search for: revectorise

Displaying 4 results from an estimated 4 matches for "revectorise".

2013 Nov 15
2
[LLVMdev] [PATCH] Add a Scalarize pass
...e be true in the other direction, for targets without vector support? (b) The situation you describe isn't the one that applies to llvmpipe. In llvmpipe the vectors are nice, known widths that are under the driver's own control. We certainly don't want to scalarise and revectorise llvmpipe IR on x86_64, or on powerpc with Altivec/VSX. The original code is already well vectorised for those targets. (And also for ARM NEON I expect.) In the llvmpipe case, codegen's type legaliser already makes a good decision about what to scalarise and what not to scalaris...
2013 Nov 15
0
[LLVMdev] [PATCH] Add a Scalarize pass
...ther direction, for targets without vector support? > > (b) The situation you describe isn't the one that applies to llvmpipe. > In llvmpipe the vectors are nice, known widths that are under the > driver's own control. We certainly don't want to scalarise and > revectorise llvmpipe IR on x86_64, or on powerpc with Altivec/VSX. > The original code is already well vectorised for those targets. > (And also for ARM NEON I expect.) > > In the llvmpipe case, codegen's type legaliser already makes a good > decision about what to scalarise and...
2013 Nov 14
0
[LLVMdev] [PATCH] Add a Scalarize pass
On Nov 14, 2013, at 2:32 PM, Richard Sandiford <rsandifo at linux.vnet.ibm.com> wrote: > Richard Sandiford <rsandifo at linux.vnet.ibm.com> writes: >> Are you worried that adding it to PMB will increase compile time? >> The pass exits very early for any target that doesn't opt-in to doing >> scalarisation at the IR level, without even looking at the function.
2013 Nov 14
2
[LLVMdev] [PATCH] Add a Scalarize pass
Richard Sandiford <rsandifo at linux.vnet.ibm.com> writes: > Are you worried that adding it to PMB will increase compile time? > The pass exits very early for any target that doesn't opt-in to doing > scalarisation at the IR level, without even looking at the function. As an alternative, adding Scalarizer and InstCombine passes to SystemZPassConfig::addIRPasses() would probably