* [PATCH v2] KVM/x86: make function emulator_do_task_switch as noinline_for_stack
@ 2024-07-12 7:50 flyingpenghao
2024-07-12 14:51 ` Sean Christopherson
0 siblings, 1 reply; 2+ messages in thread
From: flyingpenghao @ 2024-07-12 7:50 UTC (permalink / raw)
To: seanjc, pbonzini; +Cc: kvm, Peng Hao
From: Peng Hao <flyingpeng@tencent.com>
When KASAN is enabled and built with clang:
clang report
arch/x86/kvm/emulate.c:3022:5: error: stack frame size (2488) exceeds limit (2048) in 'emulator_task_switch' [-Werror,-Wframe-larger-than]
int emulator_task_switch(struct x86_emulate_ctxt *ctxt,
^
since emulator_do_task_switch() consumes a lot of stack space, mark it as
noinline_for_stack to prevent it from blowing up emulator_task_switch()'s
stack size.
Signed-off-by: Peng Hao <flyingpeng@tencent.com>
---
arch/x86/kvm/emulate.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 5d4c86133453..bbc185b9725d 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -2918,7 +2918,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, u16 old_tss_sel,
return load_state_from_tss32(ctxt, &tss_seg);
}
-static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt,
+static noinline_for_stack int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt,
u16 tss_selector, int idt_index, int reason,
bool has_error_code, u32 error_code)
{
--
2.27.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH v2] KVM/x86: make function emulator_do_task_switch as noinline_for_stack
2024-07-12 7:50 [PATCH v2] KVM/x86: make function emulator_do_task_switch as noinline_for_stack flyingpenghao
@ 2024-07-12 14:51 ` Sean Christopherson
0 siblings, 0 replies; 2+ messages in thread
From: Sean Christopherson @ 2024-07-12 14:51 UTC (permalink / raw)
To: flyingpenghao; +Cc: pbonzini, kvm, Peng Hao
On Fri, Jul 12, 2024, flyingpenghao@gmail.com wrote:
> From: Peng Hao <flyingpeng@tencent.com>
>
> When KASAN is enabled and built with clang:
> clang report
> arch/x86/kvm/emulate.c:3022:5: error: stack frame size (2488) exceeds limit (2048) in 'emulator_task_switch' [-Werror,-Wframe-larger-than]
> int emulator_task_switch(struct x86_emulate_ctxt *ctxt,
> ^
>
> since emulator_do_task_switch() consumes a lot of stack space, mark it as
> noinline_for_stack to prevent it from blowing up emulator_task_switch()'s
> stack size.
No, sprinkling noinline to combat KASAN stack frame sizes is not a maintainable
approach. Practically speaking, KASAN + -Werror + FRAME_WARN=2048 is not a sane
config.
> Signed-off-by: Peng Hao <flyingpeng@tencent.com>
> ---
> arch/x86/kvm/emulate.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
> index 5d4c86133453..bbc185b9725d 100644
> --- a/arch/x86/kvm/emulate.c
> +++ b/arch/x86/kvm/emulate.c
> @@ -2918,7 +2918,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, u16 old_tss_sel,
> return load_state_from_tss32(ctxt, &tss_seg);
> }
>
> -static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt,
> +static noinline_for_stack int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt,
> u16 tss_selector, int idt_index, int reason,
> bool has_error_code, u32 error_code)
> {
> --
> 2.27.0
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2024-07-12 14:51 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-07-12 7:50 [PATCH v2] KVM/x86: make function emulator_do_task_switch as noinline_for_stack flyingpenghao
2024-07-12 14:51 ` Sean Christopherson
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox