August 2024 - Kernel - mailweb.openeuler.org

[PATCH openEuler-22.03-LTS-SP1] drm/amd/display: Check index msg_id before read or write
by Li Nan 18 Aug '24

18 Aug '24

From: Alex Hung <alex.hung(a)amd.com> stable inclusion from stable-v5.10.222 commit b5b8837d066cc182ff69fb5629ad32ade5484567 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAGSSE CVE: CVE-2024-42121 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 59d99deb330af206a4541db0c4da8f73880fba03 ] [WHAT] msg_id is used as an array index and it cannot be a negative value, and therefore cannot be equal to MOD_HDCP_MESSAGE_ID_INVALID (-1). [HOW] Check whether msg_id is valid before reading and setting. This fixes 4 OVERRUN issues reported by Coverity. Reviewed-by: Rodrigo Siqueira <rodrigo.siqueira(a)amd.com> Acked-by: Wayne Lin <wayne.lin(a)amd.com> Signed-off-by: Alex Hung <alex.hung(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Li Nan <linan122(a)huawei.com> --- drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c index f7b5583ee609..8e9caae7c955 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c @@ -156,6 +156,10 @@ static enum mod_hdcp_status read(struct mod_hdcp *hdcp, uint32_t cur_size = 0; uint32_t data_offset = 0; + if (msg_id == MOD_HDCP_MESSAGE_ID_INVALID) { + return MOD_HDCP_STATUS_DDC_FAILURE; + } + if (is_dp_hdcp(hdcp)) { while (buf_len > 0) { cur_size = MIN(buf_len, HDCP_MAX_AUX_TRANSACTION_SIZE); @@ -215,6 +219,10 @@ static enum mod_hdcp_status write(struct mod_hdcp *hdcp, uint32_t cur_size = 0; uint32_t data_offset = 0; + if (msg_id == MOD_HDCP_MESSAGE_ID_INVALID) { + return MOD_HDCP_STATUS_DDC_FAILURE; + } + if (is_dp_hdcp(hdcp)) { while (buf_len > 0) { cur_size = MIN(buf_len, HDCP_MAX_AUX_TRANSACTION_SIZE); -- 2.39.2

2 1

[PATCH OLK-5.10] drm/amd/display: Check index msg_id before read or write
by Li Nan 18 Aug '24

18 Aug '24

From: Alex Hung <alex.hung(a)amd.com> stable inclusion from stable-v5.10.222 commit b5b8837d066cc182ff69fb5629ad32ade5484567 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAGSSE CVE: CVE-2024-42121 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 59d99deb330af206a4541db0c4da8f73880fba03 ] [WHAT] msg_id is used as an array index and it cannot be a negative value, and therefore cannot be equal to MOD_HDCP_MESSAGE_ID_INVALID (-1). [HOW] Check whether msg_id is valid before reading and setting. This fixes 4 OVERRUN issues reported by Coverity. Reviewed-by: Rodrigo Siqueira <rodrigo.siqueira(a)amd.com> Acked-by: Wayne Lin <wayne.lin(a)amd.com> Signed-off-by: Alex Hung <alex.hung(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Li Nan <linan122(a)huawei.com> --- drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c index f7b5583ee609..8e9caae7c955 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c @@ -156,6 +156,10 @@ static enum mod_hdcp_status read(struct mod_hdcp *hdcp, uint32_t cur_size = 0; uint32_t data_offset = 0; + if (msg_id == MOD_HDCP_MESSAGE_ID_INVALID) { + return MOD_HDCP_STATUS_DDC_FAILURE; + } + if (is_dp_hdcp(hdcp)) { while (buf_len > 0) { cur_size = MIN(buf_len, HDCP_MAX_AUX_TRANSACTION_SIZE); @@ -215,6 +219,10 @@ static enum mod_hdcp_status write(struct mod_hdcp *hdcp, uint32_t cur_size = 0; uint32_t data_offset = 0; + if (msg_id == MOD_HDCP_MESSAGE_ID_INVALID) { + return MOD_HDCP_STATUS_DDC_FAILURE; + } + if (is_dp_hdcp(hdcp)) { while (buf_len > 0) { cur_size = MIN(buf_len, HDCP_MAX_AUX_TRANSACTION_SIZE); -- 2.39.2

2 1

[openEuler-22.03-LTS-SP1] Kbuild: fix compiling external mods without any warning/error when there is no libelf library
by zhaoxuedong 18 Aug '24

