- Kernel - mailweb.openeuler.org

[PATCH OLK-6.6] ALSA: usb-audio: Fix out of bounds reads when finding clock sources
by Luo Gengkun 28 Dec '24

28 Dec '24

From: Takashi Iwai <tiwai(a)suse.de> stable inclusion from stable-v6.6.64 commit 74cb86e1006c5437b1d90084d22018da30fddc77 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBDHGG CVE: CVE-2024-53150 Reference: https://git.kernel.org/stable/c/74cb86e1006c5437b1d90084d22018da30fddc77 -------------------------------- commit a3dd4d63eeb452cfb064a13862fb376ab108f6a6 upstream. The current USB-audio driver code doesn't check bLength of each descriptor at traversing for clock descriptors. That is, when a device provides a bogus descriptor with a shorter bLength, the driver might hit out-of-bounds reads. For addressing it, this patch adds sanity checks to the validator functions for the clock descriptor traversal. When the descriptor length is shorter than expected, it's skipped in the loop. For the clock source and clock multiplier descriptors, we can just check bLength against the sizeof() of each descriptor type. OTOH, the clock selector descriptor of UAC2 and UAC3 has an array of bNrInPins elements and two more fields at its tail, hence those have to be checked in addition to the sizeof() check. Reported-by: Benoît Sevens <bsevens(a)google.com> Cc: <stable(a)vger.kernel.org> Link: https://lore.kernel.org/20241121140613.3651-1-bsevens@google.com Link: https://patch.msgid.link/20241125144629.20757-1-tiwai@suse.de Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Luo Gengkun <luogengkun2(a)huawei.com> --- sound/usb/clock.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) diff --git a/sound/usb/clock.c b/sound/usb/clock.c index a676ad093d18..f0f1e445cc56 100644 --- a/sound/usb/clock.c +++ b/sound/usb/clock.c @@ -36,6 +36,12 @@ union uac23_clock_multiplier_desc { struct uac_clock_multiplier_descriptor v3; }; +/* check whether the descriptor bLength has the minimal length */ +#define DESC_LENGTH_CHECK(p, proto) \ + ((proto) == UAC_VERSION_3 ? \ + ((p)->v3.bLength >= sizeof((p)->v3)) : \ + ((p)->v2.bLength >= sizeof((p)->v2))) + #define GET_VAL(p, proto, field) \ ((proto) == UAC_VERSION_3 ? (p)->v3.field : (p)->v2.field) @@ -58,6 +64,8 @@ static bool validate_clock_source(void *p, int id, int proto) { union uac23_clock_source_desc *cs = p; + if (!DESC_LENGTH_CHECK(cs, proto)) + return false; return GET_VAL(cs, proto, bClockID) == id; } @@ -65,13 +73,27 @@ static bool validate_clock_selector(void *p, int id, int proto) { union uac23_clock_selector_desc *cs = p; - return GET_VAL(cs, proto, bClockID) == id; + if (!DESC_LENGTH_CHECK(cs, proto)) + return false; + if (GET_VAL(cs, proto, bClockID) != id) + return false; + /* additional length check for baCSourceID array (in bNrInPins size) + * and two more fields (which sizes depend on the protocol) + */ + if (proto == UAC_VERSION_3) + return cs->v3.bLength >= sizeof(cs->v3) + cs->v3.bNrInPins + + 4 /* bmControls */ + 2 /* wCSelectorDescrStr */; + else + return cs->v2.bLength >= sizeof(cs->v2) + cs->v2.bNrInPins + + 1 /* bmControls */ + 1 /* iClockSelector */; } static bool validate_clock_multiplier(void *p, int id, int proto) { union uac23_clock_multiplier_desc *cs = p; + if (!DESC_LENGTH_CHECK(cs, proto)) + return false; return GET_VAL(cs, proto, bClockID) == id; } -- 2.34.1

2 1

[PATCH OLK-6.6 0/9] mm: Backport mainline patch
by Liu Shixin 28 Dec '24

28 Dec '24

Patch[8] has context conflict. David Hildenbrand (1): mm: don't install PMD mappings when THPs are disabled by the hw/process/vma Jingbo Xu (2): mm: fix arithmetic for bdi min_ratio mm: fix arithmetic for max_prop_frac when setting max_ratio Kefeng Wang (1): mm: huge_memory: add vma_thp_disabled() and thp_disabled_by_hw() Liu Shixin (4): mm: kmemleak: split __create_object into two functions mm: kmemleak: use mem_pool_free() to free object mm: kmemleak: add __find_and_remove_object() mm/kmemleak: fix partially freeing unknown object warning Xueshi Hu (1): mm/hugetlb: fix nodes huge page allocation when there are surplus pages include/linux/huge_mm.h | 18 ++++++ mm/huge_memory.c | 13 +---- mm/hugetlb.c | 4 +- mm/kmemleak.c | 126 ++++++++++++++++++++++++++++------------ mm/memory.c | 9 +++ mm/page-writeback.c | 4 +- mm/shmem.c | 7 +-- 7 files changed, 124 insertions(+), 57 deletions(-) -- 2.34.1

