Notice: This Wiki is now read only and edits are no longer possible. Please see: https://gitlab.eclipse.org/eclipsefdn/helpdesk/-/wikis/Wiki-shutdown-plan for the plan.
Difference between revisions of "Eclipse/Status meetings/Minutes"
< Eclipse | Status meetings
(→February 25, 2015) |
|||
Line 12: | Line 12: | ||
==== Platform Workspace (Szymon P) ==== | ==== Platform Workspace (Szymon P) ==== | ||
+ | * 4.5 M6 bug fixing. | ||
+ | * Performance tests seem to be fine although sometimes inexplicable spikes happen which turn out to be false alarms ({{bug|458959}}). | ||
==== Platform UI (Daniel R) ==== | ==== Platform UI (Daniel R) ==== |
Revision as of 06:44, 25 February 2015
February 25, 2015
Discussion/Reminders
- Each component lead comments on one performance test (is the test/difference accurate?)
Status
Platform Workspace (Szymon P)
- 4.5 M6 bug fixing.
- Performance tests seem to be fine although sometimes inexplicable spikes happen which turn out to be false alarms (bug 458959).
Platform UI (Daniel R)
Platform Text (Dani)
Platform Debug (Sarika)
- Mars M6 Bug fixing
- Moving to 1.7 BREE
Platform Ant (Sarika)
- No Unexpected test performance results observed
Platform SWT (Arun)
Platform Releng (David)
JDT Core (Jay)
JDT Debug (Sarika)
- Mars M6 Bug fixing
- Adapting to Platform UI IAdaptable changes
- Performance test to load Breakpoint Manager with 50 BPS is down 23%. Even though loading with 200 BPS does not seem to have regressed. In One scenario, it was reversely observed with 32% improvement with 200 BPS and 23% regression with 50 BPs. Needs some analysis.
JDT UI (Markus)
PDE (Vikas)
Equinox (Tom Watson)
Equinox p2 (Ian Bull)
February 18, 2015
Discussion/Reminders
- EGit 3.7: if you see strange selection behavoir when you click into an editor, see bug 460243
- Homework for next meeting: each component lead comments on one performance test (is the test/difference accurate?)
Status
Platform Workspace (Szymon P)
Platform UI (Daniel R)
Platform Text (Dani)
Platform Debug (Sarika)
- Mars 4.5 M6 Bug fixes
Platform Ant (Sarika)
Platform SWT (Arun)
Platform Releng (David)
JDT Core (Jay)
- Code Formatter has been replaced with a new implementation (bug 303519).
- Some follow-ups are pending in bug 458208.
- Please file bugs if you see other strange behavior in the formatter, in content assist, quick fixes, or refactorings.
JDT Debug (Sarika)
- Mars 4.5 M6 Bug fixes
JDT UI (Markus)
PDE (Vikas)
Equinox (Tom Watson)
Equinox p2 (Ian Bull)
February 11, 2015
Discussion/Reminders
Status
Platform Workspace (Szymon P)
- 4.4.2 testing
- 4.5 M6 bug fixing
Platform UI (Daniel R)
Platform Text (Dani)
Platform Debug (Sarika)
- 4.4.2 RC3 testing and pass
- Mars M6 bug fixes
Platform Ant (Sarika)
- 4.4.2 RC3 testing and pass
Platform SWT (Arun)
- 4.4.2 RC3 testing
- 4.5 M6 bug fixes
Platform Releng (David)
- Note the SDK has moved to specify -Xmx1024 (by default) instead of 512m ... which will also be used in our unit tests and performance tests (by default), effective in tonight's "nightly" tests.
- Our Platform Luna SR2 RC4 is due this week (which is our "final build" for SR2, but appears a new build is not needed. So far?
- - Unless someone feels strongly otherwise, I will rename our "RC3" as "RC4" on due date, Friday.
- Even though "same bits" (same versions, etc.) from my experience, less confusing to community to have them renamed on DL site, and will state in "announcement" that they are the same bits.
- Anyone know why so many failures in last night's Nightly?
JDT Core (Jay)
JDT Debug (Sarika)
- 4.4.2 RC3 testing and pass
- Mars M6 Bug analysis
JDT UI (Markus)
- filed CQ for JUnit 4.12
- Content Assist auto-activation delay set to zero
PDE (Vikas)
- Mars M6 bug fixes
Equinox (Tom Watson)
Equinox p2 (Ian Bull)
February 4, 2015
Discussion/Reminders
- From the Eclipse Luna SR2 (4.4.2) Endgame Plan:
- Tomorrow we have a one day test pass against Luna SR2 (4.4.2) RC3
- After RC3 all changes need PMC approval on the bug report
Status
Platform Workspace (Szymon P)
Platform UI (Daniel R)
Platform Text (Dani)
Platform Debug (Sarika)
- Mars M5 and 4.4.2 RC2 testing and pass
Platform Ant (Sarika)
- Mars M5 and 4.4.2 RC2 testing and pass
- Mars M6 bug fixes
Platform SWT (Arun)
Platform Releng (David)
- Builds, Builds, and re-builds!
- - Orbit M5a, in addition to Platform M5a
- - Infrastructure "restart" (for GHOST vulnerability fix) was unexpected "disruption"
- - That is, many "aftershocks"
JDT Core (Jay)
JDT Debug (Sarika)
- Mars M5 and 4.4.2 RC2 testing and pass
- Mars M6 bug fixes
JDT UI (Markus)
PDE (Vikas)
- Mars M5 and Luna 4.4.2 RC2 testing/pass
- Mars M6 bug fixes
Equinox (Tom Watson)
Equinox p2 (Ian Bull)
January 28, 2015
Discussion/Reminders
- Component leads, please clean up your open 4.5 M5 bugs
- [Markus] provide N&N items by Wednesday, January 28, end-of-day
- I'll be away in the M5 week, so please contact Dani if you have questions.