18 Aug '24

openEuler inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IAD5GG?from=project-issue CVE: NA ----------------------------------------------------------------------------------- When your development environment does not have the libelf library and you compile an external module with the "M=xxx" compilation option, the external module will not have ORC information and there will be no warning or errors. If a machine insmod this module, and if a process is calling a function in this module, typing kaptch will result in an error indicating that the stack of this process is not trustworthy. The worst thing is that if this module is a necessary and will be used for a long time, kpatch will be difficult to type. Revert "kbuild: fix false positive warning/error about missing libelf" This reverts commit ef7cfd00b2caf6edeb7f169682b64be2d0a798cf. Fixes: ef7cfd00b2caf6edeb ("kbuild: fix false positive warning/error about missing libelf") Reported-by: wanglong <wanglong19(a)meituan.com> Reported-by: wufan <wufan19(a)meituan.com> Reported-by: zhaoxuedong <zhaoxuedong(a)meituan.com> --- Makefile | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/Makefile b/Makefile index fbaeeb80a69e..82dcf846d07b 100644 --- a/Makefile +++ b/Makefile @@ -1100,6 +1100,11 @@ ifdef CONFIG_STACK_VALIDATION ifeq ($(has_libelf),1) objtool_target := tools/objtool FORCE else + ifdef CONFIG_UNWINDER_ORC + $(error "Cannot generate ORC metadata for CONFIG_UNWINDER_ORC=y, please install libelf-dev, libelf-devel or elfutils-libelf-devel") + else + $(warning "Cannot use CONFIG_STACK_VALIDATION=y, please install libelf-dev, libelf-devel or elfutils-libelf-devel") + endif SKIP_STACK_VALIDATION := 1 export SKIP_STACK_VALIDATION endif @@ -1249,14 +1254,6 @@ uapi-asm-generic: PHONY += prepare-objtool prepare-resolve_btfids prepare-objtool: $(objtool_target) -ifeq ($(SKIP_STACK_VALIDATION),1) -ifdef CONFIG_UNWINDER_ORC - @echo "error: Cannot generate ORC metadata for CONFIG_UNWINDER_ORC=y, please install libelf-dev, libelf-devel or elfutils-libelf-devel" >&2 - @false -else - @echo "warning: Cannot use CONFIG_STACK_VALIDATION=y, please install libelf-dev, libelf-devel or elfutils-libelf-devel" >&2 -endif -endif prepare-resolve_btfids: $(resolve_btfids_target) ifeq ($(ERROR_RESOLVE_BTFIDS),1) -- 2.43.0

1 0

[PATCH openEuler-22.03-LTS-SP1] drm/lima: fix shared irq handling on driver remove
by Cheng Yu 18 Aug '24

