linux-fbdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* Framebuffer console performance
@ 2005-03-24 22:14 Gary Thomas
  2005-04-15  4:57 ` Antonino A. Daplas
  0 siblings, 1 reply; 3+ messages in thread
From: Gary Thomas @ 2005-03-24 22:14 UTC (permalink / raw)
  To: linux-fbdev-devel

I have a non-standard framebuffer (i.e. I wrote the driver,
based on the au1000 driver) running on 2.4.27.  When I run
virtual consoles on it, I see sluggish scroll performance.
I get quite acceptable performance when the system first
comes up, e.g. during the system initialization messages,
but not long afterwards the scrolling slows down.  It can
get continually worse, eventually ending up at a crawl.

This is a memory limited embedded system, but I do not
have any swapping going on.

Ideas or hints where I might look?

Thanks for your time.

n.b. I'm not subscribed to this list, so please CC my direct
email address when replying.

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Framebuffer console performance
  2005-03-24 22:14 Framebuffer console performance Gary Thomas
@ 2005-04-15  4:57 ` Antonino A. Daplas
       [not found]   ` <1113650790.10535.42.camel@hermes>
  0 siblings, 1 reply; 3+ messages in thread
From: Antonino A. Daplas @ 2005-04-15  4:57 UTC (permalink / raw)
  To: linux-fbdev-devel, Gary Thomas

On Friday 25 March 2005 06:14, Gary Thomas wrote:
> I have a non-standard framebuffer (i.e. I wrote the driver,
> based on the au1000 driver) running on 2.4.27.  When I run
> virtual consoles on it, I see sluggish scroll performance.
> I get quite acceptable performance when the system first
> comes up, e.g. during the system initialization messages,
> but not long afterwards the scrolling slows down.  It can
> get continually worse, eventually ending up at a crawl.

When the system starts up, the scroll mode is 'redraw'.  This mode is
acceptable, not too slow and not too fast either, whatever the hardware is.

After starting up, the scroll mode changes depending on the driver.  Most
probably, it is using ypan which is not too great on systems without
accelerated rectangle moves and framebuffer reads are slow.

Try setting vyres == yres using fbset and see if that helps.

Tony




-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Framebuffer console performance
       [not found]   ` <1113650790.10535.42.camel@hermes>
@ 2005-04-18 11:07     ` Gary Thomas
  0 siblings, 0 replies; 3+ messages in thread
From: Gary Thomas @ 2005-04-18 11:07 UTC (permalink / raw)
  To: adaplas; +Cc: linux-fbdev-devel

On Sat, 2005-04-16 at 05:26 -0600, Gary Thomas wrote:
> On Fri, 2005-04-15 at 12:57 +0800, Antonino A. Daplas wrote: 
> > On Friday 25 March 2005 06:14, Gary Thomas wrote:
> > > I have a non-standard framebuffer (i.e. I wrote the driver,
> > > based on the au1000 driver) running on 2.4.27.  When I run
> > > virtual consoles on it, I see sluggish scroll performance.
> > > I get quite acceptable performance when the system first
> > > comes up, e.g. during the system initialization messages,
> > > but not long afterwards the scrolling slows down.  It can
> > > get continually worse, eventually ending up at a crawl.
> > 
> > When the system starts up, the scroll mode is 'redraw'.  This mode is
> > acceptable, not too slow and not too fast either, whatever the hardware is.
> > 
> > After starting up, the scroll mode changes depending on the driver.  Most
> > probably, it is using ypan which is not too great on systems without
> > accelerated rectangle moves and framebuffer reads are slow.
> > 
> > Try setting vyres == yres using fbset and see if that helps.

Thanks for getting back to me on this.

This does seem to have been the problem.  I made a few tweaks to
my driver's setup and now it does not switch into slow mode (I actually
think it was going into __SCROLL_YMOVE mode instead of __SCROLL_YREDRAW)

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2005-04-18 11:07 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-03-24 22:14 Framebuffer console performance Gary Thomas
2005-04-15  4:57 ` Antonino A. Daplas
     [not found]   ` <1113650790.10535.42.camel@hermes>
2005-04-18 11:07     ` Gary Thomas

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).