On Mon, Apr 14, 2025 at 2:22 PM Kuniyuki Iwashima <kuniyu@xxxxxxxxxx> wrote: > > There are some failure paths in bpf_int_jit_compile() that are not > worth triggering a warning in __bpf_prog_ret0_warn(). > > For example, if we fail to allocate memory in bpf_int_jit_compile(), > we should propagate -ENOMEM to userspace instead of attaching > __bpf_prog_ret0_warn(). > > Let's pass &err to bpf_int_jit_compile() to propagate errno. Is there any reason we are not just returning ERR_PTR() instead of the approach in this patch? That seems more canonical within BPF subsystem, if we need to return error for pointer-returning functions? > > Signed-off-by: Kuniyuki Iwashima <kuniyu@xxxxxxxxxx> > --- > arch/arc/net/bpf_jit_core.c | 2 +- > arch/arm/net/bpf_jit_32.c | 2 +- > arch/arm64/net/bpf_jit_comp.c | 2 +- > arch/loongarch/net/bpf_jit.c | 2 +- > arch/mips/net/bpf_jit_comp.c | 2 +- > arch/parisc/net/bpf_jit_core.c | 2 +- > arch/powerpc/net/bpf_jit_comp.c | 2 +- > arch/riscv/net/bpf_jit_core.c | 2 +- > arch/s390/net/bpf_jit_comp.c | 2 +- > arch/sparc/net/bpf_jit_comp_64.c | 2 +- > arch/x86/net/bpf_jit_comp.c | 2 +- > arch/x86/net/bpf_jit_comp32.c | 2 +- > include/linux/filter.h | 2 +- > kernel/bpf/core.c | 6 ++++-- > kernel/bpf/verifier.c | 21 +++++++++++++++------ > 15 files changed, 32 insertions(+), 21 deletions(-) > [...]
