Commit dfce9cb
bpf: Fix a verifier bug due to incorrect branch offset comparison with cpu=v4
Bpf cpu=v4 support is introduced in [1] and Commit 4cd58e9
("bpf: Support new 32bit offset jmp instruction") added support for new
32bit offset jmp instruction. Unfortunately, in function
bpf_adj_delta_to_off(), for new branch insn with 32bit offset, the offset
(plus/minor a small delta) compares to 16-bit offset bound
[S16_MIN, S16_MAX], which caused the following verification failure:
$ ./test_progs-cpuv4 -t verif_scale_pyperf180
...
insn 10 cannot be patched due to 16-bit range
...
libbpf: failed to load object 'pyperf180.bpf.o'
scale_test:FAIL:expect_success unexpected error: -12 (errno 12)
#405 verif_scale_pyperf180:FAIL
Note that due to recent llvm18 development, the patch [2] (already applied
in bpf-next) needs to be applied to bpf tree for testing purpose.
The fix is rather simple. For 32bit offset branch insn, the adjusted
offset compares to [S32_MIN, S32_MAX] and then verification succeeded.
[1] https://lore.kernel.org/all/20230728011143.3710005-1-yonghong.song@linux.dev
[2] https://lore.kernel.org/bpf/20231110193644.3130906-1-yonghong.song@linux.dev
Fixes: 4cd58e9 ("bpf: Support new 32bit offset jmp instruction")
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20231201024640.3417057-1-yonghong.song@linux.dev1 parent 830139e commit dfce9cb
1 file changed
+8
-4
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
371 | 371 | | |
372 | 372 | | |
373 | 373 | | |
374 | | - | |
| 374 | + | |
375 | 375 | | |
376 | | - | |
377 | 376 | | |
378 | | - | |
| 377 | + | |
379 | 378 | | |
380 | | - | |
| 379 | + | |
| 380 | + | |
| 381 | + | |
381 | 382 | | |
| 383 | + | |
| 384 | + | |
| 385 | + | |
382 | 386 | | |
383 | 387 | | |
384 | 388 | | |
| |||
0 commit comments