From: Stefan Richter <stefanr@s5r6.in-berlin.de>
To: linux1394-devel@lists.sourceforge.net
Cc: pageexec@freemail.hu, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/6] firewire: core: reduce stack usage in bus reset tasklet
Date: Mon, 07 Sep 2009 23:24:37 +0200 [thread overview]
Message-ID: <4AA57A15.1030705@s5r6.in-berlin.de> (raw)
In-Reply-To: <4AA55BCF.6080609@s5r6.in-berlin.de>
Stefan Richter wrote:
> pageexec@freemail.hu wrote:
>> On 6 Sep 2009 at 18:48, Stefan Richter wrote:
>>
>>> Index: linux-2.6.31-rc9/drivers/firewire/core-card.c
>>> ===================================================================
>>> --- linux-2.6.31-rc9.orig/drivers/firewire/core-card.c
>>> +++ linux-2.6.31-rc9/drivers/firewire/core-card.c
>>> @@ -38,16 +38,21 @@
>>>
>>> #include "core.h"
>>>
>>> -int fw_compute_block_crc(u32 *block)
>>> +int fw_compute_block_crc(u32 *block, gfp_t flags)
>>> {
>>> - __be32 be32_block[256];
>>> - int i, length;
>>> + static __be32 *be32_block;
>> ^^^^^^
>>
>> did you actually mean that to be static? if so, then you might as well
>> allocate the
>> buffer statically and not worry about a runtime allocation failure.
>
> Uh, that's a cut'n'paste mistake. It shouldn't be static. Thanks, I'll
> correct that before I put it into linux1394-2.6.git.
>
>>> + int i, length = (*block >> 16) & 0xff;
>>> +
>>> + be32_block = kmalloc(length * 4, flags);
>>> + if (WARN_ON(!be32_block))
>>> + goto out;
>>>
>>> - length = (*block >> 16) & 0xff;
>>> for (i = 0; i < length; i++)
>>> be32_block[i] = cpu_to_be32(block[i + 1]);
>>> *block |= crc_itu_t(0, (u8 *) be32_block, length * 4);
>>>
>>> + kfree(be32_block);
>>> + out:
>>> return length;
>>> }
This can be optimized further to get rid of the temporary be32_block
entirely.
There are two users of fw_compute_block_crc: The functions which
generate the local Config ROM CSR and then pass it down to the low-level
driver, and the function which which generates the local Topology Map
CSR and stores it for core-transaction.c::handle_topology_map() to read
on demand.
The low-level driver then needs to convert the Config ROM to __be32[].
Likewise, handle_topology_map needs to convert the Topology Map to
__be32[]. They would be both better off if we stored their input as
__be32[] in the first place, and of course do so before we compute the
CRC. Duh.
I'll rewrite patch 1/6 and 5/6 accordingly.
--
Stefan Richter
-=====-==--= =--= --===
http://arcgraph.de/sr/
next prev parent reply other threads:[~2009-09-07 21:25 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-09-06 16:46 [PATCH 0/6] firewire: minor fixes and cleanups Stefan Richter
2009-09-06 16:48 ` [PATCH 1/6] firewire: core: reduce stack usage in bus reset tasklet Stefan Richter
2009-09-07 19:02 ` pageexec
2009-09-07 19:15 ` Stefan Richter
2009-09-07 21:24 ` Stefan Richter [this message]
2009-09-06 16:49 ` [PATCH 2/6] firewire: ohci: fix Self ID Count register mask (safeguard against buffer overflow) Stefan Richter
2009-09-07 18:56 ` pageexec
2009-09-07 19:28 ` Stefan Richter
2009-09-06 16:49 ` [PATCH 3/6] firewire: core: header file cleanup Stefan Richter
2009-09-06 16:50 ` [PATCH 4/6] firewire: core: fix race with parallel PCI device probe Stefan Richter
2009-09-06 16:50 ` [PATCH 5/6] firewire: reduce some memsets Stefan Richter
2009-09-06 16:51 ` [PATCH 6/6] firewire: sbp2: fix status reception Stefan Richter
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=4AA57A15.1030705@s5r6.in-berlin.de \
--to=stefanr@s5r6.in-berlin.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux1394-devel@lists.sourceforge.net \
--cc=pageexec@freemail.hu \
/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®