- Kernel - mailweb.openeuler.org

[OLK-5.10] rndis_wlan: Prevent buffer overflow in rndis_query_oid
by Wang Yufen 17 Jan '23

17 Jan '23

From: Szymon Heidrich <szymon.heidrich(a)gmail.com> maillist inclusion category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I6AQJP CVE: CVE-2023-23559 Reference: https://patchwork.kernel.org/project/linux-wireless/patch/20230111175031.70… ------------------------------- Since resplen and respoffs are signed integers sufficiently large values of unsigned int len and offset members of RNDIS response will result in negative values of prior variables. This may be utilized to bypass implemented security checks to either extract memory contents by manipulating offset or overflow the data buffer via memcpy by manipulating both offset and len. Additionally assure that sum of resplen and respoffs does not overflow so buffer boundaries are kept. Fixes: 80f8c5b434f9 ("rndis_wlan: copy only useful data from rndis_command respond") Signed-off-by: Szymon Heidrich <szymon.heidrich(a)gmail.com> Signed-off-by: Wang Yufen <wangyufen(a)huawei.com> --- drivers/net/wireless/rndis_wlan.c | 19 ++++++------------- 1 file changed, 6 insertions(+), 13 deletions(-) diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c index 75b5d54..dc076d8 100644 --- a/drivers/net/wireless/rndis_wlan.c +++ b/drivers/net/wireless/rndis_wlan.c @@ -694,8 +694,8 @@ static int rndis_query_oid(struct usbnet *dev, u32 oid, void *data, int *len) struct rndis_query *get; struct rndis_query_c *get_c; } u; - int ret, buflen; - int resplen, respoffs, copylen; + int ret; + size_t buflen, resplen, respoffs, copylen; buflen = *len + sizeof(*u.get); if (buflen < CONTROL_BUFFER_SIZE) @@ -730,22 +730,15 @@ static int rndis_query_oid(struct usbnet *dev, u32 oid, void *data, int *len) if (respoffs > buflen) { /* Device returned data offset outside buffer, error. */ - netdev_dbg(dev->net, "%s(%s): received invalid " - "data offset: %d > %d\n", __func__, - oid_to_string(oid), respoffs, buflen); + netdev_dbg(dev->net, + "%s(%s): received invalid data offset: %zu > %zu\n", + __func__, oid_to_string(oid), respoffs, buflen); ret = -EINVAL; goto exit_unlock; } - if ((resplen + respoffs) > buflen) { - /* Device would have returned more data if buffer would - * have been big enough. Copy just the bits that we got. - */ - copylen = buflen - respoffs; - } else { - copylen = resplen; - } + copylen = min(resplen, buflen - respoffs); if (copylen > *len) copylen = *len; -- 1.8.3.1

1 0

[openEuler-1.0-LTS] rndis_wlan: Prevent buffer overflow in rndis_query_oid
by Wang Yufen 17 Jan '23

