September 2024 - Linux-stable-mirror

[PATCH] btrfs: drop the backref cache during relocation if we commit

by Josef Bacik

Since the inception of relocation we have maintained the backref cache across transaction commits, updating the backref cache with the new bytenr whenever we COW'ed blocks that were in the cache, and then updating their bytenr once we detected a transaction id change. This works as long as we're only ever modifying blocks, not changing the structure of the tree. However relocation does in fact change the structure of the tree. For example, if we are relocating a data extent, we will look up all the leaves that point to this data extent. We will then call do_relocation() on each of these leaves, which will COW down to the leaf and then update the file extent location. But, a key feature of do_relocation is the pending list. This is all the pending nodes that we modified when we updated the file extent item. We will then process all of these blocks via finish_pending_nodes, which calls do_relocation() on all of the nodes that led up to that leaf. The purpose of this is to make sure we don't break sharing unless we absolutely have to. Consider the case that we have 3 snapshots that all point to this leaf through the same nodes, the initial COW would have created a whole new path. If we did this for all 3 snapshots we would end up with 3x the number of nodes we had originally. To avoid this we will cycle through each of the snapshots that point to each of these nodes and update their pointers to point at the new nodes. Once we update the pointer to the new node we will drop the node we removed the link for and all of its children via btrfs_drop_subtree(). This is essentially just btrfs_drop_snapshot(), but for an arbitrary point in the snapshot. The problem with this is that we will never reflect this in the backref cache. If we do this btrfs_drop_snapshot() for a node that is in the backref tree, we will leave the node in the backref tree. This becomes a problem when we change the transid, as now the backref cache has entire subtrees that no longer exist, but exist as if they still are pointed to by the same roots. In the best case scenario you end up with "adding refs to an existing tree ref" errors from insert_inline_extent_backref(), where we attempt to link in nodes on roots that are no longer valid. Worst case you will double free some random block and re-use it when there's still references to the block. This is extremely subtle, and the consequences are quite bad. There isn't a way to make sure our backref cache is consistent between transid's. In order to fix this we need to simply evict the entire backref cache anytime we cross transid's. This reduces performance in that we have to rebuild this backref cache every time we change transid's, but fixes the bug. This has existed since relocation was added, and is a pretty critical bug. There's a lot more cleanup that can be done now that this functionality is going away, but this patch is as small as possible in order to fix the problem and make it easy for us to backport it to all the kernels it needs to be backported to. Followup series will dismantle more of this code and simplify relocation drastically to remove this functionality. We have a reproducer that reproduced the corruption within a few minutes of running. With this patch it survives several iterations/hours of running the reproducer. Fixes: 3fd0a5585eb9 ("Btrfs: Metadata ENOSPC handling for balance") Cc: stable(a)vger.kernel.org Signed-off-by: Josef Bacik <josef(a)toxicpanda.com> --- fs/btrfs/backref.c | 12 ++++--- fs/btrfs/relocation.c | 76 +++---------------------------------------- 2 files changed, 13 insertions(+), 75 deletions(-) diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index e2f478ecd7fd..f8e1d5b2c512 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -3179,10 +3179,14 @@ void btrfs_backref_release_cache(struct btrfs_backref_cache *cache) btrfs_backref_cleanup_node(cache, node); } - cache->last_trans = 0; - - for (i = 0; i < BTRFS_MAX_LEVEL; i++) - ASSERT(list_empty(&cache->pending[i])); + for (i = 0; i < BTRFS_MAX_LEVEL; i++) { + while (!list_empty(&cache->pending[i])) { + node = list_first_entry(&cache->pending[i], + struct btrfs_backref_node, + list); + btrfs_backref_cleanup_node(cache, node); + } + } ASSERT(list_empty(&cache->pending_edge)); ASSERT(list_empty(&cache->useless_node)); ASSERT(list_empty(&cache->changed)); diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index ea4ed85919ec..aaa9cac213f1 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -232,70 +232,6 @@ static struct btrfs_backref_node *walk_down_backref( return NULL; } -static void update_backref_node(struct btrfs_backref_cache *cache, - struct btrfs_backref_node *node, u64 bytenr) -{ - struct rb_node *rb_node; - rb_erase(&node->rb_node, &cache->rb_root); - node->bytenr = bytenr; - rb_node = rb_simple_insert(&cache->rb_root, node->bytenr, &node->rb_node); - if (rb_node) - btrfs_backref_panic(cache->fs_info, bytenr, -EEXIST); -} - -/* - * update backref cache after a transaction commit - */ -static int update_backref_cache(struct btrfs_trans_handle *trans, - struct btrfs_backref_cache *cache) -{ - struct btrfs_backref_node *node; - int level = 0; - - if (cache->last_trans == 0) { - cache->last_trans = trans->transid; - return 0; - } - - if (cache->last_trans == trans->transid) - return 0; - - /* - * detached nodes are used to avoid unnecessary backref - * lookup. transaction commit changes the extent tree. - * so the detached nodes are no longer useful. - */ - while (!list_empty(&cache->detached)) { - node = list_entry(cache->detached.next, - struct btrfs_backref_node, list); - btrfs_backref_cleanup_node(cache, node); - } - - while (!list_empty(&cache->changed)) { - node = list_entry(cache->changed.next, - struct btrfs_backref_node, list); - list_del_init(&node->list); - BUG_ON(node->pending); - update_backref_node(cache, node, node->new_bytenr); - } - - /* - * some nodes can be left in the pending list if there were - * errors during processing the pending nodes. - */ - for (level = 0; level < BTRFS_MAX_LEVEL; level++) { - list_for_each_entry(node, &cache->pending[level], list) { - BUG_ON(!node->pending); - if (node->bytenr == node->new_bytenr) - continue; - update_backref_node(cache, node, node->new_bytenr); - } - } - - cache->last_trans = 0; - return 1; -} - static bool reloc_root_is_dead(const struct btrfs_root *root) { /* @@ -551,9 +487,6 @@ static int clone_backref_node(struct btrfs_trans_handle *trans, struct btrfs_backref_edge *new_edge; struct rb_node *rb_node; - if (cache->last_trans > 0) - update_backref_cache(trans, cache); - rb_node = rb_simple_search(&cache->rb_root, src->commit_root->start); if (rb_node) { node = rb_entry(rb_node, struct btrfs_backref_node, rb_node); @@ -3698,10 +3631,11 @@ static noinline_for_stack int relocate_block_group(struct reloc_control *rc) break; } restart: - if (update_backref_cache(trans, &rc->backref_cache)) { - btrfs_end_transaction(trans); - trans = NULL; - continue; + if (rc->backref_cache.last_trans == 0) { + rc->backref_cache.last_trans = trans->transid; + } else if (rc->backref_cache.last_trans != trans->transid) { + btrfs_backref_release_cache(&rc->backref_cache); + rc->backref_cache.last_trans = trans->transid; } ret = find_next_extent(rc, path, &key); -- 2.43.0

