2950f9e682
- Move IRQ functions from drm_dma.h to new drm_irq.h and disentangle them
...
from __HAVE_DMA. This will be useful for adding vblank sync support to
sis and tdfx. Rename dma_service to irq_handler, which is more
accurately what it is.
- Fix the #if _HAVE_DMA_IRQ in radeon, r128, mga, i810, i830, gamma to have
the right number of underscores. This may have been a problem in the
case that the server died without doing its DRM_IOCTL_CONTROL to
uninit.
2003-10-17 05:13:48 +00:00
ff58476011
- Converted Linux drivers to initialize DRM instances based on PCI IDs, not
...
just a single instance. Moved the PCI ID lists from <card>_drv.c in BSD
to <card>.h. The PCI ID lists include a driver private field, which may
be used by drivers for chip family or other information. Based on work
by jonsmirl.
- Make tdfx_drv.c and tdfx.h match other drivers.
- Fixed up linking of sis shared files.
Tested with Radeon and SiS on Linux and FreeBSD, including a Linux setup
with
2 SiS cards in a machine, but only one head being used (with DRI)
2003-10-17 03:14:39 +00:00
355b204de0
Introduce COMMIT_RING() as in radeon DRM, stop using error prone writeback
...
for ring read pointer (Paul Mackerras)
Get rid of some superfluous stuff, minor fixes
2003-10-16 14:18:52 +00:00
a64dab1323
Try that again. It's a long.
2003-10-16 03:20:03 +00:00
a9e1a57d6d
Debug printf format fix.
2003-10-16 03:19:06 +00:00
85c16d962d
resolve merge conflicts
2003-09-12 14:24:17 +00:00
4d6d357202
Correct format in debug printfs (free is a pointer, not an int).
2003-09-07 23:56:20 +00:00
fabc64dd57
Port the SiS DRM to FreeBSD. This includes the ability for the DRM to
...
allocate framebuffer memory without sisfb, and a new ioctl to be used
by the X Server which tells the DRM what region of framebuffer memory
to allocate from. Also fixes a possibility to panic the kernel I
believe. Tested on linux with sisfb and FreeBSD (without sisfb) with
new DRI only.
2003-08-29 19:24:36 +00:00
062751ac47
Remove artificial PCI GART limitations, rename AGP to GART where
...
appropriate
2003-08-26 15:44:01 +00:00
07a9b30082
Clean up Radeon DRI resume code
2003-08-18 23:46:19 +00:00
0f094c33da
Make r128_do_wait_for_idle static, as it's only used in this file.
...
Noticed by: CScout
2003-08-18 23:42:16 +00:00
6298d1a1e0
Remove an unnecessary #define __NO_VERSION__
...
Noticed by: CScout
2003-08-18 23:41:05 +00:00
03e6674c13
Whitespace cleanup from the pageflipping commit.
2003-08-12 21:48:16 +00:00
c99acb597f
Added some information as to when (which DRM version) various queries were
...
added.
2003-08-08 21:06:44 +00:00
4b60cae90e
IRQ code cleanup suggested by Linus Torvalds
...
i830 build fix
2003-07-29 10:11:48 +00:00
c26ffeafca
Degrade uninformative error message to debug message, as in other drivers
2003-07-26 15:59:09 +00:00
aaf2105be9
Add Rage 128 pageflipping support, defaults to off. DRM version bump to
...
2.5.0. It still has some issues, including a flicker in the fps meter
in tuxracer and I've seen garbage left behind after moving/closing
windows. However, it's usable. Add the Option "EnablePageFlip" "YES" to
use it.
2003-07-26 03:25:40 +00:00
bef7017749
Compile fixes for recent 2.5/2.6 Linux kernels. I hope this doesn't break
...
the i830 driver or the BSDs. :)
2003-07-25 10:50:39 +00:00
9e7d6177d1
Possibly fix stanford checker complaints about sarea
2003-06-16 10:40:52 +00:00
0b01c70d59
Texture rectangle support for r100
2003-06-10 18:54:17 +00:00
98840144b1
Revert bogus last commit
2003-06-09 23:12:33 +00:00
1062b9930f
Don't activate blend fallbacks unless blending is enabled
2003-06-09 23:11:23 +00:00
518db771de
Restore change to _DRM_VBLANK_SIGNAL from rev. 1.41, which was reverted
...
with the documentation merge.
