[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <8496afee057d63b83a7ff02ec7f1de8c2d0e97ae.camel@intel.com>
Date:   Tue, 16 Aug 2022 17:47:14 +0800
From:   Zhang Rui <rui.zhang@...el.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     linux-kernel@...r.kernel.org, x86@...nel.org,
        linux-hwmon@...r.kernel.org, tglx@...utronix.de, mingo@...hat.com,
        bp@...en8.de, dave.hansen@...ux.intel.com, hpa@...or.com,
        corbet@....net, fenghua.yu@...el.com, jdelvare@...e.com,
        linux@...ck-us.net, len.brown@...el.com
Subject: Re: [PATCH V2 1/8] perf/x86/intel/P4: Fix smp_num_siblings usage
On Tue, 2022-08-16 at 10:27 +0200, Peter Zijlstra wrote:
> On Tue, Aug 16, 2022 at 01:16:26PM +0800, Zhang Rui wrote:
> > smp_num_siblings can be larger than 2.
> > 
> > Any value larger than 1 suggests HT is supported.
> 
> Subject and Changelog are still utterly insane.
what about this one, do I need to resend the full series?
thanks,
rui
>From 2e368e6afa83cb73e44ac8c3cf8339207097d9e1 Mon Sep 17 00:00:00 2001
From: Zhang Rui <rui.zhang@...el.com>
Date: Tue, 28 Jun 2022 11:02:12 +0800
Subject: [PATCH V3 1/8] perf/x86/intel/P4: unify logic for detecting HT
Any value larger than 1 suggests HT is supported.
Although smp_num_siblings cannot be larger than 2 on P4 platform, it is
better to keep this logic consistent across the kernel.
Reviewed-by: Len Brown <len.brown@...el.com>
Signed-off-by: Zhang Rui <rui.zhang@...el.com>
---
 arch/x86/include/asm/perf_event_p4.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/include/asm/perf_event_p4.h
b/arch/x86/include/asm/perf_event_p4.h
index 94de1a05aeba..b14e9a20a7c0 100644
--- a/arch/x86/include/asm/perf_event_p4.h
+++ b/arch/x86/include/asm/perf_event_p4.h
@@ -189,7 +189,7 @@ static inline int p4_ht_active(void)
 static inline int p4_ht_thread(int cpu)
 {
 #ifdef CONFIG_SMP
-	if (smp_num_siblings == 2)
+	if (smp_num_siblings > 1)
 		return cpu !=
cpumask_first(this_cpu_cpumask_var_ptr(cpu_sibling_map));
 #endif
 	return 0;
-- 
2.25.1
Powered by blists - more mailing lists
 
