On Sat, Jul 06, 2024 at 11:11:01AM +0800, WangYuli wrote:
This reverts commit 08f6c05feb1db21653e98ca84ea04ca032d014c7.
Upstream commit e60adf513275 ("bpf: Take return from set_memory_rox() into account with bpf_jit_binary_lock_ro()") depends on upstream commit 1dad391daef1 ("bpf, arm64: use bpf_prog_pack for memory management").
It will cause a compilation warning on the arm64 if it's not merged: arch/arm64/net/bpf_jit_comp.c: In function ‘bpf_int_jit_compile’: arch/arm64/net/bpf_jit_comp.c:1651:17: warning: ignoring return value of ‘bpf_jit_binary_lock_ro’ declared with attribute ‘warn_unused_result’ [-Wunused-result] 1651 | bpf_jit_binary_lock_ro(header); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This will prevent the kernel with the '-Werror' compile option from being compiled successfully.
We might as well revert this commit in linux-6.6.37 to solve the problem in a simple way.
This makes it sound like you are reverting this because of a build error, which is not the case here, right? Isn't this because of the powerpc issue reported here: https://lore.kernel.org/r/20240705203413.wbv2nw3747vjeibk@altlinux.org ?
If not, why not just backport the single missing arm64 commit, and why didn't this show up in testing?
confused,
greg k-h