fdo-mirrors/cairo
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/cairo/cairo.git synced 2026-04-24 11:31:43 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

[test/show-glyphs-many] Re-enable test to trigger crash in xlib.

Browse source 
Moral of this story is bugs cluster. If we made a mistake, especially in a
complicated bit of code that is interfacing with another library, then we
are likely to make a similar mistake in future. Disabling this test hid a
regression between 1.4 and 1.6.
This commit is contained in:
Chris Wilson 2008-09-30 13:29:35 +01:00
parent fade54e71a
commit 2a347a92b0
2 changed files with 10 additions and 14 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

15
test/Makefile.am
View file

@ -141,6 +141,7 @@ self-copy$(EXEEXT) \
self-copy-overlap$(EXEEXT) \
self-intersecting$(EXEEXT) \
set-source$(EXEEXT) \
show-glyphs-many$(EXEEXT) \
show-text-current-point$(EXEEXT) \
skew-extreme$(EXEEXT) \
smask$(EXEEXT) \
@ -197,19 +198,6 @@ zero-alpha$(EXEEXT)
# extend-reflect - Triggers an infinite (or near-infinite) bug in some
# X servers in some circumstances. Details and cause unknown.
#
# show-glyphs-many - This stress test was exercising a particular bug
# in the xlib surface code (exceeding the X11 protocol request
# limit) when rendering several thousand glyphs at once. The
# original xlib-surface bug is fixed now, but the test continues
# to stress some other aspects of the test suite. For example,
# when doing text as paths, the resuilting PostScript file is one
# giant path that ghostscript has a particularly hard time
# with. I'm disabling this test for now, since I don't care about
# that performance problem in ghostscript. (But, there is a
# similar performance problem when using cairo to rasterize the
# equivalen giant path---from an SVG files say---so this might be
# a useful kind of test to bring back again for performance (not
# correctness) testing.
# text-glyph-range - This test triggers the following assertion in cairo:
#
# lt-text-glyph-range: cairo-scaled-font-subsets.c:350:
@ -221,7 +209,6 @@ zero-alpha$(EXEEXT)
# that's just a bug in the test rig that should just consider
# the abort an XFAIL like any other.
DISABLED_TESTS = \
show-glyphs-many$(EXEEXT) \
text-glyph-range$(EXEEXT)
# Then we have a collection of tests that are only run if certain

9
test/show-glyphs-many.c
View file

@ -65,6 +65,15 @@
*
* Status: I replicated bug. The largest value of NUM_GLYPHS for
* which I saw success is 21842.
*
* 2008-30-08 Chris Wilson <chris@chris-wilson.co.uk>
* This is also a valid test case for:
*
* Bug 5913 crash on overlong string
* https://bugs.freedesktop.org/show_bug.cgi?id=5913
*
* which is still causing a crash in the Xlib backend - presumably, just
* a miscalculation of the length of the available request.
*/
#define TEXT_SIZE 12