From: Arnd Bergmann <arnd@arndb.de>
To: linux-kernel@vger.kernel.org,
Manfred Spraul <manfred@colorfullife.com>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>
Subject: Re: [BUG] slab debug vs. L1 alignement
Date: Sat, 16 Aug 2003 14:00:00 +0200 [thread overview]
Message-ID: <200308161205.h7GC5vlP005512@post.webmailer.de> (raw)
In-Reply-To: <l6kd.53T.1@gated-at.bofh.it>
Manfred Spraul wrote:
> Benjamin Herrenschmidt wrote:
>
>>Same for ppc32. Anyway, I don't like MUST_HWCACHE_ALIGN because it
>>just disables redzoning, I'd rather allocate more and do both redzoning
>>and cache alignement.
>>
>>
> I have a patch that creates helper functions that make that simple. The
> patch is stuck right now, because it exposes a bug in the i386 debug
> register handling. I'll add it redzoning with MUST_HWCACHE_ALIGN after
> that one is in.
I have a related problem on s390: Some of my GFP_DMA data must be 8 byte
aligned, while my cache lines are 256 bytes wide. With slab debugging,
the whole structure is only 4 byte aligned and I get addressing exceptions.
When I go to MUST_HWCACHE_ALIGN, the data already grows from ~100 Bytes to
a full cache line, adding redzoning would make it far worse.
Is it possible to make kmem_cache_create accept a user specified alignment
parameter? I suppose there are other cases where you want to force
a specific alignment that is different from the L1 cache lines.
Arnd <><
next parent reply other threads:[~2003-08-16 12:06 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <kUMe.2pd.9@gated-at.bofh.it>
[not found] ` <kWuz.41M.5@gated-at.bofh.it>
[not found] ` <kYwo.5Xr.1@gated-at.bofh.it>
[not found] ` <l5HD.4tl.21@gated-at.bofh.it>
[not found] ` <l6kd.53T.1@gated-at.bofh.it>
2003-08-16 12:00 ` Arnd Bergmann [this message]
2003-08-16 12:18 ` Manfred Spraul
2003-08-17 17:27 James Bottomley
2003-08-18 20:29 ` Kai Makisara
2003-09-30 18:40 ` Manfred Spraul
-- strict thread matches above, loose matches on Subject: below --
2003-08-15 21:50 Manfred Spraul
2003-08-15 23:41 ` Benjamin Herrenschmidt
2003-08-16 1:47 ` Manfred Spraul
2003-08-16 9:37 ` Benjamin Herrenschmidt
2003-08-16 10:09 ` Manfred Spraul
2003-08-16 10:43 ` Benjamin Herrenschmidt
2003-08-16 11:21 ` Kai Makisara
2003-08-16 11:36 ` Benjamin Herrenschmidt
2003-08-16 14:39 ` Kai Makisara
2003-08-15 14:00 Benjamin Herrenschmidt
2003-08-15 18:51 ` Philippe Elie
2003-08-15 16:54 ` Benjamin Herrenschmidt
2003-08-15 21:16 ` Andrew Morton
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=200308161205.h7GC5vlP005512@post.webmailer.de \
--to=arnd@arndb.de \
--cc=benh@kernel.crashing.org \
--cc=linux-kernel@vger.kernel.org \
--cc=manfred@colorfullife.com \
/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
all inboxes | Powered by JetHome®