18 Aug '24

From: Erico Nunes <nunes.erico(a)gmail.com> stable inclusion from stable-v6.1.98 commit 0a487e977cb8897ae4c51ecd34bbaa2b005266c9 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAGRYV CVE: CVE-2024-42127 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit a6683c690bbfd1f371510cb051e8fa49507f3f5e ] lima uses a shared interrupt, so the interrupt handlers must be prepared to be called at any time. At driver removal time, the clocks are disabled early and the interrupts stay registered until the very end of the remove process due to the devm usage. This is potentially a bug as the interrupts access device registers which assumes clocks are enabled. A crash can be triggered by removing the driver in a kernel with CONFIG_DEBUG_SHIRQ enabled. This patch frees the interrupts at each lima device finishing callback so that the handlers are already unregistered by the time we fully disable clocks. Signed-off-by: Erico Nunes <nunes.erico(a)gmail.com> Signed-off-by: Qiang Yu <yuq825(a)gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240401224329.1228468-2-nune… Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Cheng Yu <serein.chengyu(a)huawei.com> --- drivers/gpu/drm/lima/lima_gp.c | 2 ++ drivers/gpu/drm/lima/lima_mmu.c | 5 +++++ drivers/gpu/drm/lima/lima_pp.c | 4 ++++ 3 files changed, 11 insertions(+) diff --git a/drivers/gpu/drm/lima/lima_gp.c b/drivers/gpu/drm/lima/lima_gp.c index 6cf46b653e81..ca3842f71984 100644 --- a/drivers/gpu/drm/lima/lima_gp.c +++ b/drivers/gpu/drm/lima/lima_gp.c @@ -324,7 +324,9 @@ int lima_gp_init(struct lima_ip *ip) void lima_gp_fini(struct lima_ip *ip) { + struct lima_device *dev = ip->dev; + devm_free_irq(dev->dev, ip->irq, ip); } int lima_gp_pipe_init(struct lima_device *dev) diff --git a/drivers/gpu/drm/lima/lima_mmu.c b/drivers/gpu/drm/lima/lima_mmu.c index a1ae6c252dc2..8ca7047adbac 100644 --- a/drivers/gpu/drm/lima/lima_mmu.c +++ b/drivers/gpu/drm/lima/lima_mmu.c @@ -118,7 +118,12 @@ int lima_mmu_init(struct lima_ip *ip) void lima_mmu_fini(struct lima_ip *ip) { + struct lima_device *dev = ip->dev; + + if (ip->id == lima_ip_ppmmu_bcast) + return; + devm_free_irq(dev->dev, ip->irq, ip); } void lima_mmu_flush_tlb(struct lima_ip *ip) diff --git a/drivers/gpu/drm/lima/lima_pp.c b/drivers/gpu/drm/lima/lima_pp.c index 54b208a4a768..d34c9e8840f4 100644 --- a/drivers/gpu/drm/lima/lima_pp.c +++ b/drivers/gpu/drm/lima/lima_pp.c @@ -266,7 +266,9 @@ int lima_pp_init(struct lima_ip *ip) void lima_pp_fini(struct lima_ip *ip) { + struct lima_device *dev = ip->dev; + devm_free_irq(dev->dev, ip->irq, ip); } int lima_pp_bcast_resume(struct lima_ip *ip) @@ -299,7 +301,9 @@ int lima_pp_bcast_init(struct lima_ip *ip) void lima_pp_bcast_fini(struct lima_ip *ip) { + struct lima_device *dev = ip->dev; + devm_free_irq(dev->dev, ip->irq, ip); } static int lima_pp_task_validate(struct lima_sched_pipe *pipe, -- 2.25.1

2 1

[PATCH OLK-5.10 V1] drm/lima: fix shared irq handling on driver remove
by Cheng Yu 18 Aug '24

18 Aug '24

From: Erico Nunes <nunes.erico(a)gmail.com> stable inclusion from stable-v6.1.98 commit 0a487e977cb8897ae4c51ecd34bbaa2b005266c9 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAGRYV CVE: CVE-2024-42127 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit a6683c690bbfd1f371510cb051e8fa49507f3f5e ] lima uses a shared interrupt, so the interrupt handlers must be prepared to be called at any time. At driver removal time, the clocks are disabled early and the interrupts stay registered until the very end of the remove process due to the devm usage. This is potentially a bug as the interrupts access device registers which assumes clocks are enabled. A crash can be triggered by removing the driver in a kernel with CONFIG_DEBUG_SHIRQ enabled. This patch frees the interrupts at each lima device finishing callback so that the handlers are already unregistered by the time we fully disable clocks. Signed-off-by: Erico Nunes <nunes.erico(a)gmail.com> Signed-off-by: Qiang Yu <yuq825(a)gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240401224329.1228468-2-nune… Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Cheng Yu <serein.chengyu(a)huawei.com> --- drivers/gpu/drm/lima/lima_gp.c | 2 ++ drivers/gpu/drm/lima/lima_mmu.c | 5 +++++ drivers/gpu/drm/lima/lima_pp.c | 4 ++++ 3 files changed, 11 insertions(+) diff --git a/drivers/gpu/drm/lima/lima_gp.c b/drivers/gpu/drm/lima/lima_gp.c index 6cf46b653e81..ca3842f71984 100644 --- a/drivers/gpu/drm/lima/lima_gp.c +++ b/drivers/gpu/drm/lima/lima_gp.c @@ -324,7 +324,9 @@ int lima_gp_init(struct lima_ip *ip) void lima_gp_fini(struct lima_ip *ip) { + struct lima_device *dev = ip->dev; + devm_free_irq(dev->dev, ip->irq, ip); } int lima_gp_pipe_init(struct lima_device *dev) diff --git a/drivers/gpu/drm/lima/lima_mmu.c b/drivers/gpu/drm/lima/lima_mmu.c index a1ae6c252dc2..8ca7047adbac 100644 --- a/drivers/gpu/drm/lima/lima_mmu.c +++ b/drivers/gpu/drm/lima/lima_mmu.c @@ -118,7 +118,12 @@ int lima_mmu_init(struct lima_ip *ip) void lima_mmu_fini(struct lima_ip *ip) { + struct lima_device *dev = ip->dev; + + if (ip->id == lima_ip_ppmmu_bcast) + return; + devm_free_irq(dev->dev, ip->irq, ip); } void lima_mmu_flush_tlb(struct lima_ip *ip) diff --git a/drivers/gpu/drm/lima/lima_pp.c b/drivers/gpu/drm/lima/lima_pp.c index 54b208a4a768..d34c9e8840f4 100644 --- a/drivers/gpu/drm/lima/lima_pp.c +++ b/drivers/gpu/drm/lima/lima_pp.c @@ -266,7 +266,9 @@ int lima_pp_init(struct lima_ip *ip) void lima_pp_fini(struct lima_ip *ip) { + struct lima_device *dev = ip->dev; + devm_free_irq(dev->dev, ip->irq, ip); } int lima_pp_bcast_resume(struct lima_ip *ip) @@ -299,7 +301,9 @@ int lima_pp_bcast_init(struct lima_ip *ip) void lima_pp_bcast_fini(struct lima_ip *ip) { + struct lima_device *dev = ip->dev; + devm_free_irq(dev->dev, ip->irq, ip); } static int lima_pp_task_validate(struct lima_sched_pipe *pipe, -- 2.25.1

