From: Andy Whitcroft <apw@shadowen.org>
To: Andrew Morton <akpm@linux-foundation.org>,
Christoph Lameter <clameter@sgi.com>
Cc: Andy Whitcroft <apw@shadowen.org>, linux-kernel@vger.kernel.org
Subject: Re: 2.6.21-rc7-mm2 -- PPC link failure
Date: Thu, 26 Apr 2007 19:14:18 +0100 [thread overview]
Message-ID: <4630EBFA.9000207@shadowen.org> (raw)
In-Reply-To: <4630ABF2.1040305@shadowen.org>
Andy Whitcroft wrote:
> Andy Whitcroft wrote:
>> Getting a link failure on a ppc64 system:
>>
>> LD .tmp_vmlinux1
>> init/built-in.o(.init.text+0x32e4): In function `.rd_load_image':
>> : undefined reference to `.__kmalloc_size_too_large'
>> fs/built-in.o(.text+0xa6fe0): In function `.ext3_fill_super':
>> : undefined reference to `.__kmalloc_size_too_large'
>> fs/built-in.o(.text+0xc1fe0): In function `.ext2_fill_super':
>> : undefined reference to `.__kmalloc_size_too_large'
>> fs/built-in.o(.text+0xf6a1c): In function `.nfs4_proc_lookup':
>> : undefined reference to `.__kmalloc_size_too_large'
>> fs/built-in.o(.text+0x104104): In function `.nfs_idmap_new':
>> : undefined reference to `.__kmalloc_size_too_large'
>> fs/built-in.o(.text+0x105520): more undefined references to
>> `.__kmalloc_size_too_large' follow
>> make: *** [.tmp_vmlinux1] Error 1
>
> Ok, this is a SLUB related link failure. Am investigating if PPC simply
> needs larger allocs and needs CONFIG_LARGE_ALLOCS, of if this is an
> inlining issue.
Ok this is confirmed as an inlining issue. With the compiler below on
ppc64 we get the above link failure:
gcc version 3.3.3 (SuSE Linux)
What seems to happen is that although the optimiser is capable of
collapsing the kmalloc_index() call it then fails to collapse
kmalloc_slab(). This leads to the never used reference to
__kmalloc_size_too_large() and the link failure. From my testing this
seems to occur at sizes >= 32k. At 16k all of the code collapses
correctly, at 32k it does not. I am not entirely sure what to think at
this point, it is cirtainly not at all clear why the 32k version fails
and the 16k succeeds they are almost identical.
Either way it seems to me that assuming the optimiser will remove the
code is perhaps over optimistic. Perhaps it would make more sense to
put a BUG() in here. Though that points out the anomaly that the
kmalloc() for constants has different semantics to that for variable values?
Comments?
-apw
next prev parent reply other threads:[~2007-04-26 18:14 UTC|newest]
Thread overview: 116+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-04-26 5:57 2.6.21-rc7-mm2 Andrew Morton
2007-04-26 11:47 ` 2.6.21-rc7-mm2 Gabriel C
2007-04-26 20:37 ` 2.6.21-rc7-mm2 Andrew Morton
2007-04-26 20:46 ` 2.6.21-rc7-mm2 Randy Dunlap
2007-04-29 8:05 ` 2.6.21-rc7-mm2 Geert Uytterhoeven
2007-04-26 20:57 ` 2.6.21-rc7-mm2 Timur Tabi
2007-04-26 21:13 ` 2.6.21-rc7-mm2 Gabriel C
2007-04-26 21:33 ` 2.6.21-rc7-mm2 Andrew Morton
2007-04-26 12:16 ` 2.6.21-rc7-mm2 -- x86_64 VDSO compile error Andy Whitcroft
2007-04-26 13:27 ` Mel Gorman
2007-04-26 14:11 ` Andi Kleen
2007-04-26 14:31 ` Mel Gorman
2007-04-26 14:13 ` 2.6.21-rc7-mm2 -- x86_64 VDSO compile error II Andi Kleen
2007-04-26 14:39 ` Mel Gorman
2007-04-26 15:20 ` Mel Gorman
2007-04-26 15:24 ` Andi Kleen
2007-04-26 15:45 ` Mel Gorman
2007-04-27 0:39 ` Andi Kleen
2007-04-27 8:59 ` Mel Gorman
2007-04-27 15:50 ` [PATCH] Add vDSO for x86-64 with gettimeofday/clock_gettime/getcpu fix Mel Gorman
2007-04-27 16:34 ` Andi Kleen
2007-04-27 16:49 ` Mel Gorman
2007-04-26 12:30 ` 2.6.21-rc7-mm2 -- x86_64 blade hard hangs Andy Whitcroft
2007-04-26 13:51 ` Andi Kleen
2007-04-26 13:33 ` Mel Gorman
2007-04-26 14:46 ` Mel Gorman
2007-04-26 17:40 ` Mel Gorman
[not found] ` <20070426234002.GH5475@linux-os.sc.intel.com>
[not found] ` <20070427110709.GE3645@skynet.ie>
2007-04-27 17:54 ` Siddha, Suresh B
2007-04-27 23:59 ` Mel Gorman
2007-04-26 13:17 ` 2.6.21-rc7-mm2 Andy Whitcroft
