discuss-gnuradio
[Top][All Lists]
Advanced

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

Re: [Discuss-gnuradio] Make Check Failures under FC9 on Intel


From: Don Ward
Subject: Re: [Discuss-gnuradio] Make Check Failures under FC9 on Intel
Date: Tue, 1 Jul 2008 17:06:57 -0400

Eric Blossom wrote:

On Wed, Jun 18, 2008 at 01:53:33PM -0400, Suprin, Charles E. wrote:
Eric,

Looks like the current compiler in

g++ --version
g++ (GCC) 4.3.0 20080428 (Red Hat 4.3.0-8)
Copyright (C) 2008 Free Software Foundation, Inc.

I saw the earlier traffic that discussed a compile problem from around
June 2 2008.  However now we pass compile and fail the checks.


Charles

I supect a compiler bug.
Here's the list of bugs fixed from 4.3.0 to 4.3.1
http://gcc.gnu.org/bugzilla/buglist.cgi?bug_status=RESOLVED&resolution=FIXED&target_milestone=4.3.1

I suggest testing with gcc 4.3.1 and see if the problem is still
there. If it's gone, problem solved, otherwise there's more sleuthing to do.

I ran into this today with gcc 4.3.1 on Debian (testing). Line 101 of qa_gr_fxpt_nco.cc

 new_nco.sincos ((gr_complex*)new_block, SIN_COS_BLOCK_SIZE);

only fills the first 1696 members of new_block (though SIN_COS_BLOCK_SIZE = 100000). Adding printf statements to new_nco.sincos or compiling qa_gr_fxpt_nco.cc with -O1 makes the problem go away.

-- Don W.

-----Original Message-----
From: Eric Blossom [mailto:address@hidden
Sent: Wednesday, June 18, 2008 1:43 PM
To: Suprin, Charles E.
Cc: address@hidden
Subject: Re: [Discuss-gnuradio] Make Check Failures under FC9 on Intel

On Wed, Jun 18, 2008 at 11:18:47AM -0400, Suprin, Charles E. wrote:
> Hello,
>
> I am attempting to install and compile the svn version 8598 version
of
> gnuradio on a Fedora 9 Pentium 4.  It compiles fine but throws errors
> during make check.
>
> !!!FAILURES!!!
>
> Test Results:
>
> Run:  90   Failures: 1   Errors: 0
>
> 1) test: qa_gr_fxpt_nco::t1 (F) line: 101 qa_gr_fxpt_nco.cc
>
> double equality assertion failed
>
> - Expected: -0.857728600502014
>
> - Actual  : 0
>
> - Delta   : 9.99999974737875e-06
>
>
>
> It seems there is an issue both in how it computes the actual and the
> Delta.
>
>
>
> A few quick Google searches and scanning the mailing list did not
show
> any one else having the issue.
>
>
>
> Is anyone else having this issue?
> Charles

I believe I saw this or something like it when I tested with gcc 4.3.0
a few months ago.  I didn't worry too much about it since the list of
known bugs for the compiler was pretty long, and this was the first
release of 4.3.0  What version of g++ is shipping with F9?

  $ g++ --version

Eric


_______________________________________________
Discuss-gnuradio mailing list
address@hidden
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


--------------------------------------------------------------------------------



No virus found in this incoming message.
Checked by AVG.
Version: 8.0.100 / Virus Database: 270.4.0/1507 - Release Date: 6/18/2008 7:09 AM





reply via email to

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