- Kernel - mailweb.openeuler.org

[PATCH openEuler-1.0-LTS] ACPI: processor_idle: Fix memory leak in acpi_processor_power_exit()
by liwei 24 Apr '24

24 Apr '24

From: Armin Wolf <W_Armin(a)gmx.de> mainline inclusion from mainline-v6.9-rc1 commit e18afcb7b2a12b635ac10081f943fcf84ddacc51 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9HK63 CVE: CVE-2024-26894 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… ------------------------------------------------------ After unregistering the CPU idle device, the memory associated with it is not freed, leading to a memory leak: unreferenced object 0xffff896282f6c000 (size 1024): comm "swapper/0", pid 1, jiffies 4294893170 hex dump (first 32 bytes): 00 00 00 00 0b 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace (crc 8836a742): [<ffffffff993495ed>] kmalloc_trace+0x29d/0x340 [<ffffffff9972f3b3>] acpi_processor_power_init+0xf3/0x1c0 [<ffffffff9972d263>] __acpi_processor_start+0xd3/0xf0 [<ffffffff9972d2bc>] acpi_processor_start+0x2c/0x50 [<ffffffff99805872>] really_probe+0xe2/0x480 [<ffffffff99805c98>] __driver_probe_device+0x78/0x160 [<ffffffff99805daf>] driver_probe_device+0x1f/0x90 [<ffffffff9980601e>] __driver_attach+0xce/0x1c0 [<ffffffff99803170>] bus_for_each_dev+0x70/0xc0 [<ffffffff99804822>] bus_add_driver+0x112/0x210 [<ffffffff99807245>] driver_register+0x55/0x100 [<ffffffff9aee4acb>] acpi_processor_driver_init+0x3b/0xc0 [<ffffffff990012d1>] do_one_initcall+0x41/0x300 [<ffffffff9ae7c4b0>] kernel_init_freeable+0x320/0x470 [<ffffffff99b231f6>] kernel_init+0x16/0x1b0 [<ffffffff99042e6d>] ret_from_fork+0x2d/0x50 Fix this by freeing the CPU idle device after unregistering it. Fixes: 3d339dcbb56d ("cpuidle / ACPI : move cpuidle_device field out of the acpi_processor_power structure") Signed-off-by: Armin Wolf <W_Armin(a)gmx.de> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki(a)intel.com> Signed-off-by: liwei <liwei728(a)huawei.com> --- drivers/acpi/processor_idle.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 6ab492934a6f..8167c55b6c4e 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -1433,6 +1433,8 @@ int acpi_processor_power_exit(struct acpi_processor *pr) acpi_processor_registered--; if (acpi_processor_registered == 0) cpuidle_unregister_driver(&acpi_idle_driver); + + kfree(dev); } pr->flags.power_setup_done = 0; -- 2.25.1

2 1

[PATCH OLK-6.6] LoongArch: fix KASLR can not be disabled by nokaslr when boot from old BPI
by Hongchen Zhang 24 Apr '24

