linux-kernel - Re: [linus:master] [mm] c1753fd02a: stress-ng.madvise.ops_per

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]

Date:   Mon, 4 Sep 2023 06:04:28 -0400
From:   Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
To:     Yin Fengwei <fengwei.yin@...el.com>,
        kernel test robot <oliver.sang@...el.com>
Cc:     oe-lkp@...ts.linux.dev, lkp@...el.com,
        linux-kernel@...r.kernel.org,
        Andrew Morton <akpm@...ux-foundation.org>,
        kernel test robot <yujie.liu@...el.com>,
        Aaron Lu <aaron.lu@...el.com>,
        John Hubbard <jhubbard@...dia.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Olivier Dion <odion@...icios.com>,
        Feng Tang <feng.tang@...el.com>,
        Jason Gunthorpe <jgg@...dia.com>, Peter Xu <peterx@...hat.com>,
        ying.huang@...el.com
Subject: Re: [linus:master] [mm] c1753fd02a: stress-ng.madvise.ops_per_sec
 -6.5% regression

On 9/4/23 01:32, Yin Fengwei wrote:
> 
> 
> On 7/19/23 14:34, kernel test robot wrote:
>>
>> hi, Mathieu Desnoyers,
>>
>> we noticed that this commit addressed issue:
>>    "[linus:master] [sched] af7f588d8f: will-it-scale.per_thread_ops -13.9% regression"
>> we reported before on:
>>    https://lore.kernel.org/oe-lkp/202305151017.27581d75-yujie.liu@intel.com/
>>
>> we really saw a will-it-scale.per_thread_ops 92.2% improvement by this commit
>> (details are as below).
>> however, we also noticed a stress-ng regression.
>>
>> below detail report FYI.
>>
>>
>> Hello,
>>
>> kernel test robot noticed a -6.5% regression of stress-ng.madvise.ops_per_sec on:
>>
>>
>> commit: c1753fd02a0058ea43cbb31ab26d25be2f6cfe08 ("mm: move mm_count into its own cache line")
>> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
> I noticed that the struct mm_struct has following layout change after this patch.
> Without the patch:
>                  spinlock_t         page_table_lock;      /*   124     4 */
>                  /* --- cacheline 2 boundary (128 bytes) --- */
>                  struct rw_semaphore mmap_lock;           /*   128    40 */   ----> in one cache line
>                  struct list_head   mmlist;               /*   168    16 */
>                  int                mm_lock_seq;          /*   184     4 */
> 
> With the patch:
>                  spinlock_t         page_table_lock;      /*   180     4 */
>                  struct rw_semaphore mmap_lock;           /*   184    40 */   ----> cross to two cache lines
>                  /* --- cacheline 3 boundary (192 bytes) was 32 bytes ago --- */
>                  struct list_head   mmlist;               /*   224    16 */

If your intent is just to make sure that mmap_lock is entirely contained
within a cache line by forcing it to begin on a cache line boundary, you
can do:

struct mm_struct {
[...]
     struct rw_semaphore mmap_lock ____cacheline_aligned_in_smp;
     struct list_head mmlist;
[...]
};

The code above keeps mmlist on the same cache line as mmap_lock if
there happens to be enough room in the cache line after mmap_lock.

Otherwise, if your intent is to also eliminate false sharing by making
sure that mmap_lock sits alone in its cache line, you can do the following:

struct mm_struct {
[...]
     struct {
         struct rw_semaphore mmap_lock;
     } ____cacheline_aligned_in_smp;
     struct list_head mmlist;
[...]
};

The code above keeps mmlist in a separate cache line from mmap_lock;

Depending on the usage, one or the other may be better. Comparative
benchmarks of both approaches would help choosing the best way forward
here.

Thanks,

Mathieu

-- 
Mathieu Desnoyers
EfficiOS Inc.
https://www.efficios.com