[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220407155106.GL2731@worktop.programming.kicks-ass.net>
Date: Thu, 7 Apr 2022 17:51:06 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: "Edgecombe, Rick P" <rick.p.edgecombe@...el.com>
Cc: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"Poimboe, Josh" <jpoimboe@...hat.com>,
"x86@...nel.org" <x86@...nel.org>
Subject: Re: IBT related objtool warnings
On Wed, Apr 06, 2022 at 10:43:33PM +0000, Edgecombe, Rick P wrote:
> Hi,
>
> After commit:
> ed53a0d97192 x86/alternative: Use .ibt_endbr_seal to seal indirect calls
>
> I am getting objtool "unreachable instruction" warnings.
>
> On one system with gcc 8.4.0:
> vmlinux.o: warning: objtool: start_secondary()+0x10e: unreachable
> instruction
>
> On another with gcc 11.2.1:
> vmlinux.o: warning: objtool: pvh_start_xen()+0x0: unreachable
> instruction
>
> Let me know if any tests on my end would help.
This seems to make it go on gcc-11.2 tip/x86/urgent..
---
arch/x86/entry/entry_64.S | 3 +++
arch/x86/platform/pvh/head.S | 1 +
2 files changed, 4 insertions(+)
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index 4faac48ebec5..73d958522b6a 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -337,6 +337,9 @@ SYM_CODE_END(ret_from_fork)
call \cfunc
+ /* For some configurations \cfunc ends up being a noreturn. */
+ REACHABLE
+
jmp error_return
.endm
diff --git a/arch/x86/platform/pvh/head.S b/arch/x86/platform/pvh/head.S
index 72c1e42d121d..7fe564eaf228 100644
--- a/arch/x86/platform/pvh/head.S
+++ b/arch/x86/platform/pvh/head.S
@@ -50,6 +50,7 @@
#define PVH_DS_SEL (PVH_GDT_ENTRY_DS * 8)
SYM_CODE_START_LOCAL(pvh_start_xen)
+ UNWIND_HINT_EMPTY
cld
lgdt (_pa(gdt))
Powered by blists - more mailing lists