6 +386,7 @@ CONFIG_ARM64_ERRATUM_845719=y # CONFIG_ARM64_ERRATUM_843419 is not set # CONFIG_ARM64_ERRATUM_1024718 is not set # CONFIG_ARM64_ERRATUM_1463225 is not set +# CONFIG_ARM64_ERRATUM_1542419 is not set # CONFIG_CAVIUM_ERRATUM_22375 is not set # CONFIG_CAVIUM_ERRATUM_23144 is not set # CONFIG_CAVIUM_ERRATUM_23154 is not set diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 9e6f560..b943729 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -389,6 +389,7 @@ CONFIG_ARM64_ERRATUM_845719=y CONFIG_ARM64_ERRATUM_843419=y CONFIG_ARM64_ERRATUM_1024718=y # CONFIG_ARM64_ERRATUM_1463225 is not set +# CONFIG_ARM64_ERRATUM_1542419 is not set CONFIG_CAVIUM_ERRATUM_22375=y CONFIG_CAVIUM_ERRATUM_23144=y CONFIG_CAVIUM_ERRATUM_23154=y diff --git a/arch/arm64/configs/storage_ci_defconfig b/arch/arm64/configs/storage_ci_defconfig index 2ad533d..b7f23ca 100644 --- a/arch/arm64/configs/storage_ci_defconfig +++ b/arch/arm64/configs/storage_ci_defconfig @@ -373,6 +373,7 @@ CONFIG_ARM64_ERRATUM_845719=y CONFIG_ARM64_ERRATUM_843419=y CONFIG_ARM64_ERRATUM_1024718=y # CONFIG_ARM64_ERRATUM_1463225 is not set +# CONFIG_ARM64_ERRATUM_1542419 is not set CONFIG_CAVIUM_ERRATUM_22375=y CONFIG_CAVIUM_ERRATUM_23144=y CONFIG_CAVIUM_ERRATUM_23154=y diff --git a/arch/arm64/configs/syzkaller_defconfig b/arch/arm64/configs/syzkaller_defconfig index 509713a..eaf5e32 100644 --- a/arch/arm64/configs/syzkaller_defconfig +++ b/arch/arm64/configs/syzkaller_defconfig @@ -385,6 +385,7 @@ CONFIG_ARM64_ERRATUM_845719=y # CONFIG_ARM64_ERRATUM_843419 is not set # CONFIG_ARM64_ERRATUM_1024718 is not set # CONFIG_ARM64_ERRATUM_1463225 is not set +# CONFIG_ARM64_ERRATUM_1542419 is not set # CONFIG_CAVIUM_ERRATUM_22375 is not set # CONFIG_CAVIUM_ERRATUM_23144 is not set # CONFIG_CAVIUM_ERRATUM_23154 is not set -- 1.8.3

1 0

[PATCH 01/33] vhost: vsock: kick send_pkt worker once device is started
by Yang Yingliang 13 May '20

13 May '20

From: Jia He <justin.he(a)arm.com> commit 0b841030625cde5f784dd62aec72d6a766faae70 upstream. Ning Bo reported an abnormal 2-second gap when booting Kata container [1]. The unconditional timeout was caused by VSOCK_DEFAULT_CONNECT_TIMEOUT of connecting from the client side. The vhost vsock client tries to connect an initializing virtio vsock server. The abnormal flow looks like: host-userspace vhost vsock guest vsock ============== =========== ============ connect() --------> vhost_transport_send_pkt_work() initializing | vq->private_data==NULL | will not be queued V schedule_timeout(2s) vhost_vsock_start() <--------- device ready set vq->private_data wait for 2s and failed connect() again vq->private_data!=NULL recv connecting pkt Details: 1. Host userspace sends a connect pkt, at that time, guest vsock is under initializing, hence the vhost_vsock_start has not been called. So vq->private_data==NULL, and the pkt is not been queued to send to guest 2. Then it sleeps for 2s 3. After guest vsock finishes initializing, vq->private_data is set 4. When host userspace wakes up after 2s, send connecting pkt again, everything is fine. As suggested by Stefano Garzarella, this fixes it by additional kicking the send_pkt worker in vhost_vsock_start once the virtio device is started. This makes the pending pkt sent again. After this patch, kata-runtime (with vsock enabled) boot time is reduced from 3s to 1s on a ThunderX2 arm64 server. [1] https://github.com/kata-containers/runtime/issues/1917 Reported-by: Ning Bo <n.b(a)live.com> Suggested-by: Stefano Garzarella <sgarzare(a)redhat.com> Signed-off-by: Jia He <justin.he(a)arm.com> Link: https://lore.kernel.org/r/20200501043840.186557-1-justin.he@arm.com Signed-off-by: Michael S. Tsirkin <mst(a)redhat.com> Reviewed-by: Stefano Garzarella <sgarzare(a)redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/vhost/vsock.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c index 5f5c5de..bac1365 100644 --- a/drivers/vhost/vsock.c +++ b/drivers/vhost/vsock.c @@ -499,6 +499,11 @@ static int vhost_vsock_start(struct vhost_vsock *vsock) mutex_unlock(&vq->mutex); } + /* Some packets may have been queued before the device was started, + * let's kick the send worker to send them. + */ + vhost_work_queue(&vsock->dev, &vsock->send_pkt_work); + mutex_unlock(&vsock->dev.mutex); return 0; -- 1.8.3

