[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v5 23/35] target/arm: Implement SVE floating-poi
From: |
Richard Henderson |
Subject: |
Re: [Qemu-devel] [PATCH v5 23/35] target/arm: Implement SVE floating-point convert precision |
Date: |
Tue, 26 Jun 2018 21:02:33 -0700 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.8.0 |
On 06/26/2018 03:44 AM, Peter Maydell wrote:
> A comment to the effect that the SVE fp-to-fp conversion
> routines always use IEEE format halfprec (ie ignore FPCR.AHP)
> would be helpful.
Ok.
> Are you sure we have the FPCR.FZ16 handling right here? That
> is, do we need the same "use the not-fp16 fpstatus pointer,
> and temporarily clear the flush flag for the fp16 end of
> the conversion" behaviour that we have in vfp_fcvt_f16_to_f32
> and friends ? The pseudocode FPConvertSVE() calls FPConvert(),
> which is the "ignore FZ16" codepath I think. The test case would
> be (eg) a conversion where the input f16 is denormal and
> FPCR.FZ == 1: this should not do the flush-input-to-zero, right?
Yes, I read it the same way. I guess both my and Alex's RISU
test cases didn't exercise this?
r~
- [Qemu-devel] [PATCH v5 20/35] target/arm: Implement SVE Floating Point Unary Operations - Unpredicated Group, (continued)
- [Qemu-devel] [PATCH v5 20/35] target/arm: Implement SVE Floating Point Unary Operations - Unpredicated Group, Richard Henderson, 2018/06/20
- [Qemu-devel] [PATCH v5 19/35] target/arm: Implement SVE FP Fast Reduction Group, Richard Henderson, 2018/06/20
- [Qemu-devel] [PATCH v5 22/35] target/arm: Implement SVE floating-point trig multiply-add coefficient, Richard Henderson, 2018/06/20
- [Qemu-devel] [PATCH v5 21/35] target/arm: Implement SVE FP Compare with Zero Group, Richard Henderson, 2018/06/20
- [Qemu-devel] [PATCH v5 23/35] target/arm: Implement SVE floating-point convert precision, Richard Henderson, 2018/06/20
- [Qemu-devel] [PATCH v5 24/35] target/arm: Implement SVE floating-point convert to integer, Richard Henderson, 2018/06/20
- [Qemu-devel] [PATCH v5 25/35] target/arm: Implement SVE floating-point round to integral value, Richard Henderson, 2018/06/20
- [Qemu-devel] [PATCH v5 27/35] target/arm: Implement SVE MOVPRFX, Richard Henderson, 2018/06/20
- [Qemu-devel] [PATCH v5 26/35] target/arm: Implement SVE floating-point unary operations, Richard Henderson, 2018/06/20