Re: [PATCH bpf-next v3 01/11] Revert "bpf: use common instruction history across all states"

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Jun 11, 2025 at 1:08 PM Eduard Zingerman <eddyz87@xxxxxxxxx> wrote:
>
> This reverts commit 96a30e469ca1d2b8cc7811b40911f8614b558241.
> Next patches in the series modify propagate_precision() to allow
> arbitrary starting state. Precision propagation requires access to
> jump history, and arbitrary states represent history not belonging to
> `env->cur_state`.
>
> Signed-off-by: Eduard Zingerman <eddyz87@xxxxxxxxx>
> ---
>  include/linux/bpf_verifier.h |  19 +++----
>  kernel/bpf/verifier.c        | 107 ++++++++++++++++++-----------------
>  2 files changed, 63 insertions(+), 63 deletions(-)

This wasn't a clean revert. It broke the build with:

../kernel/bpf/verifier.c: In function ‘check_cond_jmp_op’:
../kernel/bpf/verifier.c:16503:23: error: implicit declaration of
function ‘push_insn_history’; did you mean ‘push_jmp_history’?
[-Wimplicit-function-declaration]
16503 |                 err = push_insn_history(env, this_branch,
insn_flags, 0);
      |                       ^~~~~~~~~~~~~~~~~
      |                       push_jmp_history


Though it's fixed later in patch 8 as:
-               err = push_insn_history(env, this_branch, insn_flags, 0);
+               err = push_jmp_history(env, this_branch, insn_flags, 0);

It would have been a bisect issue if not caught by CI:
https://netdev.bots.linux.dev/static/nipa/971030/14115002/build_clang/summary

I fixed up patch 1 and 8 while applying.
Pls pay attention to such things in the future.





[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux