On 01/25/2013 09:13 PM, Markus Armbruster wrote:
We use atomic operations to keep track of dropped events.
Inconveniently, GLib supports only int and void * atomics, but the
counter dropped_events is uint64_t. Can't stop commit 62bab732: a
quick (gint *)dropped_events bludgeons the compiler
On 01/25/2013 09:13 PM, Markus Armbruster wrote:
The GLib Reference Manual says:
It is very important that all accesses to a particular integer or
pointer be performed using only this API and that different sizes
of operation are not mixed or used on overlapping memory
On 01/25/2013 09:13 PM, Markus Armbruster wrote:
Missed when commit 4c3b5a48 moved it.
Signed-off-by: Markus Armbruster arm...@redhat.com
Reviewed-by: Harsh Prateek Bora ha...@linux.vnet.ibm.com
---
docs/tracing.txt | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git
On 01/25/2013 09:13 PM, Markus Armbruster wrote:
Signed-off-by: Markus Armbruster arm...@redhat.com
---
trace/simple.c | 18 +-
1 file changed, 5 insertions(+), 13 deletions(-)
Reviewed-by: Harsh Prateek Bora ha...@linux.vnet.ibm.com
diff --git a/trace/simple.c
On 08/08/2012 05:25 PM, Paolo Bonzini wrote:
The read-side critical sections in 9p-synth currently only include the
navigation of the list. This is incorrect; it works for two reasons,
first obviously because rcu_read_lock/unlock are still no-ops; second,
because elements of the list are never
On 08/08/2012 05:25 PM, Paolo Bonzini wrote:
Even with the fix in the previous patch, the lockless handling of paths
in 9p-synth is wrong. Paths can outlive rcu_read_unlock arbitrarily
via the V9fsPath objects that 9p-synth creates. This would require
a reference counting mechanism that is not
On 07/19/2012 04:22 PM, Stefan Hajnoczi wrote:
Simpletrace v2 has landed:
* Strings are now logged instead of their pointers
* Variable-length arguments allow for 6 trace event arguments
Thanks to Harsh Prateek Bora for this improvement!
Thanks very much to Stefan Hajnoczi for his
On 07/19/2012 05:22 PM, Stefan Hajnoczi wrote:
On Thu, Jul 19, 2012 at 12:07 PM, Harsh Prateek Bora
ha...@linux.vnet.ibm.com wrote:
Existing simpletrace backend allows to trace at max 6 args and does not
support strings. This newer tracelog format gets rid of fixed size records
and therefore
On 07/17/2012 08:51 PM, Stefan Hajnoczi wrote:
On Tue, Jul 3, 2012 at 10:20 AM, Harsh Prateek Bora
ha...@linux.vnet.ibm.com wrote:
Existing simpletrace backend allows to trace at max 6 args and does not
support strings. This newer tracelog format gets rid of fixed size records
and therefore
On 07/17/2012 08:53 PM, Stefan Hajnoczi wrote:
On Tue, Jul 3, 2012 at 10:20 AM, Harsh Prateek Bora
ha...@linux.vnet.ibm.com wrote:
Existing simpletrace backend allows to trace at max 6 args and does not
support strings. This newer tracelog format gets rid of fixed size records
and therefore
On 07/18/2012 12:31 AM, Harsh Bora wrote:
On 07/17/2012 08:51 PM, Stefan Hajnoczi wrote:
On Tue, Jul 3, 2012 at 10:20 AM, Harsh Prateek Bora
ha...@linux.vnet.ibm.com wrote:
Existing simpletrace backend allows to trace at max 6 args and does not
support strings. This newer tracelog format gets
Hi Stefan,
Please see comment below:
On 06/14/2012 07:02 PM, Harsh Prateek Bora wrote:
[...]
-static bool get_trace_record(unsigned int idx, TraceRecord *record)
+static bool get_trace_record(unsigned int idx, TraceRecord **recordptr)
{
-if (!(trace_buf[idx].event TRACE_RECORD_VALID))
[...]
Okay I think, I got it, let me dig a little further ...
