Releases: hyprwm/aquamarine
v0.4.0
A big update with a lot of fixes and patches
This update is not ABI-compatible with 0.3.3
Fixes
- drm: fix crash on null crtc in setCursor
- gbm: improve trace logging around allocs
- drm: reconfigure swapchain props after creating
- gbm: fixup good() retval
- drm: verify output format in commit
- wayland: fix possible out-of-range crash in setCursor
- gbm: improve automatic format selection
- output: always emit a destroy event in ~dtor
- gbm/drm: conform to different scanout requirements for different planes
- wayland: send 1280x720 if w/h is 0 in configure
MRs
- output avoid crashing on no crtc by @gulafaran in #58
- tearing fix by @Colt-M16A1 in #60
- drm: add missing destructor by @gulafaran in #63
- fix typo by @romanstingler in #62
- backends dont overflow on 32bit builds by @gulafaran in #64
- recheck crtcs before connecting on udev event by @ikalco in #65
- make for loops const reference by @gulafaran in #66
- don't connect before crtcs are guaranteed by @ikalco in #68
- gbm: override gbm_bo_get_modifier results for modifier-based allocation by @moetayuko in #71
- allow disabling commit, even if pageFlipPending by @ikalco in #72
New Contributors
- @Colt-M16A1 made their first contribution in #60
- @romanstingler made their first contribution in #62
Full Changelog: v0.3.3...v0.4.0
v0.3.3
Small patch to fix crashes on select drivers on 0.3.2
This release is ABI-compatible with 0.3.2
Fixes
- gbm: guard for drm renderer being present before clearing
MRs
New Contributors
Full Changelog: v0.3.2...v0.3.3
v0.3.2
A small update with quite a few fixes.
This release is ABI-compatible with 0.3.1.
Fixes
- drm/atomic: clip damage to pixel size
- drm: avoid crashes on connecting a null crtc
- drm: don't report pointer capability with no HW cursor planes
- drm: use modelines in vt restore
MRs
- cmake: link libOpenGL instead of legacy libGL by @ziyao233 in #44
- buffer align size to stride by @gulafaran in #50
- drm: clear buffer on creation for scanout DRM buffers by @vaxerski in #51
- drm: don't schedule new frame events on disabled outputs by @vaxerski in #52
New Contributors
Full Changelog: v0.3.1...v0.3.2
v0.3.1
A minor update fixing some issues with custom modelines in DRM.
This update is ABI-compatible with 0.3.0.
Fixes
- drm: fix custom mode check
- drm: fixup modeline calculation
MRs
Full Changelog: v0.3.0...v0.3.1
v0.3.0
An update bringing various improvements to the underlying explicit sync pipeline in the DRM backend alongside some minor fixes.
Fixes
- drm: don't rollback cursors on test
- drm: immediately send presentation events for tearing
- drm: avoid testing with a pf event
- drm: support explicit sync with multi-gpu destintations
- drm: add
AQ_MGPU_NO_EXPLICIT
- drm: don't use explicit in blit without explicit requested from user
MRs
New Contributors
Full Changelog: v0.2.0...v0.3.0
v0.2.0
An update with quite a few fixes, which also breaks ABI (but not API)
sover has been bumped to 1.
Fixes
- drm: avoid a crash when no cursor plane is present
- drm: reject modeless commits
- wayland: drop required dmabuf version to 4
- session/input: add missing pos to touch down events
- gbm: use flags in create_with_modifiers
- gbm: log scanout flag state in trace
- drm: conform to both renderable and scanoutable formats
MRs
- Remove redundant needsFrame & allow frame schedule skipping on mouse move by @UjinT34 in #21
- Nvidia fixes for scanout gbm allocation by @vaxerski in #27
- displaylink fix: ignore primary argument when using evdi drivers by @nktnet1 in #25
- gbm: manually set modifier for implicit bo allocation by @moetayuko in #30
- fix destroying headless outputs by @ikalco in #32
- Follow symlinks for AQ_DRM_DEVICES by @sjcobb2022 in #34
New Contributors
- @vaxerski made their first contribution in #27
- @nktnet1 made their first contribution in #25
- @moetayuko made their first contribution in #30
- @sjcobb2022 made their first contribution in #34
Full Changelog: v0.1.0...v0.2.0
v0.1.1
A minor update fixing some crashes on the DRM backend.
This release is ABI-compatible with 0.1.0.
Fixes
- drm: reject modeless commits
- drm: avoid a crash when no cursor plane is present
Full Changelog: v0.1.0...v0.1.1
v0.1.0
Initial release of aquamarine.
MRs
- formatutils: ensure we dont leak name ptr by @gulafaran in #2
- add missing virtual destructor to base class IAllocator by @gulafaran in #3
- Unbreak build on FreeBSD by @jbeich in #6
- Add scheduleFrame reasons by @UjinT34 in #7
- CMakeLists.txt: require native hyprwayland-scanner by @MarkusVolk in #8
- Fix cursor swapchain buffer creation by @UjinT34 in #9
- pass explicitOutFence pointer and do not touch the value by @UjinT34 in #10
- buffer: add a NONE flag to eBufferCapability by @gulafaran in #11
- Expose bo buffer mapping by @UjinT34 in #12
- add getting monitor render formats by @ikalco in #13
- handle monitor hotplugs and fix a few bugs that it revealed by @ikalco in #14
- Add missing include to unbreak build on FreeBSD by @jbeich in #15
- Typos in reopenDRMNode by @jbeich in #17
- use deviceID instead of fd for libseat_close_device by @ikalco in #18
- Add title/class to wayland window by @phonetic112 in #19
- drm: fix crash for connectors without a fallback mode by @JManch in #20
New Contributors
- @gulafaran made their first contribution in #2
- @jbeich made their first contribution in #6
- @UjinT34 made their first contribution in #7
- @MarkusVolk made their first contribution in #8
- @ikalco made their first contribution in #13
- @phonetic112 made their first contribution in #19
- @JManch made their first contribution in #20
Full Changelog: https://github.com/hyprwm/aquamarine/commits/v0.1.0