2 1

[PATCH OLK-6.6 0/1] secure smmu init
by Xiangkai Yang 18 Aug '24

18 Aug '24

virtCCA feature Xiangkai Yang (1): secure smmu init arch/arm64/include/asm/kvm_tmi.h | 244 +++-- arch/arm64/include/asm/kvm_tmm.h | 22 + arch/arm64/kvm/tmi.c | 128 ++- arch/arm64/kvm/virtcca_cvm.c | 10 +- drivers/iommu/arm/arm-smmu-v3/Makefile | 1 + drivers/iommu/arm/arm-smmu-v3/arm-s-smmu-v3.c | 841 ++++++++++++++++++ drivers/iommu/arm/arm-smmu-v3/arm-s-smmu-v3.h | 198 +++++ drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 168 +++- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 16 + include/linux/iommu.h | 4 + 10 files changed, 1548 insertions(+), 84 deletions(-) create mode 100644 drivers/iommu/arm/arm-smmu-v3/arm-s-smmu-v3.c create mode 100644 drivers/iommu/arm/arm-smmu-v3/arm-s-smmu-v3.h -- 2.27.0

2 2

[PATCH OLK-6.6 0/1] kvm : add coda feature
by JunBin Li 18 Aug '24

18 Aug '24

3 3

[openeuler:OLK-6.6] BUILD REGRESSION 669da8f42d5992e28de21e340a6d74abd58f83d0
by kernel test robot 18 Aug '24

