2038163 - Perma gfx/tests/mochitest/test_font_whitelist.html | unable to find a usable font (sans-serif) [@ gfxFontGroup::GetDefaultFont]

Status: RESOLVED DUPLICATE of bug 2037909

(Enterprise Products :: Firefox, defect)

Description

[Treeherder Bug Filer]

Filed by: vlopezgarcia [at] mozilla.com
Parsed log: https://treeherder.mozilla.org/logviewer?job_id=564456772&repo=enterprise-main&task=e9B5rgUMTXW0sadyUPGZig.0
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/e9B5rgUMTXW0sadyUPGZig/runs/0/artifacts/public/logs/live_backing.log

---

Comment 1

[vlopezgarcia]

It was added in here https://treeherder.mozilla.org/jobs?repo=enterprise-main&searchStr=opt-mochitest-plain%2C5&revision=60774249204178398f72d33d03a1134470c20d03

Comment 2

[vlopezgarcia]

We are seeing a crash in gfx/tests/mochitest/test_font_whitelist.html. This appears to be a regression potentially introduced by the changes in Bug 2024049.

Could you please verify if these are related? If this behavior is expected given the recent changes, should we update the test expectations or skip it for now?

https://treeherder.mozilla.org/jobs?repo=enterprise-main&searchStr=opt-mochitest-plain%2C5&fromchange=ed07f36795a5bd9548d392a0d0cc1ac56a19e341&selectedTaskRun=A1UD4BpDQy68M4uB-jRTdg.0

XUL	MOZ_CrashSequence(void*, long)	/builds/worker/checkouts/gecko/mfbt/Assertions.h:266	inline
	XUL	MOZ_Crash(char const*, int, char const*)	/builds/worker/checkouts/gecko/mfbt/Assertions.h:402	inline
0	XUL	gfxFontGroup::GetDefaultFont()	/builds/worker/checkouts/gecko/gfx/thebes/gfxTextRun.cpp:2238 +0x5e8	context
	XUL	GetMetrics(nsFontMetrics const*, nsFontMetrics::FontOrientation)	/builds/worker/checkouts/gecko/gfx/src/nsFontMetrics.cpp:214	inline
	XUL	GetMetrics(nsFontMetrics const*)	/builds/worker/checkouts/gecko/gfx/src/nsFontMetrics.cpp:219	inline
1	XUL	nsFontMetrics::ExternalLeading() const	/builds/worker/checkouts/gecko/gfx/src/nsFontMetrics.cpp:272 +0x28	cfi
	XUL	GetNormalLineHeight(nsFontMetrics*)	/builds/worker/checkouts/gecko/layout/generic/ReflowInput.cpp:2707	inline
2	XUL	ComputeLineHeight(mozilla::StyleGenericLineHeight<float, mozilla::StyleCSSPixelLength> const&, nsFont const&, nsAtom*, bool, nsPresContext*, bool, int, float)	/builds/worker/checkouts/gecko/layout/generic/ReflowInput.cpp:2759 +0xec	cfi
3	XUL	mozilla::ReflowInput::CalcLineHeight(mozilla::StyleGenericLineHeight<float, mozilla::StyleCSSPixelLength> const&, nsStyleFont const&, nsPresContext*, bool, nsIContent const*, int, float)	/builds/worker/checkouts/gecko/layout/generic/ReflowInput.cpp:2810 +0x54	cfi
4	XUL	Gecko_CalcLineHeight	/builds/worker/checkouts/gecko/layout/style/GeckoBindings.cpp:377 +0x54	cfi
	XUL	style::device::gecko::<impl style::device::Device>::calc_line_height	/builds/worker/checkouts/gecko/servo/components/style/device/gecko.rs:89	inline
	XUL	style::matching::MatchMethods::finish_restyle	/builds/worker/checkouts/gecko/servo/components/style/matching.rs:1000	inline
	XUL	style::traversal::compute_style	/builds/worker/checkouts/gecko/servo/components/style/traversal.rs:638	inline
	XUL	style::traversal::recalc_style_at	/builds/worker/checkouts/gecko/servo/components/style/traversal.rs:397	inline
	XUL	<style::gecko::traversal::RecalcStyleOnly as style::traversal::DomTraversal<style::gecko::wrapper::GeckoElement>>::process_preorder	/builds/worker/checkouts/gecko/servo/components/style/gecko/traversal.rs:37	inline
5	XUL	style::parallel::style_trees	/builds/worker/checkouts/gecko/servo/components/style/parallel.rs:171 +0x1dac	cfi

Comment 3

[Emilio Cobos Álvarez [:emilio]]

Jonathan do you know when this might be hit?

Comment 4

[Jonathan Kew [:jfkthame]]

This looks like a dupe of bug 2037909. It's not currently clear to me why we'd be failing to find any usable font, but I'll try to look into it.

I notice that the crash data shows plenty of older crashes in gfxFontGroup::GetDefaultFont(), too, so this probably isn't a new bug introduced by bug 2024049, more like just a new code path that's reaching a pre-existing issue.

Comment 5

[vlopezgarcia]

Marking as dupe, my bad!