Michel Dänzer [Wed, 14 Mar 2012 13:59:25 +0000 (14:59 +0100)]
drm/radeon: Drop radeon_gem_object_(un)pin.
Only radeon_gem_object_unpin was used anymore, in only one place.
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Michel Dänzer [Wed, 14 Mar 2012 16:12:42 +0000 (17:12 +0100)]
drm/radeon: Restrict offset for legacy display engine.
The hardware only takes 27 bits for the offset, so larger offsets are
truncated, and the display shows random bits other than the intended ones.
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Michel Dänzer [Wed, 14 Mar 2012 16:12:41 +0000 (17:12 +0100)]
drm/radeon: Restrict offset for legacy hardware cursor.
The hardware only takes 27 bits for the offset, so larger offsets are
truncated, and the hardware cursor shows random bits other than the intended
ones.
Cc: stable@vger.kernel.org Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Tvrtko Ursulin [Fri, 16 Mar 2012 16:22:11 +0000 (12:22 -0400)]
drm/radeon/kms: reduce probe latency on digital connectors
On a system with one HDMI and one VGA connector the latter
causes output polling to run every ten seconds. This causes
full EDID re-fetch on every poll and approx. 100ms rendering
stalls are experienced by full screen page-flipping applications.
Optimisation is to trust HPD sense on R600+ ASICs and to skip
doing these expensive probes unless HPD sense has changed.
Most of the error handling code in this function frees the buffers
kcmd_addr, kvb_addr, and kbox_addr allocated at the beginning of this
function. These two branches are changed to do the same.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr> Signed-off-by: Dave Airlie <airlied@redhat.com>
Julia Lawall [Sat, 17 Mar 2012 17:03:29 +0000 (18:03 +0100)]
drivers/gpu/drm/radeon/radeon_cs.c: eliminate possible double free
The function radeon_cs_parser_init is only called from two places, in
drivers/gpu/drm/radeon/radeon_cs.c and drivers/gpu/drm/radeon/r600_cs.c.
In each case, if the call fails another function is called that frees all
of the kdata and dpage information in the chunks array. So this
information should not be freed in radeon_cs_parser_init as well.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr> Reviewed-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Marek Olšák [Mon, 19 Mar 2012 02:09:33 +0000 (03:09 +0100)]
drm/radeon/kms: compute GPU addresses correctly on evergreen
There are also two fixes:
- In DRAW_INDEX_2, we read idx_value, but should have read idx+1.
- When correcting SQ_VTX_CONSTANT_WORD1_0.SIZE, we should subtract
the offset.
Signed-off-by: Marek Olšák <maraeo@gmail.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 20 Mar 2012 06:59:29 +0000 (06:59 +0000)]
drm/usb: move usb support into a separate module
In order to satisfy all the various Kconfig options between
USB and DRM, we need to split the USB code out into a separate module
and export symbols to it.
This fixes build problems in -next reported by sfr.
Dave Airlie [Tue, 14 Dec 2010 21:14:24 +0000 (07:14 +1000)]
drm/udl: initial UDL driver (v4)
This is an initial drm/kms driver for the displaylink devices.
Supports fb_defio,
supports KMS dumb interface
supports 24bpp via conversion to 16bpp, hw can do this better.
supports hot unplug using new drm core features.
On an unplug, it disables connector polling, unplugs connectors
from sysfs, unplugs fbdev layer (using Kay's API), drops all the
USB device URBs, and call the drm core to unplug the device.
This driver is based in large parts on udlfb.c so I've licensed
it under GPLv2.
Dave Airlie [Mon, 20 Feb 2012 14:18:07 +0000 (14:18 +0000)]
drm: add core support for unplugging a device (v2)
Two parts to this, one is simple unplug from sysfs for the device node.
The second adds an unplugged state, if we have device opens, we
just set the unplugged state and return, if we have no device
opens we drop the drm device.
If after a lastclose we discover we are unplugged we then
drop the drm device.
v2: use an atomic for unplugged and wrap it for users,
add checks on open + mmap + ioctl entry points.
Kay Sievers [Sat, 28 Jan 2012 19:57:46 +0000 (19:57 +0000)]
udlfb: remove sysfs framebuffer device with USB .disconnect()
The USB graphics card driver delays the unregistering of the framebuffer
device to a workqueue, which breaks the userspace visible remove uevent
sequence. Recent userspace tools started to support USB graphics card
hotplug out-of-the-box and rely on proper events sent by the kernel.
