I’ll provide my input now, since I’ll be on
vacation for the next 2 weeks, except for next Wed and Thurs.
> 2. Where do we feel we stand
with respect to the quality of 3.0? We have less dedicated test resources as
with previous releases - what are people's assesment? To Chris' point, are we
overreaching? And are people comfortable with the bugs we moved forward to
"future" (e.g. are there any bugs people feel should really be looked
at)?
From the MBS perspective, I feel that we have gotten more “community”
testing than in any prior release. The only areas of new MBS functionality
that I don’t feel comfortable with the level of testing are the new tool-chain
converter support and the new multi-version tool-chain support. I don’t
suggest that we change the schedule for that reason. If things need to be
fixed, we’ll fix them in a 3.01 or 3.1.
Regarding the other opens bugs, I agree with others than we
should not stick to the schedule irregardless of the quality. If there
are bugs that should be fixed, then we should slip the schedule to fix them. From
my perspective, it is more important to fix regressions and crashes (unless
they require uncommon circumstances) than bugs in the new functionality.
Regards,
Leo
From:
cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Sebastien Marineau
Sent: Friday, July 22, 2005 12:25
PM
To: cdt-dev@xxxxxxxxxxx
Subject: RE: [cdt-dev] CDT 3.0
Closing
To follow-up on the 3.0 closing...
1. As it stands, we have 17 open bugs
against 3.x. Looking at the bugs, the majority look like they will be moved
forward (99744 is definitely required). I just spoke with Mikhail and Alain,
they are evaluating the bugs in their name and will either fix for RC3 or move
to future. They can provide an update on where this stands.
===== current 3.x bugs =====
ID
|
Sev
|
Pri
|
Plt
|
Assignee
|
Status
|
Resolution
|
Summary
|
30365
|
nor
|
P3
|
PC
|
alain@xxxxxxx
|
NEW
|
|
Header file errors => empty tasks
|
38197
|
nor
|
P3
|
PC
|
alain@xxxxxxx
|
NEW
|
|
wrong tasks shows as result of compilation.
|
52676
|
nor
|
P3
|
PC
|
alain@xxxxxxx
|
NEW
|
|
I18N: Makefile editor input via alt keystroke
sequence cl...
|
53174
|
maj
|
P3
|
PC
|
Mikhailk@xxxxxxx
|
REOP
|
|
I18N: CDT Debug can't find a non-ascii filename
|
59193
|
nor
|
P3
|
PC
|
alain@xxxxxxx
|
NEW
|
|
Cannot debug DLLs with DOS slash in path
|
59320
|
enh
|
P3
|
PC
|
alain@xxxxxxx
|
NEW
|
|
Remove Navigator view from the default C/C++
Perspective
|
82077
|
nor
|
P3
|
PC
|
Mikhailk@xxxxxxx
|
NEW
|
|
Eclipse CDT spontaneously terminated while single
stepping
|
83566
|
enh
|
P3
|
PC
|
dinglis@xxxxxxx
|
NEW
|
|
[CPathEntry] Container content not refreshed
|
85124
|
nor
|
P3
|
PC
|
alain@xxxxxxx
|
NEW
|
|
Debugger halts with NFE on AMD64
|
92446
|
nor
|
P3
|
PC
|
alain@xxxxxxx
|
NEW
|
|
Wrong variable display if resume to breakpoint in
function
|
99744
|
nor
|
P3
|
PC
|
dschaefer@xxxxxxxxxx
|
NEW
|
|
Transition to EPL
|
102195
|
maj
|
P2
|
PC
|
cdt-core-inbox@xxxxxxxxxxx
|
NEW
|
|
StackOverFlow when updating Type Cache
|
102929
|
nor
|
P3
|
Mac
|
Mikhailk@xxxxxxx
|
NEW
|
|
race conditions when setting breakpoints at launch
|
104421
|
maj
|
P3
|
PC
|
Mikhailk@xxxxxxx
|
NEW
|
|
Register view can not show correct value when switch
betw...
|
104515
|
maj
|
P3
|
All
|
leo.treggiari@xxxxxxxxx
|
NEW
|
|
Tool converter never called
|
104605
|
nor
|
P3
|
Mac
|
cwiebe@xxxxxxxx
|
NEW
|
|
UI hangs when opening Binaries container (MachO
binary file)
|
104721
|
nor
|
P3
|
PC
|
gheorghe@xxxxxxxxxx
|
NEW
|
|
Mozilla - ctags indexer produces nothing
|
2. Where do we feel we stand with
respect to the quality of 3.0? We have less dedicated test resources as with
previous releases - what are people's assesment? To Chris' point, are we
overreaching? And are people comfortable with the bugs we moved forward to
"future" (e.g. are there any bugs people feel should really be looked
at)?
3. I've started putting the material
together for the release review (you can see the presentations from other
projects at http://www.eclipse.org/org/processes/previous-release-reviews.html).
This is something that's been part of the Eclipse process, but has not been
done in the past. For CDT 3.0, we have to go throgh this process and all
have to be comfortable with the release).
Related to this, on the question of
3-party plugin compatibility - we also do have to produce a list of APIs the
CDT provides along with their status (provisional/platform/etc) (see the BIRT
release summary above for an example); doing this systematically should make it
more clear as to the level of maturity and compatibility the different APIs
provide. I'll do an initial list but would appreciate feedback/input into it.
Can I suggest we all get together on a
conf call Monday (say at 1EST) to discuss the above and map out the
closing of 3.0?
Thanks,
Sebastien