Gareth Hughes
edfdc14215
Merge with latest trunk code. Now requires kernel 2.4.0-test11-pre5 or
...
later.
2000-11-21 07:20:28 +00:00
Gareth Hughes
c071d03e7b
r128-specific parts of sync with 2.4.0-test11-pre5. Rest is coming shortly.
2000-11-21 06:48:38 +00:00
Gareth Hughes
2a39d83ffe
- Cleanup of indirect buffer submission.
...
- Indexed vertex buffer fixes, more robust handling of indirect buffers for
the elt path.
2000-11-15 12:31:39 +00:00
Gareth Hughes
81f4398ed7
Greatly improve interactivity with MMIO-based 2D acceleration by only
...
turning CCE off when the X server actually has to do some 2D rendering.
We test if the CCE is running at the start of each accel routine, and
if so we shut it down. If we've shut it down, we start it up again in
LeaveServer(). An extra if statement in the accel routines is worth it,
IMNSHO.
Updates to the elt path code. Allow indirect buffer firing to specify start
and end parameters for buffer reuse. DRM ioctls and wrappers adjusted
accordingly. Still disabled by default.
2000-11-05 14:50:04 +00:00
Gareth Hughes
79857ad822
Add support for different primitive types in vertex buffer flush ioctl.
...
Try and fix Mobility 128 lockups. Seems to lock when shutting down the X
server from non-standard resolutions, so I've probably messed up the
CCE ioctls somewhat. Default panel resolution seems to be rock solid...
2000-10-27 01:52:53 +00:00
Gareth Hughes
05ef8effbf
Latest development work. Should be reasonably stable with the DRIScreenInit
...
locking fix. Usual caveats apply to using development code.
Includes:
- ctx->Texture.Enabled to ctx->Texture.ReallyEnabled fix
- More useful information in GL_RENDERER string
- More indirect buffer support work
2000-10-20 00:48:07 +00:00
Gareth Hughes
9b1f5e1d08
Get buffers via drmDMA. Vertex buffers are acquired one at a time, texture
...
blit buffers can be acquired in batches as needed.
Flush vertex buffers via drmR128FlushVertexBuffer. This ioctl no longer
acquires buffers, it merely flushes and/or discards them.
2000-10-16 03:57:36 +00:00
Gareth Hughes
a553ebcd12
Add support for indirect buffers. Clean up the way vertex and indirect
...
buffes are handled.
The way buffers are acquired will probably be changed back to what it was
originally, using the DRM_DMA-style ioctl. Texture uploads should grab
as many buffers as needed in a single ioctl.
2000-10-15 13:57:49 +00:00
Gareth Hughes
3c8b613ae2
Fix last of lockups. Stopping the CCE was failing when the engine was very
...
busy, and this wasn't being handled gracefully. Works great now.
2000-10-15 02:32:37 +00:00
Gareth Hughes
ce71d0abb9
- Enable client-side throttling again.
...
- Work around CCE start/stop/reset issues for now.
2000-10-11 14:13:44 +00:00
Gareth Hughes
4f1b6b9f2b
Minor updates after trunk merge:
...
- Fix front, back and depth pitch calculations
- Implement immediate mode triangle optimizations (Keith Whitwell)
- Some minor cleanups
2000-10-10 12:44:04 +00:00
Gareth Hughes
9ae70d6507
Merge trunk with ati-4-1-1-branch. Needed for Mobility M3 to work with
...
latest 3D driver development.
2000-10-10 04:22:37 +00:00
Gareth Hughes
4ca420eac5
Checking in latest development work - I need to merge the trunk into this
...
branch so I can work on my Mobility 128 laptop. This code is probably
broken, and may not even compile properly until after the merge is
done.
2000-10-09 01:55:34 +00:00
Gareth Hughes
ce6bb6dc6e
Grab latest trunk code. Sync with 2.4.0-test9-pre4 kernel.
2000-09-19 18:25:45 +00:00
Gareth Hughes
1e8cc8bfb4
Uppdate date stamps for client-side 3D driver, kernel module.
2000-09-19 17:10:20 +00:00
Gareth Hughes
04073953af
Change u32s to unsigned ints (originally unsigned longs). Will have to
...
verify that this is 64-bit safe.
2000-09-19 17:03:05 +00:00
Gareth Hughes
cc57f63b8f
Somewhat unstable development work. Use at your own risk.
2000-09-19 16:40:02 +00:00
Gareth Hughes
83253109dc
Sync with 2.4.0-test8 kernel.
2000-09-10 16:16:41 +00:00
Gareth Hughes
a5c98c730f
Get latest kernel module updates from Rik Faith. Thanks, Rik! Sync with
...
