"Michael Kerrisk" <mtk.manpages@xxxxxxxxxxxxxx> writes:
> On Fri, Jul 25, 2008 at 11:30 AM, Andreas Jaeger <aj@xxxxxxxxxx> wrote:
>> "Michael Kerrisk" <mtk.manpages@xxxxxxxxxxxxxx> writes:
>>
>>> [CC+= Andries]
>>>
>>> On Mon, Jul 21, 2008 at 12:23 PM, Michael Kerrisk
>>> <mtk.manpages@xxxxxxxxxxxxxx> wrote:
>>>> [Ooops -- wrong list CCed on previous mail; and this time, I'll update
>>>> the draft number]
>>>>
>>>> Andreas,
>>>>
>>>> The latest version of the page is below.
>>
>> This one looked ok to me,
>
> Thanks Andreas.
>
> So I've been testing all of the math functions lately (on glibc 2.8),
> and I discover the following:
>
> a) on error, many functions set errno AND raise an exception (fetestexcept()).
> b) on error, a few functions set errno but DON"T raise an exception
> (fetestexcept()).
> c) on error, a few functions DON'T set errno BUT DO raise an exception
> (fetestexcept()).
> d) on error, a very few functions pursue a mixture of all of the
> above, depending on the error.
>
> My math_error.7 page currently implies that all functions should do
> a). Clearly I'll need to amend that.
>
> But the main question is, should I raise a pile of glibc bugs for the
> functions in cases b), c), and d)?
I suggest to discuss on the libc-alpha mailing list first.
We have not done the exception handling properly. errno setting should
be correct, we have tests for these,
Andreas
--
Andreas Jaeger, Director Platform/openSUSE, aj@xxxxxxx
SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
Maxfeldstr. 5, 90409 Nürnberg, Germany
GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126
Attachment:
pgpyurhwiZRok.pgp
Description: PGP signature
