linux-kernel - Re: [PATCH v3 1/3] mm/pageblock: mitigation cmpxchg false sharing in pageblock flags

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <81d1feb1-2373-de9b-f788-adcce9647b41@linux.alibaba.com>
Date:   Thu, 3 Sep 2020 14:42:33 +0800
From:   Alex Shi <alex.shi@...ux.alibaba.com>
To:     Vlastimil Babka <vbabka@...e.cz>,
        Anshuman Khandual <anshuman.khandual@....com>,
        David Hildenbrand <david@...hat.com>,
        Matthew Wilcox <willy@...radead.org>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Mel Gorman <mgorman@...hsingularity.net>, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3 1/3] mm/pageblock: mitigation cmpxchg false sharing in
 pageblock flags



在 2020/9/2 上午1:06, Vlastimil Babka 写道:
>>
>>                    pageblock   pageblock   pageblock        rc2         rc2         rc2
>>                           16        16-2        16-3          a           b           c
>> Duration User          14.81       15.24       14.55      14.76       14.97       14.38
>> Duration System        84.44       88.38       90.64     100.43       89.15       88.89
>> Duration Elapsed       98.83       99.06       99.81     100.30       99.24       99.14
> The large variance in these numbers suggest that 3 iterations are not enough to
> conclude a statistically significant difference. You'd need more iterations and
> calculate at least mean+variance.
> 

on the machine I did seeing much variation more on Amean. but the trace event would
be more straight. It could reduce the hit_cmpxchg from thousand time to hundreds or less.

Thanks
Alex

diff --git a/include/trace/events/pageblock.h b/include/trace/events/pageblock.h
new file mode 100644
index 000000000000..003c2d716f82
--- /dev/null
+++ b/include/trace/events/pageblock.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM pageblock
+
+#if !defined(_TRACE_PAGEBLOCK_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_PAGEBLOCK_H
+
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(hit_cmpxchg,
+
+       TP_PROTO(char byte),
+
+       TP_ARGS(byte),
+
+       TP_STRUCT__entry(
+               __field(char, byte)
+       ),
+
+       TP_fast_assign(
+               __entry->byte = byte;
+       ),
+
+       TP_printk("%d", __entry->byte)
+);
+
+#endif /* _TRACE_PAGE_ISOLATION_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 60342e764090..2422dec00484 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -509,6 +509,9 @@ static __always_inline int get_pfnblock_migratetype(struct page *page, unsigned
  * @pfn: The target page frame number
  * @mask: mask of bits that the caller is interested in
  */
+#define CREATE_TRACE_POINTS
+#include <trace/events/pageblock.h>
+
 void set_pfnblock_flags_mask(struct page *page, unsigned long flags,
                                        unsigned long pfn,
                                        unsigned long mask)
@@ -536,6 +539,7 @@ void set_pfnblock_flags_mask(struct page *page, unsigned long flags,
                if (byte == old_byte)
                        break;
                byte = old_byte;
+               trace_hit_cmpxchg(byte);
        }
 }