[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID:
<173143547242.3415.16207372030310222687.stgit@skinsburskii-cloud-desktop.internal.cloudapp.net>
Date: Tue, 12 Nov 2024 18:18:01 +0000
From: Stanislav Kinsburskii <skinsburskii@...ux.microsoft.com>
To: tglx@...utronix.de, mingo@...hat.com, bp@...en8.de,
dave.hansen@...ux.intel.com
Cc: kys@...rosoft.com, haiyangz@...rosoft.com, wei.liu@...nel.org,
decui@...rosoft.com, x86@...nel.org, hpa@...or.com,
linux-hyperv@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: [PATCH] x86/hyperv: Set X86_FEATURE_TSC_RELIABLE unconditionally
Enable X86_FEATURE_TSC_RELIABLE by default as X86_FEATURE_TSC_RELIABLE is
independent from invariant TSC and should have never been gated by the
HV_ACCESS_TSC_INVARIANT privilege.
To elaborate, the HV_ACCESS_TSC_INVARIANT privilege allows certain types of
guests to opt-in to invariant TSC by writing the
HV_X64_MSR_TSC_INVARIANT_CONTROL register. Not all guests will have this
privilege and the hypervisor will automatically opt-in certain types of
guests (e.g. EXO partitions) to invariant TSC, but this functionality is
unrelated to the TSC reliability.
Signed-off-by: Stanislav Kinsburskii <skinsburskii@...ux.microsoft.com>
---
arch/x86/kernel/cpu/mshyperv.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c
index d18078834ded..14412afcc398 100644
--- a/arch/x86/kernel/cpu/mshyperv.c
+++ b/arch/x86/kernel/cpu/mshyperv.c
@@ -515,7 +515,7 @@ static void __init ms_hyperv_init_platform(void)
machine_ops.crash_shutdown = hv_machine_crash_shutdown;
#endif
#endif
- if (ms_hyperv.features & HV_ACCESS_TSC_INVARIANT) {
+ if (ms_hyperv.features & HV_ACCESS_TSC_INVARIANT)
/*
* Writing to synthetic MSR 0x40000118 updates/changes the
* guest visible CPUIDs. Setting bit 0 of this MSR enables
@@ -526,8 +526,8 @@ static void __init ms_hyperv_init_platform(void)
* is called.
*/
wrmsrl(HV_X64_MSR_TSC_INVARIANT_CONTROL, HV_EXPOSE_INVARIANT_TSC);
- setup_force_cpu_cap(X86_FEATURE_TSC_RELIABLE);
- }
+
+ setup_force_cpu_cap(X86_FEATURE_TSC_RELIABLE);
/*
* Generation 2 instances don't support reading the NMI status from
Powered by blists - more mailing lists