bug-gnu-utils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: gawk-3.1.6: fmtspcl test fail in Renesas SuperH


From: Nobuhiro Iwamatsu
Subject: Re: gawk-3.1.6: fmtspcl test fail in Renesas SuperH
Date: Tue, 16 Jun 2009 09:07:14 +0900

Hi,

2009/6/16 Aharon Robbins <address@hidden>:
> Greetings. Re this:
>
>> From: Nobuhiro Iwamatsu <address@hidden>
>> Subject: gawk-3.1.6: fmtspcl test fail in Renesas SuperH
>> Date: Mon, 15 Jun 2009 14:13:52 +0900
>> To: address@hidden
>
> Please be sure to use address@hidden for gawk bugs so that I see them...
>
Oh, sorry.

>> Hi,
>>
>> A test of fmtspcl of gawk fails on Renesas  SuperH.
>
> What kind of system is this? x86 architecture?

No, This CPU of Renesas Technology Corp.
http://america.renesas.com/fmwk.jsp?cnt=superh_family_landing.jsp&fp=/products/mpumcu/superh_family

>
>> I use gcc-.4.3.2 and glibc2.9.
>> I attached test results.
>>
>> Best regards,
>>  Nobuhiro
>
> Your results show that the two lines for the %d format are not in the output.
> I suspect something strange in glibc, but it's very hard to tell.
>
> If this is your only difference, then you probably don't need to worry
> about it too much.

I understood the cause of this problem.
SH of GCC is enable -ffinte-math-only. This is default config in SH of GCC.
And gawk use strtol in glibc. strtol check Nan and +- inf.
If we treated NaN in glibc definitely, I need to add  -mieee option
and had to compile it.

Thanks,
 Nobuhiro
-- 
Nobuhiro Iwamatsu




reply via email to

[Prev in Thread] Current Thread [Next in Thread]