17 Jan '23

From: Szymon Heidrich <szymon.heidrich(a)gmail.com> maillist inclusion category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I6AQJP CVE: CVE-2023-23559 Reference: https://patchwork.kernel.org/project/linux-wireless/patch/20230111175031.70… ------------------------------- Since resplen and respoffs are signed integers sufficiently large values of unsigned int len and offset members of RNDIS response will result in negative values of prior variables. This may be utilized to bypass implemented security checks to either extract memory contents by manipulating offset or overflow the data buffer via memcpy by manipulating both offset and len. Additionally assure that sum of resplen and respoffs does not overflow so buffer boundaries are kept. Fixes: 80f8c5b434f9 ("rndis_wlan: copy only useful data from rndis_command respond") Signed-off-by: Szymon Heidrich <szymon.heidrich(a)gmail.com> Signed-off-by: Wang Yufen <wangyufen(a)huawei.com> --- drivers/net/wireless/rndis_wlan.c | 19 ++++++------------- 1 file changed, 6 insertions(+), 13 deletions(-) diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c index 51e4e92..0bbeb61 100644 --- a/drivers/net/wireless/rndis_wlan.c +++ b/drivers/net/wireless/rndis_wlan.c @@ -712,8 +712,8 @@ static int rndis_query_oid(struct usbnet *dev, u32 oid, void *data, int *len) struct rndis_query *get; struct rndis_query_c *get_c; } u; - int ret, buflen; - int resplen, respoffs, copylen; + int ret; + size_t buflen, resplen, respoffs, copylen; buflen = *len + sizeof(*u.get); if (buflen < CONTROL_BUFFER_SIZE) @@ -748,22 +748,15 @@ static int rndis_query_oid(struct usbnet *dev, u32 oid, void *data, int *len) if (respoffs > buflen) { /* Device returned data offset outside buffer, error. */ - netdev_dbg(dev->net, "%s(%s): received invalid " - "data offset: %d > %d\n", __func__, - oid_to_string(oid), respoffs, buflen); + netdev_dbg(dev->net, + "%s(%s): received invalid data offset: %zu > %zu\n", + __func__, oid_to_string(oid), respoffs, buflen); ret = -EINVAL; goto exit_unlock; } - if ((resplen + respoffs) > buflen) { - /* Device would have returned more data if buffer would - * have been big enough. Copy just the bits that we got. - */ - copylen = buflen - respoffs; - } else { - copylen = resplen; - } + copylen = min(resplen, buflen - respoffs); if (copylen > *len) copylen = *len; -- 1.8.3.1

1 0

[PATCH] rndis_wlan: Prevent buffer overflow in rndis_query_oid
by Wang Yufen 17 Jan '23

17 Jan '23

From: Szymon Heidrich <szymon.heidrich(a)gmail.com> maillist inclusion category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I6AQJP CVE: CVE-2023-23559 Reference: https://patchwork.kernel.org/project/linux-wireless/patch/20230111175031.70… ------------------------------- Since resplen and respoffs are signed integers sufficiently large values of unsigned int len and offset members of RNDIS response will result in negative values of prior variables. This may be utilized to bypass implemented security checks to either extract memory contents by manipulating offset or overflow the data buffer via memcpy by manipulating both offset and len. Additionally assure that sum of resplen and respoffs does not overflow so buffer boundaries are kept. Fixes: 80f8c5b434f9 ("rndis_wlan: copy only useful data from rndis_command respond") Signed-off-by: Szymon Heidrich <szymon.heidrich(a)gmail.com> Signed-off-by: Wang Yufen <wangyufen(a)huawei.com> --- drivers/net/wireless/rndis_wlan.c | 19 ++++++------------- 1 file changed, 6 insertions(+), 13 deletions(-) diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c index 51e4e92..0bbeb61 100644 --- a/drivers/net/wireless/rndis_wlan.c +++ b/drivers/net/wireless/rndis_wlan.c @@ -712,8 +712,8 @@ static int rndis_query_oid(struct usbnet *dev, u32 oid, void *data, int *len) struct rndis_query *get; struct rndis_query_c *get_c; } u; - int ret, buflen; - int resplen, respoffs, copylen; + int ret; + size_t buflen, resplen, respoffs, copylen; buflen = *len + sizeof(*u.get); if (buflen < CONTROL_BUFFER_SIZE) @@ -748,22 +748,15 @@ static int rndis_query_oid(struct usbnet *dev, u32 oid, void *data, int *len) if (respoffs > buflen) { /* Device returned data offset outside buffer, error. */ - netdev_dbg(dev->net, "%s(%s): received invalid " - "data offset: %d > %d\n", __func__, - oid_to_string(oid), respoffs, buflen); + netdev_dbg(dev->net, + "%s(%s): received invalid data offset: %zu > %zu\n", + __func__, oid_to_string(oid), respoffs, buflen); ret = -EINVAL; goto exit_unlock; } - if ((resplen + respoffs) > buflen) { - /* Device would have returned more data if buffer would - * have been big enough. Copy just the bits that we got. - */ - copylen = buflen - respoffs; - } else { - copylen = resplen; - } + copylen = min(resplen, buflen - respoffs); if (copylen > *len) copylen = *len; -- 1.8.3.1

