All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Jun'ichi Nomura" <j-nomura@ce.jp.nec.com>
To: dm-devel@redhat.com
Subject: unaligned access by multipathd logging
Date: Thu, 05 May 2005 17:42:57 -0400	[thread overview]
Message-ID: <427A9361.2040009@ce.jp.nec.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 1135 bytes --]

Hi,

Logging code in multipathd cause unaligned access
which, on ia64, yields logs on both console and syslog like:
unaligned access to 0x60000000000141d1, ip=0x4000000000009d81

It's harmless except for the slight performance effect.
However, it's annoying for users and better to fix.

Attached patch aligns the la->tail to pointer size.
Patch is created on multipath-tools 0.4.4.
Please consider to apply.

This is a gdb trace of the case, for your reference:
#0  log_enqueue (prio=5, fmt=0x4000000000045970 "read /etc/multipath.conf",
    ap=0x6000000000013f70) at log.c:130
130             msg->prio = prio;
(gdb) where
#0  log_enqueue (prio=5, fmt=0x4000000000045970 "read /etc/multipath.conf",
    ap=0x6000000000013f70) at log.c:130
#1  0x400000000000a390 in log_safe (prio=5,
    fmt=0x4000000000045970 "read /etc/multipath.conf") at log_pthread.c:16
#2  0x4000000000007f60 in child (param=0x0) at main.c:883
#3  0x2000000000275930 in __clone2 () from /lib/tls/libc.so.6.1
(gdb) p msg
$1 = (struct logmsg *) 0x60000000000141d1
(gdb) p la->tail
$2 = (void *) 0x60000000000141d1
(gdb) p/x fwd
$3 = 0x31

Thanks,
Jun'ichi Nomura

[-- Attachment #2: logarea.patch --]
[-- Type: text/x-patch, Size: 1082 bytes --]

--- multipath-tools-0.4.4/multipathd/log.c.orig	2005-05-04 16:46:14.146824760 -0400
+++ multipath-tools-0.4.4/multipathd/log.c	2005-05-04 17:08:39.212237970 -0400
@@ -89,6 +89,7 @@ void log_close (void)
 	return;
 }
 
+#define ALIGN(len, s) (((len)+(s)-1)/(s)*(s))
 int log_enqueue (int prio, const char * fmt, va_list ap)
 {
 	int len, fwd;
@@ -101,10 +102,10 @@ int log_enqueue (int prio, const char * 
 	if (!la->empty) {
 		fwd = sizeof(struct logmsg) + 
 		      strlen((char *)&lastmsg->str) * sizeof(char) + 1;
-		la->tail += fwd;
+		la->tail += ALIGN(fwd, sizeof(void *));
 	}
 	vsnprintf(buff, MAX_MSG_SIZE, fmt, ap);
-	len = strlen(buff) * sizeof(char) + 1;
+	len = ALIGN(strlen(buff) * sizeof(char) + 1, sizeof(void *));
 
 	/* not enough space on tail : rewind */
 	if (la->head <= la->tail &&
@@ -128,7 +129,7 @@ int log_enqueue (int prio, const char * 
 	la->empty = 0;
 	msg = (struct logmsg *)la->tail;
 	msg->prio = prio;
-	memcpy((void *)&msg->str, buff, len);
+	memcpy((void *)&msg->str, buff, strlen(buff));
 	lastmsg->next = la->tail;
 	msg->next = la->head;
 

[-- Attachment #3: Type: text/plain, Size: 0 bytes --]



             reply	other threads:[~2005-05-05 21:42 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-05-05 21:42 Jun'ichi Nomura [this message]
2005-05-05 22:23 ` unaligned access by multipathd logging christophe varoqui

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=427A9361.2040009@ce.jp.nec.com \
    --to=j-nomura@ce.jp.nec.com \
    --cc=dm-devel@redhat.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.