This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH v2 2/2] Add tests for the long double version of ecvt and fcvt
On Tue, 06 Nov 2018, Florian Weimer wrote:
>* Gabriel F. T. Gomes:
>
>> + { L (0.0), 0, 1, "" },
>
>Do you need the L macro? I would expect you can write the literal and
>let the compiler handle the conversation.
Hrm, I don't need it, indeed.
>> #if DBL_MANT_DIG == 53
>> - { 0x1p-1074, 3, -323, "494" },
>> - { -0x1p-1074, 3, -323, "494" },
>> + { L (0x1p-1074), 3, -323, "494" },
>> + { L (-0x1p-1074), 3, -323, "494" },
>> #endif
>
>The preprocessor condition seems wrong for the long double version of
>the tests. Does something like DBL_MANT_DIG even exist for IBM long
>double?
I thought it wouldn't matter, because, if these numbers fit in a double
typed variable, they would also fit in long double. Anyhow, now that you
asked, I see that it would be very misleading.
I'm testing a new version.