Karl Mehltretter <[email protected]> writes:
> bpf_jit_supports_fsession() returns bool. Commit 92258b5bf1ec
> ("powerpc32/bpf: Add fsession support") made it return -EOPNOTSUPP on
> ppc32 until fsession is validated there. The negative value converts to
> true, so the verifier does not reject fsession programs on ppc32. Such a
> program loads, and only attaching it fails because
> __arch_prepare_bpf_trampoline() returns -EOPNOTSUPP on ppc32.
>
> Return false instead. The load then fails with "JIT does not support
> fsession". Tested with a ppc32 kernel on QEMU g3beige.
>
Fair enough.
> Fixes: 92258b5bf1ec ("powerpc32/bpf: Add fsession support")
> Reported-by: kernel test robot <[email protected]>
> Reported-by: Dan Carpenter <[email protected]>
> Closes: https://lore.kernel.org/r/[email protected]/
> Assisted-by: LLM
> Signed-off-by: Karl Mehltretter <[email protected]>
> ---
> arch/powerpc/net/bpf_jit_comp.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
Sorry, I guess, this must have falled through the cracks.
Should mostly get queued for next -rc release.
LGTM. Feel free to add:
Reviewed-by: Ritesh Harjani (IBM) <[email protected]>