2009-07-29 07:38:16

by Dave Airlie

[permalink] [raw]
Subject: [git pull] update the staging radeon kms driver


Hi Linus,

Please read below then think about pulling the 'drm-radeon-kms' branch from
ssh://master.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6.git drm-radeon-kms

The radeon kms driver has gotten a lot of bugfixes and cleanups in the run
up to using it in F12, I'm going to be pulling this tree in there, so I said I
may as well offer it up for you to include. I've been careful to not touch
any code outside the radeon kms and TTM code (I even dropped some patches
that touched the main drm code). The only non-kms file touched is adding
an export needed for TTM.

You can pull this if you want, the kms driver is under staging and I'd
like to have a better working baseline in the 2.6.31 final kernel than
what we have now.

Dave.

arch/x86/mm/highmem_32.c | 1 +
drivers/gpu/drm/radeon/Makefile | 3 +-
drivers/gpu/drm/radeon/atombios_crtc.c | 293 +++++-----
drivers/gpu/drm/radeon/r100.c | 770 +++++++++++++++++++++-
drivers/gpu/drm/radeon/r300.c | 78 ++-
drivers/gpu/drm/radeon/r300_reg.h | 4 +-
drivers/gpu/drm/radeon/r500_reg.h | 2 +
drivers/gpu/drm/radeon/r520.c | 22 +-
drivers/gpu/drm/radeon/r600.c | 5 +-
drivers/gpu/drm/radeon/radeon.h | 87 +++-
drivers/gpu/drm/radeon/radeon_asic.h | 32 +
drivers/gpu/drm/radeon/radeon_atombios.c | 9 +-
drivers/gpu/drm/radeon/radeon_benchmark.c | 4 +-
drivers/gpu/drm/radeon/radeon_cs.c | 8 +-
drivers/gpu/drm/radeon/radeon_cursor.c | 9 +-
drivers/gpu/drm/radeon/radeon_device.c | 59 +-
drivers/gpu/drm/radeon/radeon_display.c | 74 ++-
drivers/gpu/drm/radeon/radeon_drv.c | 4 +
drivers/gpu/drm/radeon/radeon_encoders.c | 359 +----------
drivers/gpu/drm/radeon/radeon_fb.c | 70 ++-
drivers/gpu/drm/radeon/radeon_fence.c | 2 +-
drivers/gpu/drm/radeon/radeon_gart.c | 2 +-
drivers/gpu/drm/radeon/radeon_gem.c | 45 ++-
drivers/gpu/drm/radeon/radeon_kms.c | 2 +
drivers/gpu/drm/radeon/radeon_legacy_crtc.c | 687 ++++++--------------
drivers/gpu/drm/radeon/radeon_legacy_encoders.c | 188 +------
drivers/gpu/drm/radeon/radeon_mode.h | 51 +-
drivers/gpu/drm/radeon/radeon_object.c | 158 ++++-
drivers/gpu/drm/radeon/radeon_ring.c | 15 +-
drivers/gpu/drm/radeon/radeon_share.h | 39 ++
drivers/gpu/drm/radeon/radeon_test.c | 209 ++++++
drivers/gpu/drm/radeon/radeon_ttm.c | 24 +-
drivers/gpu/drm/radeon/rs400.c | 30 +-
drivers/gpu/drm/radeon/rs600.c | 7 +-
drivers/gpu/drm/radeon/rs690.c | 479 ++++++++++++++-
drivers/gpu/drm/radeon/rs690r.h | 99 +++
drivers/gpu/drm/radeon/rv515.c | 798 +++++++++++++++++++----
drivers/gpu/drm/radeon/rv515r.h | 170 +++++
drivers/gpu/drm/radeon/rv770.c | 2 +-
drivers/gpu/drm/ttm/ttm_bo.c | 63 ++-
drivers/gpu/drm/ttm/ttm_bo_util.c | 69 ++-
drivers/gpu/drm/ttm/ttm_bo_vm.c | 3 +
drivers/gpu/drm/ttm/ttm_tt.c | 25 +-
include/drm/radeon_drm.h | 23 +-
include/drm/ttm/ttm_bo_driver.h | 20 +-
include/drm/ttm/ttm_module.h | 2 +-
46 files changed, 3576 insertions(+), 1529 deletions(-)
create mode 100644 drivers/gpu/drm/radeon/radeon_share.h
create mode 100644 drivers/gpu/drm/radeon/radeon_test.c
create mode 100644 drivers/gpu/drm/radeon/rs690r.h
create mode 100644 drivers/gpu/drm/radeon/rv515r.h

commit ed8f0d9e708a1a7c9222e7d0a35d97521e904223
Author: Dave Airlie <[email protected]>
Date: Wed Jul 29 17:07:38 2009 +1000

drm/radeon: set fb aperture sizes for framebuffer handoff.