+void trace_record_finish(TraceBufferRecord *rec)
{
- trace(event, x1, x2, x3, x4, x5, x6);
+ uint8_t temp_rec[sizeof(TraceRecord)];
+ TraceRecord *record = (TraceRecord *) temp_rec;
+ read_from_buffer(rec-tbuf_idx, temp_rec,
On 06/15/2012 12:09 AM, Harsh Prateek Bora wrote:
static gpointer writeout_thread(gpointer opaque)
{
-TraceRecord record;
-unsigned int writeout_idx = 0;
-unsigned int num_available, idx;
+TraceRecord *recordptr, *dropped_ptr;
+union {
+TraceRecord rec;
+
On 06/11/2012 08:25 PM, Stefan Hajnoczi wrote:
On Mon, Jun 11, 2012 at 1:31 PM, Harsh Boraha...@linux.vnet.ibm.com wrote:
On 06/07/2012 08:02 PM, Stefan Hajnoczi wrote:
On Thu, May 24, 2012 at 10:50 AM, Harsh Prateek Bora
ha...@linux.vnet.ibm.comwrote:
@@ -75,16 +96,22 @@ static char
On 06/07/2012 08:02 PM, Stefan Hajnoczi wrote:
On Thu, May 24, 2012 at 10:50 AM, Harsh Prateek Bora
ha...@linux.vnet.ibm.com wrote:
A newer tracelog format which gets rid of fixed size trace records and
therefore allows to trace multiple arguments as well as strings in trace
events.
Sample
On 05/07/2012 07:16 PM, Lluís Vilanova wrote:
Harsh Prateek Bora writes:
Going forward with simpletrace v2 variable size trace records, we cannot
have a generic function to print trace event info and therefore this
interface becomes invalid.
As per Stefan Hajnoczi:
This command is only
On 03/12/2012 08:55 PM, Lluís Vilanova wrote:
Harsh Prateek Bora writes:
diff --git a/docs/tracing.txt b/docs/tracing.txt
index 95ca16c..ab9048c 100644
--- a/docs/tracing.txt
+++ b/docs/tracing.txt
@@ -39,12 +39,12 @@ generate code for the trace events. Trace events are
invoked directly from
On 03/14/2012 01:41 AM, Lluís Vilanova wrote:
Harsh Prateek Bora writes:
This patchset introduces 2 major updates:
1) Tracetool Improvements (Conversion from shell script to python)
2) Simpletrace v2 log format (Support for variable args, strings)
Existing simple trace can log upto 6 args
On 03/01/2012 11:33 AM, Harsh Bora wrote:
On 02/29/2012 01:53 AM, Lluís Vilanova wrote:
Harsh Prateek Bora writes:
This patchset introduces 2 major updates:
1) Tracetool Improvements (Conversion from shell script to python)
2) Simpletrace v2 log format (Support for variable args, strings
On 02/29/2012 01:53 AM, Lluís Vilanova wrote:
Harsh Prateek Bora writes:
This patchset introduces 2 major updates:
1) Tracetool Improvements (Conversion from shell script to python)
2) Simpletrace v2 log format (Support for variable args, strings)
ping
Hi Lluis,
I have actually moved into
On 02/10/2012 05:25 PM, Lluís Vilanova wrote:
Signed-off-by: Lluís Vilanovavilan...@ac.upc.edu
---
scripts/tracetool.py | 190 --
1 files changed, 91 insertions(+), 99 deletions(-)
diff --git a/scripts/tracetool.py b/scripts/tracetool.py
index
Hi Lluis,
Sorry for a late response as I was out of office for last 2 weeks
because of some medical emergency at home. While trying to merge your
changes with mine, I found something (see below):
On 02/04/2012 02:41 AM, Lluís Vilanova wrote:
Signed-off-by: Lluís Vilanovavilan...@ac.upc.edu
On 02/10/2012 05:19 PM, Lluís Vilanova wrote:
Signed-off-by: Lluís Vilanovavilan...@ac.upc.edu
---
scripts/tracetool.py | 51 +++---
1 files changed, 19 insertions(+), 32 deletions(-)
diff --git a/scripts/tracetool.py b/scripts/tracetool.py
index
On 02/10/2012 05:24 PM, Lluís Vilanova wrote:
Harsh Bora writes:
Hi Lluis,
Sorry for a late response as I was out of office for last 2 weeks because of
some medical emergency at home.
Hope it turned out to be nothing serious.
Yeh, situation is better now.
While trying to merge your
On 01/10/2012 10:11 PM, Stefan Hajnoczi wrote:
On Tue, Jan 10, 2012 at 10:59 AM, Harsh Prateek Bora
ha...@linux.vnet.ibm.com wrote:
Advantages over existing simpletrace backend:
- More than 6 elements (vitually unlimited) arguments can be traced.
- This allows to trace strings (variable size
On 01/11/2012 11:35 PM, Lluís Vilanova wrote:
Signed-off-by: Lluís Vilanovavilan...@ac.upc.edu
---
scripts/tracetool.py | 14 +++---
1 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/scripts/tracetool.py b/scripts/tracetool.py
index 6874f66..80e5684 100755
---
On 01/11/2012 11:35 PM, Lluís Vilanova wrote:
This would otherwise break event numbering when actually using the disable
property.
