This fixes several bugs, in particular the GL_INVALID_OPERATION when blitting
between 3d textures, so will allow us to enable some tests.
Differential Revision: https://phabricator.services.mozilla.com/D47628
--HG--
extra : moz-landing-system : lando
This layer structure can still occur in cases where we place the RCD-RSF
scroll metadata on the root layer as a fallback.
Differential Revision: https://phabricator.services.mozilla.com/D47655
--HG--
extra : moz-landing-system : lando
Print debugging a task that runs gradle has been really annoying as gradle
reads the output of 'mach environment' and fails as soon as a debug line shows
up.
What's worse, is it redirects stderr into stdout so even printing to
'sys.stderr' fails. This fixes that so writing to stderr will at least work.
Differential Revision: https://phabricator.services.mozilla.com/D47608
--HG--
extra : moz-landing-system : lando
Note, the deferred code paths are unused until we move publication -after- bytecode emission
Differential Revision: https://phabricator.services.mozilla.com/D47659
--HG--
extra : moz-landing-system : lando
Popup window maybe fail in geckoview platform, and closing it makes IOError and crashes the test.
Differential Revision: https://phabricator.services.mozilla.com/D47599
--HG--
extra : moz-landing-system : lando
Other tests in the file explicitly call closeStream. It seems to help with the issue, but this
is a bit guess fix, since I'm not at all familiar with the relevant code. Perhaps the test
should pass without explicit closeStream? If so, there should be a comment why.
Differential Revision: https://phabricator.services.mozilla.com/D47647
--HG--
extra : moz-landing-system : lando
This patch sets the preference order for `TLS_CHACHA20_POLY1305_SHA256` over `TLS_AES_128_GCM_SHA256` for ARM builds.
As noted in the bug, this is far from an ideal way to do this. The implementation is purposefully simplistic so as to minimize any performance hit. If we want to accept doing this configuration for every new TLS connection, `SSL_CipherSuiteOrderGet` **will** return the pref-filtered (i.e. only the enabled) ciphers, but in the default NSS order. We would have to build a new list by referencing this output with another ordered list defined in PSM. If we want to leave NSS as-is (instead of offering a global reconfiguration API), we should do this.
Differential Revision: https://phabricator.services.mozilla.com/D47485
--HG--
extra : rebase_source : 0252cf321225cd644a463fd94561fd6af38b3837
extra : source : 4836c05dd2eee11bf9d836fb0505e77450b0651b
* Updates the frequency cap for the extended triplets feature
* Updates the experiment ID
Differential Revision: https://phabricator.services.mozilla.com/D47653
--HG--
extra : moz-landing-system : lando
Reading data with only one call to `readBuf()` to avoid the while loop reduces
time spent in `getHuffmanLookup()` by 26%.
Differential Revision: https://phabricator.services.mozilla.com/D47095
--HG--
extra : moz-landing-system : lando
Reading new data into `newBits` then doing the per-byte bit inversion in place
and combining with `this->bits` reduces time spent in `getHuffmanLookup()` by
about 15%.
Differential Revision: https://phabricator.services.mozilla.com/D46559
--HG--
extra : moz-landing-system : lando
To ensure that we do not block shutdown while waiting on a low-priority
untrusted modules thread:
* We switch to using `THREAD_PRIORITY_IDLE` and `THREAD_PRIORITY_NORMAL`
instead of `THREAD_MODE_BACKGROUND_BEGIN` and `THREAD_MODE_BACKGROUND_END`.
This is necessary because the latter flags only work when settting priority
on the current thread; they do not permit other threads to alter them.
* We add `BackgroundPriorityRegion::Clear` to normalize the specified thread's
priority.
* We invoke `BackgroundPriorityRegion::Clear` on the untrusted modules thread
as soon as we learn that we will be shutting down, thus boosting the thread's
priority to normal.
Differential Revision: https://phabricator.services.mozilla.com/D47495
--HG--
extra : moz-landing-system : lando
This patch sets the preference order for `TLS_CHACHA20_POLY1305_SHA256` over `TLS_AES_128_GCM_SHA256` for ARM builds.
As noted in the bug, this is far from an ideal way to do this. The implementation is purposefully simplistic so as to minimize any performance hit. If we want to accept doing this configuration for every new TLS connection, `SSL_CipherSuiteOrderGet` **will** return the pref-filtered (i.e. only the enabled) ciphers, but in the default NSS order. We would have to build a new list by referencing this output with another ordered list defined in PSM. If we want to leave NSS as-is (instead of offering a global reconfiguration API), we should do this.
Differential Revision: https://phabricator.services.mozilla.com/D47485
--HG--
extra : moz-landing-system : lando
Most of these tests have been disabled for a long time; they run well
in the current test environment.
Differential Revision: https://phabricator.services.mozilla.com/D47616
--HG--
extra : moz-landing-system : lando