[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <164734102547.16921.14808462223133394899.tip-bot2@tip-bot2>
Date: Tue, 15 Mar 2022 10:43:45 -0000
From: "tip-bot2 for Peter Zijlstra" <tip-bot2@...utronix.de>
To: linux-tip-commits@...r.kernel.org
Cc: "Peter Zijlstra (Intel)" <peterz@...radead.org>,
Josh Poimboeuf <jpoimboe@...hat.com>, x86@...nel.org,
linux-kernel@...r.kernel.org
Subject: [tip: x86/core] x86: Annotate call_on_stack()
The following commit has been merged into the x86/core branch of tip:
Commit-ID: be0075951fde739f14ee2b659e2fd6e2499c46c0
Gitweb: https://git.kernel.org/tip/be0075951fde739f14ee2b659e2fd6e2499c46c0
Author: Peter Zijlstra <peterz@...radead.org>
AuthorDate: Tue, 08 Mar 2022 16:30:50 +01:00
Committer: Peter Zijlstra <peterz@...radead.org>
CommitterDate: Tue, 15 Mar 2022 10:32:44 +01:00
x86: Annotate call_on_stack()
vmlinux.o: warning: objtool: page_fault_oops()+0x13c: unreachable instruction
0000 000000000005b460 <page_fault_oops>:
...
0128 5b588: 49 89 23 mov %rsp,(%r11)
012b 5b58b: 4c 89 dc mov %r11,%rsp
012e 5b58e: 4c 89 f2 mov %r14,%rdx
0131 5b591: 48 89 ee mov %rbp,%rsi
0134 5b594: 4c 89 e7 mov %r12,%rdi
0137 5b597: e8 00 00 00 00 call 5b59c <page_fault_oops+0x13c> 5b598: R_X86_64_PLT32 handle_stack_overflow-0x4
013c 5b59c: 5c pop %rsp
vmlinux.o: warning: objtool: sysvec_reboot()+0x6d: unreachable instruction
0000 00000000000033f0 <sysvec_reboot>:
...
005d 344d: 4c 89 dc mov %r11,%rsp
0060 3450: e8 00 00 00 00 call 3455 <sysvec_reboot+0x65> 3451: R_X86_64_PLT32 irq_enter_rcu-0x4
0065 3455: 48 89 ef mov %rbp,%rdi
0068 3458: e8 00 00 00 00 call 345d <sysvec_reboot+0x6d> 3459: R_X86_64_PC32 .text+0x47d0c
006d 345d: e8 00 00 00 00 call 3462 <sysvec_reboot+0x72> 345e: R_X86_64_PLT32 irq_exit_rcu-0x4
0072 3462: 5c pop %rsp
Both cases are due to a call_on_stack() calling a __noreturn function.
Since that's an inline asm, GCC can't do anything about the
instructions after the CALL. Therefore put in an explicit
ASM_REACHABLE annotation to make sure objtool and gcc are consistently
confused about control flow.
Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
Acked-by: Josh Poimboeuf <jpoimboe@...hat.com>
Link: https://lore.kernel.org/r/20220308154319.468805622@infradead.org
---
arch/x86/include/asm/irq_stack.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/x86/include/asm/irq_stack.h b/arch/x86/include/asm/irq_stack.h
index ae9d40f..05af249 100644
--- a/arch/x86/include/asm/irq_stack.h
+++ b/arch/x86/include/asm/irq_stack.h
@@ -99,7 +99,8 @@
}
#define ASM_CALL_ARG0 \
- "call %P[__func] \n"
+ "call %P[__func] \n" \
+ ASM_REACHABLE
#define ASM_CALL_ARG1 \
"movq %[arg1], %%rdi \n" \
Powered by blists - more mailing lists