On Wed, Dec 16, 2020 at 10:07:37AM +0000, Christophe Leroy wrote: > Implement Extended Berkeley Packet Filter on Powerpc 32 > > Test result with test_bpf module: > > test_bpf: Summary: 378 PASSED, 0 FAILED, [354/366 JIT'ed] nice! > Registers mapping: > > [BPF_REG_0] = r11-r12 > /* function arguments */ > [BPF_REG_1] = r3-r4 > [BPF_REG_2] = r5-r6 > [BPF_REG_3] = r7-r8 > [BPF_REG_4] = r9-r10 > [BPF_REG_5] = r21-r22 (Args 9 and 10 come in via the stack) > /* non volatile registers */ > [BPF_REG_6] = r23-r24 > [BPF_REG_7] = r25-r26 > [BPF_REG_8] = r27-r28 > [BPF_REG_9] = r29-r30 > /* frame pointer aka BPF_REG_10 */ > [BPF_REG_FP] = r31 > /* eBPF jit internal registers */ > [BPF_REG_AX] = r19-r20 > [TMP_REG] = r18 > > As PPC32 doesn't have a redzone in the stack, > use r17 as tail call counter. > > r0 is used as temporary register as much as possible. It is referenced > directly in the code in order to avoid misuse of it, because some > instructions interpret it as value 0 instead of register r0 > (ex: addi, addis, stw, lwz, ...) > > The following operations are not implemented: > > case BPF_ALU64 | BPF_DIV | BPF_X: /* dst /= src */ > case BPF_ALU64 | BPF_MOD | BPF_X: /* dst %= src */ > case BPF_STX | BPF_XADD | BPF_DW: /* *(u64 *)(dst + off) += src */ > > The following operations are only implemented for power of two constants: > > case BPF_ALU64 | BPF_MOD | BPF_K: /* dst %= imm */ > case BPF_ALU64 | BPF_DIV | BPF_K: /* dst /= imm */ Those are sensible limitations. MOD and DIV are rare, but XADD is common. Please consider doing it as a cmpxchg loop in the future. Also please run test_progs. It will give a lot better coverage than test_bpf.ko