24 Apr '24

LoongArch inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I9JD2Y ------------------------------------------ After this patch,KASLR is really disabled when nokaslr passed from boot parameter for old BPI. Fixes: db5bb24abc8d ("LoongArch: Old BPI compatibility") Signed-off-by: Hongchen Zhang <zhanghongchen(a)loongson.cn> Signed-off-by: Ming Wang <wangming01(a)loongson.cn> --- arch/loongarch/kernel/legacy_boot.h | 3 +++ arch/loongarch/kernel/relocate.c | 6 +++++- 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/arch/loongarch/kernel/legacy_boot.h b/arch/loongarch/kernel/legacy_boot.h index 982bf9b1de72..104d8c53bd2d 100644 --- a/arch/loongarch/kernel/legacy_boot.h +++ b/arch/loongarch/kernel/legacy_boot.h @@ -3,6 +3,7 @@ #define __LEGACY_BOOT_H_ #include <linux/screen_info.h> #include <linux/of_address.h> +#include <asm/loongson.h> #define ADDRESS_TYPE_SYSRAM 1 #define ADDRESS_TYPE_RESERVED 2 #define ADDRESS_TYPE_ACPI 3 @@ -87,4 +88,6 @@ extern int __init pch_msi_parse_madt(union acpi_subtable_headers *header, const unsigned long end); extern struct irq_domain *get_pchpic_irq_domain(void); + +extern __init void fw_init_cmdline(unsigned long argc, unsigned long cmdp); #endif diff --git a/arch/loongarch/kernel/relocate.c b/arch/loongarch/kernel/relocate.c index 1acfa704c8d0..aaa27823ea29 100644 --- a/arch/loongarch/kernel/relocate.c +++ b/arch/loongarch/kernel/relocate.c @@ -15,6 +15,7 @@ #include <asm/inst.h> #include <asm/sections.h> #include <asm/setup.h> +#include "legacy_boot.h" #define RELOCATED(x) ((void *)((long)x + reloc_offset)) #define RELOCATED_KASLR(x) ((void *)((long)x + random_offset)) @@ -172,7 +173,10 @@ unsigned long __init relocate_kernel(void) void *location_new = _text; /* Default to original kernel start */ char *cmdline = early_ioremap(fw_arg1, COMMAND_LINE_SIZE); /* Boot command line is passed in fw_arg1 */ - strscpy(boot_command_line, cmdline, COMMAND_LINE_SIZE); + if (fw_arg0 < 2) + strscpy(boot_command_line, cmdline, COMMAND_LINE_SIZE); + else + fw_init_cmdline(fw_arg0, TO_CACHE(fw_arg1)); /* OLD BPI parameters */ #ifdef CONFIG_RANDOMIZE_BASE location_new = determine_relocation_address(); -- 2.33.0

2 1

[PATCH OLK-5.10] drm/mediatek: Fix a null pointer crash in mtk_drm_crtc_finish_page_flip
by Xiongfeng Wang 24 Apr '24

24 Apr '24

From: Hsin-Yi Wang <hsinyi(a)chromium.org> mainline inclusion from mainline-v6.9-rc1 commit c958e86e9cc1b48cac004a6e245154dfba8e163b category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9HK5V CVE: CVE-2024-26874 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit?i… -------------------------------- It's possible that mtk_crtc->event is NULL in mtk_drm_crtc_finish_page_flip(). pending_needs_vblank value is set by mtk_crtc->event, but in mtk_drm_crtc_atomic_flush(), it's is not guarded by the same lock in mtk_drm_finish_page_flip(), thus a race condition happens. Consider the following case: CPU1 CPU2 step 1: mtk_drm_crtc_atomic_begin() mtk_crtc->event is not null, step 1: mtk_drm_crtc_atomic_flush: mtk_drm_crtc_update_config( !!mtk_crtc->event) step 2: mtk_crtc_ddp_irq -> mtk_drm_finish_page_flip: lock mtk_crtc->event set to null, pending_needs_vblank set to false unlock pending_needs_vblank set to true, step 2: mtk_crtc_ddp_irq -> mtk_drm_finish_page_flip called again, pending_needs_vblank is still true //null pointer Instead of guarding the entire mtk_drm_crtc_atomic_flush(), it's more efficient to just check if mtk_crtc->event is null before use. Fixes: 119f5173628a ("drm/mediatek: Add DRM Driver for Mediatek SoC MT8173.") Signed-off-by: Hsin-Yi Wang <hsinyi(a)chromium.org> Reviewed-by: CK Hu <ck.hu(a)mediatek.com> Link: https://patchwork.kernel.org/project/dri-devel/patch/20240223212404.3709690… Signed-off-by: Chun-Kuang Hu <chunkuang.hu(a)kernel.org> Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> --- drivers/gpu/drm/mediatek/mtk_drm_crtc.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c index cc3cb5b63d44..a0c8e0b30fac 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c @@ -84,11 +84,13 @@ static void mtk_drm_crtc_finish_page_flip(struct mtk_drm_crtc *mtk_crtc) struct drm_crtc *crtc = &mtk_crtc->base; unsigned long flags; - spin_lock_irqsave(&crtc->dev->event_lock, flags); - drm_crtc_send_vblank_event(crtc, mtk_crtc->event); - drm_crtc_vblank_put(crtc); - mtk_crtc->event = NULL; - spin_unlock_irqrestore(&crtc->dev->event_lock, flags); + if (mtk_crtc->event) { + spin_lock_irqsave(&crtc->dev->event_lock, flags); + drm_crtc_send_vblank_event(crtc, mtk_crtc->event); + drm_crtc_vblank_put(crtc); + mtk_crtc->event = NULL; + spin_unlock_irqrestore(&crtc->dev->event_lock, flags); + } } static void mtk_drm_finish_page_flip(struct mtk_drm_crtc *mtk_crtc) -- 2.20.1

