From: Stefan Hajnoczi <stefanha@gmail.com>
To: "Lluís Vilanova" <vilanova@ac.upc.edu>
Cc: qemu-devel@nongnu.org, "Emilio G. Cota" <cota@braap.org>,
Stefan Hajnoczi <stefanha@redhat.com>
Subject: Re: [Qemu-devel] [PATCH 00/13] instrument: Add basic event instrumentation
Date: Tue, 25 Jul 2017 14:19:31 +0100 [thread overview]
Message-ID: <20170725131931.GC23343@stefanha-x1.localdomain> (raw)
In-Reply-To: <150091574424.30739.4131793221953168474.stgit@frigg.lan>
[-- Attachment #1: Type: text/plain, Size: 1554 bytes --]
On Mon, Jul 24, 2017 at 08:02:24PM +0300, Lluís Vilanova wrote:
> This series adds a basic interface to instrument tracing events and control
> their tracing state.
>
> The instrumentation code is dynamically loaded into QEMU (either when it starts
> or later using its remote control interfaces).
>
> All events can be instrumented, but the instrumentable events must be explicitly
> specified at configure time.
>
> Signed-off-by: Lluís Vilanova <vilanova@ac.upc.edu>
Hi Lluís,
I'm concerned that the shared library interface will be abused to monkey
patch code into QEMU far beyond instrumentation use cases and/or avoid
the responsibilities of the GPL license.
Instead I suggest adding a trace backend generates calls to registered
"callback" functions:
$ cat >my-instrumentation.c
#include "trace/control.h"
static void my_cpu_in(unsigned int addr, char size, unsigned int val)
{
printf("my_cpu_in\n");
}
static void my_init(void)
{
trace_register_event_callback("cpu_in", my_cpu_in);
trace_enable_events("cpu_in");
}
trace_init(my_init);
$ ./configure --enable-trace-backends=log,callback && make -j4
This is still a clean interface that allows instrumentation code to be
kept separate from the trace event call sites.
The instrumentation code gets compiled into QEMU, but that shouldn't be
a huge burden since QEMU's Makefiles only recompile changed source
files (only the first build is slow).
Does this alternative sound reasonable to you?
Stefan
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 455 bytes --]
next prev parent reply other threads:[~2017-07-25 13:19 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-07-24 17:02 [Qemu-devel] [PATCH 00/13] instrument: Add basic event instrumentation Lluís Vilanova
2017-07-24 17:06 ` [Qemu-devel] [PATCH 01/13] instrument: Add documentation Lluís Vilanova
2017-07-24 17:10 ` [Qemu-devel] [PATCH 02/13] instrument: [none] Add null instrumentation mode Lluís Vilanova
2017-07-24 17:14 ` [Qemu-devel] [PATCH 03/13] instrument: [dynamic] Add dynamic " Lluís Vilanova
2017-07-24 17:18 ` [Qemu-devel] [PATCH 04/13] instrument: Allow adding the "instrument" property without modifying event files Lluís Vilanova
2017-07-24 17:22 ` [Qemu-devel] [PATCH 05/13] instrument: [dynamic] Add default public per-event functions Lluís Vilanova
2017-07-24 17:26 ` [Qemu-devel] [PATCH 06/13] instrument: Add event control interface Lluís Vilanova
2017-07-24 17:30 ` [Qemu-devel] [PATCH 07/13] instrument: Add generic command line library loader Lluís Vilanova
2017-07-24 17:34 ` [Qemu-devel] [PATCH 08/13] instrument: [linux-user] Add " Lluís Vilanova
2017-07-24 17:38 ` [Qemu-devel] [PATCH 09/13] instrument: [bsd-user] " Lluís Vilanova
2017-07-24 17:42 ` [Qemu-devel] [PATCH 10/13] instrument: [softmmu] " Lluís Vilanova
2017-07-24 17:46 ` [Qemu-devel] [PATCH 11/13] instrument: [qapi] Add " Lluís Vilanova
2017-07-24 18:03 ` Eric Blake
2017-07-25 8:24 ` Lluís Vilanova
2017-07-25 11:30 ` Eric Blake
2017-07-25 11:51 ` Lluís Vilanova
2017-07-24 17:50 ` [Qemu-devel] [PATCH 12/13] instrument: [hmp] " Lluís Vilanova
2017-07-24 17:54 ` [Qemu-devel] [PATCH 13/13] trace: Rename C++-specific names in event arguments Lluís Vilanova
2017-07-25 13:19 ` Stefan Hajnoczi [this message]
2017-07-25 13:30 ` [Qemu-devel] [PATCH 00/13] instrument: Add basic event instrumentation Peter Maydell
2017-07-25 15:11 ` Lluís Vilanova
2017-07-26 11:22 ` Stefan Hajnoczi
2017-07-26 12:44 ` Lluís Vilanova
2017-07-27 10:32 ` Stefan Hajnoczi
2017-07-27 10:40 ` Peter Maydell
2017-07-28 13:42 ` Stefan Hajnoczi
2017-07-28 16:21 ` Lluís Vilanova
2017-08-02 11:04 ` Stefan Hajnoczi
2017-07-26 11:26 ` Stefan Hajnoczi
2017-07-26 11:49 ` Peter Maydell
2017-07-26 12:26 ` Lluís Vilanova
2017-07-27 10:43 ` Daniel P. Berrange
2017-07-27 10:54 ` Peter Maydell
2017-07-27 14:58 ` Lluís Vilanova
2017-07-27 15:21 ` Daniel P. Berrange
2017-07-27 15:33 ` Peter Maydell
2017-07-27 15:45 ` Daniel P. Berrange
2017-07-28 13:34 ` Stefan Hajnoczi
2017-07-28 13:41 ` Peter Maydell
2017-07-28 14:06 ` Daniel P. Berrange
2017-07-28 16:05 ` Lluís Vilanova
2017-08-01 13:48 ` Stefan Hajnoczi
2017-08-01 13:54 ` Peter Maydell
2017-08-02 11:04 ` Stefan Hajnoczi
2017-08-02 11:10 ` Peter Maydell
2017-08-02 14:49 ` Stefan Hajnoczi
2017-08-02 15:19 ` Lluís Vilanova
2017-08-03 11:54 ` Stefan Hajnoczi
2017-08-26 0:14 ` Emilio G. Cota
2017-08-26 0:02 ` Emilio G. Cota
2017-08-29 9:19 ` Peter Maydell
2017-07-28 13:52 ` Daniel P. Berrange
2017-07-28 16:14 ` Lluís Vilanova
2017-08-01 13:13 ` Stefan Hajnoczi
2017-07-28 15:10 ` Lluís Vilanova
2017-07-27 19:55 ` Lluís Vilanova
2017-07-25 14:47 ` Lluís Vilanova
2017-07-26 11:29 ` Stefan Hajnoczi
2017-07-26 12:31 ` Lluís Vilanova
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=20170725131931.GC23343@stefanha-x1.localdomain \
--to=stefanha@gmail.com \
--cc=cota@braap.org \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
--cc=vilanova@ac.upc.edu \
/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).