The framebuffer device is a direct child of the USB interface which is
removed immediately after the USB .disconnect() callback. But the fb device
in /sys stays around until its final cleanup, at a time where all the parent
devices have been removed already.
To work around that, we remove the sysfs fb device directly in the USB
.disconnect() callback and leave only the cleanup of the internal fb
data to the delayed work.
Ville Syrjälä [Tue, 13 Mar 2012 10:35:49 +0000 (12:35 +0200)]
drm: Use a flexible array member for blob property data
The blob property data is always allocated immediately after the object
header. No need for the extra indirection when accessing it, just use
a flexible array member.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:48 +0000 (12:35 +0200)]
drm: Handle drm_object_get() failures
Check drm_mode_object_get() return value everywhere.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:47 +0000 (12:35 +0200)]
drm: Make drm_crtc_convert_{umode, to_umode} static and constify their params
drm_crtc_convert_umode() and drm_crtc_convert_to_umode() are never
used outside drm_crtc.c, so make them static. Also make the input
mode structure const for both functions.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:46 +0000 (12:35 +0200)]
drm: Fix drm_mode_attachmode_crtc()
Change drm_mode_attachmode_crtc() to take an "all or nothing" approach.
If an error is returned, there are no side effects visible.
Also change the function to always duplicate the mode passed in.
Also change the function to not give up when it finds the first
connector without and encoder.
A simpler approach would be to just remove the function completely as
it's unused currently.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:45 +0000 (12:35 +0200)]
drm: Check CRTC viewport against framebuffer size
Make sure the requested CRTC viewport fits inside the
framebuffer.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:44 +0000 (12:35 +0200)]
drm: Check user mode against overflows
The internal mode representation drm_display_mode uses signed data
types. When converting the user mode to internal representation,
check that the unsigned values don't overflow the signed datatypes.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:43 +0000 (12:35 +0200)]
drm: Fix memory leak in drm_mode_setcrtc()
The mode passed to the .set_config() hook was never freed. The drivers
will make a copy of the mode, so simply free it when done.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:42 +0000 (12:35 +0200)]
drm: Make drm_mode_attachmode() void
drm_mode_attachmode() always returns 0. Change the return type to void.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:41 +0000 (12:35 +0200)]
drm: Check crtc x and y coordinates
The crtc x/y panning coordinates are stored as signed integers
internally. The user provides them as unsigned, so we should check
that the user provided values actually fit in the internal datatypes.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:40 +0000 (12:35 +0200)]
drm: Warn if mode to umode conversion overflows the destination types
When converting from a drm_display_mode to drm_mode_modeinfo, print a
warning if the the timings values don't fit into the __u16 datatype.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:39 +0000 (12:35 +0200)]
drm: Change drm_display_mode::type to unsigned
The drm_display_mode type is a bitmask so it should be unsigned.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Ville Syrjälä [Tue, 13 Mar 2012 10:35:38 +0000 (12:35 +0200)]
drm: Reject mode set with current fb if no current fb is bound
When doing a mode set with the special fb id -1, reject the mode set if
no fb is currently bound to the crtc.
Also remove the pointless list traversal to find the current crtc based
on the current crtc :)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 13 Mar 2012 10:16:06 +0000 (10:16 +0000)]
Merge branch 'drm-nouveau-next' of git://anongit.freedesktop.org/git/nouveau/linux-2.6 into drm-core-next
* 'drm-nouveau-next' of git://anongit.freedesktop.org/git/nouveau/linux-2.6: (57 commits)
drm/nouveau: map first page of mmio early and determine chipset earlier
drm/nvd0/disp: disconnect encoders before reprogramming them
drm/nvd0/disp: move syncs/magic setup to or mode_set
drm/nouveau/dp: account for channel coding overhead in link training
drm/nvd0/disp: fix dcb sor link matching in supervisor handler
drm/nvd0/disp: initial implementation of displayport
drm/nouveau/dp: make dp dpms function common, call from sor code instead
drm/nv50/hwsq: some nv92 fixes
drm/nouveau/dp: move all nv50/sor-specific code out of nouveau_dp.c
drm/nouveau/dp: make functions for executing various bios tables
drm/nouveau/pm: fix oops if chipset has no pm support at all
drm/nouveau/bios: rework vbios shadowing
drm/nouveau/bios: attempt acpi rom fetch before pcirom
drm/nvd0/disp: attempt to handle more than 2 crtcs if possible
drm/nvc0/vram: get part count from PUNITS
drm/nv40/pm: fix fanspeed regression
drm/nouveau/pm: several fixes for nvc0 memory timings
drm/nvc0/pm: restrict pll mode to clocks that can actually use it
drm/nouveau/dp: fix bad comparison in dp_link_train_commit()
drm/nouveau/mxm: call mxmi to determine revision before calling mxms
...
