Re: Yet another failure on armhf and i386

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

 



On Wed, Feb 13, 2019 at 10:36:33AM +0100, Uwe Kleine-König wrote:
> Hello,
> 
> I noticed that sparse fails to check memops.h on armhf (but only
> sometimes). build2 on
> https://tests.reproducible-builds.org/debian/rb-pkg/unstable/armhf/sparse.html
> failed with:
> 
> 	env CHECK=./sparse ./cgcc -no-compile memops.c
> 	/usr/include/arm-linux-gnueabihf/gnu/stubs.h:7:12: error: unable to open 'gnu/stubs-soft.h'
> 
> (You have to click on "build2" in the left bar to see the failing log.)
> 
> It seems a define for __ARM_PCS_VFP is missing here. On the failing
> machine we have:
> 
> 	uname -m -> aarch64
> 	gcc-8 -dumpmachine -> arm-linux-gnueabihf

OK, I see. This is not supported by cgcc.

> The regular autobuild and the first builder worked fine. 

If the compiler run the native arch.
 
> On i386 there is a similar problem. It has:
> 
> 	uname -m -> x86_64
> 	gcc -dumpmachine -> i686-linux-gnu

Yes, same problem.
 
> Talking a bit to the people in #debian-reprocible the gist is that
> uname -m is a bad indicator. gcc -dumpmachine is the more reliable
> source here. Note this is not only about 64bit kernels running a 32 bit
> userspace, it also affects cross building.

Sure, once a non-nativiely configured compiler is used, information
about the compiler configration is needed. Parsing 'gcc -dumpmachine'
is certainly possible but this is really not in my current priorities.

Thanks for the information,
-- Luc



[Index of Archives]     [Newbies FAQ]     [LKML]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Trinity Fuzzer Tool]

  Powered by Linux