Platform/GFX/2011-October-11: Difference between revisions

From MozillaWiki
< Platform‎ | GFX
Jump to navigation Jump to search
(Created page with "*Exceptionally ***Tuesday*** at 2:30 PM Pacific/5:30 PM Eastern/21:30 UTC/Tuesday at 9:30 AM NZDT *+1 650 903 0800 x92 Conf# 99366 *+1 800 707 2533 (pin 369) Conf# 99366 (toll ...")
 
No edit summary
 
(3 intermediate revisions by 2 users not shown)
Line 1: Line 1:
*Exceptionally ***Tuesday*** at 2:30 PM Pacific/5:30 PM Eastern/21:30 UTC/Tuesday at 9:30 AM NZDT  
*Exceptionally '''Tuesday''' at 2:30 PM Pacific/5:30 PM Eastern/21:30 UTC/Wednesday at 9:30 AM NZDT  
*+1 650 903 0800 x92 Conf# 99366  
*+1 650 903 0800 x92 Conf# 99366  
*+1 800 707 2533 (pin 369) Conf# 99366 (toll free, Skype)  
*+1 800 707 2533 (pin 369) Conf# 99366 (toll free, Skype)  
Line 5: Line 5:
*Vidyo room 9366 (if you have LDAP and can log in at https://v.mozilla.com)
*Vidyo room 9366 (if you have LDAP and can log in at https://v.mozilla.com)
*Minutes will be recorded at http://etherpad.mozilla.com:9000/GFX-meeting then pasted on this wiki page.
*Minutes will be recorded at http://etherpad.mozilla.com:9000/GFX-meeting then pasted on this wiki page.
Agenda:


Agenda:
* BugKill aftermath (joe)
** From GFX section of today's development meeting:
** We closed 308 bugs out of the 493 total we touched - about 63% close rate. (There were bugs we didn't touch but did open; thus this is a maximum.)
** Anecdotal "evidence" says that older bugs were closed with a higher frequency than newer bugs.
** We found some patches that had languished for 4+ years, many without review requests (i.e., a failure on our part to notice them).
** We found other bugs that had languished for a long time for no good reason, and created patches for them.
** Still TBD: where we go from here for our remaining 2000+ bugs.
** Surprising/interesting bugs we found:
*** JP
**** Security bug that was open since 2005
**** Startup crash that hadn't been touched in a long time {{bug|585921}}
**** Take notes about specific bugs for further review
*** Joe
**** Bugs that talk about code cleanups that are totally out of date
**** Benoit Jacob had an example: {{bug|395585}} - OOM bugs should maybe be part of a query for making them security-sensitive
**** The other thing that surprised Joe was the ability for people to make patches for some bugs while triaging
**** Be Bold - change things! They can always be unchanged later
*** Bas
**** Having a direct line to people was great, since one specific person is often significantly more efficient than others
**** Being able to shuffle bugs to others' queues would have been very handy
***** QA Contact - perhaps?
***** Joe to edit queries to include QA Contact = "me" so they get reassigned
** Another BugKill day on the 4th of November (another Friday)
* New imagelib name (joe)
** libpr0n
** Please send your proposals to Joe
** Bobby Holley has a patch that moves imagelib to "/image"
** We can do better!
** Benoit J suggests "xxx"
* Java UI for Fennec (jpr)
** Doug Turner and a cast of thousands are building a native Java frontend for Fennec
** See Doug's blog post: http://dougt.org/wordpress/2011/09/fennec-and-native-android-uis/
** Might change some things for us with regard to GLES layers
** Might need to take some stuff off of Chris Lord's plate too
** Single process, multiple thread - oh my goodness
* GLES update (BenWa)
** Down to 12 reftest failures (!) - 2-4 are actual failures, the others have not been looked at
* Mini WebGL week
** In Toronto the week of October 24

Latest revision as of 03:33, 12 October 2011

Agenda:

  • BugKill aftermath (joe)
    • From GFX section of today's development meeting:
    • We closed 308 bugs out of the 493 total we touched - about 63% close rate. (There were bugs we didn't touch but did open; thus this is a maximum.)
    • Anecdotal "evidence" says that older bugs were closed with a higher frequency than newer bugs.
    • We found some patches that had languished for 4+ years, many without review requests (i.e., a failure on our part to notice them).
    • We found other bugs that had languished for a long time for no good reason, and created patches for them.
    • Still TBD: where we go from here for our remaining 2000+ bugs.
    • Surprising/interesting bugs we found:
      • JP
        • Security bug that was open since 2005
        • Startup crash that hadn't been touched in a long time bug 585921
        • Take notes about specific bugs for further review
      • Joe
        • Bugs that talk about code cleanups that are totally out of date
        • Benoit Jacob had an example: bug 395585 - OOM bugs should maybe be part of a query for making them security-sensitive
        • The other thing that surprised Joe was the ability for people to make patches for some bugs while triaging
        • Be Bold - change things! They can always be unchanged later
      • Bas
        • Having a direct line to people was great, since one specific person is often significantly more efficient than others
        • Being able to shuffle bugs to others' queues would have been very handy
          • QA Contact - perhaps?
          • Joe to edit queries to include QA Contact = "me" so they get reassigned
    • Another BugKill day on the 4th of November (another Friday)
  • New imagelib name (joe)
    • libpr0n
    • Please send your proposals to Joe
    • Bobby Holley has a patch that moves imagelib to "/image"
    • We can do better!
    • Benoit J suggests "xxx"
  • Java UI for Fennec (jpr)
    • Doug Turner and a cast of thousands are building a native Java frontend for Fennec
    • See Doug's blog post: http://dougt.org/wordpress/2011/09/fennec-and-native-android-uis/
    • Might change some things for us with regard to GLES layers
    • Might need to take some stuff off of Chris Lord's plate too
    • Single process, multiple thread - oh my goodness
  • GLES update (BenWa)
    • Down to 12 reftest failures (!) - 2-4 are actual failures, the others have not been looked at
  • Mini WebGL week
    • In Toronto the week of October 24