Ben Skeggs [Mon, 12 Mar 2012 01:42:20 +0000 (11:42 +1000)]
drm/nvd0/disp: move syncs/magic setup to or mode_set
NVIDIA appear to do these around the same place they do the MODE_CTRL
methods, and for DP at least we need to bash some extra bits in "syncs"
to keep EVO happy.
It's a bit of a guess as to the 6/8bpc, but i have no better idea yet.
Xi Wang [Fri, 3 Feb 2012 16:13:55 +0000 (11:13 -0500)]
drm/nouveau/dp: fix bad comparison in dp_link_train_commit()
The comparison (lpre == DP_TRAIN_PRE_EMPHASIS_9_5) is always false:
lpre is initialized as (lane & 0x0c) >> 2, which is at most 3, while
DP_TRAIN_PRE_EMPHASIS_9_5 is defined as (3 << 3).
Signed-off-by: Xi Wang <xi.wang@gmail.com> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Ben Skeggs [Wed, 1 Feb 2012 23:33:05 +0000 (09:33 +1000)]
drm/nouveau/mxm: call mxmi to determine revision before calling mxms
There's a HP laptop out there where the MXM version in the VBIOS doesn't
match what the ACPI implementation is expecting. These tables will accept
0x00 to MXMS to return latest version, but *only* if MXMI has been called
first..
Ben Skeggs [Thu, 12 Jan 2012 05:34:54 +0000 (15:34 +1000)]
drm/nv50-nvc0/vm: support unsnooped system memory
v2 (Emil Velikov <emil.l.velikov@gmail.com>):
- Fixed a regression on certain nv50 IGP due to not passing the correct
target type to nv50_vm_addr()
Signed-off-by: Ben Skeggs <bskeggs@redhat.com> Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> Tested-by: Johannes Obermayr <johannesobermayr@gmx.de>
Ben Skeggs [Tue, 24 Jan 2012 00:24:05 +0000 (10:24 +1000)]
drm/nouveau/pm: track mr2 for gddr3
There's some "extended" GDDR3 chipsets out there with EMRS2 settings that
change the layout of MRS/EMRS1 bitmaps.. Sigh.. Still need to track down
how exactly we're supposed to handle this.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com> Signed-off-by: Martin Peres <martin.peres@labri.fr>
Ben Skeggs [Mon, 23 Jan 2012 03:12:09 +0000 (13:12 +1000)]
drm/nouveau/pm: introduce ram reclocking helper
This will probably result in more lines of code, however, we're going to
have at least 3 slightly different implementations of this very soon and
I'd rather keep the ram reclocking logic separate from the hw specifics.
DDR2/DDR3/GDDR3 implemented thus far, others will be added as necessary.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com> Signed-off-by: Martin Peres <martin.peres@labri.fr>
Ben Skeggs [Tue, 17 Jan 2012 11:10:58 +0000 (21:10 +1000)]
drm/nouveau/pm: calculate memory timings at perflvl creation time
Statically generating the PFB register and MR values for each timing set
turns out to be insufficient. There's at least one (so far) known piece
of information which effects MR values which is stored in the perflvl
entry on some chipsets (and in another table on later ones), which is
disconnected from the timing table entries.