Status
Platform Workspace (Szymon P)
- M5 testing and bug verification
Platform UI (Wojtek)
Platform Text (Dani)
- Reviewed and committed a contribution that speeds up text search by parallelizing it using JobGroups (bug 441016 )
Platform Debug (Sarika)
- Mars M5 Bug fixing and verification
Platform Ant (Sarika)
- 4.4.2 Bug fixing
Platform SWT (Arun)
- Luna SR2 and Mars M5 bug fixes in SWT
- Launchers recompiled for Mars M5 458558 as well as Luna SR2 449990 to fix issues with Oracle Java 9 VMs
- Mars M5 testing and bug verification
Platform Releng (David)
- Last weeks fix for blocking 4.4.2 bug was incomplete ... so, NOW it is fixed (bug 457071).
- Rev'd ICU4J in Orbit, and Platform:
- - 52.1.1 will be in Luna SR2. 54.1.1 will be in Mars M5.
- - please sanity check, if any of you have a favorite "NL test case".
- Last minute problems with infrastructure on Tuesday night ...
- - any special considerations anyone needs? .. that impact exact schedule? Such as, does anyone need a short "delay" to finish commits that you could not finish due to lack of Git access?
JDT Core (Jay)
- 4.5 M5 bug fixes and testing
JDT Debug (Sarika)
- Mars M5 Bug fixing and verification
JDT UI (Markus)
PDE (Vikas)
Equinox (Tom Watson)
Equinox p2 (Ian Bull)
January 21, 2015
Discussion/Reminders
- M5 Stabilization Builds to begin Sunday, 1/25.
- Time to triage Luna SR2 (4.4.2) open bugs?
Status
Platform Workspace (Szymon P)
Platform UI (Wojtek)
Platform Text (Dani)
Platform Debug (Sarika)
Platform Ant (Sarika)
- Mars M5 Bug fixing and bug verifications
Platform SWT (Arun)
- Mars M5 and Luna SR2 bug fixes
- Rebuilt launchers in Luna SR2 (and soon in Mars) to allow Eclipse to run with Oracle Java 9 VMs, more details in bug 449990
Platform Releng (David)
- 4.4.2
- - All blockers fixed.
- - Still waiting for ICU4J CQ 9141 approval (needed for bug 444897
- - A few other "releng" bugs.
- 4.5
- - Want to update Jetty to 9.2.7 bug 453910 but a) its update site is not ready yet. b) not positive the desire bug was fixed.
- - Several other releng changes (such "update compiler")
JDT Core (Jay)
JDT Debug (Sarika)
- Mars M5 Bug fixing and bug verifications
JDT UI (Markus)
PDE (Vikas)
Equinox (Tom Watson)
Equinox p2 (Ian Bull)
January 14, 2015
Discussion/Reminders
- Can "web app dependency" issue be fixed by making its web app BREE 1.6? bug 457396. (Jetty and org.apache.jasper.glassfish require that anyway.)
- Reminder: This week's M-build is for Luna SR2 RC1. Build schedule has been updated for Luna SR2 and M5 ... there is some overlap, so a few "busy days".
- Architectural question: Should we increase default -Xmx? (See bug 456585) It has been 512m for 5 years. I *think* VMs now "use only what they need", not literally reserving max heap value, like they used to? Anyone know for sure? Or, who to ask? An alternative, interesting approach (at least approximately 2012) some IDEs set the default "mx" more dynamically, based on the memory they find installed on the system (such as 1/4 total RAM -- which user can still override).
- Time to triage Luna SR2 (4.4.2) open bugs?
Status
Platform Workspace (Szymon P)
Platform UI (Wojtek)
Platform Text (Dani)
Platform Debug (Sarika)
- Mars M5 Bug fixing and bug verifications
Platform Ant (Sarika)
Platform SWT (Arun)
Platform Releng (David)
- Produced the "JGit Security Fix" repo for Luna SR1a (bug 456947)
- Some tricky issues getting some teams "aggregating" on Luna (bug 457144
- Despite the problem it's causing us now, important fix in Orbit's "jasper.glassfish" bundle (with Tom's help!) (bug 442310)
- ICU4J
- - we plan to update for 4.4.2 and 4.5 (different versions). ICU team "nearly done building" (every day :)
- Inconsistencies in org.eclipes.rcp.configuration. (bug 457071)
- - in local tests, two steps forward, one step back ... but, more optimistic than I was a week ago, that it can be fixed.
- - did get some of the history of those "configuration units" from Andrew (via email) [which I'll summarize in some bug, somewhere.]
JDT Core (Jay)
JDT Debug (Sarika)
JDT UI (Markus)
PDE (Vikas)
- 442 bugs
- 45 M5 bugs
Equinox (Tom Watson)
Equinox p2 (Ian Bull)
January 7, 2015
Discussion/Reminders
- Next Wednesday, 1/14 will be the M-build we contribute to Luna SR2 RC1 (warm-up RC).
- in two weeks, 1/28 is our I- (S-) build we contribute to Mars M5.
Status
Platform Workspace (Szymon P)
Platform UI (Daniel R)
Platform Text (Dani)
Platform Debug (Sarika)
Platform Ant (Sarika)
Platform SWT (Arun)
Platform Releng (David)
- 442:
- will update ICU4J
- investigating making "rcp.config" have consistent meta-data (same for each Platform).