Am 25.10.2010 um 07:11 schrieb Carsten Haitzler (The Rasterman): > On Wed, 20 Oct 2010 22:01:26 +0200 Leif Middelschulte > <[email protected]> said: > > i don't see how that patch actually changes the logic...? first part makes no > difference. the case where res->modes[i].nameLength > 0 is unchanged, and the > alternate case is also the same (it's set to 0 -then CHANGED to es->modes > [i].nameLength if its > 0. as such name will be NULL unless length > 0... so > the 2nd part also wont change logic... i'm wondering what you truly fixed here > - other than just change some logic code that didnt fix the bug at all - it > just changed conditions on the stack? True, sorry, I missed the "ret->name = NULL;". I thought I would have checked for a NULL when freeing, even though I just used malloc when allocating the memory. I couldn't reproduce the bug described with my radeon driver and atm don't have access to an external monitor, but that will change soon and I'll be able to hunt that bug in e.
So, please forget about that non-sense patch and thanks for review. BR, Leif > > there may be an issue.. but it's somewhere/thing else. > >> Hello Thomas, >> >> at the moment I don't have access to an external monitor for further >> debugging of the randr mode setting issue. Last time I had access to >> an external monitor, I traced it down to a wrong events sent by my >> driver for some reason. >> It looks like your driver doesn't send mode names. Attached patch >> fixes their freeing. >> >> BR, >> >> Leif >> >> P.S.: Further patches for RandR follow as soon, as I have access to an >> external monitor again. >> >> 2010/10/20 Thomas Sachau <[email protected]>: >>> The first one only occured once, seems to be evas related and i cannot >>> easily reproduce it. >>> >>> The second one is randr related and i can reproduce it in a good amount of >>> cases, usually when i start an older game in wine in fullscreen mode, where >>> the aspect ration gets adjusted by the nvidia drivers (a 4:3 fullscreen >>> doesnt look that well on a 16:9 screen). If e17 does not crash during that >>> game, i can exit the game without problems. But once i try to exit e17, i >>> get the second bt, which results in e17 not reacting to any input (requires >>> a kill -9 to stop it). >>> >>> >>> -- >>> Thomas Sachau >>> >>> Gentoo Linux Developer >>> >>> >>> ------------------------------------------------------------------------------ >>> Nokia and AT&T present the 2010 Calling All Innovators-North America contest >>> Create new apps & games for the Nokia N8 for consumers in U.S. and Canada >>> $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing >>> Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store >>> http://p.sf.net/sfu/nokia-dev2dev >>> _______________________________________________ >>> enlightenment-devel mailing list >>> [email protected] >>> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel >>> >>> > > > -- > ------------- Codito, ergo sum - "I code, therefore I am" -------------- > The Rasterman (Carsten Haitzler) [email protected] > ------------------------------------------------------------------------------ Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev _______________________________________________ enlightenment-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