6 months, 3 weeks

2
5
0 0

[PATCH 6.1] mm/memory_hotplug: prevent accessing by index=-1

by Mikhail Lobanov

From: Anastasia Belova <abelova(a)astralinux.ru> commit 5958d35917e1296f46dfc8b8c959732efd6d8d5d upstream. nid may be equal to NUMA_NO_NODE=-1. Prevent accessing node_data array by invalid index with check for nid. Found by Linux Verification Center (linuxtesting.org) with SVACE. Fixes: e83a437faa62 ("mm/memory_hotplug: introduce "auto-movable" online policy") Signed-off-by: Anastasia Belova <abelova(a)astralinux.ru> Acked-by: David Hildenbrand <david(a)redhat.com> Acked-by: Oscar Salvador <osalvador(a)suse.de> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Mikhail Lobanov <m.lobanov(a)rosalinux.ru> --- mm/memory_hotplug.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index dc17618bad8b..90f0cc9a298a 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -783,7 +783,6 @@ static bool auto_movable_can_online_movable(int nid, struct memory_group *group, unsigned long kernel_early_pages, movable_pages; struct auto_movable_group_stats group_stats = {}; struct auto_movable_stats stats = {}; - pg_data_t *pgdat = NODE_DATA(nid); struct zone *zone; int i; @@ -794,6 +793,8 @@ static bool auto_movable_can_online_movable(int nid, struct memory_group *group, auto_movable_stats_account_zone(&stats, zone); } else { for (i = 0; i < MAX_NR_ZONES; i++) { + pg_data_t *pgdat = NODE_DATA(nid); + zone = pgdat->node_zones + i; if (populated_zone(zone)) auto_movable_stats_account_zone(&stats, zone); -- 2.43.0

