backport WIP queue onto master by metux · Pull Request #2172 · X11Libre/xserver

metux · 2026-04-10T18:26:21Z

Step 0 of upcoming semi-automatic backporting from Xorg's new 'main' branch.

original range: a9ee6b7..48bda9d

This queue probably still needs some more rework. Perhaps we should pick out interesting commits one by one.

stefan11111 · 2026-04-10T18:53:15Z

+                defs[i].blue = prgb[i].blue;
+                defs[i].flags = DoRed | DoGreen | DoBlue;
+            }
+            (*pmap->pScreen->StoreColors) (pmap, entries, defs);


Is the duplicates StoreColors call intended?

no, that's been a rebase artifact.
actually, the whole commit is duplicated, because we already solved that issue few lines above.

stefan11111 · 2026-04-10T18:53:36Z

    Bool async_flip = !sync_flip;

+    if (reason)
+        *reason = PRESENT_FLIP_REASON_UNKNOWN;


Duplicated assignment

thx, that was a rebase fallout. dropping this commit.

stefan11111 · 2026-04-10T18:54:30Z

Are all of these recent backports from Xorg?

I'm planning to backport them soon enough.

stefan11111 · 2026-04-10T18:56:17Z

There are still these 2 remaining from the master->main Xorg branch change:
#2136

metux · 2026-04-13T10:09:05Z

Are all of these recent backports from Xorg?

I believe so.
Note that these are coming from their new "main" branch, instead of "master".
I've started off at what I believe was their latest commit before the big history rewrite and onto-rebase'd everything above onto our maint-21.0, then 21.0, then master.

But feel free to double check, just in case something got lost.

In the next steps (once this queue is finished), my script will start off where where it left this time, so we don't need to start afresh each time (it's remembering the last commit in the tracking branch). In the future, most of it hopefully will run (almost) automatically - quite like we're already doing @Sonicde.

…pyWindow() PR: #2192

PR: #2191

…hHash() PR: #2190

PR: #2187

…ute union PR: #2182

PR: #2178

PR: #2177

PR: #2176

PR: #2175

Re-map the SELinux security classes on policy loads, as the mapping will be desynchronized (see man:selinux_set_mapping(3)) and audit messages will not show the actual class and permission names: USER_AVC pid=24283 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:xorg_t:s0 msg='avc: denied { 0x10 } for request=XFIXES:SelectSelectionInput comm=/usr/bin/python3 resid=6400001 restype=WINDOW scontext=xuser_u:xuser_r:systemd_user_instance_generic_bin_t:s0 tcontext=xuser_u:object_r:xorg_t:s0 tclass=(null) permissive=1 In addition use type-safe assignments. Signed-off-by: Christian Göttsche <cgzones@googlemail.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/801>

Only generate audit events for messages of the type avc (permission denied) and error (e.g. invalid context). For example avoid USER_SELINUX_ERR for policy load events: audit[980]: USER_SELINUX_ERR pid=980 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:xorg_t:s0 msg='avc: op=load_policy lsm=selinux seqno=8 res=1 exe="/usr/lib/xorg/Xorg" sauid=0 hostname=? addr=? terminal=?' Signed-off-by: Christian Göttsche <cgzones@googlemail.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/801>

…ts own Replaces Xwayland's server-specific implementation, but leaves Xorg's since it sets global variables in the xfree86 ddx layer. Also leaves Xephyr's differently-spelled "-verbosity" for backwards compatibility. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/2131>

Signed-off-by: Diego Viola <diego.viola@gmail.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/2134>

Fixes: e3c65cf ("xephyr: Add -title option.") Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/2137>

Fixes: f028e24 ("Bug #10016: Implement WM_CLASS hints in Xephyr.") Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/2137>

Avoids build warnings & failures, such as: ../os/xserver_poll.h:40: warning: "POLLNVAL" redefined 40 | #define POLLNVAL 0x20 | /usr/share/mingw-w64/include/winsock2.h:1190: note: this is the location of the previous definition 1190 | #define POLLNVAL 0x0004 | ../os/xserver_poll.h:42:8: error: redefinition of ‘struct pollfd’ 42 | struct pollfd | ^~~~~~ /usr/share/mingw-w64/include/winsock2.h:1192:16: note: originally defined here 1192 | typedef struct pollfd { | ^~~~~~ Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/2141>

The Render extension's Trapezoids and Triangles operations (fbTrapezoids, fbTriangles) render directly via pixman through fbShapes without going through ps->Composite, so RootlessComposite never fires for these operations. This means RootlessStartDrawing is not called and the macOS compositor is not notified of changed pixels. Add RootlessTrapezoids and RootlessTriangles wrappers following the same pattern as RootlessComposite: call RootlessStartDrawing before the operation, then compute the bounding box via miTrapezoidBounds or miTriangleBounds and call RootlessDamageBox after. The default miTriStrip and miTriFan implementations decompose their input into triangles and dispatch through ps->Triangles, so the RootlessTriangles wrapper covers TriStrip and TriFan as well. Also add a RootlessCompositeRects wrapper as defense-in-depth. The current miCompositeRects implementation already has partial rootless coverage: PictOpSrc/PictOpClear go through the GC layer, and other ops route through CompositePicture -> RootlessComposite. Wrapping at the PictureScreen level ensures robustness regardless of the underlying implementation. Fixes [1/2]: XQuartz/XQuartz#31 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>