1 0

[PATCH openEuler-22.09 0/2]fix two review issue for memcpy_mcs()
by Tong Tiangen 16 Jan '23

16 Jan '23

Tong Tiangen (2): arm64: fix return value type of memcpy_mcs() kasan: add kasan support for memcpy_mcs() arch/arm64/include/asm/string.h | 4 ++-- mm/kasan/common.c | 12 ++++++++++++ 2 files changed, 14 insertions(+), 2 deletions(-) -- 2.25.1

1 2

[PATCH OLK-5.10 0/2]fix two review issue for memcpy_mcs()
by Tong Tiangen 16 Jan '23

16 Jan '23

Tong Tiangen (2): arm64: fix return value type of memcpy_mcs() kasan: add kasan support for memcpy_mcs() arch/arm64/include/asm/string.h | 4 ++-- mm/kasan/common.c | 12 ++++++++++++ 2 files changed, 14 insertions(+), 2 deletions(-) -- 2.25.1

1 2

[PATCH openEuler-22.03-LTS] selftests/rseq: check if libc rseq support is registered
by MinMin Ren 12 Jan '23

12 Jan '23

From: Michael Jeanson <mjeanson(a)efficios.com> mainline inclusion from mainline-v6.0-rc1 commit d1a997ba4c1bf65497d956aea90de42a6398f73a category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I69PT8 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit d1a997ba4c1bf65497d956aea90de42a6398f73a upstream. When checking for libc rseq support in the library constructor, don't only depend on the symbols presence, check that the registration was completed. This targets a scenario where the libc has rseq support but it is not wired for the current architecture in 'bits/rseq.h', we want to fallback to our internal registration mechanism. Signed-off-by: Michael Jeanson <mjeanson(a)efficios.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers(a)efficios.com> Link: https://lore.kernel.org/r/20220614154830.1367382-4-mjeanson@efficios.com Signed-off-by: rminmin <rmm0811(a)gmail.com> --- tools/testing/selftests/rseq/rseq.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/rseq/rseq.c b/tools/testing/selftests/rseq/rseq.c index 986b9458efb2..4177f9507bbe 100644 --- a/tools/testing/selftests/rseq/rseq.c +++ b/tools/testing/selftests/rseq/rseq.c @@ -111,7 +111,8 @@ void rseq_init(void) libc_rseq_offset_p = dlsym(RTLD_NEXT, "__rseq_offset"); libc_rseq_size_p = dlsym(RTLD_NEXT, "__rseq_size"); libc_rseq_flags_p = dlsym(RTLD_NEXT, "__rseq_flags"); - if (libc_rseq_size_p && libc_rseq_offset_p && libc_rseq_flags_p) { + if (libc_rseq_size_p && libc_rseq_offset_p && libc_rseq_flags_p && + *libc_rseq_size_p != 0) { /* rseq registration owned by glibc */ rseq_offset = *libc_rseq_offset_p; rseq_size = *libc_rseq_size_p; -- 2.33.0

1 0

[PATCH openEuler-1.0-LTS 1/6] crypto: ecdh - avoid unaligned accesses in ecdh_set_secret()
by Yongqiang Liu 11 Jan '23

