Re: [Mesa-dev] [PATCH] gallivm: remove unused float coord wrapping for aos sampling

2018-12-07 Thread Jose Fonseca
On 07/12/2018 01:28, srol...@vmware.com wrote: From: Roland Scheidegger AoS sampling tries to use integers for coord wrapping when possible, as it should be faster. However, for AVX, this was suboptimal, because only floats can use 8x32bit vectors, whereas integers have to be split into 4x32bit

[Mesa-dev] [PATCH] gallivm: remove unused float coord wrapping for aos sampling

2018-12-06 Thread sroland
From: Roland Scheidegger AoS sampling tries to use integers for coord wrapping when possible, as it should be faster. However, for AVX, this was suboptimal, because only floats can use 8x32bit vectors, whereas integers have to be split into 4x32bit vectors. (I believe part of why it was slower wa