2:13 p.m.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
I introduced some memory leaks with my patch for handling disconnected
DBUS sends. This patch fixes it.
- --
Stephen Gallagher
RHCE 804006346421761
Delivering value year after year.
Red Hat ranks #1 in value among software vendors.
http://www.redhat.com/promo/vendor/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/
iEYEARECAAYFAkuhKe0ACgkQeiVVYja6o6NaugCeLZztH0hwwIRKfkuXccTx9r2/
04QAnRaG4NB5572AaJr+zZGPcwYqvNpe
=0Djp
-----END PGP SIGNATURE-----
2:20 p.m.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 03/17/2010 03:13 PM, Stephen Gallagher wrote:
I introduced some memory leaks with my patch for handling
disconnected
DBUS sends. This patch fixes it.
Pushed to master and 1-1-0
- --
Stephen Gallagher
RHCE 804006346421761
Delivering value year after year.
Red Hat ranks #1 in value among software vendors.
http://www.redhat.com/promo/vendor/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/
iEYEARECAAYFAkuhK24ACgkQeiVVYja6o6NxRwCgosB7L3o6CSWFrMrRBjADiACg
sUMAn1Z0cujGPagW9zZHq3uOy6ER+SPC
=iTx4
-----END PGP SIGNATURE-----
2:23 p.m.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 03/17/2010 03:20 PM, Stephen Gallagher wrote:
On 03/17/2010 03:13 PM, Stephen Gallagher wrote:
> I introduced some memory leaks with my patch for handling disconnected
> DBUS sends. This patch fixes it.
Pushed to master and 1-1-0
Sorry, I sent this reply to the wrong email. This has not yet been
reviewed or pushed upstream.
- --
Stephen Gallagher
RHCE 804006346421761
Delivering value year after year.
Red Hat ranks #1 in value among software vendors.
http://www.redhat.com/promo/vendor/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/
iEYEARECAAYFAkuhLCIACgkQeiVVYja6o6O88ACeJTDf8mXBnXdmHZGqy6Re4Hor
SNMAn2OD7hRoKsntpqf5fB+0Hyst/G6Y
=fSgZ
-----END PGP SIGNATURE-----
2:46 p.m.
Stephen Gallagher wrote:
On 03/17/2010 03:20 PM, Stephen Gallagher wrote:
> On 03/17/2010 03:13 PM, Stephen Gallagher wrote:
>> I introduced some memory leaks with my patch for handling disconnected
>> DBUS sends. This patch fixes it.
> Pushed to master and 1-1-0
Sorry, I sent this reply to the wrong email. This has not yet been
reviewed or pushed upstream.
Yeah, got me puzzled for a second.
Looking at the patch it seems pretty straightforward.
Was the messages dereferenced before your original patch and just the
dereferencing got lost or they never were dereferenced and leak was
actually always there?
_______________________________________________
sssd-devel mailing list
sssd-devel(a)lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/sssd-devel
--
Thank you,
Dmitri Pal
Engineering Manager IPA project,
Red Hat Inc.
-------------------------------
Looking to carve out IT costs?
www.redhat.com/carveoutcosts/
2:51 p.m.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 03/17/2010 03:46 PM, Dmitri Pal wrote:
Stephen Gallagher wrote:
> On 03/17/2010 03:20 PM, Stephen Gallagher wrote:
>> On 03/17/2010 03:13 PM, Stephen Gallagher wrote:
>>> I introduced some memory leaks with my patch for handling disconnected
>>> DBUS sends. This patch fixes it.
>
>
>> Pushed to master and 1-1-0
>
>
> Sorry, I sent this reply to the wrong email. This has not yet been
> reviewed or pushed upstream.
>
Yeah, got me puzzled for a second.
Looking at the patch it seems pretty straightforward.
Was the messages dereferenced before your original patch and just the
dereferencing got lost or they never were dereferenced and leak was
actually always there?
Prior to my patch, they were dereferenced (which will free the memory
once the references reach zero). When I converted the functions to use
the new routine sbus_conn_send(), I forgot to retain the dereference,
and thus caused a memory leak.
In the case of the monitor, this was a serious problem, as it pings the
other components constantly.
