On Tue, Oct 4, 2022 at 9:36 AM Dave Thaler <dthaler@xxxxxxxxxxxxx> wrote:
> > Those differences are in signed div/mod only, right?
> > Unsigned div/mod doesn't have it, right?
> > bpf has only unsigned div/mod.
>
> Ah right, will replace. However since imm is a signed integer, that leaves
> an ambiguity that is important to clarify.
>
> What is the expected value for the following 64-bit BPF_DIV operation:
> r0 = 0xFFFFFFFFFFFFFFFF
> r0 /= -10
> Is it 0x1 or 0x10000000a? i.e., is the -10 sign extended to
> 0xFFFFFFFFFFFFFFF6 or treated as 0xFFFFFFF6 when doing the unsigned
> division?
x86 and arm64 JITs treat it as imm32 is zero extended.
But looking at the interpreter:
ALU64_DIV_K:
DST = div64_u64(DST, IMM);
it looks like we have a bug there.
But we have a bunch of div_k tests in lib/test_bpf.c
including negative imm32. Hmm.
