Closed Bug 453636 Opened 14 years ago Closed 14 years ago

Make nanojit build with VC7.1

Categories

(Core :: JavaScript Engine, defect)

x86
All
defect
Not set
minor

Tracking

()

VERIFIED FIXED

People

(Reporter: neil, Assigned: neil)

References

Details

(Keywords: verified1.9.1)

Attachments

(3 files)

As hinted at in bug 451881, nanojit now uses varadic macros.

I plan to use the same "workaround" as I did for bug 450265.
Assignee: general → neil
Status: NEW → ASSIGNED
Attachment #336847 - Flags: review?(danderson)
Attachment #336847 - Flags: review?(danderson) → review+
Pushed fix as changeset ccb011438e5c.
Status: ASSIGNED → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED
Blocks: 455210
Flags: in-testsuite-
Flags: in-litmus-
Bug 457786 block this bug again.
Attached patch patchSplinter Review
fix for comment 3
Attachment #345269 - Flags: review?
Comment on attachment 345269 [details] [diff] [review]
patch

The nanojit patch fixes debug builds. I haven't built nondebug recently so I can't speak for the Fragmento.cpp changes.
Attachment #345269 - Flags: review? → review?(danderson)
(In reply to comment #5)
> (From update of attachment 345269 [details] [diff] [review])
> The nanojit patch fixes debug builds. I haven't built nondebug recently so I
> can't speak for the Fragmento.cpp changes.

As pageCount is wrapped with #ifdef NJ_VERBOSE, the NanoAssert lines that use pageCount should be wrapped with #ifdef NJ_VERBOSE too.
Attachment #345269 - Flags: review?(danderson) → review+
Status: RESOLVED → REOPENED
Keywords: checkin-needed
Resolution: FIXED → ---
Attachment #336847 - Attachment description: Proposed patch → Proposed patch [Checkin: Comment 2]
Whiteboard: [c-n: wait for comment 6, or not ?]
Fragmento.cpp compiles for me with VC7.1 non-debug without the #ifdefs
(In reply to comment #7)
> Fragmento.cpp compiles for me with VC7.1 non-debug without the #ifdefs

but I got undefined references about pageCount.
Attachment #345269 - Flags: approval1.9.1b2?
Keywords: checkin-needed
Attachment #345269 - Flags: approval1.9.1b2?
Dão, why the flag change?
Feel free to re-request yourself. I canceled my request because release drivers are trying to limit "the number of checkins required before we can finish this beta", and this bug doesn't seem to be critical for the beta.
Attachment #345269 - Flags: approval1.9.1?
Comment on attachment 345269 [details] [diff] [review]
patch

a191=beltzner
Attachment #345269 - Flags: approval1.9.1? → approval1.9.1+
Keywords: checkin-needed
Whiteboard: [c-n: wait for comment 6, or not ?]
Pushed changeset 035f57be79c0 to mozilla-central.
Status: REOPENED → RESOLVED
Closed: 14 years ago14 years ago
Keywords: checkin-needed
Resolution: --- → FIXED
I've to reopen this bug as changes in mozilla-central revert some of changes of this bug.
(In reply to comment #13)
> Created an attachment (id=352817) [details]
> patch for fix msvc 7.1 build
> 
> I've to reopen this bug as changes in mozilla-central revert some of changes of
> this bug.
Actually I filed bug 469289 on the new bustage.
Keywords: fixed1.9.1
pushed http://hg.mozilla.org/releases/mozilla-1.9.1/rev/035f57be79c0

and

http://hg.mozilla.org/mozilla-central/rev/035f57be79c0


also, seeing as there hasn't been any discussions about this bug for 4 months, I'm assuming there aren't any residual issues. I'm moving this to verified as a result. If anyone has any qualms, feel free to bring them up.
Status: RESOLVED → VERIFIED
It breaks again.
I'm thinking if Bug 494864 is blocking this bug.
(In reply to comment #16)
> It breaks again.
Is that covered by my patch in bug 501609?
(Note: I don't know whether I picked the right reviewer there.)
(In reply to comment #17)
> (In reply to comment #16)
> > It breaks again.
> Is that covered by my patch in bug 501609?
> (Note: I don't know whether I picked the right reviewer there.)

seems so.
I'm getting following errors, after applied your patch:
c:\moz\mozilla-central\js\src\jstracer.cpp(2974) : error C3861: 'getExitName': identifier not found, even with argument-dependent lookup
c:\moz\mozilla-central\js\src\jstracer.cpp(3896) : warning C4805: '==' : unsafe mix of type 'int32_t' and type 'bool' in operation
c:\moz\mozilla-central\js\src\jstracer.cpp(4804) : error C3861: 'getExitName': identifier not found, even with argument-dependent lookup
c:\moz\mozilla-central\js\src\jstracer.cpp(5359) : error C2065: 'cycles' : undeclared identifier
c:\moz\mozilla-central\js\src\jstracer.cpp(5356) : error C3861: 'getExitName': identifier not found, even with argument-dependent lookup
c:\moz\mozilla-central\js\src\jstracer.cpp(6911) : warning C4800: 'JSBool' : forcing value to bool 'true' or 'false' (performance warning)
c:\moz\mozilla-central\js\src\jstracer.cpp(6932) : warning C4800: 'JSBool' : forcing value to bool 'true' or 'false' (performance warning)
c:\moz\mozilla-central\js\src\jstracer.cpp(6992) : warning C4800: 'JSBool' : forcing value to bool 'true' or 'false' (performance warning)
c:\moz\mozilla-central\js\src\jstracer.cpp(8435) : warning C4800: 'uintN' : forcing value to bool 'true' or 'false' (performance warning)
c:\moz\mozilla-central\js\src\jstracer.cpp(8620) : warning C4018: '>' : signed/unsigned mismatch
make[4]: *** [jstracer.obj] Error 2
You need to log in before you can comment on or make changes to this bug.