From: Vlastimil Babka <vbabka@suse.cz>
To: Hanjun Guo <guohanjun@huawei.com>, Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: "Leizhen (ThunderTown)" <thunder.leizhen@huawei.com>,
Laura Abbott <labbott@redhat.com>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Sasha Levin <sasha.levin@oracle.com>,
Laura Abbott <lauraa@codeaurora.org>,
qiuxishi <qiuxishi@huawei.com>,
Catalin Marinas <Catalin.Marinas@arm.com>,
Will Deacon <will.deacon@arm.com>, Arnd Bergmann <arnd@arndb.de>,
dingtinahong <dingtianhong@huawei.com>,
chenjie6@huawei.com, "linux-mm@kvack.org" <linux-mm@kvack.org>
Subject: Re: Suspicious error for CMA stress test
Date: Thu, 17 Mar 2016 16:43:48 +0100 [thread overview]
Message-ID: <56EAD0B4.2060807@suse.cz> (raw)
In-Reply-To: <56EA77BC.2090702@huawei.com>
On 03/17/2016 10:24 AM, Hanjun Guo wrote:
> On 2016/3/17 14:54, Joonsoo Kim wrote:
>> On Wed, Mar 16, 2016 at 05:44:28PM +0800, Hanjun Guo wrote:
>>> On 2016/3/14 15:18, Joonsoo Kim wrote:
>>>> On Mon, Mar 14, 2016 at 08:06:16AM +0100, Vlastimil Babka wrote:
>>>>> On 03/14/2016 07:49 AM, Joonsoo Kim wrote:
>>>>>> On Fri, Mar 11, 2016 at 06:07:40PM +0100, Vlastimil Babka wrote:
>>>>>>> On 03/11/2016 04:00 PM, Joonsoo Kim wrote:
>>>>>>>
>>>>>>> How about something like this? Just and idea, probably buggy (off-by-one etc.).
>>>>>>> Should keep away cost from <pageblock_order iterations at the expense of the
>>>>>>> relatively fewer >pageblock_order iterations.
>>>>>> Hmm... I tested this and found that it's code size is a little bit
>>>>>> larger than mine. I'm not sure why this happens exactly but I guess it would be
>>>>>> related to compiler optimization. In this case, I'm in favor of my
>>>>>> implementation because it looks like well abstraction. It adds one
>>>>>> unlikely branch to the merge loop but compiler would optimize it to
>>>>>> check it once.
>>>>> I would be surprised if compiler optimized that to check it once, as
>>>>> order increases with each loop iteration. But maybe it's smart
>>>>> enough to do something like I did by hand? Guess I'll check the
>>>>> disassembly.
>>>> Okay. I used following slightly optimized version and I need to
>>>> add 'max_order = min_t(unsigned int, MAX_ORDER, pageblock_order + 1)'
>>>> to yours. Please consider it, too.
>>> Hmm, this one is not work, I still can see the bug is there after applying
>>> this patch, did I miss something?
>> I may find that there is a bug which was introduced by me some time
>> ago. Could you test following change in __free_one_page() on top of
>> Vlastimil's patch?
>>
>> -page_idx = pfn & ((1 << max_order) - 1);
>> +page_idx = pfn & ((1 << MAX_ORDER) - 1);
>
> I tested Vlastimil's patch + your change with stress for more than half hour, the bug
> I reported is gone :)
Oh, ok, will try to send proper patch, once I figure out what to write
in the changelog :)
next prev parent reply other threads:[~2016-03-17 15:43 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-02 13:52 Suspicious error for CMA stress test Hanjun Guo
2016-03-03 1:25 ` Laura Abbott
2016-03-03 6:07 ` Hanjun Guo
2016-03-03 7:42 ` Joonsoo Kim
2016-03-03 7:58 ` Hanjun Guo
2016-03-03 12:49 ` Hanjun Guo
2016-03-03 18:52 ` Laura Abbott
2016-03-04 2:09 ` Joonsoo Kim
2016-03-04 6:09 ` Hanjun Guo
2016-03-04 2:02 ` Joonsoo Kim
2016-03-04 4:32 ` Joonsoo Kim
2016-03-04 6:05 ` Hanjun Guo
2016-03-04 6:38 ` Joonsoo Kim
2016-03-04 7:35 ` Hanjun Guo
2016-03-07 4:34 ` Joonsoo Kim
2016-03-07 8:16 ` Leizhen (ThunderTown)
2016-03-07 18:42 ` Laura Abbott
2016-03-08 1:54 ` Leizhen (ThunderTown)
2016-03-09 1:23 ` Leizhen (ThunderTown)
2016-03-11 15:00 ` Joonsoo Kim
2016-03-11 17:07 ` Vlastimil Babka
2016-03-14 6:49 ` Joonsoo Kim
2016-03-14 7:06 ` Vlastimil Babka
2016-03-14 7:18 ` Joonsoo Kim
2016-03-14 12:30 ` Vlastimil Babka
2016-03-14 14:10 ` Joonsoo Kim
2016-03-16 12:03 ` Vlastimil Babka
2016-03-16 9:44 ` Hanjun Guo
2016-03-17 6:54 ` Joonsoo Kim
2016-03-17 9:24 ` Hanjun Guo
2016-03-17 15:31 ` Joonsoo Kim
2016-03-18 2:03 ` Hanjun Guo
2016-03-17 15:43 ` Vlastimil Babka [this message]
2016-03-17 15:52 ` Joonsoo Kim
2016-03-18 13:32 ` Lucas Stach
2016-03-21 4:42 ` Joonsoo Kim
2016-03-22 14:56 ` Lucas Stach
2016-03-23 4:42 ` Joonsoo Kim
2016-03-18 14:10 ` Vlastimil Babka
2016-03-18 14:42 ` Lucas Stach
2016-03-18 20:58 ` Vlastimil Babka
2016-03-22 14:47 ` Lucas Stach
2016-03-19 7:24 ` Hanjun Guo
2016-03-19 22:11 ` Vlastimil Babka
2016-03-23 4:44 ` Joonsoo Kim
2016-03-23 8:26 ` Vlastimil Babka
2016-03-23 8:32 ` Joonsoo Kim
2016-03-18 12:29 ` Vlastimil Babka
2016-03-08 4:03 ` Hanjun Guo
2016-03-07 12:59 ` Vlastimil Babka
2016-03-08 7:48 ` Joonsoo Kim
2016-03-08 10:45 ` Xishi Qiu
2016-03-08 15:36 ` Joonsoo Kim
2016-03-09 2:18 ` Xishi Qiu
2016-03-04 5:33 ` Hanjun Guo
2016-03-08 1:42 ` Xishi Qiu
2016-03-08 8:09 ` Joonsoo Kim
2016-03-04 6:59 ` Hanjun Guo
2016-03-07 4:40 ` Joonsoo Kim
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=56EAD0B4.2060807@suse.cz \
--to=vbabka@suse.cz \
--cc=Catalin.Marinas@arm.com \
--cc=akpm@linux-foundation.org \
--cc=arnd@arndb.de \
--cc=chenjie6@huawei.com \
--cc=dingtianhong@huawei.com \
--cc=guohanjun@huawei.com \
--cc=iamjoonsoo.kim@lge.com \
--cc=labbott@redhat.com \
--cc=lauraa@codeaurora.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=qiuxishi@huawei.com \
--cc=sasha.levin@oracle.com \
--cc=thunder.leizhen@huawei.com \
--cc=will.deacon@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).