2007-04-26 13:41 ` 2.6.21-rc7-mm2 -- PPC link failure Andy Whitcroft
2007-04-26 18:14 ` Andy Whitcroft [this message]
2007-04-26 18:27 ` Christoph Lameter
2007-04-26 18:40 ` Andy Whitcroft
2007-04-26 18:49 ` Christoph Lameter
2007-04-26 19:12 ` Christoph Lameter
2007-04-26 19:48 ` Andy Whitcroft
2007-04-26 20:23 ` Christoph Lameter
2007-04-27 16:55 ` Andy Whitcroft
2007-04-27 16:58 ` Christoph Lameter
2007-04-26 18:25 ` [PATCH] mm/memory.c: remove warning from an uninitialized spinlock. was: Re: 2.6.21-rc7-mm2 Borislav Petkov
2007-04-28 0:22 ` Andrew Morton
2007-04-28 0:27 ` Jeremy Fitzhardinge
2007-04-28 5:57 ` Borislav Petkov
2007-04-28 6:25 ` Borislav Petkov
2007-04-28 6:54 ` Andrew Morton
2007-04-28 7:03 ` Jeremy Fitzhardinge
2007-04-28 23:48 ` Andy Whitcroft
2007-04-29 3:25 ` Andrew Morton
2007-04-29 6:50 ` Borislav Petkov
2007-04-29 8:19 ` Andrew Morton
2007-04-29 9:24 ` Andrew Morton
2007-04-29 21:36 ` Dave Jones
2007-04-29 21:45 ` Andrew Morton
2007-11-07 19:20 ` Steven Rostedt
2007-11-08 5:15 ` Borislav Petkov
2007-04-26 23:47 ` [-mm patch] make drivers/hwmon/applesmc.c:backlight_work static Adrian Bunk
2007-04-26 23:47 ` [-mm patch] unexport highlevel_host_reset Adrian Bunk
2007-04-27 0:16 ` Stefan Richter
2007-04-27 2:31 ` 2.6.21-rc7-mm2 breaks 'lvm vgscan' Valdis.Kletnieks
2007-04-27 2:55 ` Andrew Morton
2007-05-05 18:04 ` Valdis.Kletnieks
2007-04-28 17:56 ` 2.6.21-rc7-mm2 crash: Eeek! page_mapcount(page) went negative! (-1) Tilman Schmidt
2007-04-28 21:10 ` Andrew Morton
2007-04-28 22:06 ` Hugh Dickins
2007-04-30 17:17 ` Tilman Schmidt
2007-04-30 18:21 ` Andrew Morton
2007-04-30 19:28 ` Tilman Schmidt
2007-04-30 19:46 ` Andrew Morton
2007-04-30 21:32 ` Tilman Schmidt
2007-05-01 11:26 ` Tilman Schmidt
2007-05-02 3:10 ` Greg KH
2007-05-02 7:01 ` Tilman Schmidt
2007-05-02 7:02 ` Greg KH
2007-05-02 7:10 ` Andrew Morton
2007-05-02 7:28 ` Greg KH
2007-05-02 7:43 ` Greg KH
2007-05-02 9:41 ` Tilman Schmidt
2007-05-02 22:06 ` Greg KH
2007-05-02 12:14 ` Kay Sievers
2007-05-02 7:52 ` Greg KH
2007-05-02 17:36 ` Tilman Schmidt
2007-05-02 20:07 ` Andrew Morton
2007-05-02 21:22 ` Tilman Schmidt
2007-05-02 7:10 ` Nick Piggin
2007-04-28 19:19 ` [-mm patch] make drivers/misc/thinkpad_acpi:fan_mutex static Adrian Bunk
2007-04-28 19:58 ` Henrique de Moraes Holschuh
2007-04-29 1:53 ` Len Brown
2007-04-29 2:50 ` Adrian Bunk
2007-04-29 4:09 ` Henrique de Moraes Holschuh
2007-04-28 19:19 ` [-mm patch] MMC: make tifm_sd_set_dma_data() static Adrian Bunk
2007-05-01 14:14 ` Pierre Ossman
2007-04-28 19:20 ` [-mm patch] drivers/rtc/rtc-dev.c should #include "rtc-core.h" Adrian Bunk
2007-04-28 19:20 ` [-mm patch] scsi/lpfc/lpfc_init.c: remove unused variable Adrian Bunk
2007-04-29 19:19 ` 2.6.21-rc7-mm2 Dan Kruchinin
2007-04-29 19:48 ` 2.6.21-rc7-mm2 Andrew Morton
2007-04-30 5:01 ` 2.6.21-rc7-mm2 hangs in boot Randy Dunlap
2007-04-30 5:23 ` Andrew Morton
2007-04-30 15:16 ` Randy Dunlap
2007-04-30 23:51 ` 2.6.21-rc7-mm2 hangs in boot (netconsole) Randy Dunlap
2007-05-01 0:12 ` Andrew Morton
2007-05-01 0:45 ` Randy Dunlap
2007-05-01 1:08 ` Andrew Morton
2007-05-01 3:43 ` Andi Kleen
2007-05-01 5:16 ` Randy Dunlap
2007-05-01 5:23 ` Andrew Morton
2007-05-01 6:24 ` Andi Kleen
2007-05-01 5:38 ` Andrew Morton
2007-05-01 16:15 ` Randy Dunlap
2007-05-01 6:22 ` Andi Kleen
2007-05-01 16:22 ` Randy Dunlap
2007-05-01 17:26 ` Randy Dunlap
2007-05-04 11:38 ` [PATCH] Re: 2.6.21-rc7-mm2 -- hvsi console driver registration failure Andy Whitcroft
2007-05-04 19:04 ` Linas Vepstas
2007-05-04 20:09 ` Andrew Morton
2007-05-04 21:21 ` Christoph Lameter
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=4630EBFA.9000207@shadowen.org \
--to=apw@shadowen.org \
--cc=akpm@linux-foundation.org \
--cc=clameter@sgi.com \
--cc=linux-kernel@vger.kernel.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