2 1

[PATCH OLK-5.10] [Backport] media: rc: bpf attach/detach requires write permission
by Zhenzeng Su 24 Apr '24

24 Apr '24

From: Sean Young <sean(a)mess.org> stable inclusion from stable-v5.10.210 commit 93d8109bf182510629bbefc8cd45296d2393987f category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9HJVU CVE: CVE-2023-52642 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 6a9d552483d50953320b9d3b57abdee8d436f23f upstream. Note that bpf attach/detach also requires CAP_NET_ADMIN. Cc: stable(a)vger.kernel.org Signed-off-by: Sean Young <sean(a)mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Zhenzeng Su <suzhenzeng(a)huawei.com> --- drivers/media/rc/bpf-lirc.c | 6 +++--- drivers/media/rc/lirc_dev.c | 5 ++++- drivers/media/rc/rc-core-priv.h | 2 +- 3 files changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/media/rc/bpf-lirc.c b/drivers/media/rc/bpf-lirc.c index afae0afe3f81..a8c55e4bfaee 100644 --- a/drivers/media/rc/bpf-lirc.c +++ b/drivers/media/rc/bpf-lirc.c @@ -249,7 +249,7 @@ int lirc_prog_attach(const union bpf_attr *attr, struct bpf_prog *prog) if (attr->attach_flags) return -EINVAL; - rcdev = rc_dev_get_from_fd(attr->target_fd); + rcdev = rc_dev_get_from_fd(attr->target_fd, true); if (IS_ERR(rcdev)) return PTR_ERR(rcdev); @@ -274,7 +274,7 @@ int lirc_prog_detach(const union bpf_attr *attr) if (IS_ERR(prog)) return PTR_ERR(prog); - rcdev = rc_dev_get_from_fd(attr->target_fd); + rcdev = rc_dev_get_from_fd(attr->target_fd, true); if (IS_ERR(rcdev)) { bpf_prog_put(prog); return PTR_ERR(rcdev); @@ -299,7 +299,7 @@ int lirc_prog_query(const union bpf_attr *attr, union bpf_attr __user *uattr) if (attr->query.query_flags) return -EINVAL; - rcdev = rc_dev_get_from_fd(attr->query.target_fd); + rcdev = rc_dev_get_from_fd(attr->query.target_fd, false); if (IS_ERR(rcdev)) return PTR_ERR(rcdev); diff --git a/drivers/media/rc/lirc_dev.c b/drivers/media/rc/lirc_dev.c index 9c888047fa99..14243ce03b46 100644 --- a/drivers/media/rc/lirc_dev.c +++ b/drivers/media/rc/lirc_dev.c @@ -826,7 +826,7 @@ void __exit lirc_dev_exit(void) unregister_chrdev_region(lirc_base_dev, RC_DEV_MAX); } -struct rc_dev *rc_dev_get_from_fd(int fd) +struct rc_dev *rc_dev_get_from_fd(int fd, bool write) { struct fd f = fdget(fd); struct lirc_fh *fh; @@ -840,6 +840,9 @@ struct rc_dev *rc_dev_get_from_fd(int fd) return ERR_PTR(-EINVAL); } + if (write && !(f.file->f_mode & FMODE_WRITE)) + return ERR_PTR(-EPERM); + fh = f.file->private_data; dev = fh->rc; diff --git a/drivers/media/rc/rc-core-priv.h b/drivers/media/rc/rc-core-priv.h index 62f032dffd33..dfe0352c0f0a 100644 --- a/drivers/media/rc/rc-core-priv.h +++ b/drivers/media/rc/rc-core-priv.h @@ -325,7 +325,7 @@ void lirc_raw_event(struct rc_dev *dev, struct ir_raw_event ev); void lirc_scancode_event(struct rc_dev *dev, struct lirc_scancode *lsc); int lirc_register(struct rc_dev *dev); void lirc_unregister(struct rc_dev *dev); -struct rc_dev *rc_dev_get_from_fd(int fd); +struct rc_dev *rc_dev_get_from_fd(int fd, bool write); #else static inline int lirc_dev_init(void) { return 0; } static inline void lirc_dev_exit(void) {} -- 2.25.1

2 1

[PATCH openEuler-1.0-LTS] drm/mediatek: Fix a null pointer crash in mtk_drm_crtc_finish_page_flip
by Xiongfeng Wang 24 Apr '24

24 Apr '24

From: Hsin-Yi Wang <hsinyi(a)chromium.org> mainline inclusion from mainline-v6.9-rc1 commit c958e86e9cc1b48cac004a6e245154dfba8e163b category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9HK5V CVE: CVE-2024-26874 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit?i… -------------------------------- It's possible that mtk_crtc->event is NULL in mtk_drm_crtc_finish_page_flip(). pending_needs_vblank value is set by mtk_crtc->event, but in mtk_drm_crtc_atomic_flush(), it's is not guarded by the same lock in mtk_drm_finish_page_flip(), thus a race condition happens. Consider the following case: CPU1 CPU2 step 1: mtk_drm_crtc_atomic_begin() mtk_crtc->event is not null, step 1: mtk_drm_crtc_atomic_flush: mtk_drm_crtc_update_config( !!mtk_crtc->event) step 2: mtk_crtc_ddp_irq -> mtk_drm_finish_page_flip: lock mtk_crtc->event set to null, pending_needs_vblank set to false unlock pending_needs_vblank set to true, step 2: mtk_crtc_ddp_irq -> mtk_drm_finish_page_flip called again, pending_needs_vblank is still true //null pointer Instead of guarding the entire mtk_drm_crtc_atomic_flush(), it's more efficient to just check if mtk_crtc->event is null before use. Fixes: 119f5173628a ("drm/mediatek: Add DRM Driver for Mediatek SoC MT8173.") Signed-off-by: Hsin-Yi Wang <hsinyi(a)chromium.org> Reviewed-by: CK Hu <ck.hu(a)mediatek.com> Link: https://patchwork.kernel.org/project/dri-devel/patch/20240223212404.3709690… Signed-off-by: Chun-Kuang Hu <chunkuang.hu(a)kernel.org> Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> --- drivers/gpu/drm/mediatek/mtk_drm_crtc.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c index 92ecb9bf982c..42da8bb4b7c3 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c @@ -79,11 +79,13 @@ static void mtk_drm_crtc_finish_page_flip(struct mtk_drm_crtc *mtk_crtc) struct drm_crtc *crtc = &mtk_crtc->base; unsigned long flags; - spin_lock_irqsave(&crtc->dev->event_lock, flags); - drm_crtc_send_vblank_event(crtc, mtk_crtc->event); - drm_crtc_vblank_put(crtc); - mtk_crtc->event = NULL; - spin_unlock_irqrestore(&crtc->dev->event_lock, flags); + if (mtk_crtc->event) { + spin_lock_irqsave(&crtc->dev->event_lock, flags); + drm_crtc_send_vblank_event(crtc, mtk_crtc->event); + drm_crtc_vblank_put(crtc); + mtk_crtc->event = NULL; + spin_unlock_irqrestore(&crtc->dev->event_lock, flags); + } } static void mtk_drm_finish_page_flip(struct mtk_drm_crtc *mtk_crtc) -- 2.20.1

