[Libre-soc-isa] [Bug 1057] split out all int/fp min/max ops into their own RFC ls013

[bugzilla-daemon at libre-soc.org]

https://bugs.libre-soc.org/show_bug.cgi?id=1057

--- Comment #26 from Jacob Lifshay <programmerjake at gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #25)
> (In reply to Jacob Lifshay from comment #24)
> > https://git.libre-soc.org/?p=openpower-isa.git;a=shortlog;
> > h=e3b46c6ec7911d08b8b0c4cb3c286c3786dae2ef
> > 
> > commit e3b46c6ec7911d08b8b0c4cb3c286c3786dae2ef
> > Author: Jacob Lifshay <programmerjake at gmail.com>
> > Date:   Mon Apr 24 23:49:19 2023 -0700
> > 
> >     replace min/max[su][.] with minmax[.]
> 
> looks really good
> 
> > commit 56697594f5f676339780d113a0be5ecf947c09e5
> > Author: Jacob Lifshay <programmerjake at gmail.com>
> > Date:   Mon Apr 24 23:47:20 2023 -0700
> > 
> >     add unofficial and comment2 columns to minor_19.csv
> 
> hmmm be *really* careful. don't take any shortcuts: always
> run the *entire* test suite if you make these kinds of
> severely-low-level changes.

i ran the tests after writing all the code but before splitting it into commits

> > commit 52434df9a3b7125137fe786821a598575c9f5edc
> > Author: Jacob Lifshay <programmerjake at gmail.com>
> > Date:   Mon Apr 24 23:40:35 2023 -0700
> > 
> >     fix minmax pseudo-code -- CR0 must not have lt/gt swapped
> 
> ah yeah i always get those wrong

yup, i needed to do the CR0 computation *before* swapping a/b for the min/max
> 
> @@ -180,7 +180,7 @@ class AVTestCase(TestAccumulatorBase):
>          e.intregs[3] = min(e.intregs[1], e.intregs[2])
> -        e.crregs[0] = 0x4
> +        e.crregs[0] = 0x8  # r1 <u r2
>          self.add_case(Program(lst, bige

that was because the CR0 = cmp(RA, RB) changes had previously only been applied
to maxs, now, because of the new pseudo-code, they're applied to all minmax.
modes.

-- 
You are receiving this mail because:
You are on the CC list for the bug.