lkml.org 
[lkml]   [2004]   [Oct]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Page Allocation Failures Return With 2.6.9+TSO patch.
It does not seem like they do, but they cannot be good...

I have applied the following patches

1] TSO patch
2] rollup.patch

Rebooting now and will alert the list if/when I receive more page
allocation failures.

FYI - I started getting these with 2.6.9.

(However, it was always possible on the Dell Optiplex GX1 to create page
allocation failure with: ifconfig eth0 mtu 9000), however, on a higher-end
machine (2.6GHZ, 2GB ram, etc) ifconfig eth0 mtu 9000 worked fine.

Is it something with the architecture of the box bus/box?

Why does it tend to affect one machine and not the other?


On Sat, 23 Oct 2004, Nick Piggin wrote:

> Justin Piszcz wrote:
>> Applying this patch now and I will let everyone know what happens, thanks.
>>
>> On Sat, 23 Oct 2004, Nick Piggin wrote:
>>
>>> Justin Piszcz wrote:
>>>
>>>> Kernel 2.6.9 w/TSO patch.
>>>>
>>>> (most likely do to the e1000/nic/issue)
>>>>
>>>> $ dmesg
>>>> gaim: page allocation failure. order:4, mode:0x21
>>>> [<c01391a7>] __alloc_pages+0x247/0x3b0
>>>> [<c0139328>] __get_free_pages+0x18/0x40
>>>> [<c035c33a>] sound_alloc_dmap+0xaa/0x1b0
>>>> [<c03648c0>] ad_mute+0x20/0x40
>>>> [<c035c70f>] open_dmap+0x1f/0x100
>>>> [<c035cb58>] DMAbuf_open+0x178/0x1d0
>>>> [<c035a4fa>] audio_open+0xba/0x280
>>>> [<c015d863>] cdev_get+0x53/0xc0
>>>> [<c035968c>] sound_open+0xac/0x110
>>>> [<c035898e>] soundcore_open+0x1ce/0x300
>>>> [<c03587c0>] soundcore_open+0x0/0x300
>>>> [<c015d524>] chrdev_open+0x104/0x250
>>>> [<c015d420>] chrdev_open+0x0/0x250
>>>> [<c0152d82>] dentry_open+0x1d2/0x270
>>>> [<c0152b9c>] filp_open+0x5c/0x70
>>>> [<c01049c8>] common_interrupt+0x18/0x20
>>>> [<c0152e75>] get_unused_fd+0x55/0xf0
>>>> [<c0152fd9>] sys_open+0x49/0x90
>>>> [<c010405b>] syscall_call+0x7/0xb
>>>
>>>
>>> Ouch, 64K atomic DMA allocation.
>>>
>>> The DMA zone barely even keeps that much total memory free.
>>>
>>> The caller probably wants fixing, but you could try this patch...
>>>
>>
>
> Oh... these allocation failure don't actually hurt anything, do they?
> sound_alloc_dmap would have just reverted to using a 32K buffer instead
> of a 64K one.
>
> Probably the easiest thing to do is stick a __GFP_NOWARN on that
> allocation.
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 14:07    [W:0.039 / U:0.404 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site