FWIW, nv50/nvc0 also don't have pow instructions (looks like the last
nvidia card with a pow instruction was the GeForce FX series, first
released in 2003... even nv40 doesn't have it). The nouveau codegen
will happily lower pow's though. However would be nice to avoid the
double-work.
On Fri, Jul
There already is an option: gl_shader_compiler_options::EmitNoPow, but
st/mesa never sets it to GL_TRUE.
Marek
On Sat, Jul 19, 2014 at 12:38 AM, Matt Turner wrote:
> On Fri, Jul 18, 2014 at 3:25 PM, Marek Olšák wrote:
>> BTW, I have just noticed r600g also lowers POW and there is no mention
>>
On Fri, Jul 18, 2014 at 3:25 PM, Marek Olšák wrote:
> BTW, I have just noticed r600g also lowers POW and there is no mention
> of POW in the SI ISA guide either, so I don't think radeons would
> benefit from an optimization pass that adds POW instructions.
Sure. We could of course add an option (
BTW, I have just noticed r600g also lowers POW and there is no mention
of POW in the SI ISA guide either, so I don't think radeons would
benefit from an optimization pass that adds POW instructions.
Marek
On Sat, Jul 19, 2014 at 12:15 AM, Marek Olšák wrote:
> All Gallium drivers must support POW
All Gallium drivers must support POW, but some drivers like r300-r500
fragment shaders lower it to LG2+MUL+EX2.
Marek
On Sun, Jul 13, 2014 at 7:50 PM, Thomas Helland
wrote:
> Hi all,
>
> I've been looking at some shaders from Portal recently.
> A lot of them seem to hand-roll a pow-function with
On Wed, Jul 16, 2014 at 4:14 PM, Thomas Helland
wrote:
> 2014-07-13 20:13 GMT+02:00 Matt Turner :
>>
>> On Sun, Jul 13, 2014 at 10:50 AM, Thomas Helland
>> wrote:
>> > I've considered writing an algebraic optimization to convert
>> > this into an ir_binop_pow. If my understanding is correct the b
2014-07-13 20:13 GMT+02:00 Matt Turner :
>
> On Sun, Jul 13, 2014 at 10:50 AM, Thomas Helland
> wrote:
> > I've considered writing an algebraic optimization to convert
> > this into an ir_binop_pow. If my understanding is correct the backend
> > will then implement this in a similar fashion as abo
On Sun, Jul 13, 2014 at 10:50 AM, Thomas Helland
wrote:
> I've considered writing an algebraic optimization to convert
> this into an ir_binop_pow. If my understanding is correct the backend
> will then implement this in a similar fashion as above if it does not
> have a native pow() instruction.
Hi all,
I've been looking at some shaders from Portal recently.
A lot of them seem to hand-roll a pow-function with
log2, multiply, and exp2, like this:
r3.x = log2( r3.x );
r3.y = log2( r3.y );
r3.z = log2( r3.z );
r3.xyz = r3.xyz * vd2.zzz;
r3.x = exp2( r3.x );
r3.y = exp2( r3.y );
r3.z = exp2(
