[Mesa-dev] [PATCH 3/3] glsl: relax loop unrolling restrictions

[Timothy Arceri]

On 22/06/17 18:46, Samuel Pitoiset wrote:
> Here's a shader-db report:
> 
> https://pastebin.com/raw/QBMnF2pv
> 
> This doesn't sound like a total win actually...

I'm surprised to see this. I thought we pretty much unrolled everything 
already, although maybe that was what happened after NIR unrolling.

I've done a full shader-db run and I'm getting very different results 
from you:

https://pastebin.com/XRH7Vbvv

There are still a little mixed but they look much more positive.


> 
> For now, patches 1&2 are:
> 
> Reviewed-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
> 
> On 06/21/2017 12:12 PM, Timothy Arceri wrote:
>> The main reason these restriction exist is because glsl the loop
>> unrolling pass is super slow with large loops.
>>
>> be5f27a84d0d fixed things so that expression trees were counted
>> against the limit, however it left the limit as
>> max_iterations * 5 which is actually fine for most shaders but
>> probably over conservative.
>>
>> This change relaxes the limit to allow more loops to unroll in the
>> Unigine Superposition benchmark.
>>
>> Results from Unigine Superposition @ 1920x1080 - High - Fullscreen
>> On radeonsi (RX480)
>>
>> Before:
>>
>> Average: 28.20 Frames Per Second
>>
>> After:
>>
>> Average: 28.60 Frames Per Second
>> ---
>>
>>   The increase is small but it seems to be consistent, I would be
>>   interested in the results if others were interested in testing it.
>>
>>   src/compiler/glsl/loop_unroll.cpp | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/src/compiler/glsl/loop_unroll.cpp 
>> b/src/compiler/glsl/loop_unroll.cpp
>> index bc377df..64ebf0f 100644
>> --- a/src/compiler/glsl/loop_unroll.cpp
>> +++ b/src/compiler/glsl/loop_unroll.cpp
>> @@ -350,21 +350,21 @@ loop_unroll_visitor::visit_leave(ir_loop *ir)
>>      /* Don't try to unroll loops that have zillions of iterations 
>> either.
>>       */
>>      if (iterations > max_iterations)
>>         return visit_continue;
>>      /* Don't try to unroll nested loops and loops with a huge body.
>>       */
>>      loop_unroll_count count(&ir->body_instructions, ls, options);
>>      bool loop_too_large =
>> -      count.nested_loop || count.nodes * iterations > max_iterations 
>> * 5;
>> +      count.nested_loop || count.nodes * iterations > max_iterations 
>> * 10;
>>      if (loop_too_large && !count.unsupported_variable_indexing &&
>>          !count.array_indexed_by_induction_var_with_exact_iterations)
>>         return visit_continue;
>>      /* Note: the limiting terminator contributes 1 to 
>> ls->num_loop_jumps.
>>       * We'll be removing the limiting terminator before we unroll.
>>       */
>>      assert(ls->num_loop_jumps > 0);
>>      unsigned predicted_num_loop_jumps = ls->num_loop_jumps - 1;
>>