2003-05-28 02:03:37 +00:00
d2443b2186
Merged DRM documentation.
2003-05-27 00:37:33 +00:00
aeb4bc3f5b
Add support to r128 for MESA_ycbcr_texture (Ian Romanick, Leif Delgass)
2003-05-26 20:04:53 +00:00
c0efa1a777
DRM part of Radeon DRI suspend/resume support (Charl Botha).
2003-05-20 22:43:39 +00:00
2134577e31
Support for building DRM module on OpenBSD (Wilbern Cobb, Matthieu Herrb).
...
(resync with XFree86 trunk)
2003-05-20 22:42:24 +00:00
e5d3c7f260
Support AGP bridges where the AGP aperture can't be accessed directly by
...
the CPU (David Mosberger, Benjamin Herrenschmidt, myself, Paul
Mackerras, Jeff Wiedemeier)
2003-05-16 23:41:27 +00:00
1d5bf7a7de
fix warning on machines where sizeof(drm_addr_t) != 4 (Randy Dunlap)
2003-05-06 21:10:33 +00:00
285b1cdc39
Merged texmem-0-0-1
2003-04-30 01:51:00 +00:00
766a1da2e5
Remove the map argument from DRM_*MEMORYBARRIER. Not all of the uses of
...
DRM_*MEMORYBARRIER we had were related to an MMIO space. This means
arch-specific code on the BSDs, unfortunately. Also add
DRM_MEMORYBARRIER() and change the DRM_READMEMORYBARRIER()s that used
to be read/write barriers to it.
2003-04-26 23:32:00 +00:00
f2a0c5438d
Ensure driver has been initialized (dev_private != NULL) before installing
...
irq handler in DRM(irq_install). Modify all drivers to ensure irq
handler is removed before cleanup and cleanup is called at takedown.
Remove unused buffer private struct fields in i810, i830. Check for
lock on init/cleanup in all drivers except i810/i830. The current DDX
for i810 and i830 doesn't hold the lock on kernel init (FIXME?).
2003-04-26 22:28:56 +00:00
d5db1144dd
get rid of superfluous fields in struct drm_radeon_ring_buffer
...
use correct address for ring read pointer writeback (yes, we seem to have
been running with bogus values for the ring read pointer, which
'worked' because the return value of radeon_wait_ring() is never
checked and the ring usually never fills up)
2003-04-22 21:45:06 +00:00
5ee61c18f4
Remove AGP dependency in kernel config for radeon, sis. Remove
...
PCIGART_ENABLED define for radeon, pcigart support now included for any
arch.
2003-04-22 21:30:24 +00:00
13211ad82c
add more get_param queries for embedded project
2003-04-22 09:49:14 +00:00
46e06192a8
Check for NULL map before calling DRM(ioremapfree) on cleanup. Prevents an
...
oops if a map wasn't found (e.g. XFree86 Bugzilla #108 )
2003-04-21 16:07:17 +00:00
e27d2f8c7c
buf->filp is a pointer, so make printf format args be %p not %d
2003-03-29 03:30:21 +00:00
1728bc637d
merged drm-filp-0-1-branch
2003-03-28 14:27:37 +00:00
c14006ba9f
XFree86 4.3.0 merge
2003-03-25 00:29:14 +00:00
51e5f73d2a
DRM_FREE/2 patch from Philip Brown
2003-03-04 11:41:12 +00:00
cfa778af9c
Merge from bsd-4-0-0-branch.
