lwsync() is a full read/write-barrier. That's all we need, never
need sync(). I'm not sure why an isync() was used in fetch_and_add,
but since that's a read-modify-write, I just changed it to have
lwsync() on both sides.
Originally, glib's atomic_get was implemented as "memory_barrier; load".
I copied this into cairo, fontconfig, and harfbuzz. However, that's
wrong. Correct way is "load; memory_barrier". The details are long
and hard to fully grasp. Best to read:
https://www.kernel.org/doc/Documentation/memory-barriers.txt
Also see my report against GNOME:
https://gitlab.gnome.org/GNOME/glib/issues/1449
Note that this is irrelevant if C++11-like atomic ops are available.
This reverts commit b1e5650c67.
After lots of head-scratching and finally finding the only truly
readable source to be the good old:
https://www.kernel.org/doc/Documentation/memory-barriers.txt
I've convinced myself that we need consume memory-ordering on get().
The location of memory-barrier in a load should be after, not before
the load. That needs fixing. I'll do that separately.
Oriya went down from 9 to 2.
BENGALI: 353725 out of 354188 tests passed. 463 failed (0.130722%)
DEVANAGARI: 707311 out of 707394 tests passed. 83 failed (0.0117332%)
GUJARATI: 366355 out of 366457 tests passed. 102 failed (0.0278341%)
GURMUKHI: 60729 out of 60747 tests passed. 18 failed (0.0296311%)
KANNADA: 951300 out of 951913 tests passed. 613 failed (0.0643966%)
MALAYALAM: 1048136 out of 1048334 tests passed. 198 failed (0.0188871%)
MYANMAR: 1115830 out of 1123883 tests passed. 8053 failed (0.716534%)
ORIYA: 42327 out of 42329 tests passed. 2 failed (0.00472489%)
SINHALA: 271596 out of 271847 tests passed. 251 failed (0.0923313%)
TAMIL: 1091754 out of 1091754 tests passed. 0 failed (0%)
TELUGU: 970555 out of 970573 tests passed. 18 failed (0.00185457%)
Behdad Esfahbod
Ebrahim Byagowi
Garret Rieger