January 2024 - Kernel - mailweb.openeuler.org

[PATCH OLK-6.6 v3] kabi: reserve space for cpu cgroup and cpuset cgroup related structures
by Xiang Yang 29 Jan '24

29 Jan '24

From: Lu Jialin <lujialin4(a)huawei.com> hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8SWPP --------------------------------------------- We reserve some fields beforehand for cpu cgroup and cpuset related structures prone to change, therefore, we can hot add/change features of cpu cgroup cpuset and cgroup with this enhancement. After reserving, normally cache does not matter as the reserved fields are not accessed at all. Signed-off-by: Lu Jialin <lujialin4(a)huawei.com> Signed-off-by: Xiang Yang <xiangyang3(a)huawei.com> --- kernel/cgroup/cpuset.c | 6 ++++++ kernel/sched/sched.h | 10 ++++++++++ 2 files changed, 16 insertions(+) diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index cfdca8aeabda..ea78008dd899 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -43,6 +43,7 @@ #include <linux/sched/isolation.h> #include <linux/cgroup.h> #include <linux/wait.h> +#include <linux/kabi.h> DEFINE_STATIC_KEY_FALSE(cpusets_pre_enable_key); DEFINE_STATIC_KEY_FALSE(cpusets_enabled_key); @@ -186,6 +187,11 @@ struct cpuset { /* Handle for cpuset.cpus.partition */ struct cgroup_file partition_file; + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; /* diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 88dbcbe3d7c1..bd5d45ba0de8 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -68,6 +68,7 @@ #include <linux/wait_api.h> #include <linux/wait_bit.h> #include <linux/workqueue_api.h> +#include <linux/kabi.h> #include <trace/events/power.h> #include <trace/events/sched.h> @@ -455,6 +456,15 @@ struct task_group { #if defined(CONFIG_QOS_SCHED_SMART_GRID) && !defined(__GENKSYMS__) struct auto_affinity *auto_affinity; #endif + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) + KABI_RESERVE(5) + KABI_RESERVE(6) + KABI_RESERVE(7) + KABI_RESERVE(8) }; #ifdef CONFIG_FAIR_GROUP_SCHED -- 2.34.1

1 0

[PATCH openEuler-1.0-LTS] drm/atomic: Fix potential use-after-free in nonblocking commits
by Guo Mengqi 29 Jan '24

29 Jan '24

From: Daniel Vetter <daniel.vetter(a)ffwll.ch> stable inclusion from stable-v4.19.291 commit 73a82b22963defa87204f0f9f44a534adf7f831a category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7V6NJ CVE: CVE-2023-51043 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 4e076c73e4f6e90816b30fcd4a0d7ab365087255 upstream. This requires a bit of background. Properly done a modeset driver's unload/remove sequence should be drm_dev_unplug(); drm_atomic_helper_shutdown(); drm_dev_put(); The trouble is that the drm_dev_unplugged() checks are by design racy, they do not synchronize against all outstanding ioctl. This is because those ioctl could block forever (both for modeset and for driver specific ioctls), leading to deadlocks in hotunplug. Instead the code sections that touch the hardware need to be annotated with drm_dev_enter/exit, to avoid accessing hardware resources after the unload/remove has finished. To avoid use-after-free issues all the involved userspace visible objects are supposed to hold a reference on the underlying drm_device, like drm_file does. The issue now is that we missed one, the atomic modeset ioctl can be run in a nonblocking fashion, and in that case it cannot rely on the implied drm_device reference provided by the ioctl calling context. This can result in a use-after-free if an nonblocking atomic commit is carefully raced against a driver unload. Fix this by unconditionally grabbing a drm_device reference for any drm_atomic_state structures. Strictly speaking this isn't required for blocking commits and TEST_ONLY calls, but it's the simpler approach. Thanks to shanzhulig for the initial idea of grabbing an unconditional reference, I just added comments, a condensed commit message and fixed a minor potential issue in where exactly we drop the final reference. Reported-by: shanzhulig <shanzhulig(a)gmail.com> Suggested-by: shanzhulig <shanzhulig(a)gmail.com> Reviewed-by: Maxime Ripard <mripard(a)kernel.org> Cc: Maarten Lankhorst <maarten.lankhorst(a)linux.intel.com> Cc: Thomas Zimmermann <tzimmermann(a)suse.de> Cc: David Airlie <airlied(a)gmail.com> Cc: stable(a)kernel.org Signed-off-by: Daniel Vetter <daniel.vetter(a)intel.com> Signed-off-by: Daniel Vetter <daniel.vetter(a)ffwll.ch> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Guo Mengqi <guomengqi3(a)huawei.com> --- drivers/gpu/drm/drm_atomic.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index 1a4b44923aec..e703341e4cb2 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -91,6 +91,12 @@ drm_atomic_state_init(struct drm_device *dev, struct drm_atomic_state *state) if (!state->planes) goto fail; + /* + * Because drm_atomic_state can be committed asynchronously we need our + * own reference and cannot rely on the on implied by drm_file in the + * ioctl call. + */ + drm_dev_get(dev); state->dev = dev; DRM_DEBUG_ATOMIC("Allocated atomic state %p\n", state); @@ -250,7 +256,8 @@ EXPORT_SYMBOL(drm_atomic_state_clear); void __drm_atomic_state_free(struct kref *ref) { struct drm_atomic_state *state = container_of(ref, typeof(*state), ref); - struct drm_mode_config *config = &state->dev->mode_config; + struct drm_device *dev = state->dev; + struct drm_mode_config *config = &dev->mode_config; drm_atomic_state_clear(state); @@ -262,6 +269,8 @@ void __drm_atomic_state_free(struct kref *ref) drm_atomic_state_default_release(state); kfree(state); } + + drm_dev_put(dev); } EXPORT_SYMBOL(__drm_atomic_state_free); -- 2.17.1

