Commit Graph

273 Commits (34563921dd0b41d4ccf08374227e31d765b40353)

Author SHA1 Message Date
Keith Whitwell 47c178da08 Copy HC_ParaType_Auto change to shared-core. Bump version numbers and
dates.
2004-12-20 12:03:02 +00:00
Thomas Hellstrom ae0db704b4 via changes: Fixed typecasting bug in via_dma.c, and possible short-circuit
bug in the cmdbuf_size ioctl. Modified ring-buffer jump code AGAIN, due
    to new oddities discovered on the Unichrome Pro with faster processors.
    Bumped patchlevel and driver date.
2004-12-19 19:13:47 +00:00
Thomas Hellstrom aff0a25485 VIA drm updates:
1. Improved security check of AGP texture adresses.
2. Hopefully last fix of ring-buffer jump oddities.
3. Added ioctl to check available space and command regulator lag in
    ring-buffer. This is needed for 3D application responsiveness.
2004-12-13 13:53:12 +00:00
Roland Scheidegger c4a87c6883 (Stephane Marchesin, me) add hyperz support to radeon drm. Only fast z
clear and z buffer compression are working correctly, hierarchical-z is
    not.
2004-12-08 16:43:00 +00:00
Thomas Hellstrom 1fbfd9eb32 Security and optimization fixes for the via drm:
1. The command verifier was never initialized in the non-core source tree.
2. Check added that the AGP ring buffer has been initialized before
    accepting command buffer.
3. Free space check in the AGP buffer is moved to after command
    verification, which is more optimal in most cases.
2004-12-06 11:19:23 +00:00
Thomas Hellstrom 267e064527 Added 3D functionality to the via command verifier. Modified the via
ring-buffer code somewhat to workaround hardware problems. Bumped via
    minor version number.
2004-12-03 23:03:36 +00:00
Dave Airlie f197110e07 Make 1-bit fields be unsigned (no sign bit :). sparse complains about them:
drivers/char/drm/sis_ds.h:88:12: warning: dubious one-bit signed
    bitfield drivers/char/drm/sis_ds.h:89:16: warning: dubious one-bit
    signed bitfield
Signed-off-by: Randy Dunlap <rddunlap@osdl.org> Signed-off-by: Dave Airlie
    <airlied@linux.ie>
2004-12-03 10:22:15 +00:00
Thomas Hellstrom 4f8fa60286 Reworked PCI MMIO command buffer parser, and imported code from the Mesa
driver. It can now handle the 3D OpenGL commands from the Mesa
    unichrome driver.
Added vsync frequency detection support. This will be used in the future
    for XvMC and better frame timing.
Bumped minor version number and driver date.
2004-11-27 22:55:31 +00:00
Dave Airlie f0a86288fa patch from bug 1803 - will try and push to kernel soon 2004-11-11 11:09:11 +00:00
Dave Airlie c5bededa51 add some more r300 pci ids 2004-11-07 02:19:58 +00:00
Eric Anholt cb5aaa8987 Convert more drivers for bsd-core, moving the ioctl definitions to shared
code. Remove the "drv" from sisdrv, as it's unnecessary. Use the
    drm_pci functions in i915 instead of per-os implementations of the
    same. Avoid whitespace within fields in drm_pciids.txt (one of the r300
    definitions), since it breaks the bsd pciids script. Tested on sis,
    mga, r128. i915 needs more work.
2004-11-06 23:02:07 +00:00
Jon Smirl 069f53a93b Move radeon i2c include to top of file 2004-11-06 16:55:41 +00:00
Jon Smirl 642a8106d2 Export missing r128 ioctl symbol 2004-11-06 16:51:36 +00:00
Eric Anholt c9202c8965 Commit WIP of BSD conversion to core model. Compiles for r128, radeon, but
doesn't run yet. Moves the ioctl definitions for these two drivers back
    to the shared code -- they aren't OS-specific.
2004-11-06 01:41:47 +00:00
Thomas Hellstrom f08a01c0e2 Fixed bug in via_dma.c. The code didn't check that the lock was held by the
caller. Just that it was held.
2004-11-03 13:37:37 +00:00
Thomas Hellstrom 9409d8231c Reworked the jump-code in via_dma.c. The command regulator now seems to
pause correctly after a jump. Removed the debug message from within the
    interrupt handler of via_irq.c
2004-11-01 20:48:49 +00:00
Keith Whitwell 08758b2fb7 correct historic mis-attribution of copyright 2004-11-01 10:52:18 +00:00
Thomas Hellstrom 8264e2c8aa Some stabilizing work to the DMA ring-buffer code. Temporarily replaced the
rewind jump with a reinitialization. This makes the code stable on
    CLE266 and KM400.
2004-10-30 13:01:48 +00:00
Ville Syrjala d403173005 Fixed off by one errors in clipping. 2004-10-16 11:21:56 +00:00
Ville Syrjala 52fdf10fd7 Fixed bad formatting. 2004-10-16 10:54:58 +00:00
Thomas Hellstrom bbfe18e3bf Via updates. Fixed unlikely but possible uint32_t overflow in ring-buffer
code. bumped patchlevel.
2004-10-12 18:46:26 +00:00
Jon Smirl ad70dc676e Breakout heads into their own data structures. 2004-10-12 03:59:17 +00:00
Jon Smirl ad549c5ae6 Rename fn_tbl to driver. Core driver now uses pci_driver name which
reflects the personality name.
2004-10-10 22:54:55 +00:00
Dave Airlie 3f02a79351 Vladimir requested support so we can at least load r300 microcode for
helping
2D operations.
Ups radeon to version 1.12.0, Vladimir, you might want to add any extra
    pciids...
Approved-by: Dave Airlie <airlied@linux.ie>
2004-10-10 05:52:19 +00:00
Dave Airlie 8727326371 cleanup VIA driver to look a bit like others before kernel merge 2004-10-09 12:42:52 +00:00
Dave Airlie fec94a8274 Lindent the via stuff so I can include it in kernel 2004-10-09 11:12:24 +00:00
Thomas Hellstrom d24194e904 Changed unsigned to uint32_t in some ioctl parameters. Introduced first
rudimentary command verifier for dma buffers. Changed the decoder futex
    ioctl parameters. Bumped the via major version number.
2004-10-08 21:11:02 +00:00
Jon Smirl 61d36f6179 Revert back to drm_order() instead of using kernel get_order(). The
functions are not identical.
2004-10-06 16:27:55 +00:00
Jon Smirl aba12cfc0e janitor-list_for_each-drivers-char-drm-radeon_memc.patch from mm kernel 2004-10-02 19:11:30 +00:00
Jon Smirl 9f9a8f1382 Lindent of core build. Drivers checked for no binary diffs. A few files
weren't Lindent's because their comments didn't convert very well. A
    bunch of other minor clean up with no code implact included.
2004-09-30 21:12:10 +00:00
Jon Smirl 1c0a437fa2 Move things around to reduce public symbols and even out files. Switch to
get_order from drm_order.
2004-09-30 19:26:35 +00:00
Jon Smirl 3aef3841d0 Make fops per driver instead of global, remove default flush, poll, read
functions
2004-09-30 18:13:33 +00:00
Jon Smirl fa6b1d129e First check in for DRM that splits core from personality modules 2004-09-27 19:51:38 +00:00
Jon Smirl 74f063fc9d Create new linux-2.6 build. Move all gpl files into the 2.6 build. If you
edit files for 2.6 be sure and break the link to the 2.4 directory and
    copy the cvs history.
2004-09-24 03:12:17 +00:00
Jon Smirl c158a36c4c 1) switches from class_sysfs to drm sysfs implementation to allow
customization
2) compiles again on 2.4, but doesn't work
2004-09-23 05:39:15 +00:00
Felix Kuehling af326f6f0c Create permanent maps of framebuffer, aperture and MMIO registers. Added
chipset-type information in driver data field of Savage PCI-IDs. Added
    missing PCI-ID 0x8d03 (ProSavageDDR on Pentium boards). Don't require
    AGP.
2004-09-22 22:51:18 +00:00
Jon Smirl 27fc998f7d Remove hotplug reset support from DRM driver. This will be handled by the
VGA driver when it gets written.
2004-09-22 19:13:02 +00:00
Jon Smirl 6f31c42fe9 Remove size restriction on permanent addmap 2004-09-20 19:23:27 +00:00
Dave Airlie 5654880eee remove HAVE_COUNTERS 2004-09-20 11:29:16 +00:00
Jon Smirl 0c6fb0fbe1 Add linux sysfs i2c support to radeon driver. This patch adds GPL licensed
files to the linux build but not to the BSD directories.
2004-09-17 04:02:28 +00:00
Jon Smirl c7c9d3ef7b Let's try adding the dyn-minor patch again. This patch will reuse minor
numbers if a card is hotplugged in/out instead of just having them
    increase.
