On 3/22/24 05:29, Jeff Law wrote: >> Another option is to enable -fsched-pressure which should help with >> this issue. > In theory we're already using that by default -- it's part of what makes > me so curious to understand what's going on. We are actually using it in practice :-) Its the default for RISC-V port since Aug of last year. -Vineet
- Re: [gcc-15 0/3] RISC-V improve stack/array access by const... Jeff Law
- Re: [gcc-15 0/3] RISC-V improve stack/array access by ... Vineet Gupta
- scheduler queue flush (was Re: [gcc-15 0/3] RISC-V imp... Vineet Gupta
- Re: scheduler queue flush (was Re: [gcc-15 0/3] RI... Jeff Law
- Re: scheduler queue flush (was Re: [gcc-15 0/3... Vineet Gupta
- Re: scheduler queue flush (was Re: [gcc-15... Jeff Law
- Re: scheduler queue flush Vineet Gupta
- Re: scheduler queue flush (was Re: [g... Richard Biener
- Re: scheduler queue flush (was Re... Jeff Law
- Re: scheduler queue flush (was Re... Vineet Gupta
- Re: scheduler queue flush (was Re: [gcc-15... Jeff Law