| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <Yyw5JqLtiXcqid3K@hirez.programming.kicks-ass.net>
Date: Thu, 22 Sep 2022 12:29:58 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Pawan Gupta <pawan.kumar.gupta@...ux.intel.com>
Cc: Thomas Gleixner <tglx@...utronix.de>, linux-kernel@...r.kernel.org,
x86@...nel.org, Linus Torvalds <torvalds@...ux-foundation.org>,
Tim Chen <tim.c.chen@...ux.intel.com>,
Josh Poimboeuf <jpoimboe@...nel.org>,
Andrew Cooper <Andrew.Cooper3@...rix.com>,
Johannes Wikner <kwikner@...z.ch>,
Alyssa Milburn <alyssa.milburn@...ux.intel.com>,
Jann Horn <jannh@...gle.com>, "H.J. Lu" <hjl.tools@...il.com>,
Joao Moreira <joao.moreira@...el.com>,
Joseph Nuzman <joseph.nuzman@...el.com>,
Steven Rostedt <rostedt@...dmis.org>,
Juergen Gross <jgross@...e.com>,
Masami Hiramatsu <mhiramat@...nel.org>,
Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
K Prateek Nayak <kprateek.nayak@....com>,
Eric Dumazet <edumazet@...gle.com>
Subject: Re: [PATCH v3 32/59] objtool: Allow STT_NOTYPE -> STT_FUNC+0
tail-calls
On Wed, Sep 21, 2022 at 10:27:50PM -0700, Pawan Gupta wrote:
> On Thu, Sep 15, 2022 at 01:11:11PM +0200, Peter Zijlstra wrote:
> > From: Peter Zijlstra <peterz@...radead.org>
> >
> > Allow STT_NOTYPE to tail-call into STT_FUNC, per definition STT_NOTYPE
> > is not a sub-function of the STT_FUNC.
> >
> > Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
> > Signed-off-by: Thomas Gleixner <tglx@...utronix.de>
> > Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
> > ---
> > tools/objtool/check.c | 29 ++++++++++++++++++++---------
> > 1 file changed, 20 insertions(+), 9 deletions(-)
> >
> > --- a/tools/objtool/check.c
> > +++ b/tools/objtool/check.c
> > @@ -1370,6 +1370,16 @@ static void add_return_call(struct objto
> >
> > static bool same_function(struct instruction *insn1, struct instruction *insn2)
> > {
> > + if (!insn1->func && !insn2->func)
> > + return true;
> > +
> > + /* Allow STT_NOTYPE -> STT_FUNC+0 tail-calls */
> > + if (!insn1->func && insn1->func != insn2->func)
> > + return false;
>
> Looks like this check is triggering below warning:
>
> vmlinux.o: warning: objtool: ftrace_replace_code.cold+0x0: unreachable instruction
>
I managed to reproduce with 12.2.0 -- my 12.1.0 compiler didn't
reproduce.
$ ./scripts/objdump-func vmlinux.o ftrace_replace_code
0000 00000000001032f0 <ftrace_replace_code>:
0000 1032f0: f3 0f 1e fa endbr64
0004 1032f4: 41 57 push %r15
0006 1032f6: 41 56 push %r14
0008 1032f8: 41 55 push %r13
000a 1032fa: 41 89 fd mov %edi,%r13d
000d 1032fd: 41 54 push %r12
000f 1032ff: 55 push %rbp
0010 103300: 53 push %rbx
0011 103301: 48 83 ec 10 sub $0x10,%rsp
