Mir

mir:cpu-sample-from-gl-texture

Last commit made on 2020-11-24
Get this branch:
git clone -b cpu-sample-from-gl-texture https://git.launchpad.net/mir

Branch merges

Branch information

Name:
cpu-sample-from-gl-texture
Repository:
lp:mir

Recent commits

df304cf... by Chris Halse Rogers

graphics: Support* CPU sampling from mg::Texture

*: With the notable caveat that it only works with a current EGL
context.

Fixes: #1811

d1ea950... by "bors[bot]" <26634292+bors[bot]@users.noreply.github.com>

Merge #1812

1812: Use standard key repeat values r=AlanGriffiths a=wmww

This solves a minor annoyance I've had for a while. Both Weston and Sway use these values by default. GNOME-on-Wayland, on the other hand uses 33 and 500, which I would also be open to. Mostly, the delay just needs to go up.

Of course this will eventually need be configurable for accessability reasons and general preference, but building an input configuration system is a job for another day,

Co-authored-by: William Wold <email address hidden>

0941ae2... by "bors[bot]" <26634292+bors[bot]@users.noreply.github.com>

Merge #1796

1796: Release 2.2 r=RAOF a=wmww

Co-authored-by: William Wold <email address hidden>
Co-authored-by: Alan Griffiths <email address hidden>
Co-authored-by: bors[bot] <26634292+bors[bot]@users.noreply.github.com>

78b98b7... by "bors[bot]" <26634292+bors[bot]@users.noreply.github.com>

Merge #1816

1816: Fix wlcs integration r=wmww a=AlanGriffiths

As highlighted by #1815 this is broken

Co-authored-by: Alan Griffiths <email address hidden>

bc83e06... by Alan Griffiths

Deduplicate wlcs related logic

b4e6b46... by Alan Griffiths

Move test_wlcs_display_server.* to the wlcs integration module

bb5c137... by William Wold <email address hidden>

Use key repeat values from Weston and Sway

56624c1... by "bors[bot]" <26634292+bors[bot]@users.noreply.github.com>

Merge #1809

1809: gbm-kms: Disable bypass by default. r=AlanGriffiths a=RAOF

This triggers a GPU OOM (issue #1801) on the Pi3 when used with default settings.
There may be due to us hanging on to buffers longer than they need to be, but
it's not clear yet.

Users who want the performance boost and are running on known hardware can
set `--bypass=true`, but by default use the safe option of `false`.

Co-authored-by: Christopher James Halse Rogers <email address hidden>

b52c0aa... by "bors[bot]" <26634292+bors[bot]@users.noreply.github.com>

Merge #1809

1809: gbm-kms: Disable bypass by default. r=AlanGriffiths a=RAOF

This triggers a GPU OOM (issue #1801) on the Pi3 when used with default settings.
There may be due to us hanging on to buffers longer than they need to be, but
it's not clear yet.

Users who want the performance boost and are running on known hardware can
set `--bypass=true`, but by default use the safe option of `false`.

Co-authored-by: Christopher James Halse Rogers <email address hidden>

8d7bea9... by Chris Halse Rogers

gbm-kms: Disable bypass by default.

This triggers a GPU OOM (issue #1801) on the Pi3 when used with default settings.
There may be due to us hanging on to buffers longer than they need to be, but
it's not clear yet.

Users who want the performance boost and are running on known hardware can
set `--bypass=true`, but by default use the safe option of `false`.