Re: [PULL 08/13] softfloat: Fix BAD_SHIFT from normalizeFloatx80Subnormal

2020-04-11 Thread Alex Bennée
Peter Maydell writes: > On Fri, 10 Apr 2020 at 16:17, Richard Henderson > wrote: >> Although why Alex didn't add his own R-b to my patch when merging it to his >> branch, I don't know. > > I think this is one of those areas where different submaintainers > have different work practices. Person

Re: [PULL 08/13] softfloat: Fix BAD_SHIFT from normalizeFloatx80Subnormal

2020-04-10 Thread Aleksandar Markovic
18:14 Pet, 10.04.2020. Peter Maydell је написао/ла: > But I think some submaintainers don't bother to add r-by tags > for things they review in the process of assembling their > tree because they see it as implicit in the process. > I think that was precisely the case in this patch. May I wish

Re: [PULL 08/13] softfloat: Fix BAD_SHIFT from normalizeFloatx80Subnormal

2020-04-10 Thread Peter Maydell
On Fri, 10 Apr 2020 at 16:17, Richard Henderson wrote: > Although why Alex didn't add his own R-b to my patch when merging it to his > branch, I don't know. I think this is one of those areas where different submaintainers have different work practices. Personally I distinguish "did I actually re

Re: [PULL 08/13] softfloat: Fix BAD_SHIFT from normalizeFloatx80Subnormal

2020-04-10 Thread Aleksandar Markovic
17:17 Pet, 10.04.2020. Richard Henderson је написао/ла: > > On 4/10/20 2:38 AM, Aleksandar Markovic wrote: > > 17:55 Uto, 07.04.2020. Alex Bennée > > је написао/ла: > >> > >> From: Richard Henderson > > > >> > >> All other calls

Re: [PULL 08/13] softfloat: Fix BAD_SHIFT from normalizeFloatx80Subnormal

2020-04-10 Thread Richard Henderson
On 4/10/20 2:38 AM, Aleksandar Markovic wrote: > 17:55 Uto, 07.04.2020. Alex Bennée > је написао/ла: >> >> From: Richard Henderson > >> >> All other calls to normalize*Subnormal detect zero input before >> the call -- this is the

Re: [PULL 08/13] softfloat: Fix BAD_SHIFT from normalizeFloatx80Subnormal

2020-04-10 Thread Aleksandar Markovic
17:55 Uto, 07.04.2020. Alex Bennée је написао/ла: > > From: Richard Henderson > > All other calls to normalize*Subnormal detect zero input before > the call -- this is the only outlier. This case can happen with > +0.0 + +0.0 = +0.0 or -0.0 + -0.0 = -0.0, so return a zero of > the correct sign.