floating point operations in gcc 3.3.3, glibc 2.3.2

Kunjan Naik KNaik@ixiacom.com
Tue Jun 22 21:54:00 GMT 2004


>Can you check whether the .s files produced by 'powerpc-750-linux-gnu-gcc
-save-temps struct-ret-1.c' differ between the 
>good and bad toolchains?  And if so, send me both?

Target: ppc750
Toolchain 1: crosstool-0.28-rc19, gcc3.3.3 glibc-2.3.2
Toolchain 2: crosstool-0.28-rc19, gcc3.3.3 glibc-2.3.2 without
gcc-3.3.2-arm-softfloat.patch
Sample Testcase: modified struct-ret-1.c with printf statements (see
attached)
Complier flag: --static.

Both the toolchains output the same .s file for the testcase. 
When executed on target, toolchain 1 binary prints garbage and toolchain 2
binary prints expected output.
For toolchain 1, the testcase compares garbage values and succeeds or aborts
and for toolchain 2 the testcase
compares valid output and always succeeds.

Thanks
-kunjan
 


-----Original Message-----
From: Dan Kegel [mailto:dank@kegel.com] 
Sent: Tuesday, June 22, 2004 8:08 AM
To: Dan Kegel
Cc: Kunjan Naik; 'crossgcc@sources.redhat.com'
Subject: Re: floating point operations in gcc 3.3.3, glibc 2.3.2


Dan Kegel wrote:
> Daniel Kegel wrote:
> 
>> Kunjan Naik wrote:
>>
>>>>> I bet gcc-3.3.2-arm-softfloat.patch is the culprit.
>>>
>>> Indeed it is.
> 
> 
> I don't understand :-(
> I finally looked inside that patch, and it only affects 
> gcc/config/arm,
> there's no
> generic part.  How the heck could it cause ppc750 failures?
> 
> OK, now I'm curious.  I assume the testcase that was failing is 
> gcc/testsuite/gcc.c-torture/execute/struct-ret-1.c
> I'm going to build gcc-3.3.3 with and without that patch,
> and look at the output of 'powerpc-750-linux-gnu-gcc -S 
> struct-ret-1.c' to see if it changes.

There were no changes.  Thus I am uncertain whether this patch was the
problem.

Kunjan, how did you verify the toolchain no longer has the problem?

Can you check whether the .s files produced by 'powerpc-750-linux-gnu-gcc
-save-temps struct-ret-1.c' differ between the good and bad toolchains?  And
if so, send me both?
- Dan

-- 
My technical stuff: http://kegel.com
My politics: see http://www.misleader.org for examples of why I'm for regime
change

-------------- next part --------------
A non-text attachment was scrubbed...
Name: struct-ret-1.c
Type: application/octet-stream
Size: 1117 bytes
Desc: not available
URL: <http://sourceware.org/pipermail/crossgcc/attachments/20040622/141b8da9/attachment.obj>
-------------- next part --------------
------
Want more information?  See the CrossGCC FAQ, http://www.objsw.com/CrossGCC/
Want to unsubscribe? Send a note to crossgcc-unsubscribe@sources.redhat.com


More information about the crossgcc mailing list