From: "George Spelvin" <linux@horizon.com>
To: jack@suse.cz, linux@horizon.com
Cc: linux-ext4@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-kernel@vger.kernel.org, tytso@mit.edu,
viro@ZenIV.linux.org.uk
Subject: Re: 3.11.4: kernel BUG at fs/buffer.c:1268
Date: 10 Dec 2013 19:57:51 -0500 [thread overview]
Message-ID: <20131211005751.9957.qmail@science.horizon.com> (raw)
In-Reply-To: <20131210152701.GE1543@quack.suse.cz>
> Hum, can you try disabling the HW support of CRC32C implementation
> (CRYPTO_CRC32C_INTEL)? If the problem disappears, we know there's some
> problem in the HW support code...
To isolate it even better, I left in the hardware support, but commented
out the CLMUL code. I could have just upped the threshold, but opted
for commenting it out completely.
commit 8ebf85f2c151e4568d948bda29f39396ccc11096
Author: George Spelvin <linux@horizon.com>
Date: Tue Dec 10 21:06:40 2013 +0000
Disable Intel crc32c PCLMULQ usage
It's a great hack, but uses the FPU, and that appears to cause problems.
diff --git a/arch/x86/crypto/crc32c-intel_glue.c b/arch/x86/crypto/crc32c-intel_glue.c
index 0c8b502..e7b44fd 100644
--- a/arch/x86/crypto/crc32c-intel_glue.c
+++ b/arch/x86/crypto/crc32c-intel_glue.c
@@ -183,11 +183,11 @@ static int crc32c_pcl_intel_update(struct shash_desc *desc, const u8 *data,
* use faster PCL version if datasize is large enough to
* overcome kernel fpu state save/restore overhead
*/
- if (len >= crc32c_pcl_breakeven && irq_fpu_usable()) {
- kernel_fpu_begin();
- *crcp = crc_pcl(data, len, *crcp);
- kernel_fpu_end();
- } else
+// if (len >= crc32c_pcl_breakeven && irq_fpu_usable()) {
+// kernel_fpu_begin();
+// *crcp = crc_pcl(data, len, *crcp);
+// kernel_fpu_end();
+// } else
*crcp = crc32c_intel_le_hw(*crcp, data, len);
return 0;
}
@@ -195,11 +195,11 @@ static int crc32c_pcl_intel_update(struct shash_desc *desc, const u8 *data,
static int __crc32c_pcl_intel_finup(u32 *crcp, const u8 *data, unsigned int len,
u8 *out)
{
- if (len >= crc32c_pcl_breakeven && irq_fpu_usable()) {
- kernel_fpu_begin();
- *(__le32 *)out = ~cpu_to_le32(crc_pcl(data, len, *crcp));
- kernel_fpu_end();
- } else
+// if (len >= crc32c_pcl_breakeven && irq_fpu_usable()) {
+// kernel_fpu_begin();
+// *(__le32 *)out = ~cpu_to_le32(crc_pcl(data, len, *crcp));
+// kernel_fpu_end();
+// } else
*(__le32 *)out =
~cpu_to_le32(crc32c_intel_le_hw(*crcp, data, len));
return 0;
next prev parent reply other threads:[~2013-12-11 0:57 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-09 11:55 3.11.4: kernel BUG at fs/buffer.c:1268 George Spelvin
2013-10-09 15:18 ` Jan Kara
2013-10-09 17:23 ` Al Viro
2013-10-17 21:11 ` George Spelvin
2013-10-17 21:31 ` Jan Kara
2013-10-31 9:58 ` George Spelvin
2013-10-31 14:25 ` Jan Kara
2013-10-31 16:30 ` George Spelvin
2013-10-31 20:37 ` Jan Kara
2013-10-31 20:43 ` Jan Kara
2013-11-01 0:50 ` George Spelvin
2013-11-28 5:09 ` George Spelvin
2013-11-28 15:34 ` Jan Kara
2013-12-10 9:35 ` George Spelvin
2013-12-10 15:27 ` Jan Kara
2013-12-10 16:21 ` Jan Kara
2013-12-11 0:57 ` George Spelvin [this message]
2013-10-17 22:14 ` Al Viro
2013-10-31 18:33 ` Andreas Dilger
2013-10-31 19:43 ` George Spelvin
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=20131211005751.9957.qmail@science.horizon.com \
--to=linux@horizon.com \
--cc=jack@suse.cz \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tytso@mit.edu \
--cc=viro@ZenIV.linux.org.uk \
/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).