2.4.0-test7, other fixes currently on the trunk.
2000-08-27 16:55:07 +00:00
Gareth Hughes
ae24dece16
Apply Rik's 2.4.0-test7-pre4 sync patch.
2000-08-22 16:37:24 +00:00
Gareth Hughes
d8ed3693a8
New ATI development branch. Resync with my unstable development work, only
...
the client-side 3D driver code is present here (ie. no new kernel
module yet). This code is probably broken, but it will at least build
correctly.
Many, many, many updates - too numerous to mention. PCI flickering is fixed
in this code, however.
2000-08-18 06:44:35 +00:00
Rik Faith
f75ec3010a
Bug #112196 : auth.c uses semaphores while holding spinlocks Make spinlocked
...
region smaller to avoid semaphore.
2000-08-17 20:26:11 +00:00
Rik Faith
761ffd63f3
Bug #112197 Made sis driver compile under 2.4.0-test7/pre4
2000-08-17 20:20:33 +00:00
Jeff Hartmann
3e42514360
Fix for bug #111744 which caused any application to never render and dma to
...
stall on certain mga cards
2000-08-16 21:13:24 +00:00
David Dawes
cba4d3504c
first part of 4.0.1b merge
2000-08-16 18:29:06 +00:00
David Dawes
0cbca9cca2
Import of XFree86 4.0.1b
2000-08-16 17:23:02 +00:00
David Dawes
1df8afa8ff
Initial revision
2000-08-16 17:23:02 +00:00
Rik Faith
8b9363d1b1
Sync with Linux 2.4.0-test6-pre8
2000-08-08 16:04:21 +00:00
Jeff Hartmann
9199075a6a
Fix for multiple sarea bug + agp built into kernel segfault
2000-08-07 16:55:10 +00:00
Rik Faith
f3ae6786f8
Sync with Linux 2.4.0-test6-pre2
2000-08-04 10:04:55 +00:00
Rik Faith
bcad118182
Fix typo
2000-07-22 01:08:00 +00:00
Rik Faith
1e54e55431
Move to new, denser, easier-to-read Linux kernel Makefile format
2000-07-22 01:04:22 +00:00
Rik Faith
3bcc55a74d
Revert some changes and try alternative way to clean up AGP handling.
2000-07-21 18:28:01 +00:00
Rik Faith
6dd96a15ed
Revert some changes and try alternative way to clean up AGP handling.
2000-07-21 16:57:00 +00:00
Rik Faith
e38b34e54a
Fixes for building in the kernel tree
2000-07-21 15:48:24 +00:00
Rik Faith
81da79d1ed
Changes to make AGP optional for in-kernel builds
2000-07-21 14:20:57 +00:00
Rik Faith
6d1ce8fe77
kfree_s was deprecated in 2.4.0-test?
2000-07-20 20:21:10 +00:00
Rik Faith
b4722ec943
More fixups for kernel build: EXPORT_SYMTAB warning removal
...
put_module_symbol clean up of tdfx tdfx_cleanup routine Makefile.kernel
fix
2000-07-20 05:17:49 +00:00
Rik Faith
0bd1fa795a
Fixes for building in the kernel tree
2000-07-20 01:52:00 +00:00
Rik Faith
e8c9ed03f8
Fixed for monolithic kernel build
2000-07-20 01:14:58 +00:00
Rik Faith
742c33bf61
Fix signature for *_options function
2000-07-20 00:59:40 +00:00
Rik Faith
c459c9ead3
Added support for building as modules or as part of monolithic kernel
2000-07-20 00:48:12 +00:00
Rik Faith
1d3c4d2ed2
Bump driver dates and add descriptions
2000-07-19 20:59:35 +00:00
Rik Faith
ee9514b63b
Allow SAREA > 1 page in size
2000-07-19 20:41:51 +00:00
Rik Faith
aed8549e91
Sync with Linux 2.4.0-test4 kernel
2000-07-19 18:38:41 +00:00
Brian Paul
7b888f87fa
applied Jeff's xf86cvs-I810copy.patch
2000-07-13 18:47:48 +00:00
Alan Hourihane
4509385d37
Final misc changes to bring upto 4.0.1
2000-07-12 21:37:17 +00:00
Alan Hourihane
8751b6774d
Merge XFree86 4.0.1
2000-07-11 11:41:07 +00:00
Alan Hourihane
8bc39d01ca
Import of XFree86 4.0.1
2000-07-10 21:32:06 +00:00
Gareth Hughes
62838b948c
Fix for agpgart module detection. If the kernel module has been configured
...
to use AGP, and the module is not present, fail gracefully instead of
doing evil things.
2000-07-01 01:50:00 +00:00