GSoC 2019 : POSIX Compliance - Weekly Report

Joel Sherrill joel at rtems.org
Tue Jul 30 22:39:02 UTC 2019


On Tue, Jul 30, 2019 at 12:29 PM Vaibhav Gupta <vaibhavgupta40 at gmail.com>
wrote:

>
>
> On Tue, Jul 30, 2019 at 7:23 PM Joel Sherrill <joel at rtems.org> wrote:
>
>>
>>
>> On Tue, Jul 30, 2019 at 7:44 AM Vaibhav Gupta <vaibhavgupta40 at gmail.com>
>> wrote:
>>
>>>
>>>
>>> On Tue, Jul 30, 2019 at 3:25 AM Gedare Bloom <gedare at rtems.org> wrote:
>>>
>>>> On Wed, Jul 24, 2019 at 10:43 AM Vaibhav Gupta <
>>>> vaibhavgupta40 at gmail.com> wrote:
>>>> >
>>>> > Hello,
>>>> > Progress so far:
>>>> > - NDBM patch is pushed by newlib.
>>>> > - NDBM Testsuite is sent to devel ([PATCH V6] ndbm test suite), its
>>>> execution will be verified once newlib has required files. Although I have
>>>> tested it by manually generating required file using 'autoreconf -fvi' on
>>>> my local system.
>>>> >
>>>> > Blocker:
>>>> > - My pace got slow last week as my college started, I informed about
>>>> it on devel.
>>>> >
>>>> > Next:
>>>> > - Implement fenv for SPARC first, then move on to ARM, PPC and others.
>>>> >
>>>> What about fenv testcase?
>>>>
>>> I guess after Joel had a discussion in newlib, we will be switching code
>>> structure for fenv, so implementation of riscv will also need to be
>>> updated.
>>> .
>>> Hence, I will implement fenv for SPARC and make test case for same.
>>>
>>
>> The test case can be added now if you have a RISC-V toolchain.  I expect
>> the test code will
>> have to be conditionalized on whether the target architecture supports
>> the fenv.h methods.
>> This would be best implemented with  like AC_CHECK_FUNC and pick a method
>> like
>> fegetround() to probe for.
>>
> Or I can use AC_CHECK_HEADER like you used for NDBM Testsuite.
>

The current proposal is to always have non-functional stubs so that won't
work this time.

I'm hoping to work things out with Craig Howland and drop the code in.

--joel


> That way as you add support for other architectures, the test(s)
>> will automatically get enabled.
>>
>> The alternative is like how libdebugger tests have to list specific
>> architectures that are
>> supported. That requires manual intervention (but is the only option for
>> libdebugger).
>>
>>>
>>> -Vaibhav Gupta
>>>
>>>>
>>>> >
>>>> > Thank You
>>>> > Vaibhav Gupta
>>>> > _______________________________________________
>>>> > devel mailing list
>>>> > devel at rtems.org
>>>> > http://lists.rtems.org/mailman/listinfo/devel
>>>>
>>> _______________________________________________
>>> devel mailing list
>>> devel at rtems.org
>>> http://lists.rtems.org/mailman/listinfo/devel
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20190730/24f2cf3d/attachment.html>


More information about the devel mailing list