From: Josh Poimboeuf <jpoimboe@redhat.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: x86@kernel.org, andrew.cooper3@citrix.com,
linux-kernel@vger.kernel.org, alexei.starovoitov@gmail.com,
ndesaulniers@google.com
Subject: Re: [PATCH v2 14/14] bpf,x86: Respect X86_FEATURE_RETPOLINE*
Date: Wed, 20 Oct 2021 09:56:55 -0700 [thread overview]
Message-ID: <20211020165655.632slp4ujsajis4j@treble> (raw)
In-Reply-To: <YW/4/7MjUf3hWfjz@hirez.programming.kicks-ass.net>
On Wed, Oct 20, 2021 at 01:09:51PM +0200, Peter Zijlstra wrote:
> Alexei; could the above not be further improved with something like the
> below?
>
> Despite several hours trying and Song helping, I can't seem to run
> anything bpf, that stuff is cursed. So I've no idea if the below
> actually works, but it seems reasonable.
The below fix gets it to run with test_verififer.
I do like it, it does seem less fiddly and more robust against future
changes, though it probably needs a comment for 'out_label' clarifying
it only works because this function is always called at least twice for
a given bpf_tail_call emission.
---
diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index 08f32c9fceaa..c9230c5bbca5 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -457,7 +457,7 @@ static void emit_bpf_tail_call_indirect(u8 **pprog, bool *callee_regs_used,
EMIT3(0x39, 0x56, /* cmp dword ptr [rsi + 16], edx */
offsetof(struct bpf_array, map.max_entries));
- offset = out_label - (prog - start) + 2;
+ offset = out_label - (prog - start + 2);
EMIT2(X86_JBE, offset); /* jbe out */
/*
@@ -467,7 +467,7 @@ static void emit_bpf_tail_call_indirect(u8 **pprog, bool *callee_regs_used,
EMIT2_off32(0x8B, 0x85, tcc_off); /* mov eax, dword ptr [rbp - tcc_off] */
EMIT3(0x83, 0xF8, MAX_TAIL_CALL_CNT); /* cmp eax, MAX_TAIL_CALL_CNT */
- offset = out_label - (prog - start) + 2;
+ offset = out_label - (prog - start + 2);
EMIT2(X86_JA, offset); /* ja out */
EMIT3(0x83, 0xC0, 0x01); /* add eax, 1 */
EMIT2_off32(0x89, 0x85, tcc_off); /* mov dword ptr [rbp - tcc_off], eax */
@@ -482,7 +482,7 @@ static void emit_bpf_tail_call_indirect(u8 **pprog, bool *callee_regs_used,
*/
EMIT3(0x48, 0x85, 0xC9); /* test rcx,rcx */
- offset = out_label - (prog - start) + 2;
+ offset = out_label - (prog - start + 2);
EMIT2(X86_JE, offset); /* je out */
*pprog = prog;
next prev parent reply other threads:[~2021-10-20 16:57 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-20 10:44 [PATCH v2 00/14] x86: Rewrite the retpoline rewrite logic Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 01/14] objtool: Tag retpoline thunk symbols Peter Zijlstra
2021-10-20 15:17 ` Josh Poimboeuf
2021-10-26 7:55 ` Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 02/14] objtool: Explicitly avoid self modifying code in .altinstr_replacement Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 03/14] objtool: Shrink struct instruction Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 04/14] objtool,x86: Replace alternatives with .retpoline_sites Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 05/14] x86/retpoline: Remove unused replacement symbols Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 06/14] x86/asm: Fix register order Peter Zijlstra
2021-10-22 19:27 ` David Laight
2021-10-25 14:09 ` Borislav Petkov
2021-10-20 10:44 ` [PATCH v2 07/14] x86/asm: Fixup odd GEN-for-each-reg.h usage Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 08/14] x86/retpoline: Create a retpoline thunk array Peter Zijlstra
2021-10-20 15:57 ` Josh Poimboeuf
2021-10-20 16:46 ` Andrew Cooper
2021-10-20 17:09 ` Josh Poimboeuf
2021-10-20 19:22 ` Peter Zijlstra
2021-10-20 19:43 ` Josh Poimboeuf
2021-10-20 19:34 ` Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 09/14] x86/alternative: Implement .retpoline_sites support Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 10/14] x86/alternative: Handle Jcc __x86_indirect_thunk_\reg Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 11/14] x86/alternative: Try inline spectre_v2=retpoline,amd Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 12/14] x86/alternative: Add debug prints to apply_retpolines() Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 13/14] x86,bugs: Unconditionally allow spectre_v2=retpoline,amd Peter Zijlstra
2021-10-20 10:44 ` [PATCH v2 14/14] bpf,x86: Respect X86_FEATURE_RETPOLINE* Peter Zijlstra
2021-10-20 11:09 ` Peter Zijlstra
2021-10-20 16:56 ` Josh Poimboeuf [this message]
2021-10-20 19:23 ` Peter Zijlstra
2021-10-21 0:05 ` Alexei Starovoitov
2021-10-21 8:47 ` Peter Zijlstra
2021-10-21 18:03 ` Alexei Starovoitov
2021-10-21 22:37 ` Peter Zijlstra
2021-10-21 23:24 ` Alexei Starovoitov
2021-10-21 23:38 ` Josh Poimboeuf
2021-10-21 23:42 ` Alexei Starovoitov
2021-10-22 11:31 ` Peter Zijlstra
2021-10-22 15:22 ` Alexei Starovoitov
2021-10-25 13:44 ` Maciej Fijalkowski
2021-10-25 12:42 ` Peter Zijlstra
2021-10-21 23:51 ` Zvi Effron
2021-10-22 8:33 ` Peter Zijlstra
2021-10-22 21:06 ` Zvi Effron
2021-10-21 0:07 ` Alexei Starovoitov
2021-10-21 0:18 ` Josh Poimboeuf
2021-10-21 8:53 ` Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20211020165655.632slp4ujsajis4j@treble \
--to=jpoimboe@redhat.com \
--cc=alexei.starovoitov@gmail.com \
--cc=andrew.cooper3@citrix.com \
--cc=linux-kernel@vger.kernel.org \
--cc=ndesaulniers@google.com \
--cc=peterz@infradead.org \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).