0015 103305: 65 48 8b 04 25 28 00 00 00 mov %gs:0x28,%rax
001e 10330e: 48 89 44 24 08 mov %rax,0x8(%rsp)
0023 103313: 31 c0 xor %eax,%eax
0025 103315: e8 00 00 00 00 call 10331a <ftrace_replace_code+0x2a> 103316: R_X86_64_PLT32 ftrace_rec_iter_start-0x4
002a 10331a: 45 85 ed test %r13d,%r13d
002d 10331d: 41 0f 95 c4 setne %r12b
0031 103321: 48 85 c0 test %rax,%rax
0034 103324: 0f 84 e1 00 00 00 je 10340b <ftrace_replace_code+0x11b>
003a 10332a: 48 89 c3 mov %rax,%rbx
003d 10332d: 45 0f b6 e4 movzbl %r12b,%r12d
0041 103331: 48 c7 c5 00 00 00 00 mov $0x0,%rbp 103334: R_X86_64_32S .bss+0x16538
0048 103338: 48 89 df mov %rbx,%rdi
004b 10333b: e8 00 00 00 00 call 103340 <ftrace_replace_code+0x50> 10333c: R_X86_64_PLT32 ftrace_rec_iter_record-0x4
0050 103340: 44 89 e6 mov %r12d,%esi
0053 103343: 48 89 c7 mov %rax,%rdi
0056 103346: 49 89 c6 mov %rax,%r14
0059 103349: e8 00 00 00 00 call 10334e <ftrace_replace_code+0x5e> 10334a: R_X86_64_PLT32 ftrace_test_record-0x4
005e 10334e: 83 f8 01 cmp $0x1,%eax
0061 103351: 0f 84 61 01 00 00 je 1034b8 <ftrace_replace_code+0x1c8>
0067 103357: 83 e8 02 sub $0x2,%eax
006a 10335a: 83 f8 01 cmp $0x1,%eax
006d 10335d: 0f 87 94 00 00 00 ja 1033f7 <ftrace_replace_code+0x107>
0073 103363: 4c 89 f7 mov %r14,%rdi
0076 103366: 49 c7 c7 00 00 00 00 mov $0x0,%r15 103369: R_X86_64_32S .bss+0x16538
007d 10336d: e8 00 00 00 00 call 103372 <ftrace_replace_code+0x82> 10336e: R_X86_64_PLT32 ftrace_get_addr_curr-0x4
0082 103372: 49 8b 16 mov (%r14),%rdx
0085 103375: c6 45 00 e8 movb $0xe8,0x0(%rbp)
0089 103379: 48 83 c2 05 add $0x5,%rdx
008d 10337d: 48 29 d0 sub %rdx,%rax
0090 103380: 89 45 01 mov %eax,0x1(%rbp)
0093 103383: 49 8b 36 mov (%r14),%rsi
0096 103386: ba 05 00 00 00 mov $0x5,%edx
009b 10338b: 48 8d 7c 24 03 lea 0x3(%rsp),%rdi
00a0 103390: c7 44 24 03 00 00 00 00 movl $0x0,0x3(%rsp)
00a8 103398: c6 44 24 07 00 movb $0x0,0x7(%rsp)
00ad 10339d: e8 00 00 00 00 call 1033a2 <ftrace_replace_code+0xb2> 10339e: R_X86_64_PLT32 copy_from_kernel_nofault-0x4
00b2 1033a2: 48 85 c0 test %rax,%rax
00b5 1033a5: 0f 85 3e 01 00 00 jne 1034e9 <ftrace_replace_code+0x1f9>
00bb 1033ab: 41 8b 07 mov (%r15),%eax
00be 1033ae: 39 44 24 03 cmp %eax,0x3(%rsp)
00c2 1033b2: 74 38 je 1033ec <ftrace_replace_code+0xfc>
00c4 1033b4: 4c 89 3d 00 00 00 00 mov %r15,0x0(%rip) # 1033bb <ftrace_replace_code+0xcb> 1033b7: R_X86_64_PC32 ftrace_expected-0x4
00cb 1033bb: 0f 0b ud2
00cd 1033bd: bf ea ff ff ff mov $0xffffffea,%edi
00d2 1033c2: 48 8b 44 24 08 mov 0x8(%rsp),%rax
00d7 1033c7: 65 48 2b 04 25 28 00 00 00 sub %gs:0x28,%rax
00e0 1033d0: 0f 85 1f 01 00 00 jne 1034f5 <ftrace_replace_code+0x205>