11 Jan '23

From: Ard Biesheuvel <ardb(a)kernel.org> stable inclusion from stable-v4.19.164 commit 85637bc064f4fd5539d4173798d8cb55dde7fc0a category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I65ZD2 CVE: NA -------------------------------- commit 17858b140bf49961b71d4e73f1c3ea9bc8e7dda0 upstream. ecdh_set_secret() casts a void* pointer to a const u64* in order to feed it into ecc_is_key_valid(). This is not generally permitted by the C standard, and leads to actual misalignment faults on ARMv6 cores. In some cases, these are fixed up in software, but this still leads to performance hits that are entirely avoidable. So let's copy the key into the ctx buffer first, which we will do anyway in the common case, and which guarantees correct alignment. Cc: <stable(a)vger.kernel.org> Signed-off-by: Ard Biesheuvel <ardb(a)kernel.org> Signed-off-by: Herbert Xu <herbert(a)gondor.apana.org.au> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: GUO Zihua <guozihua(a)huawei.com> Reviewed-by: Wang Weiyang <wangweiyang2(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- crypto/ecdh.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/crypto/ecdh.c b/crypto/ecdh.c index bf6300175b9c..a6e1a5d43fa7 100644 --- a/crypto/ecdh.c +++ b/crypto/ecdh.c @@ -57,12 +57,13 @@ static int ecdh_set_secret(struct crypto_kpp *tfm, const void *buf, return ecc_gen_privkey(ctx->curve_id, ctx->ndigits, ctx->private_key); - if (ecc_is_key_valid(ctx->curve_id, ctx->ndigits, - (const u64 *)params.key, params.key_size) < 0) - return -EINVAL; - memcpy(ctx->private_key, params.key, params.key_size); + if (ecc_is_key_valid(ctx->curve_id, ctx->ndigits, + ctx->private_key, params.key_size) < 0) { + memzero_explicit(ctx->private_key, params.key_size); + return -EINVAL; + } return 0; } -- 2.25.1

1 5

[PATCH openEuler-1.0-LTS 01/13] audit: fix undefined behavior in bit shift for AUDIT_BIT
by Yongqiang Liu 11 Jan '23

11 Jan '23

From: Gaosheng Cui <cuigaosheng1(a)huawei.com> stable inclusion from stable-v4.19.268 commit bbd0427fc0241bccb2c56235cbaad9fd63cd615b category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I69KBO CVE: NA -------------------------------- [ Upstream commit 986d93f55bdeab1cac858d1e47b41fac10b2d7f6 ] Shifting signed 32-bit value by 31 bits is undefined, so changing significant bit to unsigned. The UBSAN warning calltrace like below: UBSAN: shift-out-of-bounds in kernel/auditfilter.c:179:23 left shift of 1 by 31 places cannot be represented in type 'int' Call Trace: <TASK> dump_stack_lvl+0x7d/0xa5 dump_stack+0x15/0x1b ubsan_epilogue+0xe/0x4e __ubsan_handle_shift_out_of_bounds+0x1e7/0x20c audit_register_class+0x9d/0x137 audit_classes_init+0x4d/0xb8 do_one_initcall+0x76/0x430 kernel_init_freeable+0x3b3/0x422 kernel_init+0x24/0x1e0 ret_from_fork+0x1f/0x30 </TASK> Signed-off-by: Gaosheng Cui <cuigaosheng1(a)huawei.com> [PM: remove bad 'Fixes' tag as issue predates git, added in v2.6.6-rc1] Signed-off-by: Paul Moore <paul(a)paul-moore.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- include/uapi/linux/audit.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/uapi/linux/audit.h b/include/uapi/linux/audit.h index 818ae690ab79..b163911b1d39 100644 --- a/include/uapi/linux/audit.h +++ b/include/uapi/linux/audit.h @@ -177,7 +177,7 @@ #define AUDIT_MAX_KEY_LEN 256 #define AUDIT_BITMASK_SIZE 64 #define AUDIT_WORD(nr) ((__u32)((nr)/32)) -#define AUDIT_BIT(nr) (1 << ((nr) - AUDIT_WORD(nr)*32)) +#define AUDIT_BIT(nr) (1U << ((nr) - AUDIT_WORD(nr)*32)) #define AUDIT_SYSCALL_CLASSES 16 #define AUDIT_CLASS_DIR_WRITE 0 -- 2.25.1

1 12

[PATCH openEuler-1.0-LTS 1/2] USB: core: Prevent nested device-reset calls
by Yongqiang Liu 11 Jan '23