2 1

[PATCH openEuler-1.0-LTS] [Backport] media: rc: bpf attach/detach requires write permission
by Zhenzeng Su 24 Apr '24

24 Apr '24

From: Sean Young <sean(a)mess.org> stable inclusion from stable-v5.10.210 commit 93d8109bf182510629bbefc8cd45296d2393987f category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9HJVU CVE: CVE-2023-52642 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 6a9d552483d50953320b9d3b57abdee8d436f23f upstream. Note that bpf attach/detach also requires CAP_NET_ADMIN. Cc: stable(a)vger.kernel.org Signed-off-by: Sean Young <sean(a)mess.org> Signed-off-by: Mauro Carvalho Chehab <mchehab(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Zhenzeng Su <suzhenzeng(a)huawei.com> --- drivers/media/rc/bpf-lirc.c | 6 +++--- drivers/media/rc/lirc_dev.c | 5 ++++- drivers/media/rc/rc-core-priv.h | 2 +- 3 files changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/media/rc/bpf-lirc.c b/drivers/media/rc/bpf-lirc.c index 8b97fd1f0cea..7a83dc5fed00 100644 --- a/drivers/media/rc/bpf-lirc.c +++ b/drivers/media/rc/bpf-lirc.c @@ -218,7 +218,7 @@ int lirc_prog_attach(const union bpf_attr *attr, struct bpf_prog *prog) if (attr->attach_flags) return -EINVAL; - rcdev = rc_dev_get_from_fd(attr->target_fd); + rcdev = rc_dev_get_from_fd(attr->target_fd, true); if (IS_ERR(rcdev)) return PTR_ERR(rcdev); @@ -243,7 +243,7 @@ int lirc_prog_detach(const union bpf_attr *attr) if (IS_ERR(prog)) return PTR_ERR(prog); - rcdev = rc_dev_get_from_fd(attr->target_fd); + rcdev = rc_dev_get_from_fd(attr->target_fd, true); if (IS_ERR(rcdev)) { bpf_prog_put(prog); return PTR_ERR(rcdev); @@ -268,7 +268,7 @@ int lirc_prog_query(const union bpf_attr *attr, union bpf_attr __user *uattr) if (attr->query.query_flags) return -EINVAL; - rcdev = rc_dev_get_from_fd(attr->query.target_fd); + rcdev = rc_dev_get_from_fd(attr->query.target_fd, false); if (IS_ERR(rcdev)) return PTR_ERR(rcdev); diff --git a/drivers/media/rc/lirc_dev.c b/drivers/media/rc/lirc_dev.c index f862f1b7f996..71a7c7cdb1b8 100644 --- a/drivers/media/rc/lirc_dev.c +++ b/drivers/media/rc/lirc_dev.c @@ -848,7 +848,7 @@ void __exit lirc_dev_exit(void) unregister_chrdev_region(lirc_base_dev, RC_DEV_MAX); } -struct rc_dev *rc_dev_get_from_fd(int fd) +struct rc_dev *rc_dev_get_from_fd(int fd, bool write) { struct fd f = fdget(fd); struct lirc_fh *fh; @@ -862,6 +862,9 @@ struct rc_dev *rc_dev_get_from_fd(int fd) return ERR_PTR(-EINVAL); } + if (write && !(f.file->f_mode & FMODE_WRITE)) + return ERR_PTR(-EPERM); + fh = f.file->private_data; dev = fh->rc; diff --git a/drivers/media/rc/rc-core-priv.h b/drivers/media/rc/rc-core-priv.h index e847bdad5c51..8056be86f55f 100644 --- a/drivers/media/rc/rc-core-priv.h +++ b/drivers/media/rc/rc-core-priv.h @@ -303,7 +303,7 @@ void ir_lirc_raw_event(struct rc_dev *dev, struct ir_raw_event ev); void ir_lirc_scancode_event(struct rc_dev *dev, struct lirc_scancode *lsc); int ir_lirc_register(struct rc_dev *dev); void ir_lirc_unregister(struct rc_dev *dev); -struct rc_dev *rc_dev_get_from_fd(int fd); +struct rc_dev *rc_dev_get_from_fd(int fd, bool write); #else static inline int lirc_dev_init(void) { return 0; } static inline void lirc_dev_exit(void) {} -- 2.25.1