00e6 1033d6: 48 83 c4 10 add $0x10,%rsp
00ea 1033da: 4c 89 f6 mov %r14,%rsi
00ed 1033dd: 5b pop %rbx
00ee 1033de: 5d pop %rbp
00ef 1033df: 41 5c pop %r12
00f1 1033e1: 41 5d pop %r13
00f3 1033e3: 41 5e pop %r14
00f5 1033e5: 41 5f pop %r15
00f7 1033e7: e9 00 00 00 00 jmp 1033ec <ftrace_replace_code+0xfc> 1033e8: R_X86_64_PLT32 ftrace_bug-0x4
00fc 1033ec: 41 0f b6 47 04 movzbl 0x4(%r15),%eax
0101 1033f1: 38 44 24 07 cmp %al,0x7(%rsp)
0105 1033f5: 75 bd jne 1033b4 <ftrace_replace_code+0xc4>
0107 1033f7: 48 89 df mov %rbx,%rdi
010a 1033fa: e8 00 00 00 00 call 1033ff <ftrace_replace_code+0x10f> 1033fb: R_X86_64_PLT32 ftrace_rec_iter_next-0x4
010f 1033ff: 48 89 c3 mov %rax,%rbx
0112 103402: 48 85 c0 test %rax,%rax
0115 103405: 0f 85 2d ff ff ff jne 103338 <ftrace_replace_code+0x48>
011b 10340b: e8 00 00 00 00 call 103410 <ftrace_replace_code+0x120> 10340c: R_X86_64_PLT32 ftrace_rec_iter_start-0x4
0120 103410: 45 31 e4 xor %r12d,%r12d
0123 103413: 45 85 ed test %r13d,%r13d
0126 103416: 49 c7 c5 00 00 00 00 mov $0x0,%r13 103419: R_X86_64_32S .bss+0x16538
012d 10341d: 41 0f 95 c4 setne %r12b
0131 103421: 48 89 c3 mov %rax,%rbx
0134 103424: 48 85 c0 test %rax,%rax
0137 103427: 75 1a jne 103443 <ftrace_replace_code+0x153>
0139 103429: eb 6a jmp 103495 <ftrace_replace_code+0x1a5>
013b 10342b: 85 c0 test %eax,%eax
013d 10342d: 0f 8f 91 00 00 00 jg 1034c4 <ftrace_replace_code+0x1d4>
0143 103433: 48 89 df mov %rbx,%rdi
0146 103436: e8 00 00 00 00 call 10343b <ftrace_replace_code+0x14b> 103437: R_X86_64_PLT32 ftrace_rec_iter_next-0x4
014b 10343b: 48 89 c3 mov %rax,%rbx
014e 10343e: 48 85 c0 test %rax,%rax
0151 103441: 74 52 je 103495 <ftrace_replace_code+0x1a5>
0153 103443: 48 89 df mov %rbx,%rdi
0156 103446: e8 00 00 00 00 call 10344b <ftrace_replace_code+0x15b> 103447: R_X86_64_PLT32 ftrace_rec_iter_record-0x4
015b 10344b: 44 89 e6 mov %r12d,%esi
015e 10344e: 48 89 c7 mov %rax,%rdi
0161 103451: 48 89 c5 mov %rax,%rbp
0164 103454: e8 00 00 00 00 call 103459 <ftrace_replace_code+0x169> 103455: R_X86_64_PLT32 ftrace_test_record-0x4
0169 103459: 83 f8 02 cmp $0x2,%eax
016c 10345c: 7e cd jle 10342b <ftrace_replace_code+0x13b>
016e 10345e: 83 f8 03 cmp $0x3,%eax
0171 103461: 75 d0 jne 103433 <ftrace_replace_code+0x143>
0173 103463: 48 8b 35 00 00 00 00 mov 0x0(%rip),%rsi # 10346a <ftrace_replace_code+0x17a> 103466: R_X86_64_PC32 x86_nops+0x24
017a 10346a: 48 8b 7d 00 mov 0x0(%rbp),%rdi
017e 10346e: 31 c9 xor %ecx,%ecx
0180 103470: ba 05 00 00 00 mov $0x5,%edx
0185 103475: e8 00 00 00 00 call 10347a <ftrace_replace_code+0x18a> 103476: R_X86_64_PLT32 text_poke_queue-0x4
018a 10347a: 44 89 e6 mov %r12d,%esi
