All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chris Wright <chrisw@redhat.com>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Ingo Molnar <mingo@elte.hu>, LKML <linux-kernel@vger.kernel.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Christoph Hellwig <hch@infradead.org>,
	Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>,
	Gregory Haskins <ghaskins@novell.com>,
	Arnaldo Carvalho de Melo <acme@ghostprotocols.net>,
	"William L. Irwin" <sparclinux@vger.kernel.org>,
	Steven Rostedt <srostedt@redhat.com>,
	Chris Wright <chrisw@sous-sol.org>,
	Rusty Russell <rusty@rustcorp.com.au>,
	virtualization@lists.linux-foundation.org
Subject: Re: [RFC PATCH 01/11] Add basic support for gcc profiler
Date: Thu, 03 Jan 2008 18:16:51 +0000	[thread overview]
Message-ID: <20080103181651.GC3627@sequoia.sous-sol.org> (raw)
In-Reply-To: <Pine.LNX.4.58.0801030849180.1468@gandalf.stny.rr.com>

* Steven Rostedt (rostedt@goodmis.org) wrote:
> Hmm, I know paravirt-ops had an issue with mcount in the RT tree. I can't
> remember the exact issues, but it did have something to do with the way
> parameters were passed in.
> 
> Chris, do you remember what the issues were?

Yes, paravirt ops have a well-specified calling convention (register
based).  There was a cleanup that Andi did that caused the problem
because it removed all the "fastcall" annotations since -mregparm=3
is now always on for i386.  Since MCOUNT disables REGPARM the calling
convention changes (caller pushes to stack, callee expects register)
chaos ensues.  I sent a patch to fix that quite some months back, but
it went stale and I neglected to update it.  Would you like me to dig
it up refresh and resend?

thanks,
-chris

WARNING: multiple messages have this Message-ID (diff)
From: Chris Wright <chrisw@redhat.com>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Chris Wright <chrisw@sous-sol.org>,
	Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Gregory Haskins <ghaskins@novell.com>,
	LKML <linux-kernel@vger.kernel.org>,
	virtualization@lists.linux-foundation.org,
	Christoph Hellwig <hch@infradead.org>,
	Steven Rostedt <srostedt@redhat.com>,
	Arnaldo Carvalho de Melo <acme@ghostprotocols.net>,
	"William L. Irwin" <sparclinux@vger.kernel.org>,
	Ingo Molnar <mingo@elte.hu>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [RFC PATCH 01/11] Add basic support for gcc profiler instrumentation
Date: Thu, 3 Jan 2008 10:16:51 -0800	[thread overview]
Message-ID: <20080103181651.GC3627@sequoia.sous-sol.org> (raw)
In-Reply-To: <Pine.LNX.4.58.0801030849180.1468@gandalf.stny.rr.com>

* Steven Rostedt (rostedt@goodmis.org) wrote:
> Hmm, I know paravirt-ops had an issue with mcount in the RT tree. I can't
> remember the exact issues, but it did have something to do with the way
> parameters were passed in.
> 
> Chris, do you remember what the issues were?

Yes, paravirt ops have a well-specified calling convention (register
based).  There was a cleanup that Andi did that caused the problem
because it removed all the "fastcall" annotations since -mregparm=3
is now always on for i386.  Since MCOUNT disables REGPARM the calling
convention changes (caller pushes to stack, callee expects register)
chaos ensues.  I sent a patch to fix that quite some months back, but
it went stale and I neglected to update it.  Would you like me to dig
it up refresh and resend?

thanks,
-chris

WARNING: multiple messages have this Message-ID (diff)
From: Chris Wright <chrisw@redhat.com>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Ingo Molnar <mingo@elte.hu>, LKML <linux-kernel@vger.kernel.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Christoph Hellwig <hch@infradead.org>,
	Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>,
	Gregory Haskins <ghaskins@novell.com>,
	Arnaldo Carvalho de Melo <acme@ghostprotocols.net>,
	"William L. Irwin" <sparclinux@vger.kernel.org>,
	Steven Rostedt <srostedt@redhat.com>,
	Chris Wright <chrisw@sous-sol.org>,
	Rusty Russell <rusty@rustcorp.com.au>,
	virtualization@lists.linux-foundation.org
Subject: Re: [RFC PATCH 01/11] Add basic support for gcc profiler instrumentation
Date: Thu, 3 Jan 2008 10:16:51 -0800	[thread overview]
Message-ID: <20080103181651.GC3627@sequoia.sous-sol.org> (raw)
In-Reply-To: <Pine.LNX.4.58.0801030849180.1468@gandalf.stny.rr.com>

* Steven Rostedt (rostedt@goodmis.org) wrote:
> Hmm, I know paravirt-ops had an issue with mcount in the RT tree. I can't
> remember the exact issues, but it did have something to do with the way
> parameters were passed in.
> 
> Chris, do you remember what the issues were?