1 0

[PATCH openEuler-1.0-LTS] smb: client: fix NULL deref in asn1_ber_decoder()
by ZhaoLong Wang 29 Jan '24

29 Jan '24

From: Paulo Alcantara <pc(a)manguebit.com> stable inclusion from stable-v4.19.304 commit 832c20fc4cc82c497566db35996ea488661fc764 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8ZPJF CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 90d025c2e953c11974e76637977c473200593a46 ] If server replied SMB2_NEGOTIATE with a zero SecurityBufferOffset, smb2_get_data_area() sets @len to non-zero but return NULL, so decode_negTokeninit() ends up being called with a NULL @security_blob: BUG: kernel NULL pointer dereference, address: 0000000000000000 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 0 P4D 0 Oops: 0000 [#1] PREEMPT SMP NOPTI CPU: 2 PID: 871 Comm: mount.cifs Not tainted 6.7.0-rc4 #2 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.2-3-gd478f380-rebuilt.opensuse.org 04/01/2014 RIP: 0010:asn1_ber_decoder+0x173/0xc80 Code: 01 4c 39 2c 24 75 09 45 84 c9 0f 85 2f 03 00 00 48 8b 14 24 4c 29 ea 48 83 fa 01 0f 86 1e 07 00 00 48 8b 74 24 28 4d 8d 5d 01 <42> 0f b6 3c 2e 89 fa 40 88 7c 24 5c f7 d2 83 e2 1f 0f 84 3d 07 00 RSP: 0018:ffffc9000063f950 EFLAGS: 00010202 RAX: 0000000000000002 RBX: 0000000000000000 RCX: 000000000000004a RDX: 000000000000004a RSI: 0000000000000000 RDI: 0000000000000000 RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000002 R11: 0000000000000001 R12: 0000000000000000 R13: 0000000000000000 R14: 000000000000004d R15: 0000000000000000 FS: 00007fce52b0fbc0(0000) GS:ffff88806ba00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000000 CR3: 000000001ae64000 CR4: 0000000000750ef0 PKRU: 55555554 Call Trace: <TASK> ? __die+0x23/0x70 ? page_fault_oops+0x181/0x480 ? __stack_depot_save+0x1e6/0x480 ? exc_page_fault+0x6f/0x1c0 ? asm_exc_page_fault+0x26/0x30 ? asn1_ber_decoder+0x173/0xc80 ? check_object+0x40/0x340 decode_negTokenInit+0x1e/0x30 [cifs] SMB2_negotiate+0xc99/0x17c0 [cifs] ? smb2_negotiate+0x46/0x60 [cifs] ? srso_alias_return_thunk+0x5/0xfbef5 smb2_negotiate+0x46/0x60 [cifs] cifs_negotiate_protocol+0xae/0x130 [cifs] cifs_get_smb_ses+0x517/0x1040 [cifs] ? srso_alias_return_thunk+0x5/0xfbef5 ? srso_alias_return_thunk+0x5/0xfbef5 ? queue_delayed_work_on+0x5d/0x90 cifs_mount_get_session+0x78/0x200 [cifs] dfs_mount_share+0x13a/0x9f0 [cifs] ? srso_alias_return_thunk+0x5/0xfbef5 ? lock_acquire+0xbf/0x2b0 ? find_nls+0x16/0x80 ? srso_alias_return_thunk+0x5/0xfbef5 cifs_mount+0x7e/0x350 [cifs] cifs_smb3_do_mount+0x128/0x780 [cifs] smb3_get_tree+0xd9/0x290 [cifs] vfs_get_tree+0x2c/0x100 ? capable+0x37/0x70 path_mount+0x2d7/0xb80 ? srso_alias_return_thunk+0x5/0xfbef5 ? _raw_spin_unlock_irqrestore+0x44/0x60 __x64_sys_mount+0x11a/0x150 do_syscall_64+0x47/0xf0 entry_SYSCALL_64_after_hwframe+0x6f/0x77 RIP: 0033:0x7fce52c2ab1e Fix this by setting @len to zero when @off == 0 so callers won't attempt to dereference non-existing data areas. Reported-by: Robert Morris <rtm(a)csail.mit.edu> Cc: stable(a)vger.kernel.org Signed-off-by: Paulo Alcantara (SUSE) <pc(a)manguebit.com> Signed-off-by: Steve French <stfrench(a)microsoft.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: ZhaoLong Wang <wangzhaolong1(a)huawei.com> --- fs/cifs/smb2misc.c | 26 ++++++++++---------------- 1 file changed, 10 insertions(+), 16 deletions(-) diff --git a/fs/cifs/smb2misc.c b/fs/cifs/smb2misc.c index 94fb82077bdd..6ac33ec15583 100644 --- a/fs/cifs/smb2misc.c +++ b/fs/cifs/smb2misc.c @@ -302,6 +302,9 @@ static const bool has_smb2_data_area[NUMBER_OF_SMB2_COMMANDS] = { char * smb2_get_data_area_len(int *off, int *len, struct smb2_sync_hdr *shdr) { + const int max_off = 4096; + const int max_len = 128 * 1024; + *off = 0; *len = 0; @@ -369,29 +372,20 @@ smb2_get_data_area_len(int *off, int *len, struct smb2_sync_hdr *shdr) * Invalid length or offset probably means data area is invalid, but * we have little choice but to ignore the data area in this case. */ - if (*off > 4096) { - cifs_dbg(VFS, "offset %d too large, data area ignored\n", *off); - *len = 0; - *off = 0; - } else if (*off < 0) { - cifs_dbg(VFS, "negative offset %d to data invalid ignore data area\n", - *off); + if (unlikely(*off < 0 || *off > max_off || + *len < 0 || *len > max_len)) { + cifs_dbg(VFS, "%s: invalid data area (off=%d len=%d)\n", + __func__, *off, *len); *off = 0; *len = 0; - } else if (*len < 0) { - cifs_dbg(VFS, "negative data length %d invalid, data area ignored\n", - *len); - *len = 0; - } else if (*len > 128 * 1024) { - cifs_dbg(VFS, "data area larger than 128K: %d\n", *len); + } else if (*off == 0) { *len = 0; } /* return pointer to beginning of data area, ie offset from SMB start */ - if ((*off != 0) && (*len != 0)) + if (*off > 0 && *len > 0) return (char *)shdr + *off; - else - return NULL; + return NULL; } /* -- 2.39.2

2 1

[PATCH OLK-6.6 0/2] Expose swapcache stat for memcg v1
by Liu Shixin 29 Jan '24

29 Jan '24

The first patch expose swapcache stat for memcg v1, the second patch remote unused do_memsw_account() in memcg1_stat_format(). Liu Shixin (2): memcg: expose swapcache stat for memcg v1 memcg: remove unused do_memsw_account in memcg1_stat_format Documentation/admin-guide/cgroup-v1/memory.rst | 1 + mm/memcontrol.c | 15 ++++++++------- 2 files changed, 9 insertions(+), 7 deletions(-) -- 2.25.1