2003-02-21 23:23:09 +00:00
fac2ed4d10
fix EAGAIN handling in radeon_cp_dispatch_texture() (fixes corruption of
...
large textures), and get rid of superfluous local y variable
2003-02-06 18:20:00 +00:00
73bf29a6c1
fix PCI and AGP posting problems (based on testing by Chris Ison and
...
suggestions by Benjamin Herrenschmidt and Arjan van de Ven)
remove radeon_flush_write_combine() which has been unused for a while
2003-02-04 19:20:18 +00:00
f13af50838
only acknowledge interrupts we handle - others could be used outside the
...
DRM
2003-02-04 15:56:37 +00:00
f3751850c8
Fix size of VERTEX2 ioctl struct (Egbert Eich)
2003-02-03 14:30:32 +00:00
9b9b099471
Add cast to avoid void * arithmetic warning
2003-01-26 22:25:35 +00:00
77ea378b09
Fix radeon for BSD.
2003-01-16 06:20:44 +00:00
a1fc6af2d8
Note that radeon_do_cp_idle() can fail, cope with it.
2003-01-10 17:04:21 +00:00
4b3051d3f6
[TRIVIAL] [TRIVIAL PATCH 2.5.48] Remove unused function from radeon_mem.c
...
Submitted by Rusty Trivial Russell
2003-01-06 07:38:25 +00:00
320802ebc1
more cleanups, free mem heap data on last client exit
2003-01-03 11:22:23 +00:00
14e831baf4
Remove printk
2003-01-02 18:41:02 +00:00
208c0779b6
Make the radeon drm module better at cleaning up after itself if all the
...
clients (particularly the X server) exit without doing so for it.
2003-01-02 18:38:07 +00:00
05f761fa4f
enums should be ints
2002-12-12 22:20:30 +00:00
b03fa556b2
Rewrite radeon_cp_dispatch_texture() to avoid pingponging back to userspace
...
when issue large (multi-buffer) uploads.
2002-12-06 12:22:43 +00:00
85025d4f2a
further vertical blank interrupt cleanups: remove unused variable,
...
non-ambiguous variable names, don't express subtraction in
unnecessarily complicated ways
2002-12-04 15:39:53 +00:00
4acba63bb7
vertical blank interrupt cleanups: use spinlock instead of semaphore, send
...
signal directly from interrupt handler instead of using a taskqueue
(based on feedback by Linus Torvalds)
2002-12-03 00:43:47 +00:00
40891ac190
vertical blank ioctl can send signal instead of blocking
2002-11-30 14:24:07 +00:00
a654424ca1
Silence the radeon_freelist_get 'returning NULL' message. It's not an error
...
& just confuses people.
2002-11-25 09:34:47 +00:00
e656655a36
Kernel support for vblank syncing on Rage 128 and Matrox.
2002-10-30 06:10:34 +00:00
344c7f6b41
updated e-mail addresses for Keith, Alan and Jens
2002-10-29 20:29:05 +00:00
5e1b8ed88a
preserve CRTC{,2}_OFFSET_CNTL in 2D driver to avoid bad effects when
...
pageflipping after a mode switch
take current page into account in AdjustFrame(); writing the CRTC offset
via the CP was probably a bad idea as this can happen asynchronously,
reverted
take frame offset into account when flipping pages
handle CRTC2 as well for pageflipping (untested)
preserve GEN_INT_CNTL on mode switches to prevent interrupts from getting
disabled
2002-10-29 13:49:26 +00:00
ff25e7016c
merge from mesa-4-1-branch to get cube-map registers. bumped version to 1.7
2002-10-28 19:05:40 +00:00
516392beff
s/udelay/DRM_UDELAY/
2002-10-27 05:25:50 +00:00
3d69766ea6
Jonny Strom's mga_dma.c patch
2002-10-10 08:25:40 +00:00
381e0fe7f8
Free correct block in free_block (K. Rasche)
2002-10-02 07:55:27 +00:00
8c14585fc3
fix wait condition for vertical blank IRQs
2002-10-01 17:31:20 +00:00
899df4f530
Turn boxes off again
2002-09-29 21:22:01 +00:00
9243e642da
Move os-dependent stuff out of radeon_irq.c
2002-09-29 21:19:01 +00:00
e6901cad69
Fix up BSD irq handling.
