Re: [PATCH 1/2] sparc: remove ceil, floor, trunc sparc specific implementations

[Carlos O'Donell <carlos at redhat dot com>]

On 08/01/2016 05:50 PM, Aurelien Jarno wrote:
> On 2016-08-01 14:44, David Miller wrote:
>> From: Aurelien Jarno <aurelien@aurel32.net>
>> Date: Mon,  1 Aug 2016 15:24:22 +0200
>>
>>> The ceil, floor and trunc functions on sparc do not fully follow the
>>> standard and trigger an inexact exception when presented a value which
>>> is not an integer. Since glibc 2.24 this causes a few tests to fail,
>>> for instance:
>>
>> Carlos, this specific patch fixes most of the math failures on sparc,
>> could we get this into the release?
> 
> I think you should ask Adhemerval instead.

I have the baton right now to do the branch cut.

If you have tested these patches, and it looks like you have.

And if these patches are restricted to just sparc, and it looks like they are.

Then I'm happy for you to check them in.

However, I will not delay the release for them.

I'm doing some last minute checklist checking before I cut the branch.

So if you want to commit what you have, then please go ahead.

>> Aurelien, it looks like we have the same exact issue with nearbyint on
>> sparc, right?
> 
> I don't see the issue on nearbyint here. What is the issue exactly?
> 
>> And that leaves the fdim failures, which are probably a similar
>> problem.
> 
> Yes, I am currently working on the fdim issue, to see if we should apply
> the same strategy or fix it.

I will not hold up the release for this.

Cheers,
Carlos.