18 Aug '24

tree/branch: https://gitee.com/openeuler/kernel.git OLK-6.6 branch HEAD: 669da8f42d5992e28de21e340a6d74abd58f83d0 !10911 Fix CVE-2024-40966 Error/Warning ids grouped by kconfigs: recent_errors |-- arm64-allmodconfig | |-- arch-arm64-kvm-arm.c:warning:variable-r-is-used-uninitialized-whenever-if-condition-is-false | |-- arch-arm64-kvm-tmi.c:warning:no-previous-prototype-for-function-tmi_tmm_inf_test | |-- arch-arm64-kvm-virtcca_cvm.c:warning:no-previous-prototype-for-function-kvm_cvm_create_ttt_levels | |-- arch-arm64-kvm-virtcca_cvm.c:warning:no-previous-prototype-for-function-kvm_cvm_get_num_brps | |-- arch-arm64-kvm-virtcca_cvm.c:warning:no-previous-prototype-for-function-kvm_cvm_get_num_wrps | |-- arch-arm64-kvm-virtcca_cvm.c:warning:no-previous-prototype-for-function-kvm_cvm_ipa_limit | |-- arch-arm64-kvm-virtcca_cvm.c:warning:no-previous-prototype-for-function-kvm_cvm_populate_par_region | |-- arch-arm64-kvm-virtcca_cvm.c:warning:no-previous-prototype-for-function-kvm_cvm_supports_pmu | `-- arch-arm64-kvm-virtcca_cvm.c:warning:no-previous-prototype-for-function-kvm_cvm_supports_sve |-- x86_64-buildonly-randconfig-003-20240817 | `-- kernel-sched-isolation.c:error:setup_max_cpus-undeclared-(first-use-in-this-function) |-- x86_64-buildonly-randconfig-005-20240817 | `-- kernel-sched-isolation.c:error:setup_max_cpus-undeclared-(first-use-in-this-function) `-- x86_64-buildonly-randconfig-006-20240817 `-- kernel-sched-isolation.c:error:use-of-undeclared-identifier-setup_max_cpus elapsed time: 724m configs tested: 25 configs skipped: 124 tested configs: arm64 allmodconfig clang-20 arm64 allnoconfig gcc-14.1.0 arm64 randconfig-001-20240817 clang-20 arm64 randconfig-002-20240817 gcc-14.1.0 arm64 randconfig-003-20240817 gcc-14.1.0 arm64 randconfig-004-20240817 gcc-14.1.0 loongarch allmodconfig gcc-14.1.0 loongarch allnoconfig gcc-14.1.0 loongarch randconfig-001-20240817 gcc-14.1.0 loongarch randconfig-002-20240817 gcc-14.1.0 x86_64 allnoconfig clang-18 x86_64 allyesconfig clang-18 x86_64 buildonly-randconfig-001-20240817 gcc-12 x86_64 buildonly-randconfig-002-20240817 gcc-12 x86_64 buildonly-randconfig-003-20240817 gcc-11 x86_64 buildonly-randconfig-004-20240817 gcc-12 x86_64 buildonly-randconfig-005-20240817 gcc-12 x86_64 buildonly-randconfig-006-20240817 clang-18 x86_64 defconfig gcc-11 x86_64 randconfig-001-20240817 clang-18 x86_64 randconfig-002-20240817 clang-18 x86_64 randconfig-003-20240817 gcc-12 x86_64 randconfig-004-20240817 gcc-12 x86_64 randconfig-005-20240817 clang-18 x86_64 rhel-8.3-rust clang-18 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[PATCH openEuler-1.0-LTS] libceph: fix race between delayed_work() and ceph_monc_stop()
by Dong Chenchen 17 Aug '24

17 Aug '24

From: Ilya Dryomov <idryomov(a)gmail.com> stable inclusion from stable-v4.19.318 commit 1177afeca833174ba83504688eec898c6214f4bf category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IAILFZ Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 69c7b2fe4c9cc1d3b1186d1c5606627ecf0de883 upstream. The way the delayed work is handled in ceph_monc_stop() is prone to races with mon_fault() and possibly also finish_hunting(). Both of these can requeue the delayed work which wouldn't be canceled by any of the following code in case that happens after cancel_delayed_work_sync() runs -- __close_session() doesn't mess with the delayed work in order to avoid interfering with the hunting interval logic. This part was missed in commit b5d91704f53e ("libceph: behave in mon_fault() if cur_mon < 0") and use-after-free can still ensue on monc and objects that hang off of it, with monc->auth and monc->monmap being particularly susceptible to quickly being reused. To fix this: - clear monc->cur_mon and monc->hunting as part of closing the session in ceph_monc_stop() - bail from delayed_work() if monc->cur_mon is cleared, similar to how it's done in mon_fault() and finish_hunting() (based on monc->hunting) - call cancel_delayed_work_sync() after the session is closed Cc: stable(a)vger.kernel.org Link: https://tracker.ceph.com/issues/66857 Signed-off-by: Ilya Dryomov <idryomov(a)gmail.com> Reviewed-by: Xiubo Li <xiubli(a)redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/ceph/mon_client.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/net/ceph/mon_client.c b/net/ceph/mon_client.c index a53e4fbb6319..fa2cde7c77c4 100644 --- a/net/ceph/mon_client.c +++ b/net/ceph/mon_client.c @@ -963,13 +963,19 @@ static void delayed_work(struct work_struct *work) struct ceph_mon_client *monc = container_of(work, struct ceph_mon_client, delayed_work.work); - dout("monc delayed_work\n"); mutex_lock(&monc->mutex); + dout("%s mon%d\n", __func__, monc->cur_mon); + if (monc->cur_mon < 0) { + goto out; + } + if (monc->hunting) { dout("%s continuing hunt\n", __func__); reopen_session(monc); } else { int is_auth = ceph_auth_is_authenticated(monc->auth); + + dout("%s is_authed %d\n", __func__, is_auth); if (ceph_con_keepalive_expired(&monc->con, CEPH_MONC_PING_TIMEOUT)) { dout("monc keepalive timeout\n"); @@ -994,6 +1000,8 @@ static void delayed_work(struct work_struct *work) } } __schedule_delayed(monc); + +out: mutex_unlock(&monc->mutex); } @@ -1106,13 +1114,15 @@ EXPORT_SYMBOL(ceph_monc_init); void ceph_monc_stop(struct ceph_mon_client *monc) { dout("stop\n"); - cancel_delayed_work_sync(&monc->delayed_work); mutex_lock(&monc->mutex); __close_session(monc); + monc->hunting = false; monc->cur_mon = -1; mutex_unlock(&monc->mutex); + cancel_delayed_work_sync(&monc->delayed_work); + /* * flush msgr queue before we destroy ourselves to ensure that: * - any work that references our embedded con is finished. -- 2.25.1

