Re: [PATCH] rs6000: Enable generate const through pli+pli+rldimi

2022-08-11 Thread Segher Boessenkool
Hi! On Thu, Aug 11, 2022 at 08:52:49PM +0800, Jiufu Guo wrote: > Segher Boessenkool writes: > > On Wed, Aug 10, 2022 at 03:11:23PM +0800, Jiufu Guo wrote: > >> @@ -9659,7 +9659,7 @@ (define_split > >> ;; When non-easy constants can go in the TOC, this should use > >> ;; easy_fp_constant predica

Re: [PATCH] rs6000: Enable generate const through pli+pli+rldimi

2022-08-11 Thread Jiufu Guo via Gcc-patches
Hi, Segher Boessenkool writes: > Hi! > > On Wed, Aug 10, 2022 at 03:11:23PM +0800, Jiufu Guo wrote: >> As mentioned in PR106550, since pli could support 34bits immediate, we could >> use less instructions(3insn would be ok) to build 64bits constant with pli. >> >> For example, for constant 0x0

Re: [PATCH] rs6000: Enable generate const through pli+pli+rldimi

2022-08-10 Thread Segher Boessenkool
Hi! On Wed, Aug 10, 2022 at 03:11:23PM +0800, Jiufu Guo wrote: > As mentioned in PR106550, since pli could support 34bits immediate, we could > use less instructions(3insn would be ok) to build 64bits constant with pli. > > For example, for constant 0x020805006106003, we could generate it with: >

[PATCH] rs6000: Enable generate const through pli+pli+rldimi

2022-08-10 Thread Jiufu Guo via Gcc-patches
Hi, As mentioned in PR106550, since pli could support 34bits immediate, we could use less instructions(3insn would be ok) to build 64bits constant with pli. For example, for constant 0x020805006106003, we could generate it with: asm code1: pli 9,101736451 (0x6106003) sldi 9,9,32 paddi 9,9, 213000