From: Paradoxxa <paradoxxa2001@domain.hid>
To: xenomai@xenomai.org
Subject: Re: [Xenomai-help] at91sam9263 and user-space support compile fails
Date: Tue, 20 May 2008 14:43:25 +0200 [thread overview]
Message-ID: <4832C76D.3040803@domain.hid> (raw)
In-Reply-To: <2ff1a98a0805200123m2a6456d8ud384a8b9535818a8@domain.hid>
Gilles Chanteperdrix schrieb:
> On Tue, May 20, 2008 at 10:22 AM, Gilles Chanteperdrix
> <gilles.chanteperdrix@xenomai.org> wrote:
>
>> On Tue, May 20, 2008 at 10:20 AM, Gilles Chanteperdrix
>> <gilles.chanteperdrix@xenomai.org> wrote:
>>
>>> On Tue, May 20, 2008 at 10:19 AM, Paradoxxa <paradoxxa2001@domain.hid> wrote:
>>>
>>>> Gilles Chanteperdrix schrieb:
>>>>
>>>>> paradoxxa wrote:
>>>>> > Gilles Chanteperdrix schrieb:
>>>>> > > On Mon, May 19, 2008 at 4:28 PM, paradoxxa <paradoxxa2001@domain.hid>
>>>>> wrote:
>>>>> > > > >> Gilles Chanteperdrix schrieb:
>>>>> > >> > >>> On Mon, May 19, 2008 at 3:04 PM, paradoxxa
>>>>> <paradoxxa2001@domain.hid> wrote:
>>>>> > >>>
>>>>> > >>> > >>>> Gilles Chanteperdrix schrieb:
>>>>> > >>>>
>>>>> > >>>> > >>>>> On Mon, May 19, 2008 at 2:17 PM, paradoxxa
>>>>> <paradoxxa2001@domain.hid> wrote:
>>>>> > >>>>>
>>>>> > >>>>>
>>>>> > >>>>> No, you are talking about CONFIG_DEBUG_LL, I am talking about
>>>>> > >>>>> CONFIG_DEBUG_INFO. These are two different options. So, this make
>>>>> at
>>>>> > >>>>> least two options that you changed. Could you check if a kernel
>>>>> with
>>>>> > >>>>> an unmodified defconfig boots ?
>>>>> > >>>>>
>>>>> > >>>>>
>>>>> > >>>>>
>>>>> > >>>>> > >>>> The CONFIG_DEBUG_INFO is already set in the
>>>>> defconfig. The
>>>>> > >>>> CONFIG_DEBUG_LL i
>>>>> > >>>> set manually.
>>>>> > >>>> As well I had to changed the AT91 Master Clock Frequency.
>>>>> > >>>>
>>>>> > >>>> Here are the logs of a unmodified defconfig boot:
>>>>> > >>>>
>>>>> > >>>> > >>> Ok. But this is a boot with your old compiler.
>>>>> Could you try a
>>>>> > >>> defconfig build (with I-pipe AT91 master clock frequency modified)
>>>>> > >>> with CodeSourcery compiler ?
>>>>> > >>> If you can not get CodeSourcery compiler to compile correctly a
>>>>> Linux
>>>>> > >>> kernel, I can send you the 2007q3 compiler, or you could try to
>>>>> > >>> compile the kernel with your old compiler, and to compile
>>>>> user-space
>>>>> > >>> with CodeSourcery compiler.
>>>>> > >>>
>>>>> > >>> > >> I tried it with the both compiler. When i execute
>>>>> /usr/xenomai/bin/latency i
>>>>> > >> get the message "sh: /usr/xenomai/bin/latency: not found"
>>>>> > >> > >
>>>>> > > Have you tried running file, ldd, or readelf on the binary to see
>>>>> what
>>>>> > > it is made of ?
>>>>> > >
>>>>> > > > >> By the way, how i can figure out if the binary or the
>>>>> compiler is a
>>>>> > >> oabi/eabi file/compiler?
>>>>> > >> > >
>>>>> > > codesourcery compiles for eabi by default. So, you have to enable
>>>>> eabi
>>>>> > > in kernel configuration to be able to run binaries generated by
>>>>> > > codesourcery compiler.
>>>>> > >
>>>>> > > > The eabi was disabled in the kernel configuration. But to enable
>>>>> it make > no changes.
>>>>> > > > $ file usr/xenomai/bin/latency
>>>>> > usr/xenomai/bin/latency: ELF 32-bit LSB executable, ARM, version 1 >
>>>>> (SYSV), for GNU/Linux 2.6.14, dynamically linked (uses shared libs), not >
>>>>> stripped
>>>>> > > > $ ldd usr/xenomai/bin/latency
>>>>> > not a dynamic executable
>>>>>
>>>>> This is wrong. ldd should be run on the target, not on the host. Also
>>>>> check that /usr/xenomai/lib is in the ld.so.conf on your target (and
>>>>> that you have run ldconfig) or use LD_LIBRARY_PATH
>>>>>
>>>>> > > $ readelf --sections /usr/xenomai/bin/latency
>>>>>
>>>>> I was rather thinking about readelf -d /usr/xenomai/bin/latency
>>>>>
>>>>>
>>>>>
>>>> I actuall don't have ldd and readelf on the target. I did a ldconfig -p. The
>>>> Xenomai Libraries were listed there.
