linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: linux-mm@kvack.org
Cc: bugzilla-daemon@bugzilla.kernel.org, iceman_dvd@yahoo.com
Subject: Re: [Bug 56881] New: MAP_HUGETLB mmap fails for certain sizes
Date: Tue, 23 Apr 2013 13:25:22 -0700	[thread overview]
Message-ID: <20130423132522.042fa8d27668bbca6a410a92@linux-foundation.org> (raw)
In-Reply-To: <bug-56881-27@https.bugzilla.kernel.org/>


(switched to email.  Please respond via emailed reply-to-all, not via the
bugzilla web interface).

On Sat, 20 Apr 2013 03:00:30 +0000 (UTC) bugzilla-daemon@bugzilla.kernel.org wrote:

> https://bugzilla.kernel.org/show_bug.cgi?id=56881
> 
>            Summary: MAP_HUGETLB mmap fails for certain sizes
>            Product: Memory Management
>            Version: 2.5
>     Kernel Version: 3.5.0-27

Thanks.

It's a post-3.4 regression, testcase included.  Does someone want to
take a look, please?

>           Platform: All
>         OS/Version: Linux
>               Tree: Mainline
>             Status: NEW
>           Severity: high
>           Priority: P1
>          Component: Other
>         AssignedTo: akpm@linux-foundation.org
>         ReportedBy: iceman_dvd@yahoo.com
>         Regression: No
> 
> 
> This is on an Ubuntu 12.10 desktop, but the same issue has been found on 12.04
> with 3.5.0 kernel.
> See the sample program. An allocation with MAP_HUGETLB consistently fails with
> certain sizes, while it succeeds with others.
> The allocation sizes are well below the number of free huge pages.
> 
> $ uname -a Linux davide-lnx2 3.5.0-27-generic #46-Ubuntu SMP Mon Mar 25
> 19:58:17 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
> 
> 
> # echo 100 > /proc/sys/vm/nr_hugepages
> 
> # cat /proc/meminfo
> ...
> AnonHugePages:         0 kB
> HugePages_Total:     100
> HugePages_Free:      100
> HugePages_Rsvd:        0
> HugePages_Surp:        0
> Hugepagesize:       2048 kB
> 
> 
> $ ./mmappu $((5 * 2 * 1024 * 1024 - 4096))
> size=10481664    0x9ff000
> hugepage mmap: Invalid argument
> 
> 
> $ ./mmappu $((5 * 2 * 1024 * 1024 - 4095))
> size=10481665    0x9ff001
> OK!
> 
> 
> It seems the trigger point is a normal page size.
> The same binary works flawlessly in previous kernels.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

       reply	other threads:[~2013-04-23 20:25 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <bug-56881-27@https.bugzilla.kernel.org/>
2013-04-23 20:25 ` Andrew Morton [this message]
2013-04-23 22:26   ` Naoya Horiguchi
2013-04-24  8:20     ` Jianguo Wu
2013-04-24 14:17       ` Naoya Horiguchi
2013-04-24  8:14   ` Johannes Weiner
2013-04-24 15:16     ` Naoya Horiguchi
2013-04-24 15:39       ` Johannes Weiner
2013-04-24 23:05         ` Naoya Horiguchi
2013-04-24 23:13           ` Naoya Horiguchi
2013-04-24 23:44             ` Johannes Weiner
2013-04-24 23:26           ` Johannes Weiner
2013-04-25 21:00             ` Naoya Horiguchi
2013-04-26  4:35               ` [PATCH v2] hugetlbfs: fix mmap failure in unaligned size request Naoya Horiguchi
2013-04-30 16:45                 ` Johannes Weiner
2013-04-30 17:02                   ` [PATCH v3] " Naoya Horiguchi
2013-05-01  8:00                     ` Sam Ben
2013-04-25  3:02           ` [Bug 56881] New: MAP_HUGETLB mmap fails for certain sizes Jianguo Wu
2013-04-25 21:03             ` Naoya Horiguchi
2013-06-12 12:16           ` Aneesh Kumar K.V
2013-06-13 21:29             ` Andrew Morton
2013-06-18 11:14               ` Aneesh Kumar K.V

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=20130423132522.042fa8d27668bbca6a410a92@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=bugzilla-daemon@bugzilla.kernel.org \
    --cc=iceman_dvd@yahoo.com \
    --cc=linux-mm@kvack.org \
    /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