1 2

[PATCH V2 OLK-6.6 0/5] Add support for ecmdq
by Zhang Zekun 29 Jan '24

29 Jan '24

Add support for ecmdq v2: fix some code check errors Zhen Lei (5): iommu/arm-smmu-v3: Add support for ECMDQ register mode iommu/arm-smmu-v3: Ensure that a set of associated commands are inserted in the same ECMDQ iommu/arm-smmu-v3: Add arm_smmu_ecmdq_issue_cmdlist() for non-shared ECMDQ iommu/arm-smmu-v3: Add support for less than one ECMDQ per core iommu/arm-smmu-v3: Allow disabling ECMDQs at boot time drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 434 +++++++++++++++++++- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 39 ++ 2 files changed, 465 insertions(+), 8 deletions(-) -- 2.17.1

2 6

[PATCH OLK-6.6 0/4] misc for fd
by Chen Ridong 29 Jan '24

29 Jan '24

Chen Ridong (4): filescgroup: remove files of dfl_cftypes. filecgroup: move code of filescontrol.c to files-cgroup.c misc: add a type of resource to control fd. filscgroup: fix null pointer access when files_cgroup is not inited. Documentation/admin-guide/cgroup-v2.rst | 7 + fs/Makefile | 2 + fs/file.c | 16 +- fs/files-cgroup.c | 335 +++++++++++++++++++++++ fs/filescontrol.c | 344 +++--------------------- fs/misc-fd.c | 77 ++++++ include/linux/fdtable.h | 3 + include/linux/files-cgroup.h | 32 +++ include/linux/filescontrol.h | 46 ++-- include/linux/misc-fd.h | 21 ++ include/linux/misc_cgroup.h | 4 + kernel/cgroup/misc.c | 4 + 12 files changed, 556 insertions(+), 335 deletions(-) create mode 100644 fs/files-cgroup.c create mode 100644 fs/misc-fd.c create mode 100644 include/linux/files-cgroup.h create mode 100644 include/linux/misc-fd.h -- 2.34.1

1 4

[PATCH V2 PATCH 0/5] Add support for ecmdq
by Zhang Zekun 29 Jan '24

29 Jan '24

Add support for ecmdq v2: fix some code check errors Zhen Lei (5): iommu/arm-smmu-v3: Add support for ECMDQ register mode iommu/arm-smmu-v3: Ensure that a set of associated commands are inserted in the same ECMDQ iommu/arm-smmu-v3: Add arm_smmu_ecmdq_issue_cmdlist() for non-shared ECMDQ iommu/arm-smmu-v3: Add support for less than one ECMDQ per core iommu/arm-smmu-v3: Allow disabling ECMDQs at boot time drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 434 +++++++++++++++++++- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 39 ++ 2 files changed, 465 insertions(+), 8 deletions(-) -- 2.17.1

1 5

[PATCH OLK-6.6 00/20] kabi reserve patches
by Lin Yujun 29 Jan '24

29 Jan '24

*** BLURB HERE *** Chen Jiahao (1): KABI: reserve space for struct input_dev Lin Ruizhe (5): msi: Add kabi_reserve in msi.h interrupt: Add kabi_reserve in interrupt.h irq: Add kabi_reserve in irq irq_desc: Add kabi_reserve in irq_desc irqdomain: Add kabi_reserve in irqdomain Lin Yujun (13): drm: drm_atomic.h: Add kabi_reserve drm: drm_client.h: Add kabi_reserve drm: drm_connector.h: Add kabi_reserve drm: drm_crtc.h: Add kabi_reserve drm: drm_device.h: Add kabi_reserve drm: drm_drv.h: Add kabi_reserve drm: drm_fb_helper.h: Add kabi_reserve drm: drm_file.h: Add kabi_reserve drm: drm_fourcc.h: Add kabi_reserve drm: drm_gem.h: Add kabi_reserve pm: pm_domain.h: Add kabi_reserve pm: pm.h: Add kabi_reserve energy_model: Add kabi_reserve Yang Yingliang (1): kabi: reserve space for struct cpu_stop_work include/drm/drm_atomic.h | 6 ++++++ include/drm/drm_client.h | 7 +++++++ include/drm/drm_connector.h | 18 ++++++++++++++++++ include/drm/drm_crtc.h | 7 +++++++ include/drm/drm_device.h | 4 ++++ include/drm/drm_drv.h | 5 +++++ include/drm/drm_fb_helper.h | 8 ++++++++ include/drm/drm_file.h | 8 ++++++++ include/drm/drm_fourcc.h | 5 +++++ include/drm/drm_gem.h | 8 ++++++++ include/linux/energy_model.h | 5 +++++ include/linux/input.h | 4 ++++ include/linux/interrupt.h | 6 ++++++ include/linux/irq.h | 8 ++++++++ include/linux/irqdesc.h | 4 +++- include/linux/irqdomain.h | 3 +++ include/linux/msi.h | 7 +++++++ include/linux/pm.h | 11 +++++++++++ include/linux/pm_domain.h | 6 ++++++ include/linux/stop_machine.h | 3 +++ 20 files changed, 132 insertions(+), 1 deletion(-) -- 2.34.1