2002-09-29 20:48:18 +00:00
a33d42e2ba
make SW interrupts more robust: write sequence number to scratch register,
...
acknowledge any lost interrupts before waiting
2002-09-27 21:47:52 +00:00
cc9a169d08
DRM(vblank_wait) is driver specific
2002-09-26 12:49:18 +00:00
97961e2c65
BSD vblank framework.
2002-09-26 07:45:07 +00:00
f40674ea9f
change RADEON_PARAM_IRQ_ACTIVE to RADEON_PARAM_IRQ_NR
2002-09-25 19:48:51 +00:00
55acd0d5a6
common ioctl to wait for vertical blank IRQs
2002-09-25 17:18:19 +00:00
f1c8fe9557
merged r200-0-2-branch to trunk
2002-09-23 17:26:43 +00:00
4fcde1efc1
standardize use of __FUNCTION__ (Linus)
2002-08-29 07:34:49 +00:00
22c1ca1fd5
Turn boxes off.
2002-08-28 08:47:39 +00:00
48cc350e21
merged r200-0-1-branch
2002-08-26 22:16:18 +00:00
3165128fc4
TRUE and FALSE don't seem to be defined everywhere...
2002-08-21 00:09:32 +00:00
33d57137da
zero-cliprect case patch from Jacek
2002-08-12 07:26:00 +00:00
d0ac4e5ad0
test scratch register writeback before using it
2002-08-11 15:56:44 +00:00
977b420d5d
Updates from Rusty Russell to:
...
1) Remove redundant header inclusion
2) Silence bitop on non-long argument warnings (change int to long)
3) Move to ISO C (gcc 2.6) initializers (accepted by older gccs also) All
of these are syntax changes that should not impact functionality.
2002-08-06 18:00:57 +00:00
881a9b214d
fix off-by-one error for right bottom corner in radeon_emit_clip_rect()
...
(Jacek Rosik)
2002-07-18 23:17:13 +00:00
8fa8db126a
Workaround for Radeon lockups on fast machines
2002-07-17 08:30:36 +00:00
fd86ac9561
Don't read scratch registers directly, obtain the values via the GET_PARAM
...
ioctl. The DRM reads them from memory addresses the chip writes to on
updates. Fall back to reading the registers directly with an old DRM.
(Tim Smith, cleanups by myself)
2002-07-11 20:31:12 +00:00
74ef13fd00
merged bsd-3-0-0-branch
2002-07-05 08:31:11 +00:00
2dcada361d
merged tcl-0-0-branch
2002-06-12 15:50:28 +00:00
3903e5ac94
Merged drmcommand-0-0-1
2002-04-09 21:54:56 +00:00
ab87c5d0d1
First pass of mesa-4-0 branch merge into trunk.
2002-02-14 02:00:26 +00:00
44aa4d6297
First pass merge of XFree86 4.2.0 import.
2002-01-27 20:05:42 +00:00
4eafeec960
Add module version name at a lower layer of the code, allows things to be
...
more flexible and allow older module versions to still work. Might not
be final code if we move over to DRI kernel module version, but will be
if we use XFree86 version. + Only build tdfx.o instead of tdfx-4_1_0.o
since tdfx.o has never changed its API.
2001-07-18 21:02:58 +00:00
d5749d7603
Added version string to the end of the kernel module name. This allows
...
multiple versions of the kernel module to co-exist on one machine.
2001-07-16 22:42:36 +00:00
7ccc528911
Fix typo in XFREE86_VERSION().
2001-07-13 01:42:48 +00:00
d87c873df0
First pass of 4.1.0 merge.
2001-06-14 22:23:44 +00:00
18ce40c9d3
Make the SiS module work again. At least glxinfo reports it's working, yet
...
trying to run the simple apps return 'out of video memory'
2001-05-17 15:20:40 +00:00
94c6d951a2
Make SiS driver compile with the new templated format. Not tested. minor
...
cleanups
2001-05-03 14:51:01 +00:00
5d6ddbca26
Merged ati-pcigart-1-0-0
2001-04-05 22:16:12 +00:00
301a1ad315
Move #include <linux/config.h> inside the __linux__ #ifdef, so that it
...
doesn't break non-Linux builds.
