xf86-video-wsfb: Add support for 16 color mode on Amiga
- Sets up 8bpp shadow framebuffer with depth 4
- Bitplane conversion done in shadow update function
- Adds new shadow update function shadowUpdateAfb4x8,
which is a slightly modified version of existing shadowUpdateAfb4/8.
Discussion on tech-x11: http://mail-index.netbsd.org/tech-x11/2024/02/29/msg002447.html
Pull up following revision(s) (requested by tsutsui in ticket #572):
external/mit/xf86-video-wsfb/dist/src/wsfb_driver.c: revision 1.49
Handle "LSBit is left" bitorder VRAM of smg(4) and qv(4) on VAXen.
Confirmed on smg(4) on my VAXstation 3100/30.
Pull up following revision(s) (requested by jmcneill in ticket #562):
external/mit/xf86-video-wsfb/dist/src/wsfb_driver.c: revision 1.48
external/mit/xf86-video-wsfb/dist/src/wsfb.h: revision 1.11
xf86-video-wsfb: Add support for Nintendo Wii
The Wii's framebuffer is YUY2, so add a custom shadowproc that converts
from RGB565 to YUY2 on the fly.
add support for doubling a font's size, with optional smoothing
- works only on fonts up to 8 pixels wide for now
- smoothing attempts to detect pixel staircases and add pixels to smooth
them out
Pull up following revision(s) (requested by tsutsui in ticket #1766):
external/mit/xf86-input-keyboard/dist/src/ws_KbdMap.c: revision 1.1
external/mit/xf86-input-keyboard/dist/src/bsd_kbd.h: revision 1.5
external/mit/xf86-input-keyboard/dist/src/bsd_kbd.c: revision 1.22
external/mit/xf86-input-keyboard/dist/src/kbd.c: revision 1.9
xf86-input-keyboard: add suppport to get keymap via WSKBDIO_GETMAP ioctl(2).
No particular comment on tech-x11@:
https://mail-index.netbsd.org/tech-x11/2023/10/31/msg002415.htmlhttps://mail-index.netbsd.org/tech-x11/2023/11/04/msg002416.html
Tested on NWS-3260 and NWS-3470 with Xorg 1.10 server in netbsd-9.
Pull up following revision(s) (requested by tsutsui in ticket #466):
xsrc/external/mit/xf86-input-keyboard/dist/src/ws_KbdMap.c: revision 1.1
xsrc/external/mit/xf86-input-keyboard/dist/src/bsd_kbd.h: revision 1.5
xsrc/external/mit/xf86-input-keyboard/dist/src/bsd_kbd.c: revision 1.22
xsrc/external/mit/xf86-input-keyboard/dist/src/kbd.c: revision 1.9
src/external/mit/xorg/server/drivers/xf86-input-keyboard/Makefile: revision 1.23
xf86-input-keyboard: add suppport to get keymap via WSKBDIO_GETMAP ioctl(2).
No particular comment on tech-x11@:
https://mail-index.netbsd.org/tech-x11/2023/10/31/msg002415.htmlhttps://mail-index.netbsd.org/tech-x11/2023/11/04/msg002416.html
Tested on NWS-3260 and NWS-3470 with Xorg 1.10 server in netbsd-9.
Worth to pullup to netbsd-10 and netbsd-9.
xf86-input-keyboard: add build glue to enable USE_WSKBD_GETMAP.
Pull up following revision(s) (requested by jdc in ticket #465):
external/mit/xf86-video-pnozz/dist/src/pnozz_exa.c: revision 1.5
external/mit/xf86-video-pnozz/dist/src/pnozz_exa.c: revision 1.6
Revert r1.4.
The changes are minimal and they cause redraw problems (as reported by
Maxim Devaev on port-sparc).
Add NetBSD keyword.
Pull up the following, requested by mrg in ticket #436:
external/mit/xorg-server/dist/ChangeLog up to 1.1.1.25
external/mit/xorg-server/dist/Makefile.in up to 1.1.1.17
external/mit/xorg-server/dist/aclocal.m4 up to 1.1.1.20
external/mit/xorg-server/dist/config.guess up to 1.1.1.15
external/mit/xorg-server/dist/config.sub up to 1.1.1.15
external/mit/xorg-server/dist/configure up to 1.18
external/mit/xorg-server/dist/configure.ac up to 1.20
external/mit/xorg-server/dist/meson.build up to 1.1.1.15
external/mit/xorg-server/dist/Xext/Makefile.in up to 1.1.1.17
external/mit/xorg-server/dist/Xi/Makefile.in up to 1.1.1.17
external/mit/xorg-server/dist/Xi/xiproperty.c up to 1.7
external/mit/xorg-server/dist/composite/Makefile.in up to 1.1.1.17
external/mit/xorg-server/dist/config/Makefile.in up to 1.1.1.17
external/mit/xorg-server/dist/damageext/Makefile.in up to 1.1.1.17
external/mit/xorg-server/dist/dbe/Makefile.in up to 1.1.1.17
external/mit/xorg-server/dist/dix/Makefile.in up to 1.1.1.18
external/mit/xorg-server/dist/dix/enterleave.h up to 1.1.1.4
[99 lines not shown]
Use a SIGCHLD handler instead of ignoring the signal.
This avoids lockups when child processes were inherited (e.g. from .xsession)
but new children are waited for in system().
