OSchip
/
llvm-project
mirror of https://github.com/THU-DSP-LAB/llvm-project.git
13
0
Fork 6
Code Issues Packages Projects Releases Wiki Activity
397,002 Commits 31 Branches 0 Tags 2.2 GiB
Commit Graph

7 Commits

Author SHA1 Message Date
Daniel Egger 98c2e4115d [ARM] Add lowering of uadd_sat to uq{add|sub}8 and uq{add|sub}16 
This follow the lead of https://reviews.llvm.org/D68974 to add lowering
of unsigned saturated addition/subtraction.

Differential Revision: https://reviews.llvm.org/D105413
 2021-07-11 15:58:11 +01:00
David Green 32556a9832 [ARM] Remove more unused check prefixes, NFC 2020-11-14 15:37:53 +00:00
David Green e6f313b380 [Codegen] Alter the default promotion for saturating adds and subs 
The default promotion for the add_sat/sub_sat nodes currently does:
    ANY_EXTEND iN to iM
    SHL by M-N
    [US][ADD|SUB]SAT
    L/ASHR by M-N

If the promoted add_sat or sub_sat node is not legal, this can produce code
that effectively does a lot of shifting (and requiring large constants to be
materialised) just to use the overflow flag. It is simpler to just do the
saturation manually, using the higher bitwidth addition and a min/max against
the saturating bounds. That is what this patch attempts to do.

Differential Revision: https://reviews.llvm.org/D68926

llvm-svn: 375211
 2019-10-18 09:47:48 +00:00
David Green fe2d15b39b [Codegen] Adjust saturation test. NFC. 
Add some extra sat tests and adjust some of the existing tests to use signext where it would naturally be.

llvm-svn: 375009
 2019-10-16 15:50:42 +00:00
David Green 7c30af8e65 Revert 374373: [Codegen] Alter the default promotion for saturating adds and subs 
This commit is not extending the promoted integers as it should. Reverting
whilst I look into the details.

llvm-svn: 374592
 2019-10-11 20:33:03 +00:00
David Green 94d379095a [Codegen] Alter the default promotion for saturating adds and subs 
The default promotion for the add_sat/sub_sat nodes currently does:
   1. ANY_EXTEND iN to iM
   2. SHL by M-N
   3. [US][ADD|SUB]SAT
   4. L/ASHR by M-N
If the promoted add_sat or sub_sat node is not legal, this can produce code
that effectively does a lot of shifting (and requiring large constants to be
materialised) just to use the overflow flag. It is simpler to just do the
saturation manually, using the higher bitwidth addition and a min/max against
the saturating bounds. That is what this patch attempts to do.

Differential Revision: https://reviews.llvm.org/D68643

llvm-svn: 374373
 2019-10-10 16:04:49 +00:00
David Green fcc9c4627e Add and adjust saturating tests. NFC 
This adds some extra testing to the existing [su][add/sub]_sat X86 and AArch64
tests and adds equivalent tests for ARM.

llvm-svn: 374169
 2019-10-09 14:17:38 +00:00