Re: Artifacts left on screen while dragging icons around
From: Mo (marabo82_at_hotmail.com)
Date: 12/21/04
- Next message: David Liao \(MS\): "Re: PCI driver gets loaded too many times"
- Previous message: Raymond Rogers: "Re: PCI driver gets loaded too many times"
- Messages sorted by: [ date ] [ thread ]
Date: 20 Dec 2004 16:50:37 -0800
Hi Kermit,
i hope u can help me with this altough ur post is quite old
u said u hade problem with SH-based display driver, my driver crashed
sometime and i get screens of death... i checked
timming/registers/classed and so far
nothing came up
now, what SH-based processor u were using? what BSP?
i am using PFMDS11 from Renesas and they sort of gave up on windows CE.
i read that u solved ur problem by dealing with the frame buffer!!
the frame buffer is not used in my display driver... that wired ha? i
have it defined as m_pVirtualFrameBuffer, and its not being set
anywhere..
do u have access to the driver?
my env is (CE4.2, SH3 7727, PFMDS11, 16 bits per pixel, 320*240
landscape screen)
we are performing Rotate on our screen from portrait to landscape...
any help?
thanks!!
Kermit wrote:
> Hi all,
>
> It's been awhile since I've written in to the group. I guess that is
a good
> thing on my end because development has been going well. However, I
do
> apologize for being absent for so long.
>
> I am having a problem with artifacts being left on the display after
> dragging
> icons around, and I'm totally stuck on it. I've dealt with this issue
> one time before on an SH-based platform. That time it turned out that
my
> frame buffer resided in a cacheable memory area. Therefore, writes to
the
> frame buffer were being cached and the screen wasn't always being
updated
> correctly.
>
> This platform is ARM-based, a Sharp LH7A400 with a 922T-core is the
CPU.
> I have made sure that the frame buffer is non-cacheable by using the
> following
> call to map it:
>
> frame_buffer = MmMapIoSpace(phy_addr, size, FALSE);
>
> I know that the screen artifacts are _not_ due to a hardware problem.
I've
> verified this by using the Remote Zoom-in tool to check the contents
of
> the frame buffer. The artifacts are in fact in the frame buffer as
evidenced
> by the fact that they show up on the bitmap downloaded using this
tool.
>
> I've also tried different formats (16 and 8 Bpp). Both of them show
the same
> artifacts. I've even moved the frame buffer from system SDRAM into
the CPU's
> embedded SRAM just in case there was a timing issue but that didn't
work
> either.
>
> The display driver is very simple, it's all based on the GPE classes
and
> I've
> disabled any and all of the emulation accelerated blits. For
instance,
> ENABLE_TEXT_EMULATION, ENABLE_FILL_EMULATION, etc.
>
> I've searched all of the past posts in the newsgroup and didn't find
> anything
> useful, so I'd like to ask if anyone has any advice on where to go
from
> here.
>
> Any and all comments are greatly appreciated.
>
> Thanks,
> --kermit
>
> --
> It wasn't easy being Greazy ....but it was interesting.
- Next message: David Liao \(MS\): "Re: PCI driver gets loaded too many times"
- Previous message: Raymond Rogers: "Re: PCI driver gets loaded too many times"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
|
