8e1a464e6a
[CodeGen][X86] Expand UADDSAT to NOT+UMIN+ADD
...
Followup to D56636, this time handling the UADDSAT case by expanding
uadd.sat(a, b) to umin(a, ~b) + b.
Differential Revision: https://reviews.llvm.org/D56869
llvm-svn: 352409
2019-01-28 19:19:09 +00:00
a131e4e296
[TTI] Add generic UADDSAT/USUBSAT costs
...
Add generic costs calculation for UADDSAT/USUBSAT intrinsics, this fallbacks to using generic costs for uadd_with_overflow/usub_with_overflow + a select.
Differential Revision: https://reviews.llvm.org/D56907
llvm-svn: 352044
2019-01-24 12:27:10 +00:00
d3b86b79fa
Reapply "[CodeGen][X86] Expand USUBSAT to UMAX+SUB, also for vectors"
...
Related to https://bugs.llvm.org/show_bug.cgi?id=40123 .
Rather than scalarizing, expand a vector USUBSAT into UMAX+SUB,
which produces much better code for X86.
Reapplying with updated SLPVectorizer tests.
Differential Revision: https://reviews.llvm.org/D56636
llvm-svn: 351219
2019-01-15 18:43:41 +00:00
5885eec35a
Revert "[CodeGen][X86] Expand USUBSAT to UMAX+SUB, also for vectors"
...
This reverts commit r351125.
I missed test changes in an SLPVectorizer test, due to the cost model
changes. Reverting for now.
llvm-svn: 351129
2019-01-14 22:18:39 +00:00
8e9a8432a8
[CodeGen][X86] Expand USUBSAT to UMAX+SUB, also for vectors
...
Related to https://bugs.llvm.org/show_bug.cgi?id=40123 .
Rather than scalarizing, expand a vector USUBSAT into UMAX+SUB,
which produces much better code for X86.
Differential Revision: https://reviews.llvm.org/D56636
llvm-svn: 351125
2019-01-14 21:43:30 +00:00
d824f99a6c
[X86] Add ADD/SUB SSAT/USAT vector costs (PR40123)
...
Costs for real SSE2 instructions
llvm-svn: 350295
2019-01-03 11:38:42 +00:00
55ea89305d
[X86] Add ADD/SUB SSAT/USAT cost tests (PR40123)
...
llvm-svn: 350293
2019-01-03 11:29:24 +00:00
Nikita Popov