2 1

[PATCH openEuler-22.03-LTS-SP1] libceph: fix race between delayed_work() and ceph_monc_stop()
by Dong Chenchen 17 Aug '24

17 Aug '24

From: Ilya Dryomov <idryomov(a)gmail.com> stable inclusion from stable-v5.10.222 commit 34b76d1922e41da1fa73d43b764cddd82ac9733c category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IAILFZ Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 69c7b2fe4c9cc1d3b1186d1c5606627ecf0de883 upstream. The way the delayed work is handled in ceph_monc_stop() is prone to races with mon_fault() and possibly also finish_hunting(). Both of these can requeue the delayed work which wouldn't be canceled by any of the following code in case that happens after cancel_delayed_work_sync() runs -- __close_session() doesn't mess with the delayed work in order to avoid interfering with the hunting interval logic. This part was missed in commit b5d91704f53e ("libceph: behave in mon_fault() if cur_mon < 0") and use-after-free can still ensue on monc and objects that hang off of it, with monc->auth and monc->monmap being particularly susceptible to quickly being reused. To fix this: - clear monc->cur_mon and monc->hunting as part of closing the session in ceph_monc_stop() - bail from delayed_work() if monc->cur_mon is cleared, similar to how it's done in mon_fault() and finish_hunting() (based on monc->hunting) - call cancel_delayed_work_sync() after the session is closed Cc: stable(a)vger.kernel.org Link: https://tracker.ceph.com/issues/66857 Signed-off-by: Ilya Dryomov <idryomov(a)gmail.com> Reviewed-by: Xiubo Li <xiubli(a)redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/ceph/mon_client.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/net/ceph/mon_client.c b/net/ceph/mon_client.c index ef5c174102d5..82d3f1204ee4 100644 --- a/net/ceph/mon_client.c +++ b/net/ceph/mon_client.c @@ -1014,13 +1014,19 @@ static void delayed_work(struct work_struct *work) struct ceph_mon_client *monc = container_of(work, struct ceph_mon_client, delayed_work.work); - dout("monc delayed_work\n"); mutex_lock(&monc->mutex); + dout("%s mon%d\n", __func__, monc->cur_mon); + if (monc->cur_mon < 0) { + goto out; + } + if (monc->hunting) { dout("%s continuing hunt\n", __func__); reopen_session(monc); } else { int is_auth = ceph_auth_is_authenticated(monc->auth); + + dout("%s is_authed %d\n", __func__, is_auth); if (ceph_con_keepalive_expired(&monc->con, CEPH_MONC_PING_TIMEOUT)) { dout("monc keepalive timeout\n"); @@ -1045,6 +1051,8 @@ static void delayed_work(struct work_struct *work) } } __schedule_delayed(monc); + +out: mutex_unlock(&monc->mutex); } @@ -1157,13 +1165,15 @@ EXPORT_SYMBOL(ceph_monc_init); void ceph_monc_stop(struct ceph_mon_client *monc) { dout("stop\n"); - cancel_delayed_work_sync(&monc->delayed_work); mutex_lock(&monc->mutex); __close_session(monc); + monc->hunting = false; monc->cur_mon = -1; mutex_unlock(&monc->mutex); + cancel_delayed_work_sync(&monc->delayed_work); + /* * flush msgr queue before we destroy ourselves to ensure that: * - any work that references our embedded con is finished. -- 2.25.1

2 1

2024

2023

2022

2021

2020

2019

Kernel August 2024