From: Arnd Bergmann <arnd@arndb.de>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: perfmon2-devel@lists.sourceforge.net,
Christoph Hellwig <hch@infradead.org>,
Andrew Morton <akpm@linux-foundation.org>,
eranian@gmail.com, Philip Mucci <mucci@eecs.utk.edu>,
LKML <linux-kernel@vger.kernel.org>,
Andi Kleen <andi@firstfloor.org>,
Paul Mackerras <paulus@samba.org>,
Maynard Johnson <mpjohn@us.ibm.com>, Ingo Molnar <mingo@elte.hu>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [perfmon2] I.1 - System calls - ioctl
Date: Mon, 13 Jul 2009 19:53:45 +0200 [thread overview]
Message-ID: <200907131953.45786.arnd@arndb.de> (raw)
In-Reply-To: <1247506440.7500.43.camel@twins>
On Monday 13 July 2009, Peter Zijlstra wrote:
> On Mon, 2009-07-13 at 19:30 +0200, Arnd Bergmann wrote:
> > > +struct perf_counter_target {
> > > + __u32 id;
> > > + __u64 val;
> > > +};
> >
> > This structure is not compatible between 32 and 64 bit user space on x86,
> > because everything except i386 adds implicit padding between id and val.
>
> Humm, __u64 doesn't have natural alignment? That would break more than
> just this I think -- it sure surprises me.
Yes, nobody expects this, so it is a frequent source of bugs in the ABI.
Look for compat_u64 and __packed in the definition of compat ioctl and
syscall interfaces for how we had to work around this elsewhere.
> > Other than that, making it extensible sounds reasonable. How about just
> > using a '__u64 *target' and a bit in the 'flags' argument?
>
> Would there still be a point in having it a pointer in that case?, but
> yeah, that might work too?
passing u64 bit arguments directly to system calls is a bit complicated,
because some 32 bit architectures can only pass them in certain
register pairs, see the trouble we go through for llseek, sync_file_range
or preadv.
If you can directly pass an 'unsigned long' instead, that would work fine
though.
Arnd <><
next prev parent reply other threads:[~2009-07-13 17:54 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-06-16 17:42 v2 of comments on Performance Counters for Linux (PCL) stephane eranian
2009-06-22 11:48 ` Ingo Molnar
2009-06-22 11:49 ` I.1 - System calls - ioctl Ingo Molnar
2009-06-22 12:58 ` Christoph Hellwig
2009-06-22 13:56 ` Ingo Molnar
2009-06-22 17:41 ` Arnd Bergmann
2009-07-13 10:53 ` Peter Zijlstra
2009-07-13 17:30 ` [perfmon2] " Arnd Bergmann
2009-07-13 17:34 ` Peter Zijlstra
2009-07-13 17:53 ` Arnd Bergmann [this message]
2009-07-14 13:51 ` Christoph Hellwig
2009-07-30 13:58 ` stephane eranian
2009-07-30 14:13 ` Peter Zijlstra
2009-07-30 16:17 ` stephane eranian
2009-07-30 16:40 ` Arnd Bergmann
2009-07-30 16:53 ` stephane eranian
2009-07-30 17:20 ` Arnd Bergmann
2009-08-03 14:22 ` Peter Zijlstra
2009-06-22 11:50 ` I.2 - Grouping Ingo Molnar
2009-06-22 19:45 ` stephane eranian
2009-06-22 22:04 ` Corey Ashford
2009-06-23 17:51 ` stephane eranian
2009-06-22 21:38 ` Corey Ashford
2009-06-23 5:16 ` Paul Mackerras
2009-06-23 7:36 ` stephane eranian
2009-06-23 8:26 ` Paul Mackerras
2009-06-23 8:30 ` stephane eranian
2009-06-23 16:24 ` Corey Ashford
2009-06-22 11:51 ` I.3 - Multiplexing and system-wide Ingo Molnar
2009-06-22 11:51 ` I.4 - Controlling group multiplexing Ingo Molnar
2009-06-22 11:52 ` I.5 - Mmaped count Ingo Molnar
2009-06-22 12:25 ` stephane eranian
2009-06-22 12:35 ` Peter Zijlstra
2009-06-22 12:54 ` stephane eranian
2009-06-22 14:39 ` Peter Zijlstra
2009-06-23 0:41 ` Paul Mackerras
2009-06-23 0:39 ` Paul Mackerras
2009-06-23 6:13 ` Peter Zijlstra
2009-06-23 7:40 ` stephane eranian
2009-06-23 0:33 ` Paul Mackerras
2009-06-22 11:53 ` I.6 - Group scheduling Ingo Molnar
2009-06-22 11:54 ` I.7 - Group validity checking Ingo Molnar
2009-06-22 11:54 ` I.8 - Generalized cache events Ingo Molnar
2009-06-22 11:55 ` I.9 - Group reading Ingo Molnar
2009-06-22 11:55 ` I.10 - Event buffer minimal useful size Ingo Molnar
2009-06-22 11:56 ` I.11 - Missing definitions for generic events Ingo Molnar
2009-06-22 14:54 ` stephane eranian
2009-06-22 11:57 ` II.1 - Fixed counters on Intel Ingo Molnar
2009-06-22 14:27 ` stephane eranian
2009-06-22 11:57 ` II.2 - Event knowledge missing Ingo Molnar
2009-06-23 13:18 ` stephane eranian
2009-06-22 11:58 ` III.1 - Sampling period randomization Ingo Molnar
2009-06-22 11:58 ` IV.1 - Support for model-specific uncore PMU Ingo Molnar
2009-06-22 11:59 ` IV.2 - Features impacting all counters Ingo Molnar
2009-06-22 12:00 ` IV.3 - AMD IBS Ingo Molnar
2009-06-22 14:08 ` [perfmon2] " Rob Fowler
2009-06-22 17:58 ` Maynard Johnson
2009-06-23 6:19 ` Peter Zijlstra
2009-06-23 8:19 ` stephane eranian
2009-06-23 14:05 ` Ingo Molnar
2009-06-23 14:25 ` stephane eranian
2009-06-23 14:55 ` Ingo Molnar
2009-06-23 14:40 ` Rob Fowler
2009-06-22 19:17 ` stephane eranian
2009-06-22 12:00 ` IV.4 - Intel PEBS Ingo Molnar
2009-06-22 12:16 ` Andi Kleen
2009-06-22 12:01 ` IV.5 - Intel Last Branch Record (LBR) Ingo Molnar
2009-06-22 20:02 ` stephane eranian
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=200907131953.45786.arnd@arndb.de \
--to=arnd@arndb.de \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=eranian@gmail.com \
--cc=hch@infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=mpjohn@us.ibm.com \
--cc=mucci@eecs.utk.edu \
--cc=paulus@samba.org \
--cc=perfmon2-devel@lists.sourceforge.net \
--cc=tglx@linutronix.de \
/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