FB-access-wrapper?

[Thomas Winischhofer]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Torrey Lyons wrote:
| At 8:31 PM -0800 1/14/05, Keith Packard wrote:
|
|> Around 19 o'clock on Jan 14, Thomas Winischhofer wrote:
|>
|>>  But I have found a MUCH better solution. I don't refresh the areas when
|>>  RefreshArea is called. Instead, I create a box that "covers" all
|>>  submitted boxes and update this "large" box in the blockhandler.
|>
|>
|> The miext/shadow code delays until the block handler already; is there
|> something wrong with that code?
|
|
| It is worth noting that on modern Macs anyway, the block handler fires
| far too often for this purpose. That is we found that updating each time
| you hit the block handler is inefficient as you end up updating far too
| often. The generic rootless code instead ensures that a sufficient
| amount of time has passed before doing another update of all the damaged
| regions.


I assume that you refer to my posting above.

This is Linux-only driver since it requires a special kernel module.

And as regards Keith's question: No, generally not. But the shadow code
submits very small areas and refreshing small areas is painful due to
the USB transfer overhead. Now I "collect" the areas by building a
"large" area covering all the small areas RefreshArea() got in the meantime.

If there is no screen update, the block handler doesn't do anything.

Thomas

- --
Thomas Winischhofer
Vienna/Austria
thomas AT winischhofer DOT net          http://www.winischhofer.net/
twini AT xfree86 DOT org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD4DBQFB6PTUzydIRAktyUcRAmQoAJi6Hgjbr6pdags5SFvxeKbSCC8lAJ4sdEpp
DJkPN69RnBrzY4TngvaVwg==
=+HPn
-----END PGP SIGNATURE-----