2 10

[PATCH openEuler-22.03-LTS-SP1] ALSA: usb-audio: Fix out of bounds reads when finding clock sources
by Luo Gengkun 28 Dec '24

28 Dec '24

From: Takashi Iwai <tiwai(a)suse.de> stable inclusion from stable-v5.10.231 commit 45a92cbc88e4013bfed7fd2ccab3ade45f8e896b category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBDHGG CVE: CVE-2024-53150 Reference: https://git.kernel.org/stable/c/45a92cbc88e4013bfed7fd2ccab3ade45f8e896b -------------------------------- commit a3dd4d63eeb452cfb064a13862fb376ab108f6a6 upstream. The current USB-audio driver code doesn't check bLength of each descriptor at traversing for clock descriptors. That is, when a device provides a bogus descriptor with a shorter bLength, the driver might hit out-of-bounds reads. For addressing it, this patch adds sanity checks to the validator functions for the clock descriptor traversal. When the descriptor length is shorter than expected, it's skipped in the loop. For the clock source and clock multiplier descriptors, we can just check bLength against the sizeof() of each descriptor type. OTOH, the clock selector descriptor of UAC2 and UAC3 has an array of bNrInPins elements and two more fields at its tail, hence those have to be checked in addition to the sizeof() check. Reported-by: Benoît Sevens <bsevens(a)google.com> Cc: <stable(a)vger.kernel.org> Link: https://lore.kernel.org/20241121140613.3651-1-bsevens@google.com Link: https://patch.msgid.link/20241125144629.20757-1-tiwai@suse.de Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Signed-off-by: Benoît Sevens <bsevens(a)google.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Luo Gengkun <luogengkun2(a)huawei.com> --- sound/usb/clock.c | 32 ++++++++++++++++++++++++++++++-- 1 file changed, 30 insertions(+), 2 deletions(-) diff --git a/sound/usb/clock.c b/sound/usb/clock.c index 95b019f15224..6ccea6d1727f 100644 --- a/sound/usb/clock.c +++ b/sound/usb/clock.c @@ -21,6 +21,10 @@ #include "clock.h" #include "quirks.h" +/* check whether the descriptor bLength has the minimal length */ +#define DESC_LENGTH_CHECK(p) \ + (p->bLength >= sizeof(*p)) + static void *find_uac_clock_desc(struct usb_host_interface *iface, int id, bool (*validator)(void *, int), u8 type) { @@ -38,36 +42,60 @@ static void *find_uac_clock_desc(struct usb_host_interface *iface, int id, static bool validate_clock_source_v2(void *p, int id) { struct uac_clock_source_descriptor *cs = p; + if (!DESC_LENGTH_CHECK(cs)) + return false; return cs->bClockID == id; } static bool validate_clock_source_v3(void *p, int id) { struct uac3_clock_source_descriptor *cs = p; + if (!DESC_LENGTH_CHECK(cs)) + return false; return cs->bClockID == id; } static bool validate_clock_selector_v2(void *p, int id) { struct uac_clock_selector_descriptor *cs = p; - return cs->bClockID == id; + if (!DESC_LENGTH_CHECK(cs)) + return false; + if (cs->bClockID != id) + return false; + /* additional length check for baCSourceID array (in bNrInPins size) + * and two more fields (which sizes depend on the protocol) + */ + return cs->bLength >= sizeof(*cs) + cs->bNrInPins + + 1 /* bmControls */ + 1 /* iClockSelector */; } static bool validate_clock_selector_v3(void *p, int id) { struct uac3_clock_selector_descriptor *cs = p; - return cs->bClockID == id; + if (!DESC_LENGTH_CHECK(cs)) + return false; + if (cs->bClockID != id) + return false; + /* additional length check for baCSourceID array (in bNrInPins size) + * and two more fields (which sizes depend on the protocol) + */ + return cs->bLength >= sizeof(*cs) + cs->bNrInPins + + 4 /* bmControls */ + 2 /* wCSelectorDescrStr */; } static bool validate_clock_multiplier_v2(void *p, int id) { struct uac_clock_multiplier_descriptor *cs = p; + if (!DESC_LENGTH_CHECK(cs)) + return false; return cs->bClockID == id; } static bool validate_clock_multiplier_v3(void *p, int id) { struct uac3_clock_multiplier_descriptor *cs = p; + if (!DESC_LENGTH_CHECK(cs)) + return false; return cs->bClockID == id; } -- 2.34.1

2 1

[PATCH OLK-6.6] ALSA: usb-audio: Fix out of bounds reads when finding clock sources
by Luo Gengkun 28 Dec '24

28 Dec '24

From: Takashi Iwai <tiwai(a)suse.de> stable inclusion from stable-v6.6.64 commit 74cb86e1006c5437b1d90084d22018da30fddc77 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBDHGG CVE: CVE-2024-53150 Reference: https://git.kernel.org/stable/c/74cb86e1006c5437b1d90084d22018da30fddc77 commit a3dd4d63eeb452cfb064a13862fb376ab108f6a6 upstream. The current USB-audio driver code doesn't check bLength of each descriptor at traversing for clock descriptors. That is, when a device provides a bogus descriptor with a shorter bLength, the driver might hit out-of-bounds reads. For addressing it, this patch adds sanity checks to the validator functions for the clock descriptor traversal. When the descriptor length is shorter than expected, it's skipped in the loop. For the clock source and clock multiplier descriptors, we can just check bLength against the sizeof() of each descriptor type. OTOH, the clock selector descriptor of UAC2 and UAC3 has an array of bNrInPins elements and two more fields at its tail, hence those have to be checked in addition to the sizeof() check. Reported-by: Benoît Sevens <bsevens(a)google.com> Cc: <stable(a)vger.kernel.org> Link: https://lore.kernel.org/20241121140613.3651-1-bsevens@google.com Link: https://patch.msgid.link/20241125144629.20757-1-tiwai@suse.de Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Luo Gengkun <luogengkun2(a)huawei.com> --- sound/usb/clock.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) diff --git a/sound/usb/clock.c b/sound/usb/clock.c index a676ad093d18..f0f1e445cc56 100644 --- a/sound/usb/clock.c +++ b/sound/usb/clock.c @@ -36,6 +36,12 @@ union uac23_clock_multiplier_desc { struct uac_clock_multiplier_descriptor v3; }; +/* check whether the descriptor bLength has the minimal length */ +#define DESC_LENGTH_CHECK(p, proto) \ + ((proto) == UAC_VERSION_3 ? \ + ((p)->v3.bLength >= sizeof((p)->v3)) : \ + ((p)->v2.bLength >= sizeof((p)->v2))) + #define GET_VAL(p, proto, field) \ ((proto) == UAC_VERSION_3 ? (p)->v3.field : (p)->v2.field) @@ -58,6 +64,8 @@ static bool validate_clock_source(void *p, int id, int proto) { union uac23_clock_source_desc *cs = p; + if (!DESC_LENGTH_CHECK(cs, proto)) + return false; return GET_VAL(cs, proto, bClockID) == id; } @@ -65,13 +73,27 @@ static bool validate_clock_selector(void *p, int id, int proto) { union uac23_clock_selector_desc *cs = p; - return GET_VAL(cs, proto, bClockID) == id; + if (!DESC_LENGTH_CHECK(cs, proto)) + return false; + if (GET_VAL(cs, proto, bClockID) != id) + return false; + /* additional length check for baCSourceID array (in bNrInPins size) + * and two more fields (which sizes depend on the protocol) + */ + if (proto == UAC_VERSION_3) + return cs->v3.bLength >= sizeof(cs->v3) + cs->v3.bNrInPins + + 4 /* bmControls */ + 2 /* wCSelectorDescrStr */; + else + return cs->v2.bLength >= sizeof(cs->v2) + cs->v2.bNrInPins + + 1 /* bmControls */ + 1 /* iClockSelector */; } static bool validate_clock_multiplier(void *p, int id, int proto) { union uac23_clock_multiplier_desc *cs = p; + if (!DESC_LENGTH_CHECK(cs, proto)) + return false; return GET_VAL(cs, proto, bClockID) == id; } -- 2.34.1

2 1

[PATCH OLK-5.10] ALSA: usb-audio: Fix out of bounds reads when finding clock sources
by Luo Gengkun 28 Dec '24