This will allow efi/vesa to handoff to radeon.

Signed-off-by: Dave Airlie <[email protected]>

commit b42db2b12df7b4f7b2ace581a7726cb5bcb2d658
Author: Dave Airlie <[email protected]>
Date: Wed Jul 29 16:56:52 2009 +1000

drm/ttm: fix highuser vs dma32 confusion.

DMA32 and highmem are sort of exclusive.

Noticed by AndrewR on #radeon.

Signed-off-by: Dave Airlie <[email protected]>

commit ea3c13bd8c2ed1d3670bd72e60f562a427355fdf
Author: Michel D?nzer <[email protected]>
Date: Mon Jul 20 01:44:03 2009 +0200

drm/radeon: Fix size used for benchmarking BO copies.

The incorrect size caused benchmark results to be inflated by a factor of 4.

Signed-off-by: Michel D?nzer <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit ecc0b32645bf19a3a240e72be3022ab3b46ad3d0
Author: Michel D?nzer <[email protected]>
Date: Tue Jul 21 11:23:57 2009 +0200

drm/radeon: Add radeon.test parameter for running BO GPU copy tests.

If enabled, during initialization BO GTT->VRAM and VRAM->GTT GPU copies are
tested across the whole GTT aperture.

This has helped uncover the benchmark copy size bug and verify the maximum
aperture size supported by the AGP bridge in my PowerBook.

Signed-off-by: Michel D?nzer <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit 3b170c3b2e688665fbc2845ba5bb4304bf38a119
Author: Dave Airlie <[email protected]>
Date: Fri Jul 24 13:47:45 2009 +1000

drm/radeon/kms: allow interruptible waits for objects.

Blocking here isn't something the X server mouse appreciates,
avoid the block and let userspace retry the waits.

libdrm_radeon userspace library is also expecting EBUSY not ERESTART

Signed-off-by: Dave Airlie <[email protected]>

commit f121ecfebbea1452a17d57c656def7d1537440f7
Author: Thomas Hellstrom <[email protected]>
Date: Fri Jul 24 10:22:36 2009 +0200

drm/ttm: powerpc: Fix Highmem cache flushing.

Temporarily maps highmem pages while flushing to get a valid virtual
address to flush.

Signed-off-by: Thomas Hellstrom <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit 73ba651fc246fcc3e446da4155e0425b4219d2c4
Author: Thomas Hellstrom <[email protected]>
Date: Fri Jul 24 09:57:33 2009 +0200

x86: Export kmap_atomic_prot() needed for TTM.

This functionality is needed to kmap_atomic() highmem pages that may
potentially have or are about to set up other mappings with
non-standard caching attributes.

Signed-off-by: Thomas Hellstrom <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit 542c6f6df51327dbb180cf4d9b34827e147efe17
Author: Thomas Hellstrom <[email protected]>
Date: Fri Jul 24 09:57:34 2009 +0200

drm/ttm: Fix ttm in-kernel copying of pages with non-standard caching attributes.

For x86 this affected highmem pages only, since they were always kmapped
cache-coherent, and this is fixed using kmap_atomic_prot().

For other architectures that may not modify the linear kernel map we
resort to vmap() for now, since kmap_atomic_prot() generally uses the
linear kernel map for lowmem pages. This of course comes with a
performance impact and should be optimized when possible.

Signed-off-by: Thomas Hellstrom <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit 4677f15c60421d48566c48c3149474e64977f071
Author: Thomas Hellstrom <[email protected]>
Date: Tue Jul 21 17:45:13 2009 +0200

drm/ttm: Fix an oops and sync object leak.

The code was potentially dereferencig a NULL sync object pointer.
At the same time a sync object reference was potentially leaked.