2001-03-21 20:22:16 +00:00
74e19a4018
Merged sarea-1-0-0
2001-03-14 22:22:50 +00:00
3a74d3a371
Merge tdfx-3-1-0 branch.
2001-03-06 04:37:37 +00:00
88dbee54ed
Implement drmOpen* without /proc dependence (Fallback to /proc is included
...
for backward compatibility.) Move statistic-gathering drm* calls from
dristat.c to xf86drm.c
2001-02-28 09:27:44 +00:00
360475376c
Merge mga-1-0-0-branch into trunk.
2001-02-15 08:12:14 +00:00
0994e635b2
Merged ati-5-0-0
2001-01-05 22:57:55 +00:00
833af23c5d
Sync with Linux 2.4.0-prerelease
2001-01-04 18:41:41 +00:00
c7c6156aec
add blit ioctl, fix plnwt handling
2000-12-30 23:28:53 +00:00
e15a24e154
Merged ati-4-1-1-branch into trunk.
2000-12-02 06:14:18 +00:00
94071289a4
merge with 4.0.1d
2000-11-08 00:07:17 +00:00
52f0bc3c9e
More changes for sync with Linux 2.4.0-test9-pre7
2000-09-29 02:05:41 +00:00
39a659e877
Audit calls to schedule() Remove tags from files shared with Linux kernel
...
tree Remove debugging statements to make debugging more useful Other
minor cleanups in preparation for sync with Linux 2.4.0-test9-pre7
2000-09-29 01:47:11 +00:00
c7558d8fa4
commit xfree86 4.0.1d-pre update
2000-09-24 09:34:10 +00:00
7b888f87fa
applied Jeff's xf86cvs-I810copy.patch
2000-07-13 18:47:48 +00:00
d399dbcd56
Merged bsd-1-0-1
2000-06-13 17:38:09 +00:00
2fbd4bf018
Unify code with kernel: Change some spacing in comments Add #include
...
<linux/config.h> to all *_drv.c files
2000-06-13 14:34:13 +00:00
d5348efe25
Merging the ati-4-1-0 branch onto the trunk.
2000-06-12 15:47:31 +00:00
569da5a42e
Merged glxmisc-3-0-0
2000-06-08 14:38:22 +00:00
fa0a35a25f
Merged bsd-1-0-0
2000-05-30 17:13:31 +00:00
ddcb28259f
Merged mga-0-0-3-branch
...
- New security model for i810
- Enable i810 dri by default
- New indexed vertex path for mga
- Mga kernel driver rework
- Removed dead files in i810 driver
2000-05-25 21:06:02 +00:00
99efe3c247
Merged ati-4-0-1
2000-05-18 06:14:27 +00:00
ba1b1ae380
Merged mga branch with trunk
2000-04-04 22:08:14 +00:00
5bd8014f27
Merge with 4.0
2000-03-16 03:37:30 +00:00
7a9b291ab5
Import of XFree86 3.9.18
2000-02-22 15:43:59 +00:00
1c33c69095
Import of XFree86 3.9.17
2000-01-06 12:55:00 +00:00
e1dba5c3a7
Move Mesa to xc/extras Update to the latest Mesa 3.2 code Fix the Q3Demo
...
bugs (white railgun and texture mapping) Simplify driver texture
mapping routines Fix device driver for 2.3 kernels Improve performance
1999-12-07 03:37:16 +00:00
b6a28bfe98
First DRI release of 3dfx driver.
1999-12-05 23:10:37 +00:00
Eric Anholt
Michel Daenzer
Alan Hourihane
Ian Romanick
Keith Whitwell
Leif Delgass
Jose Fonseca
David Dawes
Rik Faith
Jens Owen
Brian Paul
Tim Smith
Jeff Hartmann
Gareth Hughes
Kevin E Martin
Doug Rabson
Daryll Strauss