Re: [PING^3][PATCH] optimize costly division in rtx_cost

2020-02-13 Thread Richard Biener
On Thu, Feb 13, 2020 at 2:46 PM Alexander Monakov wrote: > > Ping^3. OK. > On Sun, 5 Jan 2020, Alexander Monakov wrote: > > > Hi, > > > > I noticed there's a costly signed 64-bit division in rtx_cost on x86 as > > well as > > any other target where UNITS_PER_WORD is implemented like TARGET_64BI

[PING^3][PATCH] optimize costly division in rtx_cost

2020-02-13 Thread Alexander Monakov
Ping^3. On Sun, 5 Jan 2020, Alexander Monakov wrote: > Hi, > > I noticed there's a costly signed 64-bit division in rtx_cost on x86 as well > as > any other target where UNITS_PER_WORD is implemented like TARGET_64BIT ? 8 : > 4. > It's also evident that rtx_cost does redundant work for a SET r