Signed-off-by: Thomas Hellstrom <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit 7a50f01a4ab89d5c05eb2cf62e206ac0bfc61d2c
Author: Dave Airlie <[email protected]>
Date: Tue Jul 21 20:39:30 2009 +1000

drm/radeon/kms: vram sizing on certain r100 chips needs workaround.

If an rn50/r100/m6/m7 GPU has < 64MB RAM, i.e. 8/16/32, the
aperture used to calculate the MC_FB_LOCATION needs to be worked
out from the CONFIG_APER_SIZE register, and not the actual vram size.

TTM VRAM size was also being initialised wrong, use actual vram size
to initialise it.

Signed-off-by: Dave Airlie <[email protected]>

commit 664f86590295217b2319edf88830e87b800f6c4a
Author: Michel D?nzer <[email protected]>
Date: Tue Jul 28 12:30:57 2009 +0200

drm/radeon: Pay more attention to object placement requested by userspace.

Previously we were basically always setting the GTT and VRAM flags regardless of
what userspace requested.

Signed-off-by: Michel D?nzer <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit 1ab2e1059916b917af19e4137a4222988bd7a169
Author: Michel D?nzer <[email protected]>
Date: Tue Jul 28 12:30:56 2009 +0200

drm/radeon: Fall back to evicting BOs with memcpy if necessary.

Otherwise if there's no GTT space we would fail the eviction, leading to
cascaded failure.

Signed-off-by: Michel D?nzer <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit e46074effd5510e7a8fe34b93828d98a50835da2
Author: Michel D?nzer <[email protected]>
Date: Tue Jul 28 12:30:55 2009 +0200

drm/radeon: Don't unreserve twice on failure to validate.

This is done later in radeon_object_list_unvalidate(). Doing it twice triggers
a BUG in TTM, rendering X on KMS unusable until reboot.

Signed-off-by: Michel D?nzer <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit c93bb85b5cba3e3a06f2cad8e9bc5c23d3d10aac
Author: Jerome Glisse <[email protected]>
Date: Mon Jul 13 21:04:08 2009 +0200

drm/radeon/kms: fix bandwidth computation on avivo hardware

Fix bandwidth computation and crtc priority in memory controller
so that crtc memory request are fullfill in time to avoid display
artifact.

Signed-off-by: Jerome Glisse <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit e024e11070a0a0dc7163ce1ec2da354a638bdbed
Author: Dave Airlie <[email protected]>
Date: Wed Jun 24 09:48:08 2009 +1000

drm/radeon/kms: add initial colortiling support.

This adds new set/get tiling interfaces where the pitch
and macro/micro tiling enables can be set. Along with
a flag to decide if this object should have a surface when mapped.

The only thing we need to allocate with a mapped surface should be
the frontbuffer. Note rotate scanout shouldn't require one, and
back/depth shouldn't either, though mesa needs some fixes.

It fixes the TTM interfaces along Thomas's suggestions, and I've tested
the surface stealing code with two X servers and not seen any lockdep issues.

I've stopped tiling the fbcon frontbuffer, as I don't see there being
any advantage other than testing, I've left the testing commands in there,
just flip the fb_tiled to true in radeon_fb.c

Open: Can we integrate endian swapping in with this?

Future features:
texture tiling - need to relocate texture registers TXOFFSET* with tiling info.

This also merges Michel's cleanup surfaces regs at init time patch
even though it makes sense on its own, this patch really relies on it.

Some PowerMac firmwares set up a tiling surface at the beginning of VRAM
which messes us up otherwise.
that patch is:
Signed-off-by: Michel D?nzer <[email protected]>

Signed-off-by: Dave Airlie <[email protected]>

commit c836e862803b2aa2bd9a354e151316d2b42c44ec
Author: Alex Deucher <[email protected]>
Date: Mon Jul 13 13:51:03 2009 -0400

drm/radeon/kms: fix hotspot handling on pre-avivo chips

Need to adjust CUR_OFFSET for yorigin

Signed-off-by: Alex Deucher <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit eb1300bcd70b3bffbefb6ae0eab13a571255ee93
Author: Alex Deucher <[email protected]>
Date: Mon Jul 13 11:09:56 2009 -0400

drm/radeon/kms: enable frac fb divs on rs600/rs690/rs740

Allows us to hit dot clocks much closer, especially on
chips with non-27 Mhz reference clocks like most IGP chips.
This fixes most flickering and blanking problems with
non-exact dot clocks on these chips.

