From: Martin Peschke <mpeschke@linux.vnet.ibm.com>
To: linux-btrace@vger.kernel.org
Subject: Re: [PATCH] blkiomon: fix unaligned accesses on ia64
Date: Fri, 08 May 2009 16:09:39 +0000 [thread overview]
Message-ID: <1241798979.4084.20.camel@kitka.ibm.com> (raw)
In-Reply-To: <4A0205FD.7070004@redhat.com>
Eric,
I think you have got a valid point. But I would like to make two small
changes to you patch:
A new blkiomon version number is needed then. Not a perfect solution for
reflecting this change, but good enough at least for our tool that
consumes blkiomon data through a message queue.
Moving the __u32 device member instead of a new padding field should be
fine.
I have tested the patch below. If it's fine with you, please just push
it upstream through Jens.
Thanks,
Martin
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
---
blkiomon.c | 2 +-
blkiomon.h | 4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
--- a/blkiomon.h
+++ b/blkiomon.h
@@ -34,6 +34,7 @@ struct blkiomon_stat {
__u64 time;
__u32 size_hist[BLKIOMON_SIZE_BUCKETS];
__u32 d2c_hist[BLKIOMON_D2C_BUCKETS];
+ __u32 device;
struct minmax size_r;
struct minmax size_w;
struct minmax d2c_r;
@@ -41,8 +42,7 @@ struct blkiomon_stat {
struct minmax thrput_r;
struct minmax thrput_w;
__u64 bidir;
- __u32 device;
-} __attribute__ ((packed));
+};
static struct histlog2 size_hist = {
.first = 0,
--- a/blkiomon.c
+++ b/blkiomon.c
@@ -71,7 +71,7 @@ struct output {
int pipe;
};
-static char blkiomon_version[] = "0.2";
+static char blkiomon_version[] = "0.3";
static FILE *ifp;
static int interval = -1;
On Wed, 2009-05-06 at 16:49 -0500, Eric Sandeen wrote:
> commit 7aa3ebcec011bfe9cc60d6476252c03376a37551 packed
> the blkiomon_stat structure so that traces from one
> arch could be analyzed on another (in truth only x86
> is different, at least from x86_64/ia64/ppc/ppc64/s390/s390x)
>
> Rather than packing it, which generates unaligned access
> warnings on ia64, just pad the structure out so that it's
> naturally aligned on all arches.
>
> Martin, care to test this to be sure it still works for
> you? (I'm not sure if we might also need a 4 byte pad on
> the end of the structure to align the containing structure...)
>
> Signed-off-by: Eric Sandeen <sandeen@redhat.com>
> ---
>
> diff --git a/blkiomon.h b/blkiomon.h
> index 2e430a6..ae48b4c 100644
> --- a/blkiomon.h
> +++ b/blkiomon.h
> @@ -34,6 +34,7 @@ struct blkiomon_stat {
> __u64 time;
> __u32 size_hist[BLKIOMON_SIZE_BUCKETS];
> __u32 d2c_hist[BLKIOMON_D2C_BUCKETS];
> + __u32 pad; /* Align the structure */
> struct minmax size_r;
> struct minmax size_w;
> struct minmax d2c_r;
> @@ -42,7 +43,7 @@ struct blkiomon_stat {
> struct minmax thrput_w;
> __u64 bidir;
> __u32 device;
> -} __attribute__ ((packed));
> +};
>
> static struct histlog2 size_hist = {
> .first = 0,
>
next prev parent reply other threads:[~2009-05-08 16:09 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-05-06 21:49 [PATCH] blkiomon: fix unaligned accesses on ia64 Eric Sandeen
2009-05-08 16:09 ` Martin Peschke [this message]
2009-05-08 16:22 ` Eric Sandeen
2009-05-11 6:41 ` Jens Axboe
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=1241798979.4084.20.camel@kitka.ibm.com \
--to=mpeschke@linux.vnet.ibm.com \
--cc=linux-btrace@vger.kernel.org \
/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).