The Render extension operates on PictFormats rather than GC, so it bypasses the rootless layer's GC-based alpha protection (ROOTLESS_PROTECT_ALPHA). Depth-24 windows use PICT_x8r8g8b8 where the 'x' tells pixman the high byte is padding it may freely zero. The macOS compositor needs this byte to be 0xFF (opaque). Before each Render operation that targets a rootless window, temporarily upgrade the destination Picture's format from PICT_x8r8g8b8 to PICT_a8r8g8b8. This tells pixman that the alpha channel is significant and must not be optimized away. The format is restored after the operation completes. This parallels how ROOTLESS_PROTECT_ALPHA handles GC operations (by masking alpha out of the planemask) and how ROOTLESS_SAFEALPHA handles PaintWindow (by forcing alpha in solid fills). The save/restore approach is analogous to rootlessGC.c's GC_SAVE / GC_RESTORE pattern. Fixes [2/2]: XQuartz/XQuartz#31 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>

Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>

Reported in #1817: xwayland-24.1.6/redhat-linux-build/../os/utils.c:1108:9: warning[-Wanalyzer-possible-null-dereference]: dereference of possibly-NULL ‘result’ xwayland-24.1.6/redhat-linux-build/../os/utils.c:1108:9: danger: ‘malloc((long unsigned int)len + 18)’ could be NULL: unchecked value from (3) Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/2163>

Reported in #1817: xwayland-24.1.6/redhat-linux-build/../present/present_notify.c:83:17: warning[-Wanalyzer-malloc-leak]: leak of ‘notifies’ xwayland-24.1.6/redhat-linux-build/../present/present_notify.c:83:17: branch_false: following ‘false’ branch (when ‘i >= num_notifies’)... Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/2164>

metux requested a review from a team April 10, 2026 18:26

metux marked this pull request as draft April 10, 2026 18:26

stefan11111 reviewed Apr 10, 2026

View reviewed changes

metux force-pushed the rfc/backport-master branch 2 times, most recently from 234c862 to c28a7df Compare April 13, 2026 10:25

alanc and others added 21 commits April 13, 2026 12:40

[PR #2192] fb: quiet -Wanalyzer-out-of-bounds warnings in fbOverlayCo…

04d9972

…pyWindow() PR: #2192

[PR #2191] tests: plug leak of results in compute_expected_damage()

6530d1a

PR: #2191

[PR #2190] render: handle -Wanalyzer-null-dereference in AllocateGlyp…

36d24b6

…hHash() PR: #2190

[PR #2187] config: Fix compiler warning

aacdd73

PR: #2187

[PR #2182] rootless: Fix Glyphs damage bounding box to correctly comp…

dc000f4

…ute union PR: #2182

[PR #2178] glx: avoid null dereference in validGlxFBConfigForWindow()

80bfd26

PR: #2178

[PR #2177] dri2: little spelling fix

fc6225d

PR: #2177

[PR #2176] xv: change FOURCC_RGBA32 to AMD one

a7931e2

PR: #2176

[PR #2175] xfree86: remove leftover ev56.c source files

30c4293

PR: #2175

Fix typos

d5bcf62

Signed-off-by: Diego Viola <diego.viola@gmail.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/2134>

ephyr: add -title to Xephyr man page

f9e388b

Fixes: e3c65cf ("xephyr: Add -title option.") Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/2137>

ephyr: add -name to Xephyr man page

ce2e9e1

Fixes: f028e24 ("Bug #10016: Implement WM_CLASS hints in Xephyr.") Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/2137>

xquartz: Bump copyrights in Info.plist to 2026

5a6dcdf

Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>

metux force-pushed the rfc/backport-master branch from c28a7df to cbe66c7 Compare April 13, 2026 10:44

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

backport WIP queue onto master#2172

backport WIP queue onto master#2172
metux wants to merge 21 commits intomasterfrom
rfc/backport-master

metux commented Apr 10, 2026

Uh oh!

stefan11111 Apr 10, 2026

Uh oh!

metux Apr 13, 2026

Uh oh!

stefan11111 Apr 10, 2026

Uh oh!

metux Apr 13, 2026

Uh oh!

stefan11111 commented Apr 10, 2026

Uh oh!

stefan11111 commented Apr 10, 2026

Uh oh!

metux commented Apr 13, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

Conversation

metux commented Apr 10, 2026

Uh oh!

stefan11111 Apr 10, 2026

Choose a reason for hiding this comment

Uh oh!

metux Apr 13, 2026

Choose a reason for hiding this comment

Uh oh!

stefan11111 Apr 10, 2026

Choose a reason for hiding this comment

Uh oh!

metux Apr 13, 2026

Choose a reason for hiding this comment

Uh oh!

stefan11111 commented Apr 10, 2026

Uh oh!

stefan11111 commented Apr 10, 2026

Uh oh!

metux commented Apr 13, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants