CVE-2023-52604

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> FS:JFS:UBSAN:array-index-out-of-bounds in dbAdjTree<br /> <br /> Syzkaller reported the following issue:<br /> <br /> UBSAN: array-index-out-of-bounds in fs/jfs/jfs_dmap.c:2867:6<br /> index 196694 is out of range for type &amp;#39;s8[1365]&amp;#39; (aka &amp;#39;signed char[1365]&amp;#39;)<br /> CPU: 1 PID: 109 Comm: jfsCommit Not tainted 6.6.0-rc3-syzkaller #0<br /> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/04/2023<br /> Call Trace:<br /> <br /> __dump_stack lib/dump_stack.c:88 [inline]<br /> dump_stack_lvl+0x1e7/0x2d0 lib/dump_stack.c:106<br /> ubsan_epilogue lib/ubsan.c:217 [inline]<br /> __ubsan_handle_out_of_bounds+0x11c/0x150 lib/ubsan.c:348<br /> dbAdjTree+0x474/0x4f0 fs/jfs/jfs_dmap.c:2867<br /> dbJoin+0x210/0x2d0 fs/jfs/jfs_dmap.c:2834<br /> dbFreeBits+0x4eb/0xda0 fs/jfs/jfs_dmap.c:2331<br /> dbFreeDmap fs/jfs/jfs_dmap.c:2080 [inline]<br /> dbFree+0x343/0x650 fs/jfs/jfs_dmap.c:402<br /> txFreeMap+0x798/0xd50 fs/jfs/jfs_txnmgr.c:2534<br /> txUpdateMap+0x342/0x9e0<br /> txLazyCommit fs/jfs/jfs_txnmgr.c:2664 [inline]<br /> jfs_lazycommit+0x47a/0xb70 fs/jfs/jfs_txnmgr.c:2732<br /> kthread+0x2d3/0x370 kernel/kthread.c:388<br /> ret_from_fork+0x48/0x80 arch/x86/kernel/process.c:147<br /> ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:304<br /> <br /> ================================================================================<br /> Kernel panic - not syncing: UBSAN: panic_on_warn set ...<br /> CPU: 1 PID: 109 Comm: jfsCommit Not tainted 6.6.0-rc3-syzkaller #0<br /> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/04/2023<br /> Call Trace:<br /> <br /> __dump_stack lib/dump_stack.c:88 [inline]<br /> dump_stack_lvl+0x1e7/0x2d0 lib/dump_stack.c:106<br /> panic+0x30f/0x770 kernel/panic.c:340<br /> check_panic_on_warn+0x82/0xa0 kernel/panic.c:236<br /> ubsan_epilogue lib/ubsan.c:223 [inline]<br /> __ubsan_handle_out_of_bounds+0x13c/0x150 lib/ubsan.c:348<br /> dbAdjTree+0x474/0x4f0 fs/jfs/jfs_dmap.c:2867<br /> dbJoin+0x210/0x2d0 fs/jfs/jfs_dmap.c:2834<br /> dbFreeBits+0x4eb/0xda0 fs/jfs/jfs_dmap.c:2331<br /> dbFreeDmap fs/jfs/jfs_dmap.c:2080 [inline]<br /> dbFree+0x343/0x650 fs/jfs/jfs_dmap.c:402<br /> txFreeMap+0x798/0xd50 fs/jfs/jfs_txnmgr.c:2534<br /> txUpdateMap+0x342/0x9e0<br /> txLazyCommit fs/jfs/jfs_txnmgr.c:2664 [inline]<br /> jfs_lazycommit+0x47a/0xb70 fs/jfs/jfs_txnmgr.c:2732<br /> kthread+0x2d3/0x370 kernel/kthread.c:388<br /> ret_from_fork+0x48/0x80 arch/x86/kernel/process.c:147<br /> ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:304<br /> <br /> Kernel Offset: disabled<br /> Rebooting in 86400 seconds..<br /> <br /> The issue is caused when the value of lp becomes greater than<br /> CTLTREESIZE which is the max size of stree. Adding a simple check<br /> solves this issue.<br /> <br /> Dave:<br /> As the function returns a void, good error handling<br /> would require a more intrusive code reorganization, so I modified<br /> Osama&amp;#39;s patch at use WARN_ON_ONCE for lack of a cleaner option.<br /> <br /> The patch is tested via syzbot.