CVE-2023-54158

In the Linux kernel, the following vulnerability has been resolved: btrfs: don't free qgroup space unless specified Boris noticed in his simple quotas testing that he was getting a leak with Sweet Tea's change to subvol create that stopped doing a transaction commit. This was just a side effect of that change. In the delayed inode code we have an optimization that will free extra reservations if we think we can pack a dir item into an already modified leaf. Previously this wouldn't be triggered in the subvolume create case because we'd commit the transaction, it was still possible but much harder to trigger. It could actually be triggered if we did a mkdir && subvol create with qgroups enabled. This occurs because in btrfs_insert_delayed_dir_index(), which gets called when we're adding the dir item, we do the following: btrfs_block_rsv_release(fs_info, trans->block_rsv, bytes, NULL); if we're able to skip reserving space. The problem here is that trans->block_rsv points at the temporary block rsv for the subvolume create, which has qgroup reservations in the block rsv. This is a problem because btrfs_block_rsv_release() will do the following: if (block_rsv->qgroup_rsv_reserved >= block_rsv->qgroup_rsv_size) { qgroup_to_release = block_rsv->qgroup_rsv_reserved - block_rsv->qgroup_rsv_size; block_rsv->qgroup_rsv_reserved = block_rsv->qgroup_rsv_size; } The temporary block rsv just has ->qgroup_rsv_reserved set, ->qgroup_rsv_size == 0. The optimization in btrfs_insert_delayed_dir_index() sets ->qgroup_rsv_reserved = 0. Then later on when we call btrfs_subvolume_release_metadata() which has btrfs_block_rsv_release(fs_info, rsv, (u64)-1, &qgroup_to_release); btrfs_qgroup_convert_reserved_meta(root, qgroup_to_release); qgroup_to_release is set to 0, and we do not convert the reserved metadata space. The problem here is that the block rsv code has been unconditionally messing with ->qgroup_rsv_reserved, because the main place this is used is delalloc, and any time we call btrfs_block_rsv_release() we do it with qgroup_to_release set, and thus do the proper accounting. The subvolume code is the only other code that uses the qgroup reservation stuff, but it's intermingled with the above optimization, and thus was getting its reservation freed out from underneath it and thus leaking the reserved space. The solution is to simply not mess with the qgroup reservations if we don't have qgroup_to_release set. This works with the existing code as anything that messes with the delalloc reservations always have qgroup_to_release set. This fixes the leak that Boris was observing.

In the Linux kernel, the following vulnerability has been resolved:

btrfs: don't free qgroup space unless specified

Boris noticed in his simple quotas testing that he was getting a leak with Sweet Tea's change to subvol create that stopped doing a transaction commit. This was just a side effect of that change.

In the delayed inode code we have an optimization that will free extra reservations if we think we can pack a dir item into an already modified leaf. Previously this wouldn't be triggered in the subvolume create case because we'd commit the transaction, it was still possible but much harder to trigger. It could actually be triggered if we did a mkdir && subvol create with qgroups enabled.

This occurs because in btrfs_insert_delayed_dir_index(), which gets called when we're adding the dir item, we do the following:

btrfs_block_rsv_release(fs_info, trans->block_rsv, bytes, NULL);

if we're able to skip reserving space.

The problem here is that trans->block_rsv points at the temporary block rsv for the subvolume create, which has qgroup reservations in the block rsv.

This is a problem because btrfs_block_rsv_release() will do the following:

if (block_rsv->qgroup_rsv_reserved >= block_rsv->qgroup_rsv_size) { qgroup_to_release = block_rsv->qgroup_rsv_reserved - block_rsv->qgroup_rsv_size; block_rsv->qgroup_rsv_reserved = block_rsv->qgroup_rsv_size; }

The temporary block rsv just has ->qgroup_rsv_reserved set, ->qgroup_rsv_size == 0. The optimization in btrfs_insert_delayed_dir_index() sets ->qgroup_rsv_reserved = 0. Then later on when we call btrfs_subvolume_release_metadata() which has

btrfs_block_rsv_release(fs_info, rsv, (u64)-1, &qgroup_to_release); btrfs_qgroup_convert_reserved_meta(root, qgroup_to_release);

qgroup_to_release is set to 0, and we do not convert the reserved metadata space.

The problem here is that the block rsv code has been unconditionally messing with ->qgroup_rsv_reserved, because the main place this is used is delalloc, and any time we call btrfs_block_rsv_release() we do it with qgroup_to_release set, and thus do the proper accounting.

The subvolume code is the only other code that uses the qgroup reservation stuff, but it's intermingled with the above optimization, and thus was getting its reservation freed out from underneath it and thus leaking the reserved space.

The solution is to simply not mess with the qgroup reservations if we don't have qgroup_to_release set. This works with the existing code as anything that messes with the delalloc reservations always have qgroup_to_release set. This fixes the leak that Boris was observing.

In the Linux kernel, the following vulnerability has been resolved:

btrfs: don't free qgroup space unless specified

Boris noticed in his simple quotas testing that he was getting a leak with Sweet Tea's change to subvol create that stopped doing a transaction commit. This was just a side effect of that change.

In the delayed inode code we have an optimization that will free extra reservations if we think we can pack a dir item into an already modified leaf. Previously this wouldn't be triggered in the subvolume create case because we'd commit the transaction, it was still possible but much harder to trigger. It could actually be triggered if we did a mkdir && subvol create with qgroups enabled.

This occurs because in btrfs_insert_delayed_dir_index(), which gets called when we're adding the dir item, we do the following:

btrfs_block_rsv_release(fs_info, trans->block_rsv, bytes, NULL);

if we're able to skip reserving space.

The problem here is that trans->block_rsv points at the temporary block rsv for the subvolume create, which has qgroup reservations in the block rsv.

This is a problem because btrfs_block_rsv_release() will do the following:

if (block_rsv->qgroup_rsv_reserved >= block_rsv->qgroup_rsv_size) { qgroup_to_release = block_rsv->qgroup_rsv_reserved - block_rsv->qgroup_rsv_size; block_rsv->qgroup_rsv_reserved = block_rsv->qgroup_rsv_size; }

The temporary block rsv just has ->qgroup_rsv_reserved set, ->qgroup_rsv_size == 0. The optimization in btrfs_insert_delayed_dir_index() sets ->qgroup_rsv_reserved = 0. Then later on when we call btrfs_subvolume_release_metadata() which has

btrfs_block_rsv_release(fs_info, rsv, (u64)-1, &qgroup_to_release); btrfs_qgroup_convert_reserved_meta(root, qgroup_to_release);

qgroup_to_release is set to 0, and we do not convert the reserved metadata space.

The problem here is that the block rsv code has been unconditionally messing with ->qgroup_rsv_reserved, because the main place this is used is delalloc, and any time we call btrfs_block_rsv_release() we do it with qgroup_to_release set, and thus do the proper accounting.

The subvolume code is the only other code that uses the qgroup reservation stuff, but it's intermingled with the above optimization, and thus was getting its reservation freed out from underneath it and thus leaking the reserved space.

The solution is to simply not mess with the qgroup reservations if we don't have qgroup_to_release set. This works with the existing code as anything that messes with the delalloc reservations always have qgroup_to_release set. This fixes the leak that Boris was observing.

In the Linux kernel, the following vulnerability has been resolved:

btrfs: don't free qgroup space unless specified

Boris noticed in his simple quotas testing that he was getting a leak with Sweet Tea's change to subvol create that stopped doing a transaction commit. This was just a side effect of that change.

In the delayed inode code we have an optimization that will free extra reservations if we think we can pack a dir item into an already modified leaf. Previously this wouldn't be triggered in the subvolume create case because we'd commit the transaction, it was still possible but much harder to trigger. It could actually be triggered if we did a mkdir && subvol create with qgroups enabled.

This occurs because in btrfs_insert_delayed_dir_index(), which gets called when we're adding the dir item, we do the following:

btrfs_block_rsv_release(fs_info, trans->block_rsv, bytes, NULL);

if we're able to skip reserving space.

The problem here is that trans->block_rsv points at the temporary block rsv for the subvolume create, which has qgroup reservations in the block rsv.

This is a problem because btrfs_block_rsv_release() will do the following:

if (block_rsv->qgroup_rsv_reserved >= block_rsv->qgroup_rsv_size) { qgroup_to_release = block_rsv->qgroup_rsv_reserved - block_rsv->qgroup_rsv_size; block_rsv->qgroup_rsv_reserved = block_rsv->qgroup_rsv_size; }

The temporary block rsv just has ->qgroup_rsv_reserved set, ->qgroup_rsv_size == 0. The optimization in btrfs_insert_delayed_dir_index() sets ->qgroup_rsv_reserved = 0. Then later on when we call btrfs_subvolume_release_metadata() which has

btrfs_block_rsv_release(fs_info, rsv, (u64)-1, &qgroup_to_release); btrfs_qgroup_convert_reserved_meta(root, qgroup_to_release);

qgroup_to_release is set to 0, and we do not convert the reserved metadata space.

The problem here is that the block rsv code has been unconditionally messing with ->qgroup_rsv_reserved, because the main place this is used is delalloc, and any time we call btrfs_block_rsv_release() we do it with qgroup_to_release set, and thus do the proper accounting.

The subvolume code is the only other code that uses the qgroup reservation stuff, but it's intermingled with the above optimization, and thus was getting its reservation freed out from underneath it and thus leaking the reserved space.

The solution is to simply not mess with the qgroup reservations if we don't have qgroup_to_release set. This works with the existing code as anything that messes with the delalloc reservations always have qgroup_to_release set. This fixes the leak that Boris was observing.

In the Linux kernel, the following vulnerability has been resolved:

btrfs: don't free qgroup space unless specified

Boris noticed in his simple quotas testing that he was getting a leak with Sweet Tea's change to subvol create that stopped doing a transaction commit. This was just a side effect of that change.

In the delayed inode code we have an optimization that will free extra reservations if we think we can pack a dir item into an already modified leaf. Previously this wouldn't be triggered in the subvolume create case because we'd commit the transaction, it was still possible but much harder to trigger. It could actually be triggered if we did a mkdir && subvol create with qgroups enabled.

This occurs because in btrfs_insert_delayed_dir_index(), which gets called when we're adding the dir item, we do the following:

btrfs_block_rsv_release(fs_info, trans->block_rsv, bytes, NULL);

if we're able to skip reserving space.

The problem here is that trans->block_rsv points at the temporary block rsv for the subvolume create, which has qgroup reservations in the block rsv.

This is a problem because btrfs_block_rsv_release() will do the following:

if (block_rsv->qgroup_rsv_reserved >= block_rsv->qgroup_rsv_size) { qgroup_to_release = block_rsv->qgroup_rsv_reserved - block_rsv->qgroup_rsv_size; block_rsv->qgroup_rsv_reserved = block_rsv->qgroup_rsv_size; }

The temporary block rsv just has ->qgroup_rsv_reserved set, ->qgroup_rsv_size == 0. The optimization in btrfs_insert_delayed_dir_index() sets ->qgroup_rsv_reserved = 0. Then later on when we call btrfs_subvolume_release_metadata() which has

btrfs_block_rsv_release(fs_info, rsv, (u64)-1, &qgroup_to_release); btrfs_qgroup_convert_reserved_meta(root, qgroup_to_release);

qgroup_to_release is set to 0, and we do not convert the reserved metadata space.

The problem here is that the block rsv code has been unconditionally messing with ->qgroup_rsv_reserved, because the main place this is used is delalloc, and any time we call btrfs_block_rsv_release() we do it with qgroup_to_release set, and thus do the proper accounting.

The subvolume code is the only other code that uses the qgroup reservation stuff, but it's intermingled with the above optimization, and thus was getting its reservation freed out from underneath it and thus leaking the reserved space.

The solution is to simply not mess with the qgroup reservations if we don't have qgroup_to_release set. This works with the existing code as anything that messes with the delalloc reservations always have qgroup_to_release set. This fixes the leak that Boris was observing.

In the Linux kernel, the following vulnerability has been resolved:

btrfs: don't free qgroup space unless specified

Boris noticed in his simple quotas testing that he was getting a leak with Sweet Tea's change to subvol create that stopped doing a transaction commit. This was just a side effect of that change.

In the delayed inode code we have an optimization that will free extra reservations if we think we can pack a dir item into an already modified leaf. Previously this wouldn't be triggered in the subvolume create case because we'd commit the transaction, it was still possible but much harder to trigger. It could actually be triggered if we did a mkdir && subvol create with qgroups enabled.

This occurs because in btrfs_insert_delayed_dir_index(), which gets called when we're adding the dir item, we do the following:

btrfs_block_rsv_release(fs_info, trans->block_rsv, bytes, NULL);

if we're able to skip reserving space.

The problem here is that trans->block_rsv points at the temporary block rsv for the subvolume create, which has qgroup reservations in the block rsv.

This is a problem because btrfs_block_rsv_release() will do the following:

if (block_rsv->qgroup_rsv_reserved >= block_rsv->qgroup_rsv_size) { qgroup_to_release = block_rsv->qgroup_rsv_reserved - block_rsv->qgroup_rsv_size; block_rsv->qgroup_rsv_reserved = block_rsv->qgroup_rsv_size; }

The temporary block rsv just has ->qgroup_rsv_reserved set, ->qgroup_rsv_size == 0. The optimization in btrfs_insert_delayed_dir_index() sets ->qgroup_rsv_reserved = 0. Then later on when we call btrfs_subvolume_release_metadata() which has

btrfs_block_rsv_release(fs_info, rsv, (u64)-1, &qgroup_to_release); btrfs_qgroup_convert_reserved_meta(root, qgroup_to_release);

qgroup_to_release is set to 0, and we do not convert the reserved metadata space.

The problem here is that the block rsv code has been unconditionally messing with ->qgroup_rsv_reserved, because the main place this is used is delalloc, and any time we call btrfs_block_rsv_release() we do it with qgroup_to_release set, and thus do the proper accounting.

The subvolume code is the only other code that uses the qgroup reservation stuff, but it's intermingled with the above optimization, and thus was getting its reservation freed out from underneath it and thus leaking the reserved space.

The solution is to simply not mess with the qgroup reservations if we don't have qgroup_to_release set. This works with the existing code as anything that messes with the delalloc reservations always have qgroup_to_release set. This fixes the leak that Boris was observing.

In the Linux kernel, the following vulnerability has been resolved: btrfs: don't free qgroup space unless specified Boris noticed in his simple quotas testing that he was getting a leak with Sweet Tea's change to subvol create that stopped doing a transaction commit. This was just a side effect of that change. In the delayed inode code we have an optimization that will free extra reservations if we think we can pack a dir item into an already modified leaf. Previously this wouldn't be triggered in the subvolume create case because we'd commit the transaction, it was still possible but much harder to trigger. It could actually be triggered if we did a mkdir && subvol create with qgroups enabled. This occurs because in btrfs_insert_delayed_dir_index(), which gets called when we're adding the dir item, we do the following: btrfs_block_rsv_release(fs_info, trans->block_rsv, bytes, NULL); if we're able to skip reserving space. The problem here is that trans->block_rsv points at the temporary block rsv for the subvolume create, which has qgroup reservations in the block rsv. This is a problem because btrfs_block_rsv_release() will do the following: if (block_rsv->qgroup_rsv_reserved >= block_rsv->qgroup_rsv_size) { qgroup_to_release = block_rsv->qgroup_rsv_reserved - block_rsv->qgroup_rsv_size; block_rsv->qgroup_rsv_reserved = block_rsv->qgroup_rsv_size; } The temporary block rsv just has ->qgroup_rsv_reserved set, ->qgroup_rsv_size == 0. The optimization in btrfs_insert_delayed_dir_index() sets ->qgroup_rsv_reserved = 0. Then later on when we call btrfs_subvolume_release_metadata() which has btrfs_block_rsv_release(fs_info, rsv, (u64)-1, &qgroup_to_release); btrfs_qgroup_convert_reserved_meta(root, qgroup_to_release); qgroup_to_release is set to 0, and we do not convert the reserved metadata space. The problem here is that the block rsv code has been unconditionally messing with ->qgroup_rsv_reserved, because the main place this is used is delalloc, and any time we call btrfs_block_rsv_release() we do it with qgroup_to_release set, and thus do the proper accounting. The subvolume code is the only other code that uses the qgroup reservation stuff, but it's intermingled with the above optimization, and thus was getting its reservation freed out from underneath it and thus leaking the reserved space. The solution is to simply not mess with the qgroup reservations if we don't have qgroup_to_release set. This works with the existing code as anything that messes with the delalloc reservations always have qgroup_to_release set. This fixes the leak that Boris was observing.

The SUSE Linux Enterprise 12 SP5 kernel was updated to fix various security issues:

The following security issues were fixed:

• CVE-2023-54158: btrfs: don't free qgroup space unless specified (bsc#1256076).
• CVE-2023-54180: btrfs: handle case when repair happens with dev-replace (bsc#1255995).
• CVE-2025-68358: btrfs: fix racy bitfield write in btrfs_clear_space_info_full() (bsc#1255531).
• CVE-2026-43299: btrfs: do not ASSERT() when the fs flips RO inside btrfs_repair_io_failure() (bsc#1264851).
• CVE-2026-63948: Bluetooth: L2CAP: fix chan ref leak in l2cap_chan_timeout() on !conn (bsc#1272670).
• CVE-2026-63983: net: Introduce skb tc depth field to track packet loops (bsc#1273040).
• CVE-2026-64012: net/sched: sch_sfb: Replace direct dequeue call with peek and qdisc_dequeue_peeked (bsc#1273764).
• CVE-2026-64280: fpga: dfl-afu: validate DMA mapping length in afu_dma_map_region() (bsc#1273599).
• CVE-2026-64422: net: ipv4: bound TCP reordering sysctl writes and MTU probe sizes (bsc#1273290).
• CVE-2026-64456: hwrng: virtio: clamp device-reported used.len at copy_data() (bsc#1273288).
• CVE-2026-64593: btrfs: do not trim a device which is not writeable (bsc#1274497).
• CVE-2026-68106: drm/amdgpu: fix division by zero with invalid uvd dimensions (bsc#1274853).
• CVE-2026-68233: drm/vc4: Shut down BO cache timer before teardown (bsc#1275644).
• CVE-2026-68236: drm/amd/display: set new_stream to NULL after release (bsc#1275659).
• CVE-2026-68303: drm/vc4: hvs/v3d: Fix null dereference in unbind (bsc#1275080).
• CVE-2026-68323: tipc: serialize udp bearer replicast list updates (bsc#1274653).
• CVE-2026-68430: drm/amdgpu/gfx8: drop unecessary BUG_ON() (bsc#1276273).
• CVE-2026-72015: fs/resctrl: Fix double-add of pseudo-locked region's RMID to free list (bsc#1276791).
• CVE-2026-72236: s390/perf_cpum_cf: Add missing array_index_nospec() to __hw_perf_event_init() (bsc#1277614).
• CVE-2026-72319: ipvs: ensure inner headers in ICMP errors are in headroom (bsc#1275964).
• CVE-2026-72476: dmaengine: Fix possible use after free (bsc#1276844).
• CVE-2026-72493: net: serialize netif_running() check in enqueue_to_backlog() (bsc#1276951).
• CVE-2026-74408: wifi: ath9k: fix OOB access from firmware tx status queue ID (bsc#1276387).
• CVE-2026-74544: net/sched: cls_u32: validate offshift to prevent shift-out-of-bounds (bsc#1277894).
• CVE-2026-74578: crypto: algif_skcipher - force synchronous processing on trees without ctx->state (bsc#1275564).
• CVE-2026-74583: net/sched: cls_route: fix fastmap use-after-free on filter (bsc#1277889).
• CVE-2026-74587: sctp: fix use-after-free of cached ASCONF chunk (bsc#1277821).
• CVE-2026-74653: serial: 8250_of: clear stuck empty-FIFO RX-timeout on LPC32xx (bsc#1277310).
• CVE-2026-74675: vt: stabilize tty reference in kbd_keycode with tty_port_tty_get (bsc#1278313).
• CVE-2026-80977: net: skbuff: don't touch shared zerocopy state in skb_tx_error() (bsc#1280259).
• CVE-2026-89542: SUNRPC: harden gss_krb5_unwrap_v2 against short tokens (bsc#1280387).
• CVE-2026-89726: lib/ucs2_string.c: fix out-of-bounds read in ucs2_strnlen() (bsc#1281058).
• CVE-2026-89778: isofs: fix out-of-bounds page array access on empty zisofs block (bsc#1280966).
• CVE-2026-89987: mm/huge_memory: transfer the pmd dirty bit to the folio on zap (bsc#1281819).
• CVE-2026-90011: scsi: target: iscsi: Reserve a terminator byte for the login payload (bsc#1281280).
• CVE-2026-90283: hugetlbfs: release subpool on fill_super failure (bsc#1281771).
• CVE-2026-93252: ocfs2: make ocfs2_set_acl() static (bsc#1282784).
• CVE-2026-93287: i2c: smbus: reject oversized block transfers in the common path (bsc#1282744).
• CVE-2026-93800: btrfs: fix use-after-free on reloc root after error in insert_dirty_subvol() (bsc#1282808).
• CVE-2026-93820: PCI: rockchip: Protect root bus removal with rescan lock (bsc#1282755).
• CVE-2026-93822: PCI: iproc: Protect root bus removal with rescan lock (bsc#1282762).
• CVE-2026-97416: btrfs: balance: fix potential…