|
From: | David Hildenbrand |
Subject: | Re: [PATCH v1 00/20] s390x/tcg: Implement Vector enhancements facility and switch to z14 |
Date: | Wed, 5 May 2021 12:55:19 +0200 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.8.1 |
On 30.09.20 16:55, David Hildenbrand wrote:
This series adds support for the "Vector enhancements facility" and bumps the qemu CPU model tp to a stripped-down z14. I yet have to find a way to get more test coverage - looks like some of the functions aren't used anywhere yet (e.g., VECTOR FP MAXIMUM), writing unit tests to cover all functions and cases is just nasty. But I might be wrong - I'm planning to at least test basic functionality of all new added instructions. I have to make excessive use of c macros again to cover different element sizes (32/64/128bit). Any advise to clean things up are welcome. This is based on: "[PATCH v2 0/9] s390x/tcg: Implement some z14 facilities" "[PATCH v2 00/10] softfloat: Implement float128_muladd" Based-on: 20200928122717.30586-1-david@redhat.com Based-on: 20200925152047.709901-1-richard.henderson@linaro.org
Hi Richard,I'll have a new series ready soonish. I did what you suggested and started generating random (valid) Vector FP instructions with (valid) random parameters, executed on randomly generated vectors. It's looking pretty good by now.
I'll still have to see to which degree I can address feedback on "softfloat: Implement float128_(min|minnum|minnummag|max|maxnum|maxnummag)"
Anyhow, I'd need your "softfloat: Implement float128_muladd" series -- any idea when you might get to continue working on that? Thanks!
-- Thanks, David / dhildenb
[Prev in Thread] | Current Thread | [Next in Thread] |