[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <8735hyy3yc.ffs@tglx>
Date: Wed, 27 Apr 2022 16:51:39 +0200
From: Thomas Gleixner <tglx@...utronix.de>
To: kernel test robot <oliver.sang@...el.com>
Cc: Ammar Faizi <ammarfaizi2@...weeb.org>, lkp@...ts.01.org,
lkp@...el.com, LKML <linux-kernel@...r.kernel.org>, x86@...nel.org,
"Rafael J. Wysocki" <rafael@...nel.org>
Subject: Re: [x86/aperfmperf] adabc53c25:
WARNING:at_arch/x86/kernel/cpu/aperfmperf.c:#freq_invariance_enable
On Tue, Apr 26 2022 at 15:22, kernel test robot wrote:
> [ 11.086251][ T1] WARNING: CPU: 4 PID: 1 at arch/x86/kernel/cpu/aperfmperf.c:431 freq_invariance_enable (arch/x86/kernel/cpu/aperfmperf.c:431)
> [ 11.227178][ T1] init_freq_invariance_cppc (arch/x86/kernel/acpi/cppc.c:82 arch/x86/kernel/acpi/cppc.c:96)
> [ 11.232574][ T1] acpi_cppc_processor_probe (drivers/acpi/cppc_acpi.c:832)
Bah, the commit in question assumed that the CPPC muck is AMD only....
Fix below.
Thanks,
tglx
---
--- a/arch/x86/kernel/acpi/cppc.c
+++ b/arch/x86/kernel/acpi/cppc.c
@@ -91,6 +91,9 @@ void init_freq_invariance_cppc(void)
if (!cpu_feature_enabled(X86_FEATURE_APERFMPERF))
return;
+ if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
+ return;
+
mutex_lock(&freq_invariance_lock);
if (!init_done)
amd_set_max_freq_ratio();
Powered by blists - more mailing lists