> I think kmalloc() is the best thing for vgb. Vfb is just meant to be a sample
> driver, so the limitation to 128 KB is not that important. It's more important
> that you can actually mmap() it from user space, which is not possible with
> vmalloc().
Alan pointed out it can be done. I have seen the code in other driver to
make it possible.
> > to also show a good example of the use of par. All fbdev driver writers
>
> I'm afraid there's not much `par' to show in vfb, unless you invent some fake
> registers that have to be set up... Hmm, why not do that?
Why not? Just for the fun of it.
> Another thing: to avoid flicker and to avoid expensive PCI bus accesses, it can
> be useful to check whether you really have to write a register value. If your
> `par' looks sufficiently like the real hardware registers, you can define
>
> static inline __set_reg(unsigned long reg, u32 val, u32 *oldval)
> {
> if (val != *oldval) {
> real_set_reg(reg, val);
> *oldval = val);
> }
> }
> #define set_reg(reg, val) __set_reg(reg, par->val, &oldpar->val)
>
> and use
>
> oldpar = fb_info->par;
> set_reg(REG_XX1, reg_xx1);
> set_reg(REG_XX2, reg_xx2);
> set_reg(REG_XX3, reg_xx3);
> set_reg(REG_XX4, reg_xx4);
>
> This is also useful for accel engine setup for accelerated rectcopy/rectfill.
Meta registers :) This would be useful to emulate a graphics context
switch for cards that lack it.
Codito, ergo sum - "I code, therefore I am"
James Simmons (o_
fbdev/gfx developer (o_ (o_ //\
http://www.linux-fbdev.org (/)_ (/)_ V_/_
http://linuxgfx.sourceforge.net
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Feb 15 2000 - 21:00:16 EST