2 21

[PATCH OLK-6.6 v2] fs:/dcache.c: fix negative dentry limit not complete problem
by Long Li 29 Jan '24

29 Jan '24

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8JH5M CVE: NA -------------------------------- Restrictions of negative dentry can avoid softlock when too many dentry in memory, but restrictions before do not always take effect. For example, removing files which has been created would not enter retain_dentry(), because the dentry of file maybe has been added to lru list of superblock, it caused that fast_dput() reutrn true in last dput(). So, add restriction logic for the file which has been added to the lru list in fast_dput(), it prevents the negative dentry exceeding the limit when deleting existing file. Factor out negative dentry limit logic into limit_negative_dentry(), make the code more clean. Fixes: f3de89fc28ba ("fs/dcache.c: avoid softlock since too many negative dentry") Signed-off-by: Long Li <leo.lilong(a)huawei.com> --- fs/dcache.c | 70 ++++++++++++++++++++++++++++++++++++----------------- 1 file changed, 48 insertions(+), 22 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index c27aa8ff78f4..ab5f8e954978 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -675,10 +675,36 @@ static inline struct dentry *lock_parent(struct dentry *dentry) return __lock_parent(dentry); } -static inline bool retain_dentry(struct dentry *dentry) +/* + * Return true if dentry is negative and exceed negative dentry limit. + */ +static inline bool limit_negative_dentry(struct dentry *dentry) { struct dentry *parent; + parent = dentry->d_parent; + if (unlikely(!parent)) + return false; + + WARN_ON((atomic_read(&parent->d_neg_dnum) < 0)); + if (!dentry->d_inode) { + if (!(dentry->d_flags & DCACHE_NEGATIVE_ACCOUNT)) { + unsigned int flags = READ_ONCE(dentry->d_flags); + + flags |= DCACHE_NEGATIVE_ACCOUNT; + WRITE_ONCE(dentry->d_flags, flags); + atomic_inc(&parent->d_neg_dnum); + } + + if (atomic_read(&parent->d_neg_dnum) >= NEG_DENTRY_LIMIT) + return true; + } + + return false; +} + +static inline bool retain_dentry(struct dentry *dentry) +{ WARN_ON(d_in_lookup(dentry)); /* Unreachable? Get rid of it */ @@ -696,27 +722,9 @@ static inline bool retain_dentry(struct dentry *dentry) if (unlikely(dentry->d_flags & DCACHE_DONTCACHE)) return false; - if (unlikely(!dentry->d_parent)) - goto noparent; - - parent = dentry->d_parent; - /* Return false if it's negative */ - WARN_ON((atomic_read(&parent->d_neg_dnum) < 0)); - if (!dentry->d_inode) { - if (!(dentry->d_flags & DCACHE_NEGATIVE_ACCOUNT)) { - unsigned int flags = READ_ONCE(dentry->d_flags); - - flags |= DCACHE_NEGATIVE_ACCOUNT; - WRITE_ONCE(dentry->d_flags, flags); - atomic_inc(&parent->d_neg_dnum); - } - } - - if (!dentry->d_inode && - atomic_read(&parent->d_neg_dnum) >= NEG_DENTRY_LIMIT) + if (unlikely(limit_negative_dentry(dentry))) return false; -noparent: /* retain; LRU fodder */ dentry->d_lockref.count--; if (unlikely(!(dentry->d_flags & DCACHE_LRU_LIST))) @@ -872,8 +880,25 @@ static inline bool fast_dput(struct dentry *dentry) DCACHE_DISCONNECTED | DCACHE_DONTCACHE; /* Nothing to do? Dropping the reference was all we needed? */ - if (d_flags == (DCACHE_REFERENCED | DCACHE_LRU_LIST) && !d_unhashed(dentry)) - return true; + if (d_flags == (DCACHE_REFERENCED | DCACHE_LRU_LIST) && !d_unhashed(dentry)) { + /* + * If the dentry is not negative dentry, return true directly, + * avoid holding dentry lock in the fast put path. + */ + if (dentry->d_inode) + return true; + /* + * If dentry is negative and need to be limitted, it maybe need + * to be killed, so shouldn't fast put and retain in memory. + */ + spin_lock(&dentry->d_lock); + if (likely(!limit_negative_dentry(dentry))) { + spin_unlock(&dentry->d_lock); + return true; + } + + goto dentry_locked; + } /* * Not the fast normal case? Get the lock. We've already decremented @@ -882,6 +907,7 @@ static inline bool fast_dput(struct dentry *dentry) */ spin_lock(&dentry->d_lock); +dentry_locked: /* * Did somebody else grab a reference to it in the meantime, and * we're no longer the last user after all? Alternatively, somebody -- 2.31.1

