- 22 Mar, 2011 40 commits
-
-
Thomas Schlichter authored
with an PAT-enabled kernel, when using uvesafb or vesafb, these drivers will create uncached-minus PAT entries for the framebuffer memory because they use ioremap() (not the *_cache or *_wc variants). When the framebuffer memory intersects with the video RAM used by Xorg, the complete video RAM will be mapped uncached-minus what results in a serve performance penalty. Here are the correct MTRR entries created by uvesafb: schlicht@netbook:~$ cat /proc/mtrr reg00: base=0x000000000 ( 0MB), size= 2048MB, count=1: write-back reg01: base=0x06ff00000 ( 1791MB), size= 1MB, count=1: uncachable reg02: base=0x070000000 ( 1792MB), size= 256MB, count=1: uncachable reg03: base=0x0d0000000 ( 3328MB), size= 16MB, count=1: write-combining And here are the problematic PAT entries: schlicht@netbook:~$ sudo cat /sys/kernel/debug/x86/pat_memtype_list PAT memtype list: write-back @ 0x0-0x1000 uncached-minus @ 0x6fedd000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee3000-0x6fee4000 uncached-minus @ 0x6fee3000-0x6fee4000 uncached-minus @ 0x6fee3000-0x6fee4000 uncached-minus @ 0xd0000000-0xe0000000 <-- created by xserver-xorg uncached-minus @ 0xd0000000-0xd1194000 <-- created by uvesafb uncached-minus @ 0xf4000000-0xf4009000 uncached-minus @ 0xf4200000-0xf4400000 uncached-minus @ 0xf5000000-0xf5010000 uncached-minus @ 0xf5100000-0xf5104000 uncached-minus @ 0xf5400000-0xf5404000 uncached-minus @ 0xf5404000-0xf5405000 uncached-minus @ 0xf5404000-0xf5405000 uncached-minus @ 0xfed00000-0xfed01000 Therefore I created the attached patch for uvesafb which uses ioremap_wc() to create the correct PAT entries, as shown below: schlicht@netbook:~$ sudo cat /sys/kernel/debug/x86/pat_memtype_list PAT memtype list: write-back @ 0x0-0x1000 uncached-minus @ 0x6fedd000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee2000-0x6fee3000 uncached-minus @ 0x6fee3000-0x6fee4000 uncached-minus @ 0x6fee3000-0x6fee4000 uncached-minus @ 0x6fee3000-0x6fee4000 write-combining @ 0xd0000000-0xe0000000 write-combining @ 0xd0000000-0xd1194000 uncached-minus @ 0xf4000000-0xf4009000 uncached-minus @ 0xf4200000-0xf4400000 uncached-minus @ 0xf5000000-0xf5010000 uncached-minus @ 0xf5100000-0xf5104000 uncached-minus @ 0xf5400000-0xf5404000 uncached-minus @ 0xf5404000-0xf5405000 uncached-minus @ 0xf5404000-0xf5405000 uncached-minus @ 0xfed00000-0xfed01000 This results in a performance gain, objectively measurable with e.g. x11perf -comppixwin10 -comppixwin100 -comppixwin500: 1: x11perf_xaa.log 2: x11perf_xaa_patched.log 1 2 Operation -------- ---------------- ----------------- 124000.0 202000.0 ( 1.63) Composite 10x10 from pixmap to window 3340.0 24400.0 ( 7.31) Composite 100x100 from pixmap to window 131.0 1150.0 ( 8.78) Composite 500x500 from pixmap to window You can see the serve performance gain when composing larger pixmaps to window. The patches replace the ioremap() function with the variant matching the mtrr- parameter. To create "write-back" PAT entries, the ioremap_cache() function must be called after creating the MTRR entries, and the ioremap_cache() region must completely fit into the MTRR region, this is why the MTRR region size is now rounded up to the next power-of-two. Signed-off-by:Thomas Schlichter <thomas.schlichter@web.de> Signed-off-by:
Paul Mundt <lethal@linux-sh.org>
-
axel lin authored
Current implementation calls of_iounmap for par->fbc twice in error path. In the case of goto out_unmap_dac, we should call of_iounmap for par->dac. Signed-off-by:
Axel Lin <axel.lin@gmail.com> Acked-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
Jean Delvare authored
Some Radeon cards have an I2C-based thermal sensor chip connected to the "monid" I2C bus. Set the I2C probing class of this bus properly so that hwmon drivers can detect devices on it and bind to them. This closes kernel.org bug #26172. We exclude PPC for the time being, as Benjamin doesn't want us to mess up with them without explicit testing, and there is no evidence that this change is needed for them either. Reported-by:
Alexander Goomenyuk <emerg.reanimator@gmail.com> Signed-off-by:
Jean Delvare <khali@linux-fr.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Dan Carpenter authored
backlight_device_register() returns an ERR_PTR. It doesn't return NULL. Signed-off-by:
Dan Carpenter <error27@gmail.com> Signed-off-by:
-
axel lin authored
The size calculation is done incorrectly in request_mem_region because it should include both the start and end (end - start + 1). Signed-off-by:
-
axel lin authored
The size calculation is done incorrectly here because it should include both the start and end (end - start + 1). Signed-off-by:
-
Ondrej Zary authored
Use "new" start address register 0x69 (bits 16-20) instead of "old" 0x31 (bits 16-17) and 0x51 (bits 18-19). This is needed for panning to work correctly on Trio3D/2X cards (and does no harm on other ones). Signed-off-by:
Ondrej Zary <linux@rainbow-software.org> Acked-by:
Ondrej Zajicek <santiago@crfreenet.org> Signed-off-by:
-
Ondrej Zary authored
Enable Data Transfer Position Control (DTPC). This is needed at least on Virge/DX to correctly display at higher pixclocks. Signed-off-by:
-
Ondrej Zary authored
Enable pixel multiplexing in 15/16bpp modes when pixclock is over 115MHz on Trio3D (86C365) cards to fix artifacts on the left side of screen. Signed-off-by:
-
Ondrej Zary authored
Add support for S3 Trio3D (86C365) cards to s3fb driver. Tested with one 4MB card. Signed-off-by:
-
Ondrej Zary authored
Maximize virtual vertical framebuffer size during init to allow fast scrolling (accelerated by panning). Signed-off-by:
-
David Miller authored
This allows the driver to work in multi-domain PCI configurations. Signed-off-by:
-
David Miller authored
This allows the driver to work in multi-domain PCI configurations. Signed-off-by:
-
David Miller authored
This allows the driver to work in multi-domain PCI configurations. Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Instead of just plain NULL. Signed-off-by:
-
David Miller authored
Instead of just plain NULL. Signed-off-by:
-
David Miller authored
Instead of just plain NULL. Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
Signed-off-by:
-
David Miller authored
And use vga_{r,w}(). Signed-off-by: -
David Miller authored
Instead of magic register location constants. Signed-off-by:
-
Aaro Koskinen authored
Add support for ZGI Z9 DDR2 POST. The init sequence is from XGI's xgifb driver. Tested with ARM board using a PCI card with XGI Z9s and 32 MB DDR2 memory. After a cold reset the POST succeeds. Signed-off-by:
Aaro Koskinen <aaro.koskinen@iki.fi> Cc: Thomas Winischhofer <thomas@winischhofer.net> Signed-off-by:
-
Aaro Koskinen authored
Move DDR2 register setting code into separate subroutines. No changes in functionality. Signed-off-by:
-
Aaro Koskinen authored
Detect the XGI Z9 RAM type as "documented" by the XGI's xgifb driver. Signed-off-by:
-
Aaro Koskinen authored
Z7 and Z9 have the same PCI ID, so additional checking is needed to detect Z9. The method was "documented" in XGI's xgifb driver. Signed-off-by:
-
Aaro Koskinen authored
Move XGI POST RAM type detection into a separate subroutine to make further code changes easier. No changes in functionality Signed-off-by:
-
Aaro Koskinen authored
Currently there is no indication if R/W test fails during POST. This can happen e.g. when user plugs in a card with unsupported type of memory. Since the driver will be unusable, it's better to fail the whole POST if the memory cannot be configured properly. Signed-off-by:
-
Kirill A. Shutemov authored
Signed-off-by:
Kirill A. Shutemov <kirill@shutemov.name> Signed-off-by:
-