Signed-off-by: Alex Deucher <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit d0e275a90a81b37409a0cfbca77581e3d235f5cf
Author: Alex Deucher <[email protected]>
Date: Mon Jul 13 11:08:18 2009 -0400

drm/radeon/kms: add PLL flag to prefer frequencies <= the target freq

This is needed when using fractional feedback dividers on some IGP
chips.

Signed-off-by: Alex Deucher <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit b995e4330de0d8b1b8b9e49ce10cc6dc78e2cbba
Author: Dave Airlie <[email protected]>
Date: Tue Jul 14 02:02:32 2009 +1000

drm/radeon/kms: block RN50 from using 3D engine.

RN50/ES1000 is a cut-down rv100 chip used in the server market.
The 3D engine on these is either not there or unverified so refuse
any attempt to configure registers on it.

Signed-off-by: Dave Airlie <[email protected]>

commit 2a0f8918fc34713ecaeb900ffb9afa61df4cb08e
Author: Dave Airlie <[email protected]>
Date: Sat Jul 11 04:44:47 2009 +1000

drm/radeon/kms: fix VRAM sizing like DDX does it.

Doing this like the DDX seems like the most sure fire way to avoid
having to reinvent it slowly and painfully. At the moment we keep
getting things wrong with aper vs vram, so we know the DDX does it right.

booted on PCI r100, PCIE rv370, IGP rs400.

Signed-off-by: Dave Airlie <[email protected]>

commit ad49f501867cba87e1e45e5ebae0b12435d68bf1
Author: Dave Airlie <[email protected]>
Date: Fri Jul 10 22:36:26 2009 +1000

drm/ttm/radeon: add dma32 support.

This add support for using dma32 memory on gpus that really need it.

Currently IGPs are left without DMA32 but we might need to change
that unless we can fix rs690.

Signed-off-by: Dave Airlie <[email protected]>

commit 61b576dbbe6a19d102c025ebc102a0749e2d3c80
Author: Michel D?nzer <[email protected]>
Date: Wed Jun 24 00:12:55 2009 +1000

drm/radeon: Endianness fixes for radeondrmfb.

For now handle it via r/g/b offsets and disallow 16 bpp modes on big endian
machines.

Signed-off-by: Michel D?nzer <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit 77bd36f014bc5a3f28507a4e86a81b2b3d2439c3
Author: Dave Airlie <[email protected]>
Date: Fri Jul 10 09:33:00 2009 +1000

drm/radeon/kms: don't swap PCIEGART PTEs in VRAM.

On powerpc, since we aren't using any hw swappers, this will
get flipped around by default in hw.

tested on a G5 + rv515.

Signed-off-by: Dave Airlie <[email protected]>

commit 5176fdc4c5873e52f9cb6e166d80e843847e7eb4
Author: Dave Airlie <[email protected]>
Date: Tue Jun 30 11:47:14 2009 +1000

drm/radeon/kms: drop zero length CS indirect buffers.

If userspace sends a zero length IB, it really shouldn't have bothered
so EINVAL it.

Signed-off-by: Dave Airlie <[email protected]>

commit e7168cab5bbac0a0e5413fd55ba0e92555bf860d
Author: Dave Airlie <[email protected]>
Date: Thu Jul 9 16:01:42 2009 +1000

drm/radeon/kms: fix vram vs aper size check.

Fix this to be correct like the DDX code, looks like a typo
on transfer to the kernel.

Signed-off-by: Dave Airlie <[email protected]>

commit 2007d633d639c896396e4c4b53b38068f3831307
Author: Alex Deucher <[email protected]>
Date: Wed Jul 8 16:17:23 2009 -0400

drm/radeon/kms: get lvds info for DIG LVTMA and UNIPHY encoders

Noticed by Rafa? Mi?ecki on dri-devel. On r6xx/r7xx hardware, laptop
panels can be driven by KLDSCP_LVTMA or UNIPHY.