1 0

[PATCH OLK-5.10] net/rds: Fix UBSAN: array-index-out-of-bounds in rds_cmsg_recv
by Dong Chenchen 29 Jan '24

29 Jan '24

From: Sharath Srinivasan <sharath.srinivasan(a)oracle.com> mainline inclusion from mainline-v6.7 commit 13e788deb7348cc88df34bed736c3b3b9927ea52 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I8YC9I CVE: CVE-2024-23849 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Syzcaller UBSAN crash occurs in rds_cmsg_recv(), which reads inc->i_rx_lat_trace[j + 1] with index 4 (3 + 1), but with array size of 4 (RDS_RX_MAX_TRACES). Here 'j' is assigned from rs->rs_rx_trace[i] and in-turn from trace.rx_trace_pos[i] in rds_recv_track_latency(), with both arrays sized 3 (RDS_MSG_RX_DGRAM_TRACE_MAX). So fix the off-by-one bounds check in rds_recv_track_latency() to prevent a potential crash in rds_cmsg_recv(). Found by syzcaller: ================================================================= UBSAN: array-index-out-of-bounds in net/rds/recv.c:585:39 index 4 is out of range for type 'u64 [4]' CPU: 1 PID: 8058 Comm: syz-executor228 Not tainted 6.6.0-gd2f51b3516da #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 Call Trace: <TASK> __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x136/0x150 lib/dump_stack.c:106 ubsan_epilogue lib/ubsan.c:217 [inline] __ubsan_handle_out_of_bounds+0xd5/0x130 lib/ubsan.c:348 rds_cmsg_recv+0x60d/0x700 net/rds/recv.c:585 rds_recvmsg+0x3fb/0x1610 net/rds/recv.c:716 sock_recvmsg_nosec net/socket.c:1044 [inline] sock_recvmsg+0xe2/0x160 net/socket.c:1066 __sys_recvfrom+0x1b6/0x2f0 net/socket.c:2246 __do_sys_recvfrom net/socket.c:2264 [inline] __se_sys_recvfrom net/socket.c:2260 [inline] __x64_sys_recvfrom+0xe0/0x1b0 net/socket.c:2260 do_syscall_x64 arch/x86/entry/common.c:51 [inline] do_syscall_64+0x40/0x110 arch/x86/entry/common.c:82 entry_SYSCALL_64_after_hwframe+0x63/0x6b ================================================================== Fixes: 3289025aedc0 ("RDS: add receive message trace used by application") Reported-by: Chenyuan Yang <chenyuan0y(a)gmail.com> Closes: https://lore.kernel.org/linux-rdma/CALGdzuoVdq-wtQ4Az9iottBqC5cv9ZhcE5q8N7L… Signed-off-by: Sharath Srinivasan <sharath.srinivasan(a)oracle.com> Reviewed-by: Simon Horman <horms(a)kernel.org> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/rds/af_rds.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/rds/af_rds.c b/net/rds/af_rds.c index b239120dd9ca..0ec0ae148349 100644 --- a/net/rds/af_rds.c +++ b/net/rds/af_rds.c @@ -419,7 +419,7 @@ static int rds_recv_track_latency(struct rds_sock *rs, sockptr_t optval, rs->rs_rx_traces = trace.rx_traces; for (i = 0; i < rs->rs_rx_traces; i++) { - if (trace.rx_trace_pos[i] > RDS_MSG_RX_DGRAM_TRACE_MAX) { + if (trace.rx_trace_pos[i] >= RDS_MSG_RX_DGRAM_TRACE_MAX) { rs->rs_rx_traces = 0; return -EFAULT; } -- 2.25.1

2 1