2004-09-16 18:42:03 +00:00
Jon Smirl eeb0ef1a70 Back dyn-minor patch out for now. fops handling is broken on some cards 2004-09-15 17:44:30 +00:00
Jon Smirl 64ef12c55c Dynamic device minor support. Minor device numbers will be reused if the
device is hotplugged in/out of the system
2004-09-15 00:20:21 +00:00
Jon Smirl 03c2e674e4 Add chip family names to the radeon driver 2004-09-14 03:59:25 +00:00
Jon Smirl 36050cc958 Fix DRM to compile cleanly with recent kernel changes in PCI IO and
DRM_COPY_FROM_USER. PCI IO changes in 2.6.9-rc1 bk currently.
2004-09-12 03:23:50 +00:00
Jon Smirl e6d468ad7f More general patch to mark resources in use by all DRM drivers. Makes the
code Linux specific.
2004-09-10 16:44:28 +00:00
Jose Fonseca 1fb27632c3 Update doxygen configuration file. Minor documentation updates/fixes. 2004-09-08 20:57:39 +00:00
Thomas Hellstrom b0c73b7fcb Added IOCTL for writing 2D DMA command buffers over PCI. Bumped minor
version number.
2004-09-07 16:48:44 +00:00
Dave Airlie eeae6a0a38 merge back bunch of whitespace and misc changes from kernel 2004-09-05 10:54:59 +00:00
Dave Airlie 7809efc8c3 drm-memory patch, cleans up alloc/free and makes calloc look more libc like 2004-08-30 09:01:50 +00:00
Erdi Chen 08de6e5b04 Fix copy_from_user return value ignored warning at via_dma.c:168. 2004-08-30 04:58:24 +00:00
Dave Airlie 73e606753f run i915 through lindent 2004-08-27 09:14:30 +00:00
Dave Airlie 019fd38a3c __NO_VERSION__ hasn't been needed since 2.3 days ditch it... 2004-08-27 09:11:07 +00:00
Jon Smirl e068fcbfdd Rearrange things so that via_dma.c will compile with inline via_check_dma 2004-08-26 03:54:01 +00:00
Dave Airlie d4dbf45781 Merged drmfntbl-0-0-2 2004-08-24 11:15:53 +00:00
Erdi Chen 25e319c1ef This patch adds three new ioctl's to the VIA Unichrome/Pro DRM driver:
DRM_IOCTL_VIA_DMA_INIT DRM_IOCTL_VIA_CMDBUFFER DRM_IOCTL_VIA_FLUSH
The first ioctl sets up an area in AGP memory that will be used as the ring
    buffer. The second ioctl copies a command buffer from user space memory
    to the ring buffer. The third ioctl waits for engine idle until it
    returns.
The motivation for this patch is to avoid the wait for engine idle call
    before each buffer flush in the current DRI driver. With this patch,
    the DRI driver can continue to flush its buffer as long as there is
    free space in the ring buffer.
This patch adds an additional copy operation on the command buffer. This
    buffer copying is necessary to support multiple DRI clients rendering
    simultaneously. Otherwise, more CPU time will be spent in the busy loop
    waiting for engine idle between DRI context switch. Even in the single
    client case, the tradeoff is reasonable in comparision to the kernel
    call to check for free buffer space for the client to render directly
    to the ring buffer.
2004-08-24 01:44:37 +00:00
Dave Airlie 7fe4f60768 set pointers to NULL after freeing, remove some extra debugging 2004-08-23 10:05:01 +00:00
Dave Airlie 5c9ed83094 Merged drmfntbl-0-0-1 2004-08-17 13:10:05 +00:00
Dave Airlie 93e8c201af preparation patch for radeon permanent mapping registers/framebuffer makes
dev_priv live always, and add AGP detection in kernel patch:
    radeon-pre-2.patch From: Jon Smirl
2004-08-17 11:24:50 +00:00
Eric Anholt dd83f39f03 Add a "dev" argument to DRIVER_CTX_[CD]TOR. This will be used in an
upcoming commit for the SiS driver.
2004-08-14 00:46:15 +00:00
Eric Anholt 211c37e250 Fix apparent copy'n'paste-o of the card attributes commit that broke the
FreeBSD build.
2004-08-13 23:32:39 +00:00
Dave Airlie 9277f9eef3 Patch from Jon Smirl to add attribute field to the pciids, and use this for
certain radeon combinations - intel drivers can probably use this for
    dual head capable devices etc..
