From a89dfde3dc3c2dbf56910af75e2d8b11ec5308f6 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 12 Mar 2021 12:32:54 +0100 Subject: x86: Remove dynamic NOP selection This ensures that a NOP is a NOP and not a random other instruction that is also a NOP. It allows simplification of dynamic code patching that wants to verify existing code before writing new instructions (ftrace, jump_label, static_call, etc..). Differentiating on NOPs is not a feature. This pessimises 32bit (DONTCARE) and 32bit on 64bit CPUs (CARELESS). 32bit is not a performance target. Everything x86_64 since AMD K10 (2007) and Intel IvyBridge (2012) is fine with using NOPL (as opposed to prefix NOP). And per FEATURE_NOPL being required for x86_64, all x86_64 CPUs can use NOPL. So stop caring about NOPs, simplify things and get on with life. [ The problem seems to be that some uarchs can only decode NOPL on a single front-end port while others have severe decode penalties for excessive prefixes. All modern uarchs can handle both, except Atom, which has prefix penalties. ] [ Also, much doubt you can actually measure any of this on normal workloads. ] After this, FEATURE_NOPL is unused except for required-features for x86_64. FEATURE_K8 is only used for PTI. [ bp: Kernel build measurements showed ~0.3s slowdown on Sandybridge which is hardly a slowdown. Get rid of X86_FEATURE_K7, while at it. ] Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Borislav Petkov Acked-by: Alexei Starovoitov # bpf Acked-by: Linus Torvalds Link: https://lkml.kernel.org/r/20210312115749.065275711@infradead.org --- arch/x86/net/bpf_jit_comp.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86/net/bpf_jit_comp.c') diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index 79e7a0ec1da5..6aa29c45c3f9 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -282,7 +282,7 @@ static void emit_prologue(u8 **pprog, u32 stack_depth, bool ebpf_from_cbpf, /* BPF trampoline can be made to work without these nops, * but let's waste 5 bytes for now and optimize later */ - memcpy(prog, ideal_nops[NOP_ATOMIC5], cnt); + memcpy(prog, x86_nops[5], cnt); prog += cnt; if (!ebpf_from_cbpf) { if (tail_call_reachable && !is_subprog) @@ -330,7 +330,7 @@ static int __bpf_arch_text_poke(void *ip, enum bpf_text_poke_type t, void *old_addr, void *new_addr, const bool text_live) { - const u8 *nop_insn = ideal_nops[NOP_ATOMIC5]; + const u8 *nop_insn = x86_nops[5]; u8 old_insn[X86_PATCH_SIZE]; u8 new_insn[X86_PATCH_SIZE]; u8 *prog; @@ -560,7 +560,7 @@ static void emit_bpf_tail_call_direct(struct bpf_jit_poke_descriptor *poke, if (stack_depth) EMIT3_off32(0x48, 0x81, 0xC4, round_up(stack_depth, 8)); - memcpy(prog, ideal_nops[NOP_ATOMIC5], X86_PATCH_SIZE); + memcpy(prog, x86_nops[5], X86_PATCH_SIZE); prog += X86_PATCH_SIZE; /* out: */ @@ -881,7 +881,7 @@ static int emit_nops(u8 **pprog, int len) noplen = ASM_NOP_MAX; for (i = 0; i < noplen; i++) - EMIT1(ideal_nops[noplen][i]); + EMIT1(x86_nops[noplen][i]); len -= noplen; } -- cgit