After this change we will generate a timing set based on an input clock
frequency instead, and have this data stored in the performance level
data.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com> Signed-off-by: Martin Peres <martin.peres@labri.fr>
Roy Spliet [Mon, 9 Jan 2012 05:23:07 +0000 (15:23 +1000)]
drm/nouveau/pm: implement DDR2/DDR3/GDDR3/GDDR5 MR generation and validation
Roy Spliet:
- Implement according to specs
- Simplify
- Make array for mc latency registers
Martin Peres:
- squash and split all the commits from Roy
- rework following Ben Skeggs comments
- add a form of timings validation
- store the initial timings for later use
Ben Skeggs
- merge slightly modified tidy-up patch with this one
- remove perflvl-dropping logic for the moment
Signed-off-by: Roy Spliet <r.spliet@student.tudelft.nl> Signed-off-by: Martin Peres <martin.peres@labri.fr> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Ben Skeggs [Mon, 16 Jan 2012 12:34:03 +0000 (22:34 +1000)]
drm/nouveau/pm: restructure bios table parsing
It turns out we need access to some additional information in various VBIOS
tables to handle PFB memory timings correctly.
Rather than hack in parsing of the new stuff in some kludgy way, I've
restructured the VBIOS parsing to be more primitive, so we can use them in
more flexible ways in the future.
The perflvl->timing association code is disabled for the moment until it can
be reworked. We don't use this stuff yet anyway, so no harm done.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com> Signed-off-by: Martin Peres <martin.peres@labri.fr>
Module parameter descriptions don't take a trailing \n, otherwise it
breaks formatting of modinfo's output. Also remove trailing space.
Signed-off-by: Jean Delvare <jdelvare@suse.de> Cc: David Airlie <airlied@linux.ie> Cc: Ben Skeggs <bskeggs@redhat.com> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Roy Spliet [Fri, 25 Nov 2011 14:52:22 +0000 (15:52 +0100)]
drm/nouveau/pm: improve memory timing generation
- Rename several VBIOS entries to closer match the real world
- Add the missing 0x100238 and 0x100240 register values
- Parse bit 14 of the VBIOS timing table
- "Magic value" -> tCWL, fixing some minor bugs in the process
- Also name a few more by their name rather than their number.
- Some values seem to be dependent on the memory type. Fix
Edits by Martin Peres <martin.peres@labri.fr>:
- this is a squash commit
- reworked for fixing some style issues
Signed-off-by: Roy Spliet <r.spliet@student.tudelft.nl> Signed-off-by: Martin Peres <martin.peres@labri.fr> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Ben Skeggs [Tue, 13 Dec 2011 02:46:18 +0000 (12:46 +1000)]
drm/nvc0: add initial memory type detection
Uses only the VBIOS tables, from what I can tell this is what NVIDIA do
too, I was able to change the detected memory type by modifying this table
on a NVC1 chipset.
Ben Skeggs [Tue, 13 Dec 2011 01:57:55 +0000 (11:57 +1000)]
drm/nv50: hopefully handle the DDR2/DDR3 memtype detection somewhat better
M version 2 appears to have a table with some form of memory type info
available.
NVIDIA appear to ignore the table information except for this DDR2/DDR3
case (which has the same value in 0x100714). My guess is this is due to
some of the supported memory types not being represented in the table.
Ben Skeggs [Mon, 12 Dec 2011 23:45:25 +0000 (09:45 +1000)]
drm/nv50: add memory type detection
DDR1/DDR[23] confirmed on NVA8 (see note about DDR3 in source) by changing
the value and watching the binary driver's behaviour.
GDDR3/4 values confirmed on a NV96 via the same method above. That GDDR4
is present is interesting, as far as I can see no boards using it were ever
released.
GDDR5 value is based on VBIOS images of known GDDR5 boards.
Ben Skeggs [Mon, 12 Dec 2011 13:03:14 +0000 (23:03 +1000)]
drm/nv20-nv40: add memory type detection
NV20/NV30 is partially educated guesswork at this point, based on any
information around about available memory types and a horribly unspeakable
amount of vbios image scouring. I'm not entirely certain the GDDR3 define
is correct, I have not spotted a single vbios with that value yet (though
it is mentioned in some 1218-using nv4x vbios), but there are reports that
some nv3x did use it..
NV40(100914) confirmed by switching an NV49 to DDR1/DDR2 values and making
sure that the binary driver behaviour showed it had detected DDR1/DDR2
instead of GDDR3 before dying horribly.
NV40(100474) confirmed by doing much the same task as above on an NV44,
except this was *much* easier as changing the values didn't seem to have
any noticable effect on the memory controller aside from changing the
binary driver's behaviour.
projects / linux-imx.git / log