6 months, 3 weeks

1
0
0 0

[PATCH 0/4] binder: several fixes for frozen notification

by Carlos Llamas

These are all fixes for the frozen notification patch [1], which as of today hasn't landed in mainline yet. As such, this patchset is rebased on top of the char-misc-next branch. [1] https://lore.kernel.org/all/20240709070047.4055369-2-yutingtseng@google.com/ Cc: stable(a)vger.kernel.org Cc: Yu-Ting Tseng <yutingtseng(a)google.com> Cc: Alice Ryhl <aliceryhl(a)google.com> Cc: Todd Kjos <tkjos(a)google.com> Cc: Martijn Coenen <maco(a)google.com> Cc: Arve Hjønnevåg <arve(a)android.com> Cc: Viktor Martensson <vmartensson(a)google.com> Carlos Llamas (4): binder: fix node UAF in binder_add_freeze_work() binder: fix OOB in binder_add_freeze_work() binder: fix freeze UAF in binder_release_work() binder: fix BINDER_WORK_FROZEN_BINDER debug logs drivers/android/binder.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) -- 2.46.0.792.g87dc391469-goog

6 months, 3 weeks

4
20
0 0

[PATCH v6.6] nvme-pci: qdepth 1 quirk

by Gagniuc, Alexandru

Please consider adding the following commit to v6.6: 83bdfcbdbe5d ("nvme-pci: qdepth 1 quirk") This resolves filesystem corruption and random drive drop-outs on machines with O2micro NVME drives, including a number of HP Thinclients. Alex

6 months, 3 weeks

1
0
0 0

[PATCH 5.15] Revert "wifi: cfg80211: check wiphy mutex is held for wdev mutex"

by Fedor Pchelkin

This reverts commit 89795eeba6d13b5ba432425dd43c34c66f2cebde. The reverted commit is based on implementation of wiphy locking that isn't planned to redo on a stable kernel, so revert it to avoid warnings in different parts of wireless stack where wdev_lock() is held, like: WARNING: CPU: 0 PID: 4296 at net/wireless/core.h:220 cfg80211_is_all_idle net/wireless/sme.c:662 [inline] WARNING: CPU: 0 PID: 4296 at net/wireless/core.h:220 disconnect_work+0x236/0x320 net/wireless/sme.c:676 Modules linked in: CPU: 0 PID: 4296 Comm: kworker/0:4 Not tainted 5.15.166-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/06/2024 Workqueue: events disconnect_work RIP: 0010:wdev_lock net/wireless/core.h:220 [inline] Call Trace: <TASK> process_one_work+0x8a1/0x10c0 kernel/workqueue.c:2310 worker_thread+0xaca/0x1280 kernel/workqueue.c:2457 kthread+0x3f6/0x4f0 kernel/kthread.c:334 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:287 Signed-off-by: Fedor Pchelkin <pchelkin(a)ispras.ru> --- net/wireless/core.h | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/net/wireless/core.h b/net/wireless/core.h index be186b5a15f3..1720abf36f92 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -217,7 +217,6 @@ void cfg80211_register_wdev(struct cfg80211_registered_device *rdev, static inline void wdev_lock(struct wireless_dev *wdev) __acquires(wdev) { - lockdep_assert_held(&wdev->wiphy->mtx); mutex_lock(&wdev->mtx); __acquire(wdev->mtx); } @@ -225,16 +224,11 @@ static inline void wdev_lock(struct wireless_dev *wdev) static inline void wdev_unlock(struct wireless_dev *wdev) __releases(wdev) { - lockdep_assert_held(&wdev->wiphy->mtx); __release(wdev->mtx); mutex_unlock(&wdev->mtx); } -static inline void ASSERT_WDEV_LOCK(struct wireless_dev *wdev) -{ - lockdep_assert_held(&wdev->wiphy->mtx); - lockdep_assert_held(&wdev->mtx); -} +#define ASSERT_WDEV_LOCK(wdev) lockdep_assert_held(&(wdev)->mtx) static inline bool cfg80211_has_monitors_only(struct cfg80211_registered_device *rdev) { -- 2.39.5

6 months, 3 weeks

1
0
0 0

[PATCH 6.1] Revert "wifi: cfg80211: check wiphy mutex is held for wdev mutex"