2 1

[PATCH V5 OLK-6.6] mm, fs: Add BPF_READAHEAD build option for bpf readhead
by ZhaoLong Wang 24 Apr '24

24 Apr '24

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I7Y9JD CVE: NA --------------------------- This patch introduces a new configuration option called BPF_READAHEAD, which is designed to optimize the read performance in Spark SQL scenarios using eBPF to implement a programmable kernel. The changes include: - Add CONFIG_BPF_READAHEAD to mm/Kconfig, which depends on CONFIG_TRACEPOINTS. - Add conditional compilation directives to fs/ext4/file.c, fs/read_write.c, fs/xfs/xfs_file.c, and include/linux/fs.h to include tracepoint-related headers and functions only when BPF_READAHEAD is enabled. - Miodify page_cache_sync_ra() in mm/readahead.c to disable forced readahead when BPF_READAHEAD is not enabled. V2: Explicitly set CONFIG_ in openeuler_defconfig. Do not add redundant macros to mm/readahead. V3: Place all macro isolation switches in the .h header file. V4: Change the macro name BPF_READAHEAD_OPTIMIZATION to BPF_READAHEAD V5: Correcting macro name replacement mistakes Signed-off-by: ZhaoLong Wang <wangzhaolong1(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 1 + arch/powerpc/configs/openeuler_defconfig | 1 + arch/riscv/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + fs/read_write.c | 4 ++-- include/linux/fs.h | 18 +++++++++++++----- include/trace/events/fs.h | 6 ++++++ mm/Kconfig | 9 +++++++++ 8 files changed, 34 insertions(+), 7 deletions(-) diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 9a6ef6175717..f3ab196f00ab 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -1192,6 +1192,7 @@ CONFIG_DYNAMIC_POOL=y CONFIG_ETMEM_SCAN=m CONFIG_ETMEM_SWAP=m CONFIG_ETMEM=y +# CONFIG_BPF_READAHEAD is not set # # Data Access Monitoring diff --git a/arch/powerpc/configs/openeuler_defconfig b/arch/powerpc/configs/openeuler_defconfig index 152b941da1d5..19754209e3c8 100644 --- a/arch/powerpc/configs/openeuler_defconfig +++ b/arch/powerpc/configs/openeuler_defconfig @@ -860,6 +860,7 @@ CONFIG_ARCH_HAS_HUGEPD=y CONFIG_USERFAULTFD=y # CONFIG_LRU_GEN is not set CONFIG_LOCK_MM_AND_FIND_VMA=y +# CONFIG_BPF_READAHEAD is not set # # Data Access Monitoring diff --git a/arch/riscv/configs/openeuler_defconfig b/arch/riscv/configs/openeuler_defconfig index cb132f4576da..90642dc106e2 100644 --- a/arch/riscv/configs/openeuler_defconfig +++ b/arch/riscv/configs/openeuler_defconfig @@ -825,6 +825,7 @@ CONFIG_ARCH_SUPPORTS_PER_VMA_LOCK=y CONFIG_PER_VMA_LOCK=y CONFIG_LOCK_MM_AND_FIND_VMA=y # CONFIG_PAGE_CACHE_LIMIT is not set +# CONFIG_BPF_READAHEAD is not set # # Data Access Monitoring diff --git a/arch/x86/configs/openeuler_defconfig b/arch/x86/configs/openeuler_defconfig index 1d20beb4bb6d..f49fc3ae7713 100644 --- a/arch/x86/configs/openeuler_defconfig +++ b/arch/x86/configs/openeuler_defconfig @@ -1179,6 +1179,7 @@ CONFIG_DYNAMIC_POOL=y CONFIG_ETMEM_SCAN=m CONFIG_ETMEM_SWAP=m CONFIG_ETMEM=y +# CONFIG_BPF_READAHEAD is not set # # Data Access Monitoring diff --git a/fs/read_write.c b/fs/read_write.c index 3d69fb284d10..dd5c90675f51 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -1721,7 +1721,7 @@ int generic_file_rw_checks(struct file *file_in, struct file *file_out) return 0; } -#ifdef CONFIG_TRACEPOINTS +#ifdef CONFIG_BPF_READAHEAD static void fs_file_read_ctx_init(struct fs_file_read_ctx *ctx, struct file *filp, loff_t pos) { @@ -1752,7 +1752,7 @@ void fs_file_read_update_args_by_trace(struct kiocb *iocb) filp->f_ctl_mode &= ~(ctx.clr_f_mode & FS_FILE_READ_MODE_MASK); } EXPORT_SYMBOL_GPL(fs_file_read_update_args_by_trace); -#endif EXPORT_TRACEPOINT_SYMBOL_GPL(fs_file_read); EXPORT_TRACEPOINT_SYMBOL_GPL(fs_file_release); +#endif diff --git a/include/linux/fs.h b/include/linux/fs.h index d74314a8fa94..55d2844f7fc6 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -43,7 +43,9 @@ #include <linux/cred.h> #include <linux/mnt_idmapping.h> #include <linux/slab.h> +#ifdef CONFIG_BPF_READAHEAD #include <linux/tracepoint-defs.h> +#endif #include <linux/kabi.h> #include <asm/byteorder.h> @@ -190,11 +192,16 @@ typedef int (dio_iodone_t)(struct kiocb *iocb, loff_t offset, /* File supports async nowait buffered writes */ #define FMODE_BUF_WASYNC ((__force fmode_t)0x80000000) +#ifdef CONFIG_BPF_READAHEAD /* File mode control flag, expect random access pattern */ #define FMODE_CTL_RANDOM ((__force fmode_t)0x1000) /* File mode control flag, will try to read head of the file into pagecache */ #define FMODE_CTL_WILLNEED ((__force fmode_t)0x400000) +#else +#define FMODE_CTL_RANDOM 0 +#define FMODE_CTL_WILLNEED 0 +#endif /* * Attribute flags. These should be or-ed together to figure out what @@ -3524,16 +3531,17 @@ struct fs_file_read_ctx { long long index; }; -#ifdef CONFIG_TRACEPOINTS +#ifdef CONFIG_BPF_READAHEAD DECLARE_TRACEPOINT(fs_file_read); extern void fs_file_read_update_args_by_trace(struct kiocb *iocb); -#else -static inline void fs_file_read_update_args_by_trace(struct kiocb *iocb) {} -#endif - static inline void fs_file_read_do_trace(struct kiocb *iocb) { if (tracepoint_enabled(fs_file_read)) fs_file_read_update_args_by_trace(iocb); } +#else +static inline void fs_file_read_update_args_by_trace(struct kiocb *iocb) {} +static inline void fs_file_read_do_trace(struct kiocb *iocb) {} +#endif + #endif /* _LINUX_FS_H */ diff --git a/include/trace/events/fs.h b/include/trace/events/fs.h index ee82dad9d9da..801aad6cb74d 100644 --- a/include/trace/events/fs.h +++ b/include/trace/events/fs.h @@ -1,4 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 */ +#ifdef CONFIG_BPF_READAHEAD + #undef TRACE_SYSTEM #define TRACE_SYSTEM fs @@ -31,3 +33,7 @@ DECLARE_TRACE(fs_file_release, /* This part must be outside protection */ #include <trace/define_trace.h> +#else +#define trace_fs_file_release(...) +#define trace_fs_file_read(...) +#endif /* CONFIG_BPF_READAHEAD */ diff --git a/mm/Kconfig b/mm/Kconfig index cdbb1ceaa554..45d4139c959c 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -1424,6 +1424,15 @@ config ETMEM high-performance storage media to release memory space and reduce memory costs. +config BPF_READAHEAD + bool "Enable bpf readahead optimization" + select TRACEPOINTS + default n + help + EBPF is used to implement a programmable kernel. The readahead behavior + of the kernel is adjusted based on the application read mode to optimize + the read performance in the Spark SQL scenario, + source "mm/damon/Kconfig" endmenu -- 2.39.2

