* [PATCH] x86/xen: fix xsave capability setting
@ 2017-04-25 6:47 Juergen Gross
2017-04-26 13:51 ` Boris Ostrovsky
0 siblings, 1 reply; 2+ messages in thread
From: Juergen Gross @ 2017-04-25 6:47 UTC (permalink / raw)
To: linux-kernel, xen-devel; +Cc: boris.ostrovsky, Juergen Gross
Commit 690b7f10b4f9f ("x86/xen: use capabilities instead of fake cpuid
values for xsave") introduced a regression as it tried to make use of
the fixup feature before it being available.
Fall back to the old variant testing via cpuid().
Signed-off-by: Juergen Gross <jgross@suse.com>
---
arch/x86/xen/enlighten_pv.c | 30 ++++++++----------------------
1 file changed, 8 insertions(+), 22 deletions(-)
diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c
index a732bc2b9dfc..a1af4f68278f 100644
--- a/arch/x86/xen/enlighten_pv.c
+++ b/arch/x86/xen/enlighten_pv.c
@@ -277,26 +277,15 @@ static bool __init xen_check_mwait(void)
static bool __init xen_check_xsave(void)
{
- unsigned int err, eax, edx;
+ unsigned int cx, xsave_mask;
- /*
- * Xen 4.0 and older accidentally leaked the host XSAVE flag into guest
- * view, despite not being able to support guests using the
- * functionality. Probe for the actual availability of XSAVE by seeing
- * whether xgetbv executes successfully or raises #UD.
- */
- asm volatile("1: .byte 0x0f,0x01,0xd0\n\t" /* xgetbv */
- "xor %[err], %[err]\n"
- "2:\n\t"
- ".pushsection .fixup,\"ax\"\n\t"
- "3: movl $1,%[err]\n\t"
- "jmp 2b\n\t"
- ".popsection\n\t"
- _ASM_EXTABLE(1b, 3b)
- : [err] "=r" (err), "=a" (eax), "=d" (edx)
- : "c" (0));
+ cx = cpuid_ecx(1);
- return err == 0;
+ xsave_mask = (1 << (X86_FEATURE_XSAVE % 32)) |
+ (1 << (X86_FEATURE_OSXSAVE % 32));
+
+ /* Xen will set CR4.OSXSAVE if supported and not disabled by force */
+ return (cx & xsave_mask) == xsave_mask;
}
static void __init xen_init_capabilities(void)
@@ -317,10 +306,7 @@ static void __init xen_init_capabilities(void)
else
setup_clear_cpu_cap(X86_FEATURE_MWAIT);
- if (xen_check_xsave()) {
- setup_force_cpu_cap(X86_FEATURE_XSAVE);
- setup_force_cpu_cap(X86_FEATURE_OSXSAVE);
- } else {
+ if (!xen_check_xsave()) {
setup_clear_cpu_cap(X86_FEATURE_XSAVE);
setup_clear_cpu_cap(X86_FEATURE_OSXSAVE);
}
--
2.12.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] x86/xen: fix xsave capability setting
2017-04-25 6:47 [PATCH] x86/xen: fix xsave capability setting Juergen Gross
@ 2017-04-26 13:51 ` Boris Ostrovsky
0 siblings, 0 replies; 2+ messages in thread
From: Boris Ostrovsky @ 2017-04-26 13:51 UTC (permalink / raw)
To: Juergen Gross, linux-kernel, xen-devel
On 04/25/2017 02:47 AM, Juergen Gross wrote:
> Commit 690b7f10b4f9f ("x86/xen: use capabilities instead of fake cpuid
> values for xsave") introduced a regression as it tried to make use of
> the fixup feature before it being available.
>
> Fall back to the old variant testing via cpuid().
>
> Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Boris Ostrovsky <boris.osrtovsky@oracle.com>
(This, BTW, was apparently broken on Intel as well, at least if you go
back to Xen 4.1).
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2017-04-26 13:52 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-04-25 6:47 [PATCH] x86/xen: fix xsave capability setting Juergen Gross
2017-04-26 13:51 ` Boris Ostrovsky
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).