1257919 - Stop using ThreadsafeAutoJSContext in DOM exception code

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Boris Zbarsky [:bzbarsky]]

This is going to be more exciting than one might think, because it involves using [explicit_jscontext] and hence not JS-implementing nsIException and the like.

Comment 1

[Boris Zbarsky [:bzbarsky]]

Attachment: part 1. Kill off nsIXSLTException

Comment 2

[Boris Zbarsky [:bzbarsky]]

Attachment: part 2. Make nsIException and nsIStackFrame builtinclass, so we can start using [implicit_jscontext] on them

Comment 3

[Boris Zbarsky [:bzbarsky]]

Attachment: part 3. Make the various toString methods on exceptions take an explicit JSContext

Comment 4

[Boris Zbarsky [:bzbarsky]]

Attachment: part 4. Stop returning StackFrame instances from exceptions::CreateStack. C++ callers of GetCurrentJSStack or exceptions::CreateStack always check for null anyway, and none of them seem to want this non-JS thing

Comment 5

[Boris Zbarsky [:bzbarsky]]

Attachment: part 5. Get rid of the now-unused StackFrame class

Comment 6

[Boris Zbarsky [:bzbarsky]]

Attachment: part 6. Make the filename getter on JSStackFrame take an explicit JSContext

Comment 7

[Boris Zbarsky [:bzbarsky]]

Attachment: part 7. Make the name getter on JSStackFrame take an explicit JSContext

Comment 8

[Boris Zbarsky [:bzbarsky]]

Attachment: part 8. Make the line/column number getters on JSStackFrame take an explicit JSContext

Comment 9

[Boris Zbarsky [:bzbarsky]]

Attachment: part 9. Make the asyncCause/asyncCaller getters on JSStackFrame take an explicit JSContext

Comment 10

[Boris Zbarsky [:bzbarsky]]

Attachment: part 10. Make the caller and formattedStack getters on JSStackFrame take an explicit JSContext

Comment 11

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Comment on attachment 8733056 [details] [diff] [review]
part 5.  Get rid of the now-unused StackFrame class

Review of attachment 8733056 [details] [diff] [review]:
-----------------------------------------------------------------

::: dom/bindings/Exceptions.cpp
@@ +262,5 @@
>  {
>    MOZ_ASSERT(mStack);
>  
>    mozilla::HoldJSObjects(this);
>    mLineno = 0;

You're initializing mLineno twice.

Comment 12

[Boris Zbarsky [:bzbarsky]]

> You're initializing mLineno twice.

Good catch, fixed.

Comment 13

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/2603e80db18f
https://hg.mozilla.org/integration/mozilla-inbound/rev/8bb1cce4804d
https://hg.mozilla.org/integration/mozilla-inbound/rev/6403c45366e0
https://hg.mozilla.org/integration/mozilla-inbound/rev/1b5db6edd791
https://hg.mozilla.org/integration/mozilla-inbound/rev/83dfa9e03d0e
https://hg.mozilla.org/integration/mozilla-inbound/rev/6e95ee3cd4c6
https://hg.mozilla.org/integration/mozilla-inbound/rev/c4faeb0be959
https://hg.mozilla.org/integration/mozilla-inbound/rev/08f1c7239cdf
https://hg.mozilla.org/integration/mozilla-inbound/rev/cf816f733296
https://hg.mozilla.org/integration/mozilla-inbound/rev/ff81c52375ba

Comment 14

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

Backed out bug 1257919, bug 1256376 and bug 1256342 for Reftest, Jit and Spidermonkey failures.

https://hg.mozilla.org/integration/mozilla-inbound/rev/42c21bf6657b
https://hg.mozilla.org/integration/mozilla-inbound/rev/da683a9c88cf
https://hg.mozilla.org/integration/mozilla-inbound/rev/b9462b6b2785
https://hg.mozilla.org/integration/mozilla-inbound/rev/f5962f5a4e3f
https://hg.mozilla.org/integration/mozilla-inbound/rev/e22b4e5d45b6
https://hg.mozilla.org/integration/mozilla-inbound/rev/037fc169fe07
https://hg.mozilla.org/integration/mozilla-inbound/rev/421122394970
https://hg.mozilla.org/integration/mozilla-inbound/rev/692110787614
https://hg.mozilla.org/integration/mozilla-inbound/rev/3946d06e581e
https://hg.mozilla.org/integration/mozilla-inbound/rev/5dc49e4fb827
https://hg.mozilla.org/integration/mozilla-inbound/rev/d93dc0d5266f
https://hg.mozilla.org/integration/mozilla-inbound/rev/952832165a9b

Push with failures: https://treeherder.mozilla.org/#/jobs?repo=mozilla-inbound&revision=ff81c52375ba

Comment 15

[Boris Zbarsky [:bzbarsky]]

The problem was the fix for bug 1256376.

Comment 16

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/1695f5baaf53
https://hg.mozilla.org/integration/mozilla-inbound/rev/163552af0458
https://hg.mozilla.org/integration/mozilla-inbound/rev/818a9b425574
https://hg.mozilla.org/integration/mozilla-inbound/rev/555822dca0fd
https://hg.mozilla.org/integration/mozilla-inbound/rev/0e68f8507f91
https://hg.mozilla.org/integration/mozilla-inbound/rev/f899bc5270d4
https://hg.mozilla.org/integration/mozilla-inbound/rev/eb6d3712cf62
https://hg.mozilla.org/integration/mozilla-inbound/rev/5481e58d8735
https://hg.mozilla.org/integration/mozilla-inbound/rev/1091b24a1f23
https://hg.mozilla.org/integration/mozilla-inbound/rev/681f1c98fcfb

Comment 17

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/1695f5baaf53
https://hg.mozilla.org/mozilla-central/rev/163552af0458
https://hg.mozilla.org/mozilla-central/rev/818a9b425574
https://hg.mozilla.org/mozilla-central/rev/555822dca0fd
https://hg.mozilla.org/mozilla-central/rev/0e68f8507f91
https://hg.mozilla.org/mozilla-central/rev/f899bc5270d4
https://hg.mozilla.org/mozilla-central/rev/eb6d3712cf62
https://hg.mozilla.org/mozilla-central/rev/5481e58d8735
https://hg.mozilla.org/mozilla-central/rev/1091b24a1f23
https://hg.mozilla.org/mozilla-central/rev/681f1c98fcfb