11 Jan '23

From: Alan Stern <stern(a)rowland.harvard.edu> stable inclusion from stable-v4.19.258 commit cc9a12e12808af178c600cc485338bac2e37d2a8 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I675RE CVE: CVE-2022-4662 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 9c6d778800b921bde3bff3cff5003d1650f942d1 upstream. Automatic kernel fuzzing revealed a recursive locking violation in usb-storage: ============================================ WARNING: possible recursive locking detected 5.18.0 #3 Not tainted -------------------------------------------- kworker/1:3/1205 is trying to acquire lock: ffff888018638db8 (&us_interface_key[i]){+.+.}-{3:3}, at: usb_stor_pre_reset+0x35/0x40 drivers/usb/storage/usb.c:230 but task is already holding lock: ffff888018638db8 (&us_interface_key[i]){+.+.}-{3:3}, at: usb_stor_pre_reset+0x35/0x40 drivers/usb/storage/usb.c:230 ... stack backtrace: CPU: 1 PID: 1205 Comm: kworker/1:3 Not tainted 5.18.0 #3 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014 Workqueue: usb_hub_wq hub_event Call Trace: <TASK> __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106 print_deadlock_bug kernel/locking/lockdep.c:2988 [inline] check_deadlock kernel/locking/lockdep.c:3031 [inline] validate_chain kernel/locking/lockdep.c:3816 [inline] __lock_acquire.cold+0x152/0x3ca kernel/locking/lockdep.c:5053 lock_acquire kernel/locking/lockdep.c:5665 [inline] lock_acquire+0x1ab/0x520 kernel/locking/lockdep.c:5630 __mutex_lock_common kernel/locking/mutex.c:603 [inline] __mutex_lock+0x14f/0x1610 kernel/locking/mutex.c:747 usb_stor_pre_reset+0x35/0x40 drivers/usb/storage/usb.c:230 usb_reset_device+0x37d/0x9a0 drivers/usb/core/hub.c:6109 r871xu_dev_remove+0x21a/0x270 drivers/staging/rtl8712/usb_intf.c:622 usb_unbind_interface+0x1bd/0x890 drivers/usb/core/driver.c:458 device_remove drivers/base/dd.c:545 [inline] device_remove+0x11f/0x170 drivers/base/dd.c:537 __device_release_driver drivers/base/dd.c:1222 [inline] device_release_driver_internal+0x1a7/0x2f0 drivers/base/dd.c:1248 usb_driver_release_interface+0x102/0x180 drivers/usb/core/driver.c:627 usb_forced_unbind_intf+0x4d/0xa0 drivers/usb/core/driver.c:1118 usb_reset_device+0x39b/0x9a0 drivers/usb/core/hub.c:6114 This turned out not to be an error in usb-storage but rather a nested device reset attempt. That is, as the rtl8712 driver was being unbound from a composite device in preparation for an unrelated USB reset (that driver does not have pre_reset or post_reset callbacks), its ->remove routine called usb_reset_device() -- thus nesting one reset call within another. Performing a reset as part of disconnect processing is a questionable practice at best. However, the bug report points out that the USB core does not have any protection against nested resets. Adding a reset_in_progress flag and testing it will prevent such errors in the future. Link: https://lore.kernel.org/all/CAB7eexKUpvX-JNiLzhXBDWgfg2T9e9_0Tw4HQ6keN==voR… Cc: stable(a)vger.kernel.org Reported-and-tested-by: Rondreis <linhaoguo86(a)gmail.com> Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu> Link: https://lore.kernel.org/r/YwkflDxvg0KWqyZK@rowland.harvard.edu Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yuyao Lin <linyuyao1(a)huawei.com> Reviewed-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> Reviewed-by: Xiu Jianfeng <xiujianfeng(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- drivers/usb/core/hub.c | 10 ++++++++++ include/linux/usb.h | 2 ++ 2 files changed, 12 insertions(+) diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index adc8e3a0138f..b983a333ca32 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -5747,6 +5747,11 @@ static int usb_reset_and_verify_device(struct usb_device *udev) * the reset is over (using their post_reset method). * * Return: The same as for usb_reset_and_verify_device(). + * However, if a reset is already in progress (for instance, if a + * driver doesn't have pre_ or post_reset() callbacks, and while + * being unbound or re-bound during the ongoing reset its disconnect() + * or probe() routine tries to perform a second, nested reset), the + * routine returns -EINPROGRESS. * * Note: * The caller must own the device lock. For example, it's safe to use @@ -5780,6 +5785,10 @@ int usb_reset_device(struct usb_device *udev) return -EISDIR; } + if (udev->reset_in_progress) + return -EINPROGRESS; + udev->reset_in_progress = 1; + port_dev = hub->ports[udev->portnum - 1]; /* @@ -5844,6 +5853,7 @@ int usb_reset_device(struct usb_device *udev) usb_autosuspend_device(udev); memalloc_noio_restore(noio_flag); + udev->reset_in_progress = 0; return ret; } EXPORT_SYMBOL_GPL(usb_reset_device); diff --git a/include/linux/usb.h b/include/linux/usb.h index 3e1e31f73d0e..317eedae1c10 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -580,6 +580,7 @@ struct usb3_lpm_parameters { * @level: number of USB hub ancestors * @can_submit: URBs may be submitted * @persist_enabled: USB_PERSIST enabled for this device + * @reset_in_progress: the device is being reset * @have_langid: whether string_langid is valid * @authorized: policy has said we can use it; * (user space) policy determines if we authorize this device to be @@ -664,6 +665,7 @@ struct usb_device { unsigned can_submit:1; unsigned persist_enabled:1; + unsigned reset_in_progress:1; unsigned have_langid:1; unsigned authorized:1; unsigned authenticated:1; -- 2.25.1

1 1

[PATCH openEuler-1.0-LTS 1/8] mm/hwpoison: do not lock page again when me_huge_page() successfully recovers
by Yongqiang Liu 11 Jan '23

11 Jan '23

From: Naoya Horiguchi <naoya.horiguchi(a)nec.com> mainline inclusion from mainline-v5.13 commit ea6d0630100b285f059d0a8d8e86f38a46407536 category: bugfix bugzilla: 188222, https://gitee.com/openeuler/kernel/issues/I69SHF CVE: NA -------------------------------- Currently me_huge_page() temporary unlocks page to perform some actions then locks it again later. My testcase (which calls hard-offline on some tail page in a hugetlb, then accesses the address of the hugetlb range) showed that page allocation code detects this page lock on buddy page and printed out "BUG: Bad page state" message. check_new_page_bad() does not consider a page with __PG_HWPOISON as bad page, so this flag works as kind of filter, but this filtering doesn't work in this case because the "bad page" is not the actual hwpoisoned page. So stop locking page again. Actions to be taken depend on the page type of the error, so page unlocking should be done in ->action() callbacks. So let's make it assumed and change all existing callbacks that way. Link: https://lkml.kernel.org/r/20210609072029.74645-1-nao.horiguchi@gmail.com Fixes: commit 78bb920344b8 ("mm: hwpoison: dissolve in-use hugepage in unrecoverable memory error") Signed-off-by: Naoya Horiguchi <naoya.horiguchi(a)nec.com> Cc: Oscar Salvador <osalvador(a)suse.de> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Tony Luck <tony.luck(a)intel.com> Cc: "Aneesh Kumar K.V" <aneesh.kumar(a)linux.vnet.ibm.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Ma Wupeng <mawupeng1(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- mm/memory-failure.c | 44 ++++++++++++++++++++++++++++++-------------- 1 file changed, 30 insertions(+), 14 deletions(-) diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 55c175f57223..9a816fdf812d 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -662,6 +662,7 @@ static int truncate_error_page(struct page *p, unsigned long pfn, */ static int me_kernel(struct page *p, unsigned long pfn) { + unlock_page(p); return MF_IGNORED; } @@ -671,6 +672,7 @@ static int me_kernel(struct page *p, unsigned long pfn) static int me_unknown(struct page *p, unsigned long pfn) { pr_err("Memory failure: %#lx: Unknown page state\n", pfn); + unlock_page(p); return MF_FAILED; } @@ -679,6 +681,7 @@ static int me_unknown(struct page *p, unsigned long pfn) */ static int me_pagecache_clean(struct page *p, unsigned long pfn) { + int ret; struct address_space *mapping; delete_from_lru_cache(p); @@ -687,8 +690,10 @@ static int me_pagecache_clean(struct page *p, unsigned long pfn) * For anonymous pages we're done the only reference left * should be