Yes, paravirt ops have a well-specified calling convention (register
based).  There was a cleanup that Andi did that caused the problem
because it removed all the "fastcall" annotations since -mregparm=3
is now always on for i386.  Since MCOUNT disables REGPARM the calling
convention changes (caller pushes to stack, callee expects register)
chaos ensues.  I sent a patch to fix that quite some months back, but
it went stale and I neglected to update it.  Would you like me to dig
it up refresh and resend?

thanks,
-chris

  reply	other threads:[~2008-01-03 18:16 UTC|newest]

Thread overview: 85+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-01-03  7:16 [RFC PATCH 00/11] mcount tracing utility Steven Rostedt
2008-01-03  7:16 ` Steven Rostedt
2008-01-03  7:16 ` [RFC PATCH 01/11] Add basic support for gcc profiler instrumentation Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03  8:31   ` Sam Ravnborg
2008-01-03  8:31     ` Sam Ravnborg
2008-01-03 14:03     ` Steven Rostedt
2008-01-03 14:03       ` Steven Rostedt
2008-01-03  9:21   ` [RFC PATCH 01/11] Add basic support for gcc profiler Ingo Molnar
2008-01-03  9:21     ` [RFC PATCH 01/11] Add basic support for gcc profiler instrumentation Ingo Molnar
2008-01-03 13:58     ` Steven Rostedt
2008-01-03 13:58       ` Steven Rostedt
2008-01-03 18:16       ` Chris Wright [this message]
2008-01-03 18:16         ` Chris Wright
2008-01-03 18:16         ` Chris Wright
2008-01-03 19:15         ` Steven Rostedt
2008-01-03 19:15         ` Steven Rostedt
2008-01-03 19:15           ` Steven Rostedt
2008-01-03 19:17           ` Chris Wright
2008-01-03 19:17           ` [RFC PATCH 01/11] Add basic support for gcc profiler Chris Wright
2008-01-03 19:17             ` [RFC PATCH 01/11] Add basic support for gcc profiler instrumentation Chris Wright
2008-01-03 19:18         ` Jeremy Fitzhardinge
2008-01-03 19:18         ` Jeremy Fitzhardinge
2008-01-03 19:18           ` Jeremy Fitzhardinge
2008-01-03 13:58     ` Steven Rostedt
2008-01-03 16:01   ` [RFC PATCH 01/11] Add basic support for gcc profiler Daniel Walker
2008-01-03 16:01     ` [RFC PATCH 01/11] Add basic support for gcc profiler instrumentation Daniel Walker
2008-01-03 17:35   ` [RFC PATCH 01/11] Add basic support for gcc profiler Mathieu Desnoyers
2008-01-03 17:35     ` [RFC PATCH 01/11] Add basic support for gcc profiler instrumentation Mathieu Desnoyers
2008-01-03 17:55     ` Steven Rostedt
2008-01-03 17:55       ` Steven Rostedt
2008-01-03  7:16 ` [RFC PATCH 02/11] Add fastcall to do_IRQ for i386 Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03 17:36   ` Mathieu Desnoyers
2008-01-03 17:36     ` Mathieu Desnoyers
2008-01-03 17:47     ` Steven Rostedt
2008-01-03 17:47       ` Steven Rostedt
2008-01-07  4:50       ` H. Peter Anvin
2008-01-07  4:50         ` H. Peter Anvin
2008-01-07 12:42         ` Steven Rostedt
2008-01-07 12:42           ` Steven Rostedt
2008-01-03  7:16 ` [RFC PATCH 03/11] Annotate core code that should not be traced Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03 17:42   ` Mathieu Desnoyers
2008-01-03 17:42     ` Mathieu Desnoyers
2008-01-03 18:07     ` Steven Rostedt
2008-01-03 18:07       ` Steven Rostedt
2008-01-03 18:34       ` Mathieu Desnoyers
2008-01-03 18:34         ` Mathieu Desnoyers
2008-01-03  7:16 ` [RFC PATCH 04/11] i386: notrace annotations Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03 17:52   ` Mathieu Desnoyers
2008-01-03 17:52     ` Mathieu Desnoyers
2008-01-03  7:16 ` [RFC PATCH 05/11] x86_64: " Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03  7:16 ` [RFC PATCH 06/11] add notrace annotations to vsyscall Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03  7:16 ` [RFC PATCH 07/11] mcount based trace in the form of a header file library Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03  7:16 ` [RFC PATCH 08/11] tracer add debugfs interface Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03  7:16 ` [RFC PATCH 09/11] mcount tracer output file Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03  7:16 ` [RFC PATCH 10/11] mcount tracer show task comm and pid Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03 17:56   ` Mathieu Desnoyers
2008-01-03 17:56     ` Mathieu Desnoyers
2008-01-06 15:37     ` Ingo Molnar
2008-01-06 15:37       ` Ingo Molnar
2008-01-07  4:45       ` Mathieu Desnoyers
2008-01-07  4:45         ` Mathieu Desnoyers
2008-01-09 16:45         ` Ingo Molnar
2008-01-09 16:45           ` Ingo Molnar
2008-01-03  7:16 ` [RFC PATCH 11/11] Add a symbol only trace output Steven Rostedt
2008-01-03  7:16   ` Steven Rostedt
2008-01-03 17:22 ` [RFC PATCH 00/11] mcount tracing utility Mathieu Desnoyers
2008-01-03 17:22   ` Mathieu Desnoyers
2008-01-03 17:42   ` Steven Rostedt
2008-01-03 17:42     ` Steven Rostedt
2008-01-03 18:05 ` Andi Kleen
2008-01-03 18:05   ` Andi Kleen
2008-01-04  6:42 ` Frank Ch. Eigler
2008-01-04  6:42   ` Frank Ch. Eigler
2008-01-08 20:35 ` Tim Bird
2008-01-08 20:35   ` Tim Bird

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=20080103181651.GC3627@sequoia.sous-sol.org \
    --to=chrisw@redhat.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=acme@ghostprotocols.net \
    --cc=akpm@linux-foundation.org \
    --cc=chrisw@sous-sol.org \
    --cc=ghaskins@novell.com \
    --cc=hch@infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@polymtl.ca \
    --cc=mingo@elte.hu \
    --cc=rostedt@goodmis.org \
    --cc=rusty@rustcorp.com.au \
    --cc=sparclinux@vger.kernel.org \
    --cc=srostedt@redhat.com \
    --cc=torvalds@linux-foundation.org \
    --cc=virtualization@lists.linux-foundation.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 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.