1 32

[PATCH 01/38] drm/edid: Fix off-by-one in DispID DTD pixel clock
by Yang Yingliang 12 May '20

12 May '20

From: Ville Syrjälä <ville.syrjala(a)linux.intel.com> commit 6292b8efe32e6be408af364132f09572aed14382 upstream. The DispID DTD pixel clock is documented as: "00 00 00 h → FF FF FF h | Pixel clock ÷ 10,000 0.01 → 167,772.16 Mega Pixels per Sec" Which seems to imply that we to add one to the raw value. Reality seems to agree as there are tiled displays in the wild which currently show a 10kHz difference in the pixel clock between the tiles (one tile gets its mode from the base EDID, the other from the DispID block). Cc: stable(a)vger.kernel.org References: https://gitlab.freedesktop.org/drm/intel/-/issues/27 Signed-off-by: Ville Syrjälä <ville.syrjala(a)linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200423151743.18767-1-ville.… Reviewed-by: Manasi Navare <manasi.d.navare(a)intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/gpu/drm/drm_edid.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index f5926bf5..d5dcee7 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -4706,7 +4706,7 @@ static struct drm_display_mode *drm_mode_displayid_detailed(struct drm_device *d struct drm_display_mode *mode; unsigned pixel_clock = (timings->pixel_clock[0] | (timings->pixel_clock[1] << 8) | - (timings->pixel_clock[2] << 16)); + (timings->pixel_clock[2] << 16)) + 1; unsigned hactive = (timings->hactive[0] | timings->hactive[1] << 8) + 1; unsigned hblank = (timings->hblank[0] | timings->hblank[1] << 8) + 1; unsigned hsync = (timings->hsync[0] | (timings->hsync[1] & 0x7f) << 8) + 1; -- 1.8.3

1 75

[PATCH 01/48] remoteproc: Fix wrong rvring index computation
by Yang Yingliang 12 May '20

12 May '20

From: Clement Leger <cleger(a)kalray.eu> commit 00a0eec59ddbb1ce966b19097d8a8d2f777e726a upstream. Index of rvring is computed using pointer arithmetic. However, since rvring->rvdev->vring is the base of the vring array, computation of rvring idx should be reversed. It previously lead to writing at negative indices in the resource table. Signed-off-by: Clement Leger <cleger(a)kalray.eu> Link: https://lore.kernel.org/r/20191004073736.8327-1-cleger@kalray.eu Signed-off-by: Bjorn Andersson <bjorn.andersson(a)linaro.org> Cc: Doug Anderson <dianders(a)chromium.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/remoteproc/remoteproc_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/remoteproc/remoteproc_core.c b/drivers/remoteproc/remoteproc_core.c index abbef17..d5ff272 100644 --- a/drivers/remoteproc/remoteproc_core.c +++ b/drivers/remoteproc/remoteproc_core.c @@ -289,7 +289,7 @@ void rproc_free_vring(struct rproc_vring *rvring) { int size = PAGE_ALIGN(vring_size(rvring->len, rvring->align)); struct rproc *rproc = rvring->rvdev->rproc; - int idx = rvring->rvdev->vring - rvring; + int idx = rvring - rvring->rvdev->vring; struct fw_rsc_vdev *rsc; dma_free_coherent(rproc->dev.parent, size, rvring->va, rvring->dma); -- 1.8.3

1 95

[PATCH 001/124] ext4: fix extent_status fragmentation for plain files
by Yang Yingliang 12 May '20

