[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 08/19] target-alpha: use new float64_unordered()
From: |
Richard Henderson |
Subject: |
Re: [Qemu-devel] [PATCH 08/19] target-alpha: use new float64_unordered() function |
Date: |
Thu, 14 Apr 2011 09:27:10 -0700 |
User-agent: |
Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.15) Gecko/20110307 Fedora/3.1.9-0.39.b3pre.fc14 Thunderbird/3.1.9 |
On 04/14/2011 08:39 AM, Peter Maydell wrote:
>> Exponent underflow and disabled:
>> Supply +0, no exception delivered to user.
>>
>> Exponent underflow and enabled:
>> Supply +-MIN denorm, Underflow delivered to user.
>>
>> Footnote 3, Overflow and Underflow have priority over Inexact.
>
> Thanks. Does "no exception delivered to user" mean also
> "and do not set FPCR bit UNF" ?
Yes.
> The reason I thought it might set Inexact is that I was looking
> at page 4-79, which says:
>
> "If both the UNFD (underflow disable) bit and the UNDZ (underflow
> to zero) bit are set in the FPCR, the implementation sets the
> result of an underflow operation to a true zero result. The
> zeroing of a denormal result by UNDZ must also be treated as an
> inexact result."
Hum. It looks like we can choose between these results then,
depending on the intersection of the FPCR disable bits, and
the per-instruction trapping mode bits (see section 4.7.7.2).
I *think* what would be best for Alpha is if, within softfloat,
both conditions are signaled, and then we can filter the result
that is actually needed via helper_fp_exc_raise? It's hard to
say without actually doing the work...
Unfortunately, I suspect that the Correct result on real HW
also depends on the OS completion handler, and I know that at
least for Linux that code was written before UNDZ was added.
So I don't know if even real HW produces the correct result
when considering Underflow priority over Inexact.
r~
- Re: [Qemu-devel] [PATCH 15/19] target-arm: fix wrong usage of floatx80_eq_quiet(), (continued)
- [Qemu-devel] [PATCH 08/19] target-alpha: use new float64_unordered() function, Aurelien Jarno, 2011/04/12
- Re: [Qemu-devel] [PATCH 08/19] target-alpha: use new float64_unordered() function, Peter Maydell, 2011/04/14
- Re: [Qemu-devel] [PATCH 08/19] target-alpha: use new float64_unordered() function, Richard Henderson, 2011/04/14
- Re: [Qemu-devel] [PATCH 08/19] target-alpha: use new float64_unordered() function, Peter Maydell, 2011/04/14
- Re: [Qemu-devel] [PATCH 08/19] target-alpha: use new float64_unordered() function,
Richard Henderson <=
- Re: [Qemu-devel] [PATCH 08/19] target-alpha: use new float64_unordered() function, Peter Maydell, 2011/04/14
- Re: [Qemu-devel] [PATCH 08/19] target-alpha: use new float64_unordered() function, Richard Henderson, 2011/04/14
[Qemu-devel] [PATCH 11/19] softfloat: rename float*_eq() into float*_eq_quiet(), Aurelien Jarno, 2011/04/12
[Qemu-devel] [PATCH 19/19] target-mips: don't hardcode softfloat exception bits, Aurelien Jarno, 2011/04/12
[Qemu-devel] [PATCH 13/19] softfloat: move float*_eq and float*_eq_quiet, Aurelien Jarno, 2011/04/12
[Qemu-devel] [PATCH 18/19] target-mips: simplify FP comparisons, Aurelien Jarno, 2011/04/12
[Qemu-devel] [PATCH 09/19] target-mips: use new float*_unordered*() functions, Aurelien Jarno, 2011/04/12
[Qemu-devel] [PATCH 16/19] target-alpha: fix wrong usage of float64_eq_quiet(), Aurelien Jarno, 2011/04/12
[Qemu-devel] [PATCH 10/19] target-i386: fix CMPUNORDPS/D and CMPORDPS/D instructions, Aurelien Jarno, 2011/04/12