Mailinglist Archive: opensuse-bugs (11322 mails)
| < Previous | Next > |
[Bug 444597] order-0 page allocation error
- From: bugzilla_noreply@xxxxxxxxxx
- Date: Tue, 2 Dec 2008 23:35:29 -0700 (MST)
- Message-id: <20081203063529.57F32245391@xxxxxxxxxxxxxxxxxxxxxx>
https://bugzilla.novell.com/show_bug.cgi?id=444597
User nfbrown@xxxxxxxxxx added comment
https://bugzilla.novell.com/show_bug.cgi?id=444597#c40
--- Comment #40 from Neil Brown <nfbrown@xxxxxxxxxx> 2008-12-02 23:35:26 MST
---
Hi Miklos.
Looks like you have found the problem despite your skepticism. Well done.
I've been trying to figure out how to implement your idea of only
blocking on a single page allocation.
I'll attach the patch that I've come up with, however I'm
not very confident of it.
I need to call cpu_cache_get both directly and via slab_force_alloc
in a context where we could conceivably be rescheduled onto another
CPU. This feels wrong, but I cannot see that it will actually be
a problem.
Have you thought about an implementation?
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
User nfbrown@xxxxxxxxxx added comment
https://bugzilla.novell.com/show_bug.cgi?id=444597#c40
--- Comment #40 from Neil Brown <nfbrown@xxxxxxxxxx> 2008-12-02 23:35:26 MST
---
Hi Miklos.
Looks like you have found the problem despite your skepticism. Well done.
I've been trying to figure out how to implement your idea of only
blocking on a single page allocation.
I'll attach the patch that I've come up with, however I'm
not very confident of it.
I need to call cpu_cache_get both directly and via slab_force_alloc
in a context where we could conceivably be rescheduled onto another
CPU. This feels wrong, but I cannot see that it will actually be
a problem.
Have you thought about an implementation?
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
| < Previous | Next > |