Signed-off-by: Alex Deucher <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit 848577ee27f704231b1860ae987a1be78b88b06e
Author: Alex Deucher <[email protected]>
Date: Wed Jul 8 16:15:30 2009 -0400

drm/radeon/kms: fix quirk for MSI laptop

The line mux for the connector in the bios tables
is used for enumerating drm connectors. Since
this laptop has a quirk where the same line much is
listed for both VGA and LVDS, the connectors get
combined. Setting the line mux on LVDS to an unused
value prevents both encoders from being combined into
the same connector. This should fix bko bug 13720.

Signed-off-by: Alex Deucher <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit ae3e8122cbf8f9301369f276f4179aa6ec1b5b9c
Author: Thomas Hellstrom <[email protected]>
Date: Wed Jun 24 19:57:34 2009 +0200

ttm: Fix caching mode selection.

A bug caused a new caching state to be selected on each buffer object
validation regardless of the current caching state.
Moreover, a caching state could be selected that wasn't supported by
the memory type.

Signed-off-by: Thomas Hellstrom <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit d1724078d6a01177c1db4ea0b75fda1ca8a73d57
Author: Thomas Hellstrom <[email protected]>
Date: Wed Jun 24 19:57:35 2009 +0200

ttm: Make messages more readable.

Signed-off-by: Thomas Hellstrom <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

commit 531369e62649bb8f31217cc0bf33ee6f89f1dff6
Author: Dave Airlie <[email protected]>
Date: Mon Jun 29 11:21:25 2009 +1000

drm/radeon: fix support for vline relocations.

Userspace sends us a special relocation type to sync video/exa
to vlines to avoid tearing, this deals with the relocation
in the kernel, it picks the correct crtc and avoids issues
where crtcs are disabled.

This version also parses the wait until to make sure it isn't
trying to do anything evil.

Signed-off-by: Dave Airlie <[email protected]>

commit 3e43d82125952826202a8cd20ba84a66f3ff8808
Author: Dave Airlie <[email protected]>
Date: Thu Jul 9 15:04:18 2009 +1000

drm/radeon/kms: respect TOM on rs100->rs480 IGP variants.

Normally we are free to place VRAM where we want in the GPUs
memory address space, however on IGP chips the VRAM is actual RAM,
and no special translation or aperture is used inside the GPU MC.

So when you move the VRAM aperture away from the TOM register,
you actually move it into main memory and can trash things quite badly.

This commit makes the code respect the TOM location for MC_FB_LOCATION.

Signed-off-by: Dave Airlie <[email protected]>

commit 4162338a1dab388474d4115289d1d7071623f04d
Author: Dave Airlie <[email protected]>
Date: Thu Jul 9 15:04:19 2009 +1000

drm/radeon/kms: set crtc and cursor offsets correctly on legacy chips.

The crtc and cursor offsets on the legacy chips are offset from
DISPLAY_BASE_ADDR. The code worked if display base addr was at 0,
but otherwise falls to pieces.

Signed-off-by: Dave Airlie <[email protected]>

commit 6cdf65855cf884712532fc72770baaef7bdf1b9a
Author: Dave Airlie <[email protected]>
Date: Mon Jun 29 18:29:13 2009 +1000

drm/radeon/kms: remove IB flushing trick.

If there is a problem then this is hiding it, we shouldn't
ever need to flush the IB. Either the buffers are:

WB - caching just works.
WC - no need to do explicit flush, the MB + readback will do it

Signed-off-by: Dave Airlie <[email protected]>

commit 4c9bc75cbc6f2f447a38a123aa6e0605fab3cb7a
Author: Dave Airlie <[email protected]>
Date: Mon Jun 29 18:29:12 2009 +1000

drm/radeon/kms: mmio base/size should be resource_size_t.

Unsigned long is incorrect for 64-bit resources on 32-bit hw.

Signed-off-by: Dave Airlie <[email protected]>

commit ed10f95d60d41033d356fdcf88c240d7065bd5b4
Author: Dave Airlie <[email protected]>
Date: Mon Jun 29 18:29:11 2009 +1000

drm/radeon/kms: fix some GART table entry bugs.

1. rv370 can accept 40-bit addresses - also at 24-bit shift not 4 bits
2. rs480 table can be in 40-bit space. - 4 bit shift for top 8 bits
3. rs480 table entries can be in 40-bit space.

Signed-off-by: Dave Airlie <[email protected]>