From: bugs at busybox.net <bugs@busybox.net>
To: buildroot@busybox.net
Subject: [Buildroot] [buildroot 0000759]: FLOAT: Building fails when using ARM920t and software Floating Point
Date: Mon, 12 Feb 2007 05:33:18 -0800 [thread overview]
Message-ID: <0411a2db3d52da32ce5d40dc724ba9ec@bugs.busybox.net> (raw)
The following issue has been ASSIGNED.
======================================================================
http://busybox.net/bugs/view.php?id=759
======================================================================
Reported By: agusfusa
Assigned To: buildroot
======================================================================
Project: buildroot
Issue ID: 759
Category: Architecture Specific
Reproducibility: always
Severity: block
Priority: normal
Status: assigned
======================================================================
Date Submitted: 02-28-2006 09:14 PST
Last Modified: 02-12-2007 05:33 PST
======================================================================
Summary: FLOAT: Building fails when using ARM920t and
software Floating Point
Description:
When building the toolchain using as a target an ARM920t and selecting that
the device does not have an floating point unit, this error ocurrs:
(...)
make -C /root/buildroot/toolchain_build_arm/uClibc \
PREFIX= \
DEVEL_PREFIX=/ \
RUNTIME_PREFIX=/ \
HOSTCC="gcc" \
all
make[1]: Entering directory `/root/buildroot/toolchain_build_arm/uClibc'
Current kernel version is 2.4.31
Using kernel headers from 2.4.31 for architecture 'arm'
provided in directory /root/buildroot/toolchain_build_arm/linux/
CC ldso/ldso/ldso.oS
AS ldso/ldso/arm/resolve.oS
STRIP -x -R .note -R .comment ldso/ldso/ld-uClibc_so.a
AR cr ldso/ldso/ld-uClibc_so.a
LD ld-uClibc-0.9.28.so
/root/buildroot/build_arm/staging_dir/bin/arm-linux-uclibc-ld: ERROR:
/root/buil
droot/build_arm/staging_dir/lib/gcc/arm-linux-uclibc/3.4.2/libgcc.a(_udivsi3.o)
uses hardware FP, whereas lib/ld-uClibc-0.9.28.so uses software FP
/root/buildroot/build_arm/staging_dir/bin/arm-linux-uclibc-ld: failed to
merge t arget specific data of file
/root/buildroot/build_arm/staging_dir/lib/gcc/arm-li
nux-uclibc/3.4.2/libgcc.a(_udivsi3.o)
/root/buildroot/build_arm/staging_dir/bin/arm-linux-uclibc-ld: ERROR:
/root/buil
droot/build_arm/staging_dir/lib/gcc/arm-linux-uclibc/3.4.2/libgcc.a(_dvmd_lnx.o)
uses hardware FP, whereas lib/ld-uClibc-0.9.28.so uses software FP
/root/buildroot/build_arm/staging_dir/bin/arm-linux-uclibc-ld: failed to
merge t arget specific data of file
/root/buildroot/build_arm/staging_dir/lib/gcc/arm-li
nux-uclibc/3.4.2/libgcc.a(_dvmd_lnx.o)
make[1]: *** [lib/ld-uClibc.so] Error 1
make[1]: Leaving directory `/root/buildroot/toolchain_build_arm/uClibc'
make: *** [/root/buildroot/toolchain_build_arm/uClibc/lib/libc.a] Error 2
Has anybody had this problem? Thanks very much!!
======================================================================
----------------------------------------------------------------------
prpplague - 03-02-06 05:11
----------------------------------------------------------------------
please provide the .config you are using for this build so that i can try
replicating the problem.
----------------------------------------------------------------------
agusfusa - 03-02-06 06:06
----------------------------------------------------------------------
Here's the .config file as you requested.
When I'm asked I select (If not specified I go with the default answers):
Arm Target (option 2)
Use BX in function return: NO
Arm 920t (option 6)
Little Endian (option 2)
Taget CPU has FPU: NO
Hope you can replicate this situation!
Thx a lot!
----------------------------------------------------------------------
trav - 04-05-06 00:26
----------------------------------------------------------------------
I've also been having this issue. I'm currently working with
buildroot-20060404, but the bug seems to be present in all recent versions
of buildroot. It is apparently triggered by this option:
[*] Use a daily snapshot of uClibc?
I've uploaded my config file as well, titled "config_trav". It's
basically a default build. The only non-default options I'm using are:
Target Architecture (arm)
Target Architecture Variant (generic_arm)
and then, during the build, I select "Little Endian" and answer NO to
"Target has an FPU".
Thanks!
Issue History
Date Modified Username Field Change
======================================================================
02-28-06 09:14 agusfusa New Issue
02-28-06 09:14 agusfusa Status new => assigned
02-28-06 09:14 agusfusa Assigned To => uClibc
03-02-06 05:11 prpplague Note Added: 0001158
03-02-06 05:11 prpplague Status assigned => feedback
03-02-06 06:06 agusfusa Note Added: 0001159
03-02-06 06:07 agusfusa File Added: .config
04-05-06 00:20 trav File Added: config_trav
04-05-06 00:23 trav Note Added: 0001223
04-05-06 00:25 trav Note Edited: 0001223
04-05-06 00:26 trav Note Edited: 0001223
10-09-06 12:25 jruere Issue Monitored: jruere
02-04-07 09:14 bernhardf Summary Building fails when
using ARM920t and software Floating Point => FLOAT: Building fails when using
ARM920t and software Floating Point
02-12-07 05:33 vapier Status feedback => assigned
02-12-07 05:33 vapier Assigned To uClibc => buildroot
======================================================================
reply other threads:[~2007-02-12 13:33 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=0411a2db3d52da32ce5d40dc724ba9ec@bugs.busybox.net \
--to=bugs@busybox.net \
--cc=buildroot@busybox.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox