CVE-2025-40246

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

xfs: fix out of bounds memory read error in symlink repair

xfs/286 produced this report on my test fleet:

================================================================== BUG: KFENCE: out-of-bounds read in memcpy_orig+0x54/0x110

Out-of-bounds read at 0xffff88843fe9e038 (184B right of kfence-#184): memcpyorig+0x54/0x110 xrepsymlinksalvageinline+0xb3/0xf0 [xfs] xrepsymlinksalvage+0x100/0x110 [xfs] xrepsymlink+0x2e/0x80 [xfs] xrepattempt+0x61/0x1f0 [xfs] xfsscrubmetadata+0x34f/0x5c0 [xfs] xfsiocscrubvmetadata+0x387/0x560 [xfs] xfsfile_ioctl+0xe23/0x10e0 [xfs] __x64sysioctl+0x76/0xc0 dosyscall64+0x4e/0x1e0 entrySYSCALL64afterhwframe+0x4b/0x53

kfence-#184: 0xffff88843fe9df80-0xffff88843fe9dfea, size=107, cache=kmalloc-128

allocated by task 3470 on cpu 1 at 263329.131592s (192823.508886s ago): xfsinitlocalfork+0x79/0xe0 [xfs] xfsiformatlocal+0xa4/0x170 [xfs] xfsiformatdatafork+0x148/0x180 [xfs] xfsinodefromdisk+0x2cd/0x480 [xfs] xfsiget+0x450/0xd60 [xfs] xfsbulkstatoneint+0x6b/0x510 [xfs] xfsbulkstatiwalk+0x1e/0x30 [xfs] xfsiwalkagrecs+0xdf/0x150 [xfs] xfsiwalkruncallbacks+0xb9/0x190 [xfs] xfsiwalkag+0x1dc/0x2f0 [xfs] xfsiwalkargs.constprop.0+0x6a/0x120 [xfs] xfsiwalk+0xa4/0xd0 [xfs] xfsbulkstat+0xfa/0x170 [xfs] xfsiocfsbulkstat.isra.0+0x13a/0x230 [xfs] xfsfile_ioctl+0xbf2/0x10e0 [xfs] __x64sysioctl+0x76/0xc0 dosyscall64+0x4e/0x1e0 entrySYSCALL64afterhwframe+0x4b/0x53

CPU: 1 UID: 0 PID: 1300113 Comm: xfs_scrub Not tainted 6.18.0-rc4-djwx #rc4 PREEMPT(lazy) 3d744dd94e92690f00a04398d2bd8631dcef1954 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.0-4.module+el8.8.0+21164+ed375313 04/01/2014 ==================================================================

On further analysis, I realized that the second parameter to min() is not correct. xfsifork::ifbytes is the size of the xfsifork::ifdata buffer. if_bytes can be smaller than the data fork size because:

(a) the forkoff code tries to keep the data area as large as possible (b) for symbolic links, if_bytes is the ondisk file size + 1 (c) forkoff is always a multiple of 8.

Case in point: for a single-byte symlink target, forkoff will be 8 but the buffer will only be 2 bytes long.

In other words, the logic here is wrong and we walk off the end of the incore buffer. Fix that.