12 May '20

From: Dmitry Monakhov <dmonakhov(a)gmail.com> [ Upstream commit 4068664e3cd2312610ceac05b74c4cf1853b8325 ] Extents are cached in read_extent_tree_block(); as a result, extents are not cached for inodes with depth == 0 when we try to find the extent using ext4_find_extent(). The result of the lookup is cached in ext4_map_blocks() but is only a subset of the extent on disk. As a result, the contents of extents status cache can get very badly fragmented for certain workloads, such as a random 4k read workload. File size of /mnt/test is 33554432 (8192 blocks of 4096 bytes) ext: logical_offset: physical_offset: length: expected: flags: 0: 0.. 8191: 40960.. 49151: 8192: last,eof $ perf record -e 'ext4:ext4_es_*' /root/bin/fio --name=t --direct=0 --rw=randread --bs=4k --filesize=32M --size=32M --filename=/mnt/test $ perf script | grep ext4_es_insert_extent | head -n 10 fio 131 [000] 13.975421: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [494/1) mapped 41454 status W fio 131 [000] 13.975939: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [6064/1) mapped 47024 status W fio 131 [000] 13.976467: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [6907/1) mapped 47867 status W fio 131 [000] 13.976937: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [3850/1) mapped 44810 status W fio 131 [000] 13.977440: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [3292/1) mapped 44252 status W fio 131 [000] 13.977931: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [6882/1) mapped 47842 status W fio 131 [000] 13.978376: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [3117/1) mapped 44077 status W fio 131 [000] 13.978957: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [2896/1) mapped 43856 status W fio 131 [000] 13.979474: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [7479/1) mapped 48439 status W Fix this by caching the extents for inodes with depth == 0 in ext4_find_extent(). [ Renamed ext4_es_cache_extents() to ext4_cache_extents() since this newly added function is not in extents_cache.c, and to avoid potential visual confusion with ext4_es_cache_extent(). -TYT ] Signed-off-by: Dmitry Monakhov <dmonakhov(a)gmail.com> Link: https://lore.kernel.org/r/20191106122502.19986-1-dmonakhov@gmail.com Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- fs/ext4/extents.c | 47 +++++++++++++++++++++++++++-------------------- 1 file changed, 27 insertions(+), 20 deletions(-) diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 90862ee..0f17c54 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -510,6 +510,30 @@ int ext4_ext_check_inode(struct inode *inode) return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode), 0); } +static void ext4_cache_extents(struct inode *inode, + struct ext4_extent_header *eh) +{ + struct ext4_extent *ex = EXT_FIRST_EXTENT(eh); + ext4_lblk_t prev = 0; + int i; + + for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) { + unsigned int status = EXTENT_STATUS_WRITTEN; + ext4_lblk_t lblk = le32_to_cpu(ex->ee_block); + int len = ext4_ext_get_actual_len(ex); + + if (prev && (prev != lblk)) + ext4_es_cache_extent(inode, prev, lblk - prev, ~0, + EXTENT_STATUS_HOLE); + + if (ext4_ext_is_unwritten(ex)) + status = EXTENT_STATUS_UNWRITTEN; + ext4_es_cache_extent(inode, lblk, len, + ext4_ext_pblock(ex), status); + prev = lblk + len; + } +} + static struct buffer_head * __read_extent_tree_block(const char *function, unsigned int line, struct inode *inode, ext4_fsblk_t pblk, int depth, @@ -544,26 +568,7 @@ int ext4_ext_check_inode(struct inode *inode) */ if (!(flags & EXT4_EX_NOCACHE) && depth == 0) { struct ext4_extent_header *eh = ext_block_hdr(bh); - struct ext4_extent *ex = EXT_FIRST_EXTENT(eh); - ext4_lblk_t prev = 0; - int i; - - for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) { - unsigned int status = EXTENT_STATUS_WRITTEN; - ext4_lblk_t lblk = le32_to_cpu(ex->ee_block); - int len = ext4_ext_get_actual_len(ex); - - if (prev && (prev != lblk)) - ext4_es_cache_extent(inode, prev, - lblk - prev, ~0, - EXTENT_STATUS_HOLE); - - if (ext4_ext_is_unwritten(ex)) - status = EXTENT_STATUS_UNWRITTEN; - ext4_es_cache_extent(inode, lblk, len, - ext4_ext_pblock(ex), status); - prev = lblk + len; - } + ext4_cache_extents(inode, eh); } return bh; errout: @@ -911,6 +916,8 @@ struct ext4_ext_path * path[0].p_bh = NULL; i = depth; + if (!(flags & EXT4_EX_NOCACHE) && depth == 0) + ext4_cache_extents(inode, eh); /* walk through the tree */ while (i) { ext_debug("depth %d: num %d, max %d\n", -- 1.8.3

1 247