Re: [PATCH] Fix a 'memory leak' in the image compositor

On Sun, Mar 11, 2018 at 10:18:47AM +0100, Uli Schlachter wrote:
> Hi Bryce,
> 
> On 10.02.2018 20:08, Uli Schlachter wrote:
> > On 19.01.2018 00:54, Bryce Harrington wrote:
> >> On Sat, Jan 13, 2018 at 02:49:52PM +0100, Uli Schlachter wrote:
> >>> There is a global pixman_glyph_cache_t instance that is initialized on
> >>> first use and shows up in valgrind output as a relatively large leak (I
> >>> think it was about 200 KiB). The reason for this is that this cache is
> >>> not freed by cairo_debug_reset_static_data().
> >>>
> >>> This commit wires up freeing the cache to
> >>> cairo_debug_reset_static_data().
> >>>
> >>> This cache was introduced in commit 615205cf0729 from 2012.
> >>>
> >>> Signed-off-by: Uli Schlachter <[email protected]>
> >>
> >> Offhand looks good to me.  Have you run this against the test suite?
> > 
> > Not really, no. I think I only ran this against one specific test to see
> > if valgrind is content. Any reason that you expect this to go wrong?
> 
> Okay, I ran this against the test suite with CAIRO_TEST_TARGET=image.
> 
> Results with and without this patch on latest master:
> 
> 491 Passed, 60 Failed [0 crashed, 14 expected], 27 Skipped
> Preamble: 1 failed - font-variations
> image (argb32): 43 failed - clip-operator clip-text coverage-rhombus
> extended-blend-alpha extended-blend-mask extended-blend-alpha-mask
> extended-blend-solid-alpha fallback halo halo-transform operator-clear
> operator-source radial-gradient radial-gradient-source
> radial-gradient-mask-source record-select-font-face
> record-text-transform record1414x-select-font-face
> record1414x-text-transform record2x-select-font-face
> record2x-text-transform record90-select-font-face
> record90-text-transform recordflip-whole-select-font-face
> recordflip-whole-text-transform recordflip-select-font-face
> recordflip-text-transform smask smask-text text-antialias-subpixel
> text-antialias-subpixel-rgb text-antialias-subpixel-bgr
> text-antialias-subpixel-vrgb text-antialias-subpixel-vbgr text-pattern
> text-unhinted-metrics unbounded-operator user-font-proxy
> user-font-rescale pthread-same-source pthread-show-text
> ft-text-vertical-layout-type1 ft-text-antialias-none
> image (rgb24): 45 failed - clip-operator clip-text coverage-rhombus
> extended-blend extended-blend-alpha extended-blend-mask
> extended-blend-alpha-mask extended-blend-solid
> extended-blend-solid-alpha fallback halo halo-transform operator-clear
> operator-source radial-gradient radial-gradient-source
> radial-gradient-mask-source record-select-font-face
> record-text-transform record1414x-select-font-face
> record1414x-text-transform record2x-select-font-face
> record2x-text-transform record90-select-font-face
> record90-text-transform recordflip-whole-select-font-face
> recordflip-whole-text-transform recordflip-select-font-face
> recordflip-text-transform smask smask-text text-antialias-subpixel
> text-antialias-subpixel-rgb text-antialias-subpixel-bgr
> text-antialias-subpixel-vrgb text-antialias-subpixel-vbgr text-pattern
> text-unhinted-metrics unbounded-operator user-font-proxy
> user-font-rescale pthread-same-source pthread-show-text
> ft-text-vertical-layout-type1 ft-text-antialias-none
> 
> Is this enough or do you want something more?

Thanks Uli, yes lets go ahead and land this (let me know if you'd prefer
I land it).  I didn't have any specific concerns just figured if there
were any issues the testsuite would (or at least should?) catch them.

Bryce
-- 
cairo mailing list
[email protected]
https://lists.cairographics.org/mailman/listinfo/cairo