28 Dec '24

From: Takashi Iwai <tiwai(a)suse.de> stable inclusion from stable-v5.10.231 commit 45a92cbc88e4013bfed7fd2ccab3ade45f8e896b category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IBDHGG CVE: CVE-2024-53150 Reference: https://git.kernel.org/stable/c/45a92cbc88e4013bfed7fd2ccab3ade45f8e896b -------------------------------- commit a3dd4d63eeb452cfb064a13862fb376ab108f6a6 upstream. The current USB-audio driver code doesn't check bLength of each descriptor at traversing for clock descriptors. That is, when a device provides a bogus descriptor with a shorter bLength, the driver might hit out-of-bounds reads. For addressing it, this patch adds sanity checks to the validator functions for the clock descriptor traversal. When the descriptor length is shorter than expected, it's skipped in the loop. For the clock source and clock multiplier descriptors, we can just check bLength against the sizeof() of each descriptor type. OTOH, the clock selector descriptor of UAC2 and UAC3 has an array of bNrInPins elements and two more fields at its tail, hence those have to be checked in addition to the sizeof() check. Reported-by: Benoît Sevens <bsevens(a)google.com> Cc: <stable(a)vger.kernel.org> Link: https://lore.kernel.org/20241121140613.3651-1-bsevens@google.com Link: https://patch.msgid.link/20241125144629.20757-1-tiwai@suse.de Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Signed-off-by: Benoît Sevens <bsevens(a)google.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Luo Gengkun <luogengkun2(a)huawei.com> --- sound/usb/clock.c | 32 ++++++++++++++++++++++++++++++-- 1 file changed, 30 insertions(+), 2 deletions(-) diff --git a/sound/usb/clock.c b/sound/usb/clock.c index 95b019f15224..6ccea6d1727f 100644 --- a/sound/usb/clock.c +++ b/sound/usb/clock.c @@ -21,6 +21,10 @@ #include "clock.h" #include "quirks.h" +/* check whether the descriptor bLength has the minimal length */ +#define DESC_LENGTH_CHECK(p) \ + (p->bLength >= sizeof(*p)) + static void *find_uac_clock_desc(struct usb_host_interface *iface, int id, bool (*validator)(void *, int), u8 type) { @@ -38,36 +42,60 @@ static void *find_uac_clock_desc(struct usb_host_interface *iface, int id, static bool validate_clock_source_v2(void *p, int id) { struct uac_clock_source_descriptor *cs = p; + if (!DESC_LENGTH_CHECK(cs)) + return false; return cs->bClockID == id; } static bool validate_clock_source_v3(void *p, int id) { struct uac3_clock_source_descriptor *cs = p; + if (!DESC_LENGTH_CHECK(cs)) + return false; return cs->bClockID == id; } static bool validate_clock_selector_v2(void *p, int id) { struct uac_clock_selector_descriptor *cs = p; - return cs->bClockID == id; + if (!DESC_LENGTH_CHECK(cs)) + return false; + if (cs->bClockID != id) + return false; + /* additional length check for baCSourceID array (in bNrInPins size) + * and two more fields (which sizes depend on the protocol) + */ + return cs->bLength >= sizeof(*cs) + cs->bNrInPins + + 1 /* bmControls */ + 1 /* iClockSelector */; } static bool validate_clock_selector_v3(void *p, int id) { struct uac3_clock_selector_descriptor *cs = p; - return cs->bClockID == id; + if (!DESC_LENGTH_CHECK(cs)) + return false; + if (cs->bClockID != id) + return false; + /* additional length check for baCSourceID array (in bNrInPins size) + * and two more fields (which sizes depend on the protocol) + */ + return cs->bLength >= sizeof(*cs) + cs->bNrInPins + + 4 /* bmControls */ + 2 /* wCSelectorDescrStr */; } static bool validate_clock_multiplier_v2(void *p, int id) { struct uac_clock_multiplier_descriptor *cs = p; + if (!DESC_LENGTH_CHECK(cs)) + return false; return cs->bClockID == id; } static bool validate_clock_multiplier_v3(void *p, int id) { struct uac3_clock_multiplier_descriptor *cs = p; + if (!DESC_LENGTH_CHECK(cs)) + return false; return cs->bClockID == id; } -- 2.34.1

2 1

[openeuler:OLK-5.10 2589/2589] mm/page_counter.o: warning: objtool: page_counter_cancel()+0x84: unreachable instruction
by kernel test robot 28 Dec '24

28 Dec '24

tree: https://gitee.com/openeuler/kernel.git OLK-5.10 head: 053a6b6f8e4c86200cdb20bc80c063c3bb119859 commit: f330d897aed4d1dfde71f7e42f47063c34e775ae [2589/2589] mm: page_counter: mitigate consequences of a page_counter underflow config: x86_64-buildonly-randconfig-002-20241218 (https://download.01.org/0day-ci/archive/20241228/202412280914.VfeVCWRu-lkp@…) compiler: clang version 19.1.3 (https://github.com/llvm/llvm-project ab51eccf88f5321e7c60591c5546b254b6afab99) reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20241228/202412280914.VfeVCWRu-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202412280914.VfeVCWRu-lkp@intel.com/ All warnings (new ones prefixed by >>): >> mm/page_counter.o: warning: objtool: page_counter_cancel()+0x84: unreachable instruction -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[openeuler:openEuler-1.0-LTS 1355/1355] mm/pin_mem.c:1075:12: sparse: sparse: symbol 'reserve_idr' was not declared. Should it be static?
by kernel test robot 28 Dec '24