by Fedor Pchelkin

This reverts commit 19d13ec00a8b1d60c5cc06bd0006b91d5bd8d46f. The reverted commit is based on implementation of wiphy locking that isn't planned to redo on a stable kernel, so revert it to avoid warnings in different parts of wireless stack where wdev_lock() is held, like: WARNING: CPU: 0 PID: 4727 at net/wireless/core.h:231 cfg80211_is_all_idle net/wireless/sme.c:692 [inline] WARNING: CPU: 0 PID: 4727 at net/wireless/core.h:231 disconnect_work+0x246/0x340 net/wireless/sme.c:706 Modules linked in: CPU: 0 PID: 4727 Comm: kworker/0:8 Not tainted 6.1.111-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/13/2024 Workqueue: events disconnect_work RIP: 0010:wdev_lock net/wireless/core.h:231 [inline] Call Trace: <TASK> process_one_work+0x8a9/0x11d0 kernel/workqueue.c:2292 worker_thread+0xa47/0x1200 kernel/workqueue.c:2439 kthread+0x28d/0x320 kernel/kthread.c:376 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295 Signed-off-by: Fedor Pchelkin <pchelkin(a)ispras.ru> --- net/wireless/core.h | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/net/wireless/core.h b/net/wireless/core.h index 8118b8614ac6..ee980965a7cf 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -228,7 +228,6 @@ void cfg80211_register_wdev(struct cfg80211_registered_device *rdev, static inline void wdev_lock(struct wireless_dev *wdev) __acquires(wdev) { - lockdep_assert_held(&wdev->wiphy->mtx); mutex_lock(&wdev->mtx); __acquire(wdev->mtx); } @@ -236,16 +235,11 @@ static inline void wdev_lock(struct wireless_dev *wdev) static inline void wdev_unlock(struct wireless_dev *wdev) __releases(wdev) { - lockdep_assert_held(&wdev->wiphy->mtx); __release(wdev->mtx); mutex_unlock(&wdev->mtx); } -static inline void ASSERT_WDEV_LOCK(struct wireless_dev *wdev) -{ - lockdep_assert_held(&wdev->wiphy->mtx); - lockdep_assert_held(&wdev->mtx); -} +#define ASSERT_WDEV_LOCK(wdev) lockdep_assert_held(&(wdev)->mtx) static inline bool cfg80211_has_monitors_only(struct cfg80211_registered_device *rdev) { -- 2.39.5

6 months, 3 weeks

1
0
0 0

Re: [PATCH stable 6.6] Revert "wifi: cfg80211: check wiphy mutex is held for wdev mutex"

by Fedor Pchelkin

The patch was also backported to 5.15 and 6.1 stables where it's causing the same splats. https://lore.kernel.org/stable/20240918-d150ee61d40b8f327d65bbf3-pchelkin@i… Need to revert it there, too. Should I send the explicit reverts to the mailing list? -- Thank you, Fedor

6 months, 3 weeks

1
1
0 0

[PATCH] mm: migrate: annotate data-race in migrate_folio_unmap()

by Jeongjun Park

I found a report from syzbot [1] This report shows that the value can be changed, but in reality, the value of __folio_set_movable() cannot be changed because it holds the folio refcount. Therefore, it is appropriate to add an annotate to make KCSAN ignore that data-race. [1] ================================================================== BUG: KCSAN: data-race in __filemap_remove_folio / migrate_pages_batch write to 0xffffea0004b81dd8 of 8 bytes by task 6348 on cpu 0: page_cache_delete mm/filemap.c:153 [inline] __filemap_remove_folio+0x1ac/0x2c0 mm/filemap.c:233 filemap_remove_folio+0x6b/0x1f0 mm/filemap.c:265 truncate_inode_folio+0x42/0x50 mm/truncate.c:178 shmem_undo_range+0x25b/0xa70 mm/shmem.c:1028 shmem_truncate_range mm/shmem.c:1144 [inline] shmem_evict_inode+0x14d/0x530 mm/shmem.c:1272 evict+0x2f0/0x580 fs/inode.c:731 iput_final fs/inode.c:1883 [inline] iput+0x42a/0x5b0 fs/inode.c:1909 dentry_unlink_inode+0x24f/0x260 fs/dcache.c:412 __dentry_kill+0x18b/0x4c0 fs/dcache.c:615 dput+0x5c/0xd0 fs/dcache.c:857 __fput+0x3fb/0x6d0 fs/file_table.c:439 ____fput+0x1c/0x30 fs/file_table.c:459 task_work_run+0x13a/0x1a0 kernel/task_work.c:228 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline] exit_to_user_mode_loop kernel/entry/common.c:114 [inline] exit_to_user_mode_prepare include/linux/entry-common.h:328 [inline] __syscall_exit_to_user_mode_work kernel/entry/common.c:207 [inline] syscall_exit_to_user_mode+0xbe/0x130 kernel/entry/common.c:218 do_syscall_64+0xd6/0x1c0 arch/x86/entry/common.c:89 entry_SYSCALL_64_after_hwframe+0x77/0x7f read to 0xffffea0004b81dd8 of 8 bytes by task 6342 on cpu 1: __folio_test_movable include/linux/page-flags.h:699 [inline] migrate_folio_unmap mm/migrate.c:1199 [inline] migrate_pages_batch+0x24c/0x1940 mm/migrate.c:1797 migrate_pages_sync mm/migrate.c:1963 [inline] migrate_pages+0xff1/0x1820 mm/migrate.c:2072 do_mbind mm/mempolicy.c:1390 [inline] kernel_mbind mm/mempolicy.c:1533 [inline] __do_sys_mbind mm/mempolicy.c:1607 [inline] __se_sys_mbind+0xf76/0x1160 mm/mempolicy.c:1603 __x64_sys_mbind+0x78/0x90 mm/mempolicy.c:1603 x64_sys_call+0x2b4d/0x2d60 arch/x86/include/generated/asm/syscalls_64.h:238 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xc9/0x1c0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f value changed: 0xffff888127601078 -> 0x0000000000000000 Reported-by: syzbot <syzkaller(a)googlegroups.com> Cc: stable(a)vger.kernel.org Fixes: 7e2a5e5ab217 ("mm: migrate: use __folio_test_movable()") Signed-off-by: Jeongjun Park <aha310510(a)gmail.com> --- mm/migrate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/migrate.c b/mm/migrate.c index 923ea80ba744..368ab3878fa6 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1118,7 +1118,7 @@ static int migrate_folio_unmap(new_folio_t get_new_folio, int rc = -EAGAIN; int old_page_state = 0; struct anon_vma *anon_vma = NULL; - bool is_lru = !__folio_test_movable(src); + bool is_lru = data_race(!__folio_test_movable(src)); bool locked = false; bool dst_locked = false; --

6 months, 3 weeks

2
1
0 0

[PATCH v3 03/10] media: qcom: camss: Fix potential crash if domain attach fails

by Vikram Sharma

Fix a potential crash in camss by ensuring detach is skipped if attach is unsuccessful. Fixes: d89751c61279 ("media: qcom: camss: Add support for named power-domains") CC: stable(a)vger.kernel.org Signed-off-by: Vikram Sharma <quic_vikramsa(a)quicinc.com> --- drivers/media/platform/qcom/camss/camss.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/media/platform/qcom/camss/camss.c b/drivers/media/platform/qcom/camss/camss.c index d64985ca6e88..b6658df37709 100644 --- a/drivers/media/platform/qcom/camss/camss.c +++ b/drivers/media/platform/qcom/camss/camss.c @@ -2131,8 +2131,7 @@ static int camss_configure_pd(struct camss *camss) camss->genpd = dev_pm_domain_attach_by_name(camss->dev, camss->res->pd_name); if (IS_ERR(camss->genpd)) { - ret = PTR_ERR(camss->genpd); - goto fail_pm; + return PTR_ERR(camss->genpd); } } @@ -2149,7 +2148,7 @@ static int camss_configure_pd(struct camss *camss) ret = -ENODEV; else ret = PTR_ERR(camss->genpd); - goto fail_pm; + return ret; } camss->genpd_link = device_link_add(camss->dev, camss->genpd, DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME | -- 2.25.1

6 months, 3 weeks

1
0
0 0

[PATCH] arm64: dts: rockchip: Move L3 cache under CPUs in RK356x SoC dtsi

by Dragan Simic

Move the "l3_cache" node under the "cpus" node in the dtsi file for Rockchip RK356x SoCs. There's no need for this cache node to be at the higher level. Fixes: 8612169a05c5 ("arm64: dts: rockchip: Add cache information to the SoC dtsi for RK356x") Cc: stable(a)vger.kernel.org Signed-off-by: Dragan Simic <dsimic(a)manjaro.org> --- arch/arm64/boot/dts/rockchip/rk356x.dtsi | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/arch/arm64/boot/dts/rockchip/rk356x.dtsi b/arch/arm64/boot/dts/rockchip/rk356x.dtsi index 4690be841a1c..9f7136e5d553 100644 --- a/arch/arm64/boot/dts/rockchip/rk356x.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk356x.dtsi @@ -113,19 +113,19 @@ cpu3: cpu@300 { d-cache-sets = <128>; next-level-cache = <&l3_cache>; }; - }; - /* - * There are no private per-core L2 caches, but only the - * L3 cache that appears to the CPU cores as L2 caches - */ - l3_cache: l3-cache { - compatible = "cache"; - cache-level = <2>; - cache-unified; - cache-size = <0x80000>; - cache-line-size = <64>; - cache-sets = <512>; + /* + * There are no private per-core L2 caches, but only the + * L3 cache that appears to the CPU cores as L2 caches + */ + l3_cache: l3-cache { + compatible = "cache"; + cache-level = <2>; + cache-unified; + cache-size = <0x80000>; + cache-line-size = <64>; + cache-sets = <512>; + }; }; cpu0_opp_table: opp-table-0 {

6 months, 3 weeks

3
5
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror September 2024