the one m_f() holds. */ - if (PageAnon(p)) - return MF_RECOVERED; + if (PageAnon(p)) { + ret = MF_RECOVERED; + goto out; + } /* * Now truncate the page in the page cache. This is really @@ -702,7 +707,8 @@ static int me_pagecache_clean(struct page *p, unsigned long pfn) /* * Page has been teared down in the meanwhile */ - return MF_FAILED; + ret = MF_FAILED; + goto out; } /* @@ -710,7 +716,10 @@ static int me_pagecache_clean(struct page *p, unsigned long pfn) * * Open: to take i_mutex or not for this? Right now we don't. */ - return truncate_error_page(p, pfn, mapping); + ret = truncate_error_page(p, pfn, mapping); +out: + unlock_page(p); + return ret; } /* @@ -786,24 +795,26 @@ static int me_pagecache_dirty(struct page *p, unsigned long pfn) */ static int me_swapcache_dirty(struct page *p, unsigned long pfn) { + int ret; + ClearPageDirty(p); /* Trigger EIO in shmem: */ ClearPageUptodate(p); - if (!delete_from_lru_cache(p)) - return MF_DELAYED; - else - return MF_FAILED; + ret = delete_from_lru_cache(p) ? MF_FAILED : MF_DELAYED; + unlock_page(p); + return ret; } static int me_swapcache_clean(struct page *p, unsigned long pfn) { + int ret; + delete_from_swap_cache(p); - if (!delete_from_lru_cache(p)) - return MF_RECOVERED; - else - return MF_FAILED; + ret = delete_from_lru_cache(p) ? MF_FAILED : MF_RECOVERED; + unlock_page(p); + return ret; } /* @@ -824,6 +835,7 @@ static int me_huge_page(struct page *p, unsigned long pfn) mapping = page_mapping(hpage); if (mapping) { res = truncate_error_page(hpage, pfn, mapping); + unlock_page(hpage); } else { res = MF_FAILED; unlock_page(hpage); @@ -838,7 +850,6 @@ static int me_huge_page(struct page *p, unsigned long pfn) page_ref_inc(p); res = MF_RECOVERED; } - lock_page(hpage); } return res; @@ -871,6 +882,8 @@ static struct page_state { unsigned long mask; unsigned long res; enum mf_action_page_type type; + + /* Callback ->action() has to unlock the relevant page inside it. */ int (*action)(struct page *p, unsigned long pfn); } error_states[] = { { reserved, reserved, MF_MSG_KERNEL, me_kernel }, @@ -935,6 +948,7 @@ static int page_action(struct page_state *ps, struct page *p, int result; int count; + /* page p should be unlocked after returning from ps->action(). */ result = ps->action(p, pfn); count = page_count(p) - 1; @@ -1235,7 +1249,7 @@ static int memory_failure_hugetlb(unsigned long pfn, int flags) goto out; } - res = identify_page_state(pfn, p, page_flags); + return identify_page_state(pfn, p, page_flags); out: unlock_page(head); return res; @@ -1533,6 +1547,8 @@ int memory_failure(unsigned long pfn, int flags) identify_page_state: res = identify_page_state(pfn, p, page_flags); + mutex_unlock(&mf_mutex); + return res; unlock_page: unlock_page(p); unlock_mutex: -- 2.25.1

1 7