IIUC, event numbering does include disabled events too, are you
proposing not to include disabled events in event numbering ? That will
affect interpreting
On 01/18/2012 04:01 PM, Stefan Hajnoczi wrote:
On Wed, Jan 18, 2012 at 9:14 AM, Harsh Boraha...@linux.vnet.ibm.com wrote:
On 01/10/2012 10:11 PM, Stefan Hajnoczi wrote:
+unused = fwrite(record, ST_V2_REC_HDR_LEN, 1, trace_fp);
writeout_idx += num_available;
}
On 01/18/2012 04:11 PM, Harsh Bora wrote:
On 01/18/2012 04:01 PM, Stefan Hajnoczi wrote:
On Wed, Jan 18, 2012 at 9:14 AM, Harsh Boraha...@linux.vnet.ibm.com
wrote:
On 01/10/2012 10:11 PM, Stefan Hajnoczi wrote:
+ unused = fwrite(record, ST_V2_REC_HDR_LEN, 1, trace_fp);
writeout_idx
On 01/18/2012 04:29 PM, Stefan Hajnoczi wrote:
On Wed, Jan 18, 2012 at 10:52 AM, Harsh Boraha...@linux.vnet.ibm.com wrote:
On 01/18/2012 04:11 PM, Harsh Bora wrote:
On 01/18/2012 04:01 PM, Stefan Hajnoczi wrote:
On Wed, Jan 18, 2012 at 9:14 AM, Harsh Boraha...@linux.vnet.ibm.com
wrote
On 01/11/2012 12:08 PM, Harsh Bora wrote:
On 01/11/2012 04:21 AM, Lluís Vilanova wrote:
Harsh Prateek Bora writes:
Signed-off-by: Harsh Prateek Boraha...@linux.vnet.ibm.com
---
Makefile.objs | 6 +-
Makefile.target | 10 +-
configure | 7 +-
scripts/tracetool | 643
On 01/10/2012 05:44 AM, Mathieu Desnoyers wrote:
* Harsh Bora (ha...@linux.vnet.ibm.com) wrote:
On 01/09/2012 09:31 PM, Mathieu Desnoyers wrote:
* Harsh Prateek Bora (ha...@linux.vnet.ibm.com) wrote:
Existing simple trace can log upto 6 args per trace event and does not
support strings
On 01/10/2012 12:47 PM, Mathieu Desnoyers wrote:
* Harsh Bora (ha...@linux.vnet.ibm.com) wrote:
On 01/10/2012 05:44 AM, Mathieu Desnoyers wrote:
* Harsh Bora (ha...@linux.vnet.ibm.com) wrote:
On 01/09/2012 09:31 PM, Mathieu Desnoyers wrote:
* Harsh Prateek Bora (ha...@linux.vnet.ibm.com
On 01/10/2012 02:36 AM, Andreas Färber wrote:
Am 09.01.2012 16:46, schrieb Harsh Prateek Bora:
Note: Upstream Qemu have build issues with LTTng ust backend, as the trace
event APIs are not yet stable in ust and therefore ust backend is not
supported in this script as of now. Once the ust API
On 01/10/2012 02:36 PM, Harsh Bora wrote:
On 01/10/2012 12:47 PM, Mathieu Desnoyers wrote:
* Harsh Bora (ha...@linux.vnet.ibm.com) wrote:
On 01/10/2012 05:44 AM, Mathieu Desnoyers wrote:
* Harsh Bora (ha...@linux.vnet.ibm.com) wrote:
On 01/09/2012 09:31 PM, Mathieu Desnoyers wrote:
* Harsh
Hi Stefan,
Thanks for an early review, I shall address your comments in next version.
Also, please confirm, whether I should work on top for qemu.git or your
tracing branch on repo.or.cz/stefanha.git ?
regards,
Harsh
On 01/10/2012 08:20 PM, Stefan Hajnoczi wrote:
On Tue, Jan 10, 2012 at
On 01/11/2012 04:21 AM, Lluís Vilanova wrote:
Harsh Prateek Bora writes:
Signed-off-by: Harsh Prateek Boraha...@linux.vnet.ibm.com
---
Makefile.objs|6 +-
Makefile.target | 10 +-
configure|7 +-
scripts/tracetool| 643
On 01/09/2012 09:31 PM, Mathieu Desnoyers wrote:
* Harsh Prateek Bora (ha...@linux.vnet.ibm.com) wrote:
Existing simple trace can log upto 6 args per trace event and does not
support strings in trace record format. Introducing new trace format as
discussed earlier on list to support variable
On 01/02/2012 06:20 PM, Lluís Vilanova wrote:
Harsh Prateek Bora writes:
Converting tracetool.sh to tracetool.py
Excellent! I was thinking about switching the script language to python, but
wasn't sure if python was an acceptable build dependency for the project.
Thanks, Python appears
Hi,
Please ignore this duplicate patch in the series which was sent by
mistake, consider the one with subject '[RFC PATCH 4/4] simpletrace.py:
updated log reader script to handle new log format'. Sorry for
inconvenience.