28 Dec '24

tree: https://gitee.com/openeuler/kernel.git openEuler-1.0-LTS head: a8a9daf97ebc8e303e039402fe784e7bd4e5a9bb commit: 892de178c180d2924ad5f3dcf55baa3b8ab2272d [1355/1355] pid: add pid reserve method for checkpoint and restore config: arm64-randconfig-r133-20241227 (https://download.01.org/0day-ci/archive/20241228/202412280946.xMuoN3LG-lkp@…) compiler: aarch64-linux-gcc (GCC) 14.2.0 reproduce: (https://download.01.org/0day-ci/archive/20241228/202412280946.xMuoN3LG-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202412280946.xMuoN3LG-lkp@intel.com/ sparse warnings: (new ones prefixed by >>) mm/pin_mem.c:34:14: sparse: sparse: symbol 'max_pin_pid_num' was not declared. Should it be static? mm/pin_mem.c:35:15: sparse: sparse: symbol 'redirect_space_size' was not declared. Should it be static? mm/pin_mem.c:171:6: sparse: sparse: symbol 'reserve_page_from_buddy' was not declared. Should it be static? mm/pin_mem.c:254:6: sparse: sparse: symbol 'free_user_map_pages' was not declared. Should it be static? mm/pin_mem.c:311:6: sparse: sparse: symbol 'check_redirect_end_valid' was not declared. Should it be static? mm/pin_mem.c:392:5: sparse: sparse: symbol 'calculate_pin_mem_digest' was not declared. Should it be static? mm/pin_mem.c:481:5: sparse: sparse: symbol 'collect_pmd_huge_pages' was not declared. Should it be static? mm/pin_mem.c:544:5: sparse: sparse: symbol 'collect_normal_pages' was not declared. Should it be static? mm/pin_mem.c:610:6: sparse: sparse: symbol 'free_pin_pages' was not declared. Should it be static? mm/pin_mem.c:770:12: sparse: sparse: symbol 'remap_normal_pages' was not declared. Should it be static? mm/pin_mem.c:857:12: sparse: sparse: symbol 'remap_huge_pmd_pages' was not declared. Should it be static? >> mm/pin_mem.c:1075:12: sparse: sparse: symbol 'reserve_idr' was not declared. Should it be static? mm/pin_mem.c:844:63: sparse: sparse: not addressable mm/pin_mem.c:846:63: sparse: sparse: not addressable mm/pin_mem.c:848:71: sparse: sparse: not addressable mm/pin_mem.c:851:65: sparse: sparse: not addressable mm/pin_mem.c:171:6: warning: no previous prototype for 'reserve_page_from_buddy' [-Wmissing-prototypes] 171 | void reserve_page_from_buddy(unsigned long nr_pages, struct page *page) | ^~~~~~~~~~~~~~~~~~~~~~~ mm/pin_mem.c:254:6: warning: no previous prototype for 'free_user_map_pages' [-Wmissing-prototypes] 254 | void free_user_map_pages(unsigned int pid_index, unsigned int entry_index, unsigned int page_index) | ^~~~~~~~~~~~~~~~~~~ mm/pin_mem.c:311:6: warning: no previous prototype for 'check_redirect_end_valid' [-Wmissing-prototypes] 311 | bool check_redirect_end_valid(struct redirect_info *redirect_start, | ^~~~~~~~~~~~~~~~~~~~~~~~ mm/pin_mem.c:392:5: warning: no previous prototype for 'calculate_pin_mem_digest' [-Wmissing-prototypes] 392 | int calculate_pin_mem_digest(struct pin_mem_dump_info *pmdi, char *digest) | ^~~~~~~~~~~~~~~~~~~~~~~~ mm/pin_mem.c:481:5: warning: no previous prototype for 'collect_pmd_huge_pages' [-Wmissing-prototypes] 481 | int collect_pmd_huge_pages(struct task_struct *task, | ^~~~~~~~~~~~~~~~~~~~~~ mm/pin_mem.c:544:5: warning: no previous prototype for 'collect_normal_pages' [-Wmissing-prototypes] 544 | int collect_normal_pages(struct task_struct *task, | ^~~~~~~~~~~~~~~~~~~~ mm/pin_mem.c: In function 'collect_normal_pages': mm/pin_mem.c:549:26: warning: variable 'nr_pages' set but not used [-Wunused-but-set-variable] 549 | unsigned long i, nr_pages; | ^~~~~~~~ mm/pin_mem.c: At top level: mm/pin_mem.c:610:6: warning: no previous prototype for 'free_pin_pages' [-Wmissing-prototypes] 610 | void free_pin_pages(struct page_map_entry *pme) | ^~~~~~~~~~~~~~ mm/pin_mem.c:770:12: warning: no previous prototype for 'remap_normal_pages' [-Wmissing-prototypes] 770 | vm_fault_t remap_normal_pages(struct mm_struct *mm, struct vm_area_struct *vma, | ^~~~~~~~~~~~~~~~~~ mm/pin_mem.c: In function 'get_hugepage_gfpmask': mm/pin_mem.c:844:63: error: lvalue required as unary '&' operand 844 | if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_DIRECT_FLAG, &transparent_hugepage_flags)) | ^ mm/pin_mem.c:846:63: error: lvalue required as unary '&' operand 846 | if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_FLAG, &transparent_hugepage_flags)) | ^ mm/pin_mem.c:848:71: error: lvalue required as unary '&' operand 848 | if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_OR_MADV_FLAG, &transparent_hugepage_flags)) | ^ mm/pin_mem.c:851:65: error: lvalue required as unary '&' operand 851 | if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG, &transparent_hugepage_flags)) | ^ mm/pin_mem.c: At top level: mm/pin_mem.c:857:12: warning: no previous prototype for 'remap_huge_pmd_pages' [-Wmissing-prototypes] 857 | vm_fault_t remap_huge_pmd_pages(struct mm_struct *mm, struct vm_area_struct *vma, | ^~~~~~~~~~~~~~~~~~~~ vim +/reserve_idr +1075 mm/pin_mem.c 1073 1074 #ifdef CONFIG_PID_RESERVE > 1075 struct idr *reserve_idr; 1076 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[PATCH OLK-6.6] mm: avoid unconditional one-tick sleep when swapcache_prepare fails
by Jinjiang Tu 28 Dec '24