2004-08-10 11:14:07 +00:00
Dave Airlie 33b740ad99 forgot to check these in.. thanks to Jon for reminding me .. 2004-08-02 11:45:23 +00:00
Dave Airlie bd71ba6428 patch from RH xorg-x11 tree ported to drm.h 2004-07-31 08:38:39 +00:00
Dave Airlie eb3d0635d4 fixes for using userspace pointers found by sparse utility
From: Dave Airlie
2004-07-31 07:26:52 +00:00
Dave Airlie 0b02bf9d45 initial port of i915 to BSD, not finished doesn't work.. no idea why... 2004-07-29 11:09:22 +00:00
Dave Airlie 862fcee057 check for __user if not there define it 2004-07-25 08:57:15 +00:00
Dave Airlie 02df04d71d sync up with current 2.6 kernel bk tree - mostly __user annotations 2004-07-25 08:47:38 +00:00
Keith Whitwell 1e5354e867 Correct a couple of packet length calculations. 2004-07-23 16:12:27 +00:00
Dave Airlie b35cba47d5 Add NULLs instead of 0 for i915 2004-07-20 11:48:25 +00:00
Dave Airlie 8efddd01e8 sparse cleanups from kernel: Al Viro 2004-07-15 13:03:55 +00:00
Dave Airlie 4cfd0d5cee whitespace align with kernel 2004-07-05 11:44:30 +00:00
Keith Whitwell 291ffeae9a i915.o drm driver 2004-06-10 12:45:38 +00:00
Ian Romanick 52e3a8a5d9 Added some comments copied from xf86drm.h. 2004-06-02 17:41:52 +00:00
Ian Romanick f994b1f31b Replace size_t with an ugly, ugly hack. This was done so that code in the
core X-server would compile. Hopefully, this will soon be replaced with
    explicitly sized types (i.e., uint32_t) and everyone will be happy.
2004-06-01 16:17:32 +00:00
Roland Scheidegger 43c244ebba fix whitespace issue in previous patch 2004-05-18 23:30:46 +00:00
Roland Scheidegger aa142ff1b5 add R200_EMIT_RB3D_BLENDCOLOR state packet to support GL_EXT_blend_color,
GL_EXT_blend_func_separate and GL_EXT_blend_equation_separate on r200
2004-05-18 23:03:22 +00:00
Erdi Chen 13894755a7 Add PCI id entry for VIA CN400 (UnichromePro) chip. XFree86 bug: Reported
by: Submitted by: Reviewed by: Obtained from:
2004-05-14 23:54:31 +00:00
Eric Anholt 853adb8be3 Merge from FreeBSD-current. Mostly 64-bit cleanliness fixes, but a few
driver interface changes from -current.
2004-05-11 04:43:43 +00:00
Eric Anholt 485b259b44 Add missing DRM_ERR()s. 2004-05-11 04:23:02 +00:00
Dave Airlie 06332524b5 add another tdfx 2004-04-26 06:43:53 +00:00
Dave Airlie 0edc2abf84 add new files to generate pci ids 2004-04-21 23:45:18 +00:00
Dave Airlie 93bd67ef62 centralise pci ids into one place and use scripts to generate files for
kernel
2004-04-21 12:13:31 +00:00
Thomas Hellstrom 6f6d2a553a 1. Added a PCI ID.
2. Big change to the XvMC part of the SAREA. OpenGL clients will not suffer
    from this, and via XvMC is still alpha. Needed to make future additions
    to XvMC (More decoders and overlays) possible.