- Harsh
On 01/02/2012 12:20 PM, Harsh Prateek Bora wrote:
Currently, Qemu provides an in-built simple trace backend which is
simple and easy to use (no additional/external dependencies) and allows
developers to trace events in Qemu code, however, it suffers from
limitations like unability to trace more than 6 elements per trace
event, lack of string
On 11/29/2011 06:26 PM, Stefan Hajnoczi wrote:
On Tue, Nov 29, 2011 at 12:35 PM, Markus Armbrusterarm...@redhat.com wrote:
Stefan Hajnoczistefa...@gmail.com writes:
[...]
So forget I said self-describing :). I think the only changes from
the v1 format we need are:
1. New magic number to
Hi,
I tried to explore LTTng UST with Qemu and built ust 0.15 (with urcu
0.16 as reqd) and observed that Qemu gives compilation errors when built
for trace-backend = ust. See this:
[harsh@harshbora v9fs]$ ./configure '--target-list=x86_64-softmmu'
'--enable-debug' '--enable-kvm'
On 11/09/2011 01:43 PM, Aneesh Kumar K.V wrote:
On Wed, 9 Nov 2011 11:55:51 +0530, Harsh Prateek
Boraha...@linux.vnet.ibm.com wrote:
Currently, we just print the numerical value of 9p operation identifier in
case of RERROR which is less meaningful for readability. Mapping 9p
operation ids to
On 10/14/2011 02:05 AM, Harsh Prateek Bora wrote:
Use QLIST_INSERT_HEAD_RCU and rcu_read_lock/unlock instead of rwlocks.
Use v9fs_synth_mutex as a write-only mutex to handle concurrent writers.
This patch is based on top of:
On 09/30/2011 04:06 PM, Harsh Prateek Bora wrote:
This python script allows to pretty print 9p simpletrace logs and can be
further enhanced to filter 9p logs based on command line arguments.
Sample ouput of this analysis script (will be added in commit log):
Pretty printing 9p
On 09/29/2011 10:46 PM, Stefan Hajnoczi wrote:
On Thu, Sep 29, 2011 at 06:09:47PM +0530, Harsh Prateek Bora wrote:
This patchset introduces Qemu Tracing to 9p pdu handlers and removes the
existing debug infrastructure which becomes less meaningful after the
introduction of coroutines. Parallel
On 09/22/2011 11:12 PM, Daniel P. Berrange wrote:
I've noticed that if you use a virtio 9p filesystem with a mount_tag
property value that is longer than 32 bytes, it gets silently truncated.
In virtio-9p-device.c
len = strlen(conf-tag);
if (len MAX_TAG_LEN) {
len =
On 08/10/2011 08:47 PM, Stefan Hajnoczi wrote:
On Fri, Aug 5, 2011 at 1:53 PM, Stefan Hajnoczistefa...@gmail.com wrote:
On Fri, Aug 5, 2011 at 12:32 PM, Aneesh Kumar K.V
aneesh.ku...@linux.vnet.ibm.com wrote:
On Fri, 5 Aug 2011 10:24:42 +0100, Stefan Hajnoczistefa...@gmail.com wrote:
On
On 04/06/2011 02:00 AM, Anthony Liguori wrote:
On 04/05/2011 03:25 PM, Peter Maydell wrote:
On 5 April 2011 14:14, Stefan Hajnoczistefa...@gmail.com wrote:
This stems from the fact that development is centered around the
mailing list. Some folks have put technical documentation on the wiki
but
On 04/05/2011 02:24 PM, Stefan Hajnoczi wrote:
On Tue, Apr 5, 2011 at 6:56 AM, Harsh Prateek Bora
ha...@linux.vnet.ibm.com wrote:
The nwnames field in TWALK message is assumed to be=0 and= MAXWELEM
which is defined as macro P9_MAXWELEM (16) in virtio-9p.h as per 9p2000
RFC. Appropriate changes
Hi,
Review comments / ACK please ?
regards,
Harsh
On 03/22/2011 11:21 AM, Harsh Prateek Bora wrote:
The nwnames field in TWALK message is assumed to be=0 and= MAXWELEM
which is defined as macro P9_MAXWELEM (16) in virtio-9p.h as per 9p2000 RFC.
Appropriate changes are required in V9fsWalkState
On 03/15/2011 04:08 PM, Arun R Bharadwaj wrote:
* Arun R Bharadwaja...@linux.vnet.ibm.com [2011-03-15 16:04:53]:
Author: Arun R Bharadwaja...@linux.vnet.ibm.com
Date: Thu Mar 10 15:11:49 2011 +0530
Helper routines to use GLib threadpool infrastructure in 9pfs.
This patch creates
54 matches
Mail list logo