28 Dec '24

From: Barry Song <v-songbaohua(a)oppo.com> mainline inclusion from mainline-v6.12-rc6 commit 01626a18230246efdcea322aa8f067e60ffe5ccd category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IB7294 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Commit 13ddaf26be32 ("mm/swap: fix race when skipping swapcache") introduced an unconditional one-tick sleep when `swapcache_prepare()` fails, which has led to reports of UI stuttering on latency-sensitive Android devices. To address this, we can use a waitqueue to wake up tasks that fail `swapcache_prepare()` sooner, instead of always sleeping for a full tick. While tasks may occasionally be woken by an unrelated `do_swap_page()`, this method is preferable to two scenarios: rapid re-entry into page faults, which can cause livelocks, and multiple millisecond sleeps, which visibly degrade user experience. Oven's testing shows that a single waitqueue resolves the UI stuttering issue. If a 'thundering herd' problem becomes apparent later, a waitqueue hash similar to `folio_wait_table[PAGE_WAIT_TABLE_SIZE]` for page bit locks can be introduced. [v-songbaohua(a)oppo.com: wake_up only when swapcache_wq waitqueue is active] Link: https://lkml.kernel.org/r/20241008130807.40833-1-21cnbao@gmail.com Link: https://lkml.kernel.org/r/20240926211936.75373-1-21cnbao@gmail.com Fixes: 13ddaf26be32 ("mm/swap: fix race when skipping swapcache") Signed-off-by: Barry Song <v-songbaohua(a)oppo.com> Reported-by: Oven Liyang <liyangouwen1(a)oppo.com> Tested-by: Oven Liyang <liyangouwen1(a)oppo.com> Cc: Kairui Song <kasong(a)tencent.com> Cc: "Huang, Ying" <ying.huang(a)intel.com> Cc: Yu Zhao <yuzhao(a)google.com> Cc: David Hildenbrand <david(a)redhat.com> Cc: Chris Li <chrisl(a)kernel.org> Cc: Hugh Dickins <hughd(a)google.com> Cc: Johannes Weiner <hannes(a)cmpxchg.org> Cc: Matthew Wilcox (Oracle) <willy(a)infradead.org> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Minchan Kim <minchan(a)kernel.org> Cc: Yosry Ahmed <yosryahmed(a)google.com> Cc: SeongJae Park <sj(a)kernel.org> Cc: Kalesh Singh <kaleshsingh(a)google.com> Cc: Suren Baghdasaryan <surenb(a)google.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Conflicts: mm/memory.c [Context conflicts.] Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> --- mm/memory.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index badaf096a344..ab52538cf68d 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3963,6 +3963,8 @@ static vm_fault_t handle_pte_marker(struct vm_fault *vmf) return VM_FAULT_SIGBUS; } +static DECLARE_WAIT_QUEUE_HEAD(swapcache_wq); + /* * We enter with non-exclusive mmap_lock (to exclude vma changes, * but allow concurrent faults), and pte mapped but not yet locked. @@ -3975,6 +3977,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) { struct vm_area_struct *vma = vmf->vma; struct folio *swapcache, *folio = NULL; + DECLARE_WAITQUEUE(wait, current); struct page *page; struct swap_info_struct *si = NULL; rmap_t rmap_flags = RMAP_NONE; @@ -4064,7 +4067,9 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) */ if (swapcache_prepare(entry, 1)) { /* Relax a bit to prevent rapid repeated page faults */ + add_wait_queue(&swapcache_wq, &wait); schedule_timeout_uninterruptible(1); + remove_wait_queue(&swapcache_wq, &wait); goto out; } need_clear_cache = true; @@ -4368,8 +4373,11 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) pte_unmap_unlock(vmf->pte, vmf->ptl); out: /* Clear the swap cache pin for direct swapin after PTL unlock */ - if (need_clear_cache) + if (need_clear_cache) { swapcache_clear(si, entry, 1); + if (waitqueue_active(&swapcache_wq)) + wake_up(&swapcache_wq); + } if (si) put_swap_device(si); return ret; @@ -4384,8 +4392,11 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) folio_unlock(swapcache); folio_put(swapcache); } - if (need_clear_cache) + if (need_clear_cache) { swapcache_clear(si, entry, 1); + if (waitqueue_active(&swapcache_wq)) + wake_up(&swapcache_wq); + } if (si) put_swap_device(si); return ret; -- 2.34.1

