Status

()

task
P3
normal
RESOLVED FIXED
3 months ago
2 months ago

People

(Reporter: RyanVM, Assigned: RyanVM)

Tracking

unspecified
mozilla69
Points:
---
Dependency tree / graph

Firefox Tracking Flags

(firefox69 fixed)

Details

(Whiteboard: [third-party-lib-audit][gfx-noted])

Attachments

(1 attachment)

Overview of changes leading to 2.5.0
Friday, May 24, 2019

  • This release does not include much functional changes, but includes major internal
    code-base changes. We now require C++11. Support for gcc 4.8 and earlier has been
    dropped.
  • New hb-config.hh facility for compiling smaller library for embedded and web usecases.
  • New Unicode Character Databse implementation that is half the size of previously-used
    UCDN.
  • Subsetter improvements.
  • Improved documentation, thanks to Nathan Willis.
  • Misc shaping fixes.

Hrm, I'm getting local build failures at the moment with a recent clang:

hb-atomic.hh:258:78: error: empty expression statement has no effect; remove unnecessary ';' to silence this warning [-Werror,-Wextra-semi-stmt]
inline void hb_atomic_int_impl_set (int *AI, int v) { _hb_memory_w_barrier (); *AI = v; }

hb-atomic.hh:261:89: error: empty expression statement has no effect; remove unnecessary ';' to silence this warning [-Werror,-Wextra-semi-stmt]
inline int hb_atomic_int_impl_get (const int *AI) { int v = *AI; _hb_memory_r_barrier (); return v; }

hb-atomic.hh:264:94: error: empty expression statement has no effect; remove unnecessary ';' to silence this warning [-Werror,-Wextra-semi-stmt]
inline void *hb_atomic_ptr_impl_get (void ** const P) { void *v = *P; _hb_memory_r_barrier (); return v; }

I'm fixing those.

Also, you might want to add -DHB_NO_UCD to your build config if Firefox provides its own unicode data.

I added the hb-atomic.hh change but am still hitting the same build errors as before :(. This is with clang 8.0.0 if it helps.

Bah, of course we were on the wrong side of that ifdef for the first fix to work :-). Confirmed that this builds locally now!

Behdad, Coverity noticed a couple possible issues in the patch too, but also mentioning that there was risk of them being false positives:
https://phabricator.services.mozilla.com/D32776#968750

Hard to know what it means there exactly, but I wouldn't worry about it. We have oss-fuzz on this codebase and it catches anything less than perfectly-right in a couple hours...

Overview of changes leading to 2.5.1
Friday, May 31, 2019

  • Fix build with various versions of Visual Studio.
  • Improved documentation, thanks to Nathan Willis.
  • Bugfix in subsetting glyf table.
  • Improved scripts for cross-compiling for Windows using mingw.
  • Rename HB_MATH_GLYPH_PART_FLAG_EXTENDER to HB_OT_MATH_GLYPH_PART_FLAG_EXTENDER.
    A deprecated macro is added for backwards-compatibility.

Try push:
https://treeherder.mozilla.org/#/jobs?repo=try&revision=c19429212b3f0d2b4d892d80d0e91a97867617c3

Summary: Update HarfBuzz to 2.5.0 → Update HarfBuzz to 2.5.1
Attachment #9067906 - Attachment description: Bug 1554306 - Update HarfBuzz to version 2.5.0. r=jfkthame → Bug 1554306 - Update HarfBuzz to version 2.5.1. r=jfkthame
Pushed by rvandermeulen@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/038735236230
Update HarfBuzz to version 2.5.1. r=jfkthame
Status: NEW → RESOLVED
Closed: 3 months ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla69
Blocks: 1560439
You need to log in before you can comment on or make changes to this bug.