2 1

[PATCH openEuler-22.03-LTS 0/1] openEuler-22.03-LTS CVE-2024-26908
by Yuntao Liu 24 Apr '24

24 Apr '24

fixup CVE-2024-26908 Kunwu Chan (1): x86/xen: Add some null pointer checking to smp.c arch/x86/xen/smp.c | 10 ++++++++++ 1 file changed, 10 insertions(+) -- 2.34.1

2 2

[PATCH OLK-5.10] [Backport] IB/hfi1: Fix a memleak in init_credit_return
by Zhenzeng Su 24 Apr '24

24 Apr '24

From: Zhipeng Lu <alexious(a)zju.edu.cn> stable inclusion from stable-v5.10.211 commit 3fa240bb6b2dbb3e7a3ee1440a4889cbb6207eb7 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9HJPD CVE: CVE-2024-26839 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 809aa64ebff51eb170ee31a95f83b2d21efa32e2 ] When dma_alloc_coherent fails to allocate dd->cr_base[i].va, init_credit_return should deallocate dd->cr_base and dd->cr_base[i] that allocated before. Or those resources would be never freed and a memleak is triggered. Fixes: 7724105686e7 ("IB/hfi1: add driver files") Signed-off-by: Zhipeng Lu <alexious(a)zju.edu.cn> Link: https://lore.kernel.org/r/20240112085523.3731720-1-alexious@zju.edu.cn Acked-by: Dennis Dalessandro <dennis.dalessandro(a)cornelisnetworks.com> Signed-off-by: Leon Romanovsky <leon(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Zhenzeng Su <suzhenzeng(a)huawei.com> --- drivers/infiniband/hw/hfi1/pio.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/hfi1/pio.c b/drivers/infiniband/hw/hfi1/pio.c index 60eb3a64518f..969004258692 100644 --- a/drivers/infiniband/hw/hfi1/pio.c +++ b/drivers/infiniband/hw/hfi1/pio.c @@ -2131,7 +2131,7 @@ int init_credit_return(struct hfi1_devdata *dd) "Unable to allocate credit return DMA range for NUMA %d\n", i); ret = -ENOMEM; - goto done; + goto free_cr_base; } } set_dev_node(&dd->pcidev->dev, dd->node); @@ -2139,6 +2139,10 @@ int init_credit_return(struct hfi1_devdata *dd) ret = 0; done: return ret; + +free_cr_base: + free_credit_return(dd); + goto done; } void free_credit_return(struct hfi1_devdata *dd) -- 2.25.1

2 1

[PATCH OLK-5.10 0/1] CVE-2024-26896
by Yongqiang Liu 24 Apr '24

24 Apr '24

Jérôme Pouiller (1): wifi: wfx: fix memory leak when starting AP drivers/staging/wfx/sta.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) -- 2.34.1

2 2