RE: [PATCH][GCC][AArch64] Add support for SVE stack clash probing [patch (2/7)]

2018-10-09 Thread Tamar Christina
org; nd ; James Greenhalgh > ; Richard Earnshaw > ; Marcus Shawcroft > > Subject: Re: [PATCH][GCC][AArch64] Add support for SVE stack clash probing > [patch (2/7)] > > Tamar Christina writes: > > Hi Richard, > > > > Here's the updated patch with all the feedbac

Re: [PATCH][GCC][AArch64] Add support for SVE stack clash probing [patch (2/7)]

2018-09-28 Thread Richard Sandiford
Tamar Christina writes: > Hi Richard, > > Here's the updated patch with all the feedback processed. > > I have also run the compile tests through with -mabi=ilp32 as well. > > Ok for trunk? OK. Thanks for your patience through all the reviews. Richard

Re: [PATCH][GCC][AArch64] Add support for SVE stack clash probing [patch (2/7)]

2018-09-28 Thread Tamar Christina
Hi Richard, Here's the updated patch with all the feedback processed. I have also run the compile tests through with -mabi=ilp32 as well. Ok for trunk? Thanks, Tamar The 09/27/2018 12:11, Richard Sandiford wrote: > > It turns out the testsuite didn't have a case in it which would cause a > >

Re: [PATCH][GCC][AArch64] Add support for SVE stack clash probing [patch (2/7)]

2018-09-27 Thread Richard Sandiford
> It turns out the testsuite didn't have a case in it which would cause a > significant enough spill to enter the loop. After creating one I noticed a > bug > in the loop and fixed it. > > The loops are now > > .cfi_startproc > mov x15, sp > cntbx16, all, mul #11

Re: [PATCH][GCC][AArch64] Add support for SVE stack clash probing [patch (2/7)]

2018-09-27 Thread Tamar Christina
Hi All, It turns out the testsuite didn't have a case in it which would cause a significant enough spill to enter the loop. After creating one I noticed a bug in the loop and fixed it. The loops are now .cfi_startproc mov x15, sp cntbx16, all, mul #11

Re: [PATCH][GCC][AArch64] Add support for SVE stack clash probing [patch (2/7)]

2018-09-26 Thread Tamar Christina
Hi Richard, I've added a new loop that should also exit early as described in my previous email. An example sequence is: .cfi_startproc mov x15, sp cntbx16, all, mul #11 add x16, x16, 304 .cfi_def_cfa_register 15 cmp x16, 61440

Re: [PATCH][GCC][AArch64] Add support for SVE stack clash probing [patch (2/7)]

2018-09-20 Thread Tamar Christina
Hi Richard, The 09/11/2018 16:20, Richard Sandiford wrote: > Tamar Christina writes: > >> > + > >> > + /* No probe leave. */ > >> > + ASM_OUTPUT_INTERNAL_LABEL (asm_out_file, loop_end_lab); > >> > + return ""; > >> > >> With the CFA stuff and constant load, I think this works out as: > >>

Re: [PATCH][GCC][AArch64] Add support for SVE stack clash probing [patch (2/7)]

2018-09-11 Thread Richard Sandiford
Tamar Christina writes: >> > + >> > + /* No probe leave. */ >> > + ASM_OUTPUT_INTERNAL_LABEL (asm_out_file, loop_end_lab); >> > + return ""; >> >> With the CFA stuff and constant load, I think this works out as: >> >> - >> # 12 insns >> mov

Re: [PATCH][GCC][AArch64] Add support for SVE stack clash probing [patch (2/7)]

2018-09-07 Thread Tamar Christina
Hi Richard, Here's the updated patch and some comments inline below. An example sequence is: .cfi_startproc mov x15, sp cntbx16, all, mul #11 add x16, x16, 304 .cfi_def_cfa_register 15 .SVLPSRL0: cmp x16, 65536 b.lt

Re: [PATCH][GCC][AArch64] Add support for SVE stack clash probing [patch (2/7)]

2018-08-28 Thread Richard Sandiford
I'll leave the AArch64 maintainers to review, but some comments. Tamar Christina writes: > diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c > index > 06451f38b11822ea77323438fe8c7e373eb9e614..e7efde79bb111e820f4df44a276f6f73070ecd17 > 100644 > ---