018d 10347d: 48 89 ef mov %rbp,%rdi
0190 103480: e8 00 00 00 00 call 103485 <ftrace_replace_code+0x195> 103481: R_X86_64_PLT32 ftrace_update_record-0x4
0195 103485: 48 89 df mov %rbx,%rdi
0198 103488: e8 00 00 00 00 call 10348d <ftrace_replace_code+0x19d> 103489: R_X86_64_PLT32 ftrace_rec_iter_next-0x4
019d 10348d: 48 89 c3 mov %rax,%rbx
01a0 103490: 48 85 c0 test %rax,%rax
01a3 103493: 75 ae jne 103443 <ftrace_replace_code+0x153>
01a5 103495: 48 8b 44 24 08 mov 0x8(%rsp),%rax
01aa 10349a: 65 48 2b 04 25 28 00 00 00 sub %gs:0x28,%rax
01b3 1034a3: 75 50 jne 1034f5 <ftrace_replace_code+0x205>
01b5 1034a5: 48 83 c4 10 add $0x10,%rsp
01b9 1034a9: 5b pop %rbx
01ba 1034aa: 5d pop %rbp
01bb 1034ab: 41 5c pop %r12
01bd 1034ad: 41 5d pop %r13
01bf 1034af: 41 5e pop %r14
01c1 1034b1: 41 5f pop %r15
01c3 1034b3: e9 00 00 00 00 jmp 1034b8 <ftrace_replace_code+0x1c8> 1034b4: R_X86_64_PLT32 text_poke_finish-0x4
01c8 1034b8: 4c 8b 3d 00 00 00 00 mov 0x0(%rip),%r15 # 1034bf <ftrace_replace_code+0x1cf> 1034bb: R_X86_64_PC32 x86_nops+0x24
01cf 1034bf: e9 bf fe ff ff jmp 103383 <ftrace_replace_code+0x93>
01d4 1034c4: 48 89 ef mov %rbp,%rdi
01d7 1034c7: e8 00 00 00 00 call 1034cc <ftrace_replace_code+0x1dc> 1034c8: R_X86_64_PLT32 ftrace_get_addr_new-0x4
01dc 1034cc: 48 8b 55 00 mov 0x0(%rbp),%rdx
01e0 1034d0: 48 c7 c6 00 00 00 00 mov $0x0,%rsi 1034d3: R_X86_64_32S .bss+0x16538
01e7 1034d7: 41 c6 45 00 e8 movb $0xe8,0x0(%r13)
01ec 1034dc: 48 83 c2 05 add $0x5,%rdx
01f0 1034e0: 48 29 d0 sub %rdx,%rax
01f3 1034e3: 41 89 45 01 mov %eax,0x1(%r13)
01f7 1034e7: eb 81 jmp 10346a <ftrace_replace_code+0x17a>
01f9 1034e9: 0f 0b ud2
01fb 1034eb: bf f2 ff ff ff mov $0xfffffff2,%edi
0200 1034f0: e9 cd fe ff ff jmp 1033c2 <ftrace_replace_code+0xd2>
0205 1034f5: e8 00 00 00 00 call 1034fa <ftrace_replace_code+0x20a> 1034f6: R_X86_64_PLT32 __stack_chk_fail-0x4
020a 1034fa: 66 0f 1f 44 00 00 nopw 0x0(%rax,%rax,1)
0210 103500: 90 nop
0211 103501: 90 nop
0212 103502: 90 nop
0213 103503: 90 nop
0214 103504: 90 nop
0215 103505: 90 nop
0216 103506: 90 nop
0217 103507: 90 nop
0218 103508: 90 nop
0219 103509: 90 nop
021a 10350a: 90 nop
021b 10350b: 90 nop
021c 10350c: 90 nop
021d 10350d: 90 nop
021e 10350e: 90 nop
021f 10350f: 90 nop
0000 0000000000012a83 <ftrace_replace_code.cold>:
0000 12a83: 48 89 de mov %rbx,%rsi
0003 12a86: 89 c7 mov %eax,%edi
0005 12a88: 5b pop %rbx
0006 12a89: 5d pop %rbp
0007 12a8a: 41 5c pop %r12
0009 12a8c: 41 5d pop %r13
000b 12a8e: 41 5e pop %r14
000d 12a90: e9 62 fd ff ff jmp 127f7 <ftrace_bug>
Seems to suggest objtool is actually right; I cannot find a reference to
that cold symbol.
Powered by blists - more mailing lists