From: Jack Steiner <steiner@sgi.com>
To: Nick Piggin <npiggin@suse.de>, Robin Holt <holt@sgi.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Linux Memory Management List <linux-mm@kvack.org>
Subject: Re: [patch 2/2] mm: remove nopfn
Date: Tue, 13 May 2008 10:48:12 -0500 [thread overview]
Message-ID: <20080513154812.GA23256@sgi.com> (raw)
In-Reply-To: <20080513074829.GC12869@wotan.suse.de>
On Tue, May 13, 2008 at 09:48:29AM +0200, Nick Piggin wrote:
> There are no users of nopfn in the tree. Remove it.
>
The SGI mspec driver use to use the nopfn callout. I see that this
was recently changed but the new code fails with:
kernel BUG at mm/memory.c:2278!
fop1[5887]: bugcheck! 0 [1]
Modules linked in:
Call Trace:
[<a000000100012740>] show_stack+0x40/0xa0
[<a000000100013050>] show_regs+0x850/0x8a0
[<a000000100036210>] die+0x1b0/0x2c0
[<a000000100036370>] die_if_kernel+0x50/0x80
[<a000000100037a50>] ia64_bad_break+0x230/0x460
[<a00000010000a2a0>] ia64_leave_kernel+0x0/0x270
[<a000000100141650>] __do_fault+0xb0/0xa20
[<a000000100145a50>] handle_mm_fault+0x2f0/0xf40
[<a000000100059160>] ia64_do_page_fault+0x220/0xa40
[<a00000010000a2a0>] ia64_leave_kernel+0x0/0x270
The mspec driver is tripping the bugcheck in __do_fault()
BUG_ON(vma->vm_flags & VM_PFNMAP);
The driver does not create pte entries at map time. Instead, it
relies on the nopfn (now fault) callout to assign resources
and create the ptes. It is intentionally done this way in order to
ensure that node-local resources are assigned.
What should the driver be doing to avoid this problem??
Also, the new GRU driver will have a similar problem. It currently
uses the nopfn callout since it needs to be able to assign resources
at fault, not mmap. The driver is not currently in-tree but will be
posted as soon as mmu_notifiers are available. I can post the current
version if it is helpful.....
--- jack
--
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>
next prev parent reply other threads:[~2008-05-13 15:48 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-05-13 7:47 [patch 1/2] spufs: convert nopfn to fault Nick Piggin
2008-05-13 7:48 ` [patch 2/2] mm: remove nopfn Nick Piggin
2008-05-13 15:48 ` Jack Steiner [this message]
[not found] ` <20080513162046.GA22407@sgi.com>
2008-05-14 0:44 ` Nick Piggin
2008-05-14 1:22 ` Jack Steiner
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=20080513154812.GA23256@sgi.com \
--to=steiner@sgi.com \
--cc=akpm@linux-foundation.org \
--cc=holt@sgi.com \
--cc=linux-mm@kvack.org \
--cc=npiggin@suse.de \
/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).