3. Bumped version number to 1.3.0.
2004-04-12 10:18:18 +00:00
Dave Airlie 873e1c4d38 Add mach64 to the trunk 2004-04-12 05:27:40 +00:00
Dave Airlie 3306abbde7 white space changes to align with kernel 2004-04-10 13:52:43 +00:00
Alan Hourihane 596cf634ae fix build problem 2004-04-08 23:22:05 +00:00
Dave Airlie bc1428035c fixes from Linux kernel 2004-04-08 12:05:25 +00:00
Thomas Hellstrom 6cd8831f7a Merged via-1-2-0 2004-03-23 21:08:48 +00:00
Jon Smirl 24115068e4 Fixes need to clean up the mess I made with the mesa merge. This code
allows the mesa drivers to use a single definition of the DRM
    sarea/IOCTLS located in the drm driver directory. Adjustments were made
    to the 2D drivers to not include these changes. Changes to the mesa
    copy of DRM were copied to the DRI copy. XFree86 bug: Reported by:
    Submitted by: Reviewed by: Obtained from:
2004-03-12 21:22:52 +00:00
Michel Daenzer 213247a441 Fix sisfb header location for 2.6 kernels
Submitted by: Andrew Morton
2004-02-18 20:56:58 +00:00
Michel Daenzer 0dea4de288 Make sure that all state packets are handled in
radeon_check_and_fixup_packets()
Fix state packet IDs of R200 cubic offsets
2004-01-10 20:59:16 +00:00
Michel Daenzer 7b62ed9aed R200_PP_CUBIC_OFFSET_F1_[0-6] state packets only contain 5 offsets, not 6
(thanks to Andreas Stenglein for spotting this)
2004-01-10 12:28:06 +00:00
Eric Anholt 9fb6986e83 Don't ioremap the framebuffer area. The ioremapped area wasn't used by
anything, and took up valuable KVA. While I'm in the area, clean up BSD
    MTRR stuff some more.
Suggested by: jonsmirl
2003-12-16 08:57:08 +00:00
Eric Anholt 66c9e3053f - Tie the DRM to a specific device: setunique no longer succeeds when given
a busid that doesn't correspond to the device the DRM is attached to.
    This is a breaking of backwards-compatibility only for the
    multiple-DRI-head case with X Servers that don't use interface 1.1.
- Move irq_busid to drm_irq.h and make it only return the IRQ for the
    current device. Retains compatibility with previous X Servers, cleans
    up unnecessary code. This means no irq_busid on !__HAVE_IRQ, but can be
    changed if necessary.
- Bump interface version to 1.2. This version when set signifies that the
    control ioctl should ignore the irq number passed in and enable the
    interrupt handler for the attached device. Otherwise it errors out when
    the passed-in irq is not equal to the device's.
- Store the highest version the interface has been set to in the device.
- Fix a recursion on DRM_LOCK in irq_uninstall on FreeBSD. This leaves
    irq_uninstall being done without the lock in some cases, but it was
    racey anyways.
2003-11-05 08:13:52 +00:00
Eric Anholt e8f5b01a40 __linux__ is spelled with a lowercase 'l' 2003-11-05 02:08:31 +00:00
Eric Anholt 5864101ab4 Repo-copy linux/drm/kernel/drm.h to shared/drm/kernel/drm.h and use it on
both Linux and *BSD.
2003-11-05 01:43:47 +00:00
Michel Daenzer 2655ccddf4 Memory layout transition:
the 2D driver initializes MC_FB_LOCATION and related registers sanely
the DRM deduces the layout from these registers
clients use the new SETPARAM ioctl to tell the DRM where they think the
    framebuffer is located in the card's address space
the DRM uses all this information to check client state and fix it up if
    necessary
This is a prerequisite for things like direct rendering with IGP chips and
    video capturing.
2003-11-04 00:46:05 +00:00
Eric Anholt 06cb132e86 - Introduce a new ioctl, DRM_IOCTL_SET_VERSION. This ioctl allows the
server or client to notify the DRM that it expects a certain version of
    the device dependent or device independent interface. If the major
    doesn't match or minor is too large, EINVAL is returned. A major of -1
    means that the requestor doesn't care about that portion of the
    interface. The ioctl returns the actual versions in the same struct.
- Introduce DRM DI interface version 1.1. If the server requests version
    1.1, then the DRM sets the unique itself according to the busid of the
    device it probed, which may then be accessed as normal using getunique.
- Request version 1.1 in libdrm's drmOpenByBusID, allowing the X Server to
    request based on a BusID. Introduce a wrapper for DRM_IOCTL_SET_VERSION
    and bump libdrm minor version.
- Pass the busid in DRIScreenInit if libdrm can handle both a busid and
    name. This allows drmOpenByBusID to be used to find the DRM instead of
    just the driver name, which allows us in the future to tie a DRM more
    strongly to the device it probed to. Introduce a function
    DRICreatePCIBusID which creates a busid in the form pci:oooo:bb:dd.f
    similar to linux's pci_name() function. This matches the format used by
    the DRM in version 1.1. libdrm knows how to match both this format and
    the old PCI🅱️d:f format.
- Use the new DRICreatePCIBusID function in the *_dri.c to request the new,
    more exact busid format.
2003-10-23 02:23:31 +00:00
Eric Anholt 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