2 1

[PATCH OLK-5.10] mm: avoid unconditional one-tick sleep when swapcache_prepare fails
by Jinjiang Tu 28 Dec '24

28 Dec '24

From: Barry Song <v-songbaohua(a)oppo.com> mainline inclusion from mainline-v6.12-rc6 commit 01626a18230246efdcea322aa8f067e60ffe5ccd category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IB7294 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Commit 13ddaf26be32 ("mm/swap: fix race when skipping swapcache") introduced an unconditional one-tick sleep when `swapcache_prepare()` fails, which has led to reports of UI stuttering on latency-sensitive Android devices. To address this, we can use a waitqueue to wake up tasks that fail `swapcache_prepare()` sooner, instead of always sleeping for a full tick. While tasks may occasionally be woken by an unrelated `do_swap_page()`, this method is preferable to two scenarios: rapid re-entry into page faults, which can cause livelocks, and multiple millisecond sleeps, which visibly degrade user experience. Oven's testing shows that a single waitqueue resolves the UI stuttering issue. If a 'thundering herd' problem becomes apparent later, a waitqueue hash similar to `folio_wait_table[PAGE_WAIT_TABLE_SIZE]` for page bit locks can be introduced. [v-songbaohua(a)oppo.com: wake_up only when swapcache_wq waitqueue is active] Link: https://lkml.kernel.org/r/20241008130807.40833-1-21cnbao@gmail.com Link: https://lkml.kernel.org/r/20240926211936.75373-1-21cnbao@gmail.com Fixes: 13ddaf26be32 ("mm/swap: fix race when skipping swapcache") Signed-off-by: Barry Song <v-songbaohua(a)oppo.com> Reported-by: Oven Liyang <liyangouwen1(a)oppo.com> Tested-by: Oven Liyang <liyangouwen1(a)oppo.com> Cc: Kairui Song <kasong(a)tencent.com> Cc: "Huang, Ying" <ying.huang(a)intel.com> Cc: Yu Zhao <yuzhao(a)google.com> Cc: David Hildenbrand <david(a)redhat.com> Cc: Chris Li <chrisl(a)kernel.org> Cc: Hugh Dickins <hughd(a)google.com> Cc: Johannes Weiner <hannes(a)cmpxchg.org> Cc: Matthew Wilcox (Oracle) <willy(a)infradead.org> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Minchan Kim <minchan(a)kernel.org> Cc: Yosry Ahmed <yosryahmed(a)google.com> Cc: SeongJae Park <sj(a)kernel.org> Cc: Kalesh Singh <kaleshsingh(a)google.com> Cc: Suren Baghdasaryan <surenb(a)google.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Conflicts: mm/memory.c [Context conflicts.] Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> --- mm/memory.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index 20869d0cd5a1..f580b9c54247 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3397,6 +3397,8 @@ void unmap_mapping_range(struct address_space *mapping, } EXPORT_SYMBOL(unmap_mapping_range); +static DECLARE_WAIT_QUEUE_HEAD(swapcache_wq); + /* * We enter with non-exclusive mmap_lock (to exclude vma changes, * but allow concurrent faults), and pte mapped but not yet locked. @@ -3408,6 +3410,7 @@ EXPORT_SYMBOL(unmap_mapping_range); vm_fault_t do_swap_page(struct vm_fault *vmf) { struct vm_area_struct *vma = vmf->vma; + DECLARE_WAITQUEUE(wait, current); struct page *page = NULL, *swapcache; struct swap_info_struct *si = NULL; bool need_clear_cache = false; @@ -3475,7 +3478,9 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) */ if (swapcache_prepare(entry)) { /* Relax a bit to prevent rapid repeated page faults */ + add_wait_queue(&swapcache_wq, &wait); schedule_timeout_uninterruptible(1); + remove_wait_queue(&swapcache_wq, &wait); goto out; } need_clear_cache = true; @@ -3650,8 +3655,11 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) pte_unmap_unlock(vmf->pte, vmf->ptl); out: /* Clear the swap cache pin for direct swapin after PTL unlock */ - if (need_clear_cache) + if (need_clear_cache) { swapcache_clear(si, entry); + if (waitqueue_active(&swapcache_wq)) + wake_up(&swapcache_wq); + } if (si) put_swap_device(si); return ret; @@ -3665,8 +3673,11 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) unlock_page(swapcache); put_page(swapcache); } - if (need_clear_cache) + if (need_clear_cache) { swapcache_clear(si, entry); + if (waitqueue_active(&swapcache_wq)) + wake_up(&swapcache_wq); + } if (si) put_swap_device(si); return ret; -- 2.34.1

2 1

[PATCH OLK-5.10] mm/khugepaged: fix ->anon_vma race
by Jinjiang Tu 28 Dec '24

28 Dec '24

From: Jann Horn <jannh(a)google.com> mainline inclusion from mainline-v6.2-rc7 commit 023f47a8250c6bdb4aebe744db4bf7f73414028b category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IBE8BO Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- If an ->anon_vma is attached to the VMA, collapse_and_free_pmd() requires it to be locked. Page table traversal is allowed under any one of the mmap lock, the anon_vma lock (if the VMA is associated with an anon_vma), and the mapping lock (if the VMA is associated with a mapping); and so to be able to remove page tables, we must hold all three of them. retract_page_tables() bails out if an ->anon_vma is attached, but does this check before holding the mmap lock (as the comment above the check explains). If we racily merged an existing ->anon_vma (shared with a child process) from a neighboring VMA, subsequent rmap traversals on pages belonging to the child will be able to see the page tables that we are concurrently removing while assuming that nothing else can access them. Repeat the ->anon_vma check once we hold the mmap lock to ensure that there really is no concurrent page table access. Hitting this bug causes a lockdep warning in collapse_and_free_pmd(), in the line "lockdep_assert_held_write(&vma->anon_vma->root->rwsem)". It can also lead to use-after-free access. Link: https://lore.kernel.org/linux-mm/CAG48ez3434wZBKFFbdx4M9j6eUwSUVPd4dxhzW_k_… Link: https://lkml.kernel.org/r/20230111133351.807024-1-jannh@google.com Fixes: f3f0e1d2150b ("khugepaged: add support of collapse for tmpfs/shmem pages") Signed-off-by: Jann Horn <jannh(a)google.com> Reported-by: Zach O'Keefe <zokeefe(a)google.com> Acked-by: Kirill A. Shutemov <kirill.shutemov(a)intel.linux.com> Reviewed-by: Yang Shi <shy828301(a)gmail.com> Cc: David Hildenbrand <david(a)redhat.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Conflicts: mm/khugepaged.c [Conflicts due to 34488399fa08 ("mm/madvise: add file and shmem support to MADV_COLLAPSE") is not merged.] Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> --- mm/khugepaged.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/mm/khugepaged.c b/mm/khugepaged.c index 1d5a985ae4cd..a3f45bca187b 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -1623,7 +1623,7 @@ static void retract_page_tables(struct address_space *mapping, pgoff_t pgoff) * has higher cost too. It would also probably require locking * the anon_vma. */ - if (vma->anon_vma) + if (READ_ONCE(vma->anon_vma)) continue; addr = vma->vm_start + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT); if (addr & ~HPAGE_PMD_MASK) @@ -1642,6 +1642,18 @@ static void retract_page_tables(struct address_space *mapping, pgoff_t pgoff) * reverse order. Trylock is a way to avoid deadlock. */ if (mmap_write_trylock(mm)) { + /* + * Re-check whether we have an ->anon_vma, because + * collapse_and_free_pmd() requires that either no + * ->anon_vma exists or the anon_vma is locked. + * We already checked ->anon_vma above, but that check + * is racy because ->anon_vma can be populated under the + * mmap lock in read mode. + */ + if (vma->anon_vma) { + mmap_write_unlock(mm); + continue; + } if (!khugepaged_test_exit(mm)) { struct mmu_notifier_range range; -- 2.34.1

2 1