| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
I've no idea what this is for, it came from an ATI codedrop and it may
fix some laptop or production system, but it looks to me more like a dirty
hack than a true fix, as I have an M7 which I need to specify CRT for the
second head and it breaks mergedfb for me.
|
|
This reverts caaed927a07ffbac68b08246185ef93c1e7bb98c commit.
See https://bugs.freedesktop.org/show_bug.cgi?id=8137 .
|
|
|
|
* [ Bug 6309 ] radeon/r128 fails to read hsync/vsync rates when rang
* descriptor is missing
* [ Bug 5473 ] Blank screen with Radeon Mobility X700 (Acer Ferrari 4005)
* Reboot should not be needed for external CRT to function when connected
later on Ferrari 4000
* Modes need to be sorted, and sorted correctly
* Provide all resolutions (instead only the largest one) in MergedFB clone
mode without config file
* Bump up clock rate for ES1000 to allow higher resolution
|
|
|
|
Rather than publish a non-existant driver name for RN50, just don't bother
initializing it in the first place.
|
|
|
|
|
|
Caching DP_WRITE_MASK causes artifacts, e.g. with evas_xrender_x11_test.
|
|
|
|
RENDER accleration is disabled by default. It is strongly recommended that the
patch from bug #6772 and the 'exa-damagetrack branch' are merged in the xserver
before enabling RENDER acceleration. To enable RENDER acceleration, add the
following in xorg.conf:
Section "Device"
[...]
Option "AccelMethod" "exa"
Option "RenderAccel" "true"
EndSection
|
|
This is inspired by a patch in XFree86 from David Dawes, it isn't the
same as we have different gamma curves.
|
|
EXA hits these limits for some operations on offscreen pixmaps and GTPRO seems
to support the reduced limits.
|
|
EXA hits more optimized paths when it does not have to fallback because of
missing UTS/DFS.
|
|
- DRI: factor out to separate function and add comments on layout
- no DRI: factor out to separate function, maxScanlines is always equal to
ATIMach64MaxY
|
|
|
|
|
|
|
|
Fix corruption associated with the engine randomly not waiting for a copy
operation to commit its results.
|
|
Fix corruption when scaling 1-pixel wide pixmaps, in which case the first CPU
read of the framebuffer will return stale data.
This bug always manifests itself in mach64 EXA (to be committed soon), with
this patch mach64 EXA passes rendercheck.
|
|
|
|
|
|
Use the corresponding texture formats with R200 family.
Unfortunately, it looks like the R100 family can't accelerate these source
formats, so we have to fall back for them.
|
|
|
|
This is a workaround for ATI Radeon 7000 erratum F1.
|
|
OTOH, flush everything when the previous engine mode is unknown, and mark the
engine mode as unknown in a couple more cases.
|
|
|
|
|
|
As a result, quite a bit of code can be removed that attempted to deal with the
EXA offscreen memory disappearing, with varying success. Because the HW cursor
memory is now always immediately after the front buffer and before the back
buffer, this also fixes bug #6808.
As the HW cursor memory is now reserved statically with EXA and XAA, change
RADEONUseHWCursor() to check info->cursor instead of info->cursor_offset, for
which 0 might become valid with HW cursor enabled in the future.
Change info->cursor_offset from unsigned long to CARD32 as the corresponding
register is 32 bits wide, and to avoid issues with printf.
|
|
|
|
This will still break with 2048 lines...
|
|
Based on patch from Weixing Zhang, but still allows enabling it with Option
"RenderAccel".
|
|
|
|
|
|
|
|
|
|
|
|
fglrx appears to do this on r300 and Jerome Glisse has spent most of his
life tracking down what caused the 9800 to lockup. This is my attempt to
fix this. Please report if it works for you.
Signed-off-by: Dave Airlie <airlied@linux.ie>
|
|
|
|
|
|
|
|
Thanks to Matthieu Herrb for pointing out some of these.
|
|
x11perf -getimage numbers only increase by about 20-30% on my PowerBook with an
M9, but by about 100 times(!) with a PCIe X550. I suspect the former could
perform better with PCI as opposed to AGP transfers, which would also remove
the need to disable this by default with AGP.
|
|
|
|
Also make the formatting of the if statements consistent.
|
|
This allows getting higher depth buffer precision and stencil hardware
acceleration in depth 16 or saving some video RAM at the cost of these features
in depth 24.
|
|
Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
Signed-off-by: Dave Airlie <airlied@linux.ie>
|
