Jonas Paulsson 31393f1f98 [SystemZ::TTI] i8/i16 operands extension costs revisited ...

Three minor changes to these extra costs:

* For ICmp instructions, instead of adding 2 all the time for extending each
  operand, this is only done if that operand is neither a load or an
  immediate.

* The operands extension costs for divides removed, because we now use a high
  cost already for the divide (20).

* The costs for lhsr/ashr extra costs removed as this did not seem useful.

Review: Ulrich Weigand
https://reviews.llvm.org/D55053

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@347961 91177308-0d34-0410-b5e6-96231b3b80d8

2018-11-30 07:09:34 +00:00

..

cmp-ext-01.ll

[SystemZ] Adjust cost functions for subtargets that use LI + LOC instead of IPM

2018-09-14 06:46:55 +00:00

cmp-ext-02.ll

[SystemZ] Adjust cost functions for subtargets that use LI + LOC instead of IPM

2018-09-14 06:46:55 +00:00

cmp-mem.ll

[SystemZ::TTI] Improved cost values for comparison against memory.

2018-11-28 08:08:05 +00:00

cmp-tofp-scalar.ll

[SystemZ::TTI] Recognize the higher cost of scalar i1 -> fp conversion

2018-11-01 09:05:32 +00:00

cmp-tofp.ll

[SystemZ::TTI] Accurate costs for i1->double vector conversions

2018-11-01 09:01:51 +00:00

cmpsel.ll

[SystemZ] TargetTransformInfo cost functions implemented.

2017-04-12 11:49:08 +00:00

divrem-const.ll

[SystemZ] Improve handling and cost estimates of vector integer div/rem

2018-10-25 21:47:22 +00:00

divrem-pow2.ll

[SystemZ] Improve handling and cost estimates of vector integer div/rem

2018-10-25 21:47:22 +00:00

divrem-reg.ll

[SystemZ::TTI] i8/i16 operands extension costs revisited

2018-11-30 07:09:34 +00:00

ext-load.ll

[SystemZ] TargetTransformInfo cost functions implemented.

2017-04-12 11:49:08 +00:00

fp-arith.ll

[SystemZ] Add support for IBM z14 processor (2/3)

2017-07-17 17:42:48 +00:00

fp-cast.ll

[SystemZ::TTI] Improve accuracy of costs for vector fp <-> int conversions

2018-11-12 15:32:27 +00:00

int-arith.ll

[SystemZ] Improve handling and cost estimates of vector integer div/rem

2018-10-25 21:47:22 +00:00

int-cast.ll

[SystemZ] TargetTransformInfo cost functions implemented.

2017-04-12 11:49:08 +00:00

int-operands-extcost.ll

[SystemZ::TTI] i8/i16 operands extension costs revisited

2018-11-30 07:09:34 +00:00

intrinsic-cost-crash.ll

…

intrinsics.ll

[SystemZ::TTI] Return zero cost for scalar load/store connected with a bswap.

2018-11-28 07:52:34 +00:00

lit.local.cfg

[SystemZ] Fix target specific tests

2017-04-12 17:14:46 +00:00

load_store.ll

[SystemZ] TargetTransformInfo cost functions implemented.

2017-04-12 11:49:08 +00:00

load-ptr-cmp-fold.ll

[SystemZ] Take better care when computing needed vector registers in TTI.

2018-10-10 07:36:27 +00:00

logical.ll

[SystemZ::TTI] i8/i16 operands extension costs revisited

2018-11-30 07:09:34 +00:00

memop-folding-int-arith.ll

[SystemZ::TTI] i8/i16 operands extension costs revisited

2018-11-30 07:09:34 +00:00

numvectorregs.ll

[SystemZ] Take better care when computing needed vector registers in TTI.

2018-10-10 07:36:27 +00:00

scalar-cmp-cmp-log-sel.ll

[SystemZ] Adjust cost functions for subtargets that use LI + LOC instead of IPM

2018-09-14 06:46:55 +00:00

shuffle.ll

[SystemZ] TargetTransformInfo cost functions implemented.

2017-04-12 11:49:08 +00:00

vectorinstrs.ll

[SystemZ] TargetTransformInfo cost functions implemented.

2017-04-12 11:49:08 +00:00