- --
Stephen Gallagher
RHCE 804006346421761
Delivering value year after year.
Red Hat ranks #1 in value among software vendors.
http://www.redhat.com/promo/vendor/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/
iEYEARECAAYFAkuhMskACgkQeiVVYja6o6OY6ACgmhMx/+cVzB9SIw6tXp34oziC
stYAn1GZ2x+SJRTh2H7UAjbvcXFjKfXM
=Pvcf
-----END PGP SIGNATURE-----
3:12 p.m.
Stephen Gallagher wrote:
On 03/17/2010 03:46 PM, Dmitri Pal wrote:
> Stephen Gallagher wrote:
>> On 03/17/2010 03:20 PM, Stephen Gallagher wrote:
>>> On 03/17/2010 03:13 PM, Stephen Gallagher wrote:
>>>> I introduced some memory leaks with my patch for handling
disconnected
>>>> DBUS sends. This patch fixes it.
>>
>>> Pushed to master and 1-1-0
>>
>> Sorry, I sent this reply to the wrong email. This has not yet been
>> reviewed or pushed upstream.
>>
> Yeah, got me puzzled for a second.
> Looking at the patch it seems pretty straightforward.
> Was the messages dereferenced before your original patch and just the
> dereferencing got lost or they never were dereferenced and leak was
> actually always there?
Prior to my patch, they were dereferenced (which will free the memory
once the references reach zero). When I converted the functions to use
the new routine sbus_conn_send(), I forgot to retain the dereference,
and thus caused a memory leak.
In the case of the monitor, this was a serious problem, as it pings the
other components constantly.
I would say ACK then but I reviewed just he diff and have not applied it.
I suspect this is not enough to give a full ACK so count as 1/2.
_______________________________________________
sssd-devel mailing list
sssd-devel(a)lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/sssd-devel
--
Thank you,
Dmitri Pal
Engineering Manager IPA project,
Red Hat Inc.
-------------------------------
Looking to carve out IT costs?
www.redhat.com/carveoutcosts/
3:41 p.m.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 03/17/2010 04:12 PM, Dmitri Pal wrote:
Stephen Gallagher wrote:
> On 03/17/2010 03:46 PM, Dmitri Pal wrote:
>> Stephen Gallagher wrote:
>>> On 03/17/2010 03:20 PM, Stephen Gallagher wrote:
>>>> On 03/17/2010 03:13 PM, Stephen Gallagher wrote:
>>>>> I introduced some memory leaks with my patch for handling
> disconnected
>>>>> DBUS sends. This patch fixes it.
>>>
>>>> Pushed to master and 1-1-0
>>>
>>> Sorry, I sent this reply to the wrong email. This has not yet been
>>> reviewed or pushed upstream.
>>>
>> Yeah, got me puzzled for a second.
>> Looking at the patch it seems pretty straightforward.
>> Was the messages dereferenced before your original patch and just the
>> dereferencing got lost or they never were dereferenced and leak was
>> actually always there?
>
>
> Prior to my patch, they were dereferenced (which will free the memory
> once the references reach zero). When I converted the functions to use
> the new routine sbus_conn_send(), I forgot to retain the dereference,
> and thus caused a memory leak.
>
> In the case of the monitor, this was a serious problem, as it pings the
> other components constantly.
>
I would say ACK then but I reviewed just he diff and have not applied it.
I suspect this is not enough to give a full ACK so count as 1/2.
George also applied, tested and acked this patch on IRC.
Pushed to master and 1-1-0
- --
Stephen Gallagher
RHCE 804006346421761
Delivering value year after year.
Red Hat ranks #1 in value among software vendors.
http://www.redhat.com/promo/vendor/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/
iEYEARECAAYFAkuhPoYACgkQeiVVYja6o6PeRACfYdEeQiC555sLiTCmTuqCf6GB
GzwAn3PzH4TIt2p46VqHIc5LZGBwAZ1M
=UoTq
-----END PGP SIGNATURE-----
5165
days inactive
5165
days old
sssd-devel@lists.fedorahosted.org
6 comments
2 participants
participants (2)
-
Dmitri Pal -
Stephen Gallagher