>>>>
>>> readelf can work on the host. But ldd on the target is the best way to
>>> know if the installation worked correctly. If you are looking for ldd,
>>> it is compiled as part of the glibc. You may even found it in the
>>> toolchain files.
>>>
>> Actually ldd is a script, it is not compiled.
>>
>
> All what it does is launching
>
> LD_TRACE_LOADED_OBJECTS=1 /usr/xenomai/bin/latency
>
> So, you do not even need the script.
>
>
> On Tue, May 20, 2008 at 10:22 AM, Gilles Chanteperdrix
> <gilles.chanteperdrix@xenomai.org> wrote:
>
>> On Tue, May 20, 2008 at 10:20 AM, Gilles Chanteperdrix
>> <gilles.chanteperdrix@xenomai.org> wrote:
>>
>>> On Tue, May 20, 2008 at 10:19 AM, Paradoxxa <paradoxxa2001@domain.hid> wrote:
>>>
>>>> Gilles Chanteperdrix schrieb:
>>>>
>>>>> paradoxxa wrote:
>>>>> > Gilles Chanteperdrix schrieb:
>>>>> > > On Mon, May 19, 2008 at 4:28 PM, paradoxxa <paradoxxa2001@domain.hid>
>>>>> wrote:
>>>>> > > > >> Gilles Chanteperdrix schrieb:
>>>>> > >> > >>> On Mon, May 19, 2008 at 3:04 PM, paradoxxa
>>>>> <paradoxxa2001@domain.hid> wrote:
>>>>> > >>>
>>>>> > >>> > >>>> Gilles Chanteperdrix schrieb:
>>>>> > >>>>
>>>>> > >>>> > >>>>> On Mon, May 19, 2008 at 2:17 PM, paradoxxa
>>>>> <paradoxxa2001@domain.hid> wrote:
>>>>> > >>>>>
>>>>> > >>>>>
>>>>> > >>>>> No, you are talking about CONFIG_DEBUG_LL, I am talking about
>>>>> > >>>>> CONFIG_DEBUG_INFO. These are two different options. So, this make
>>>>> at
>>>>> > >>>>> least two options that you changed. Could you check if a kernel
>>>>> with
>>>>> > >>>>> an unmodified defconfig boots ?
>>>>> > >>>>>
>>>>> > >>>>>
>>>>> > >>>>>
>>>>> > >>>>> > >>>> The CONFIG_DEBUG_INFO is already set in the
>>>>> defconfig. The
>>>>> > >>>> CONFIG_DEBUG_LL i
>>>>> > >>>> set manually.
>>>>> > >>>> As well I had to changed the AT91 Master Clock Frequency.
>>>>> > >>>>
>>>>> > >>>> Here are the logs of a unmodified defconfig boot:
>>>>> > >>>>
>>>>> > >>>> > >>> Ok. But this is a boot with your old compiler.
>>>>> Could you try a
>>>>> > >>> defconfig build (with I-pipe AT91 master clock frequency modified)
>>>>> > >>> with CodeSourcery compiler ?
>>>>> > >>> If you can not get CodeSourcery compiler to compile correctly a
>>>>> Linux
>>>>> > >>> kernel, I can send you the 2007q3 compiler, or you could try to
>>>>> > >>> compile the kernel with your old compiler, and to compile
>>>>> user-space
>>>>> > >>> with CodeSourcery compiler.
>>>>> > >>>
>>>>> > >>> > >> I tried it with the both compiler. When i execute
>>>>> /usr/xenomai/bin/latency i
>>>>> > >> get the message "sh: /usr/xenomai/bin/latency: not found"
>>>>> > >> > >
>>>>> > > Have you tried running file, ldd, or readelf on the binary to see
>>>>> what
>>>>> > > it is made of ?
>>>>> > >
>>>>> > > > >> By the way, how i can figure out if the binary or the
>>>>> compiler is a
>>>>> > >> oabi/eabi file/compiler?
>>>>> > >> > >
>>>>> > > codesourcery compiles for eabi by default. So, you have to enable
>>>>> eabi
>>>>> > > in kernel configuration to be able to run binaries generated by
>>>>> > > codesourcery compiler.
>>>>> > >
>>>>> > > > The eabi was disabled in the kernel configuration. But to enable
>>>>> it make > no changes.
>>>>> > > > $ file usr/xenomai/bin/latency
>>>>> > usr/xenomai/bin/latency: ELF 32-bit LSB executable, ARM, version 1 >
>>>>> (SYSV), for GNU/Linux 2.6.14, dynamically linked (uses shared libs), not >
>>>>> stripped
>>>>> > > > $ ldd usr/xenomai/bin/latency
>>>>> > not a dynamic executable
>>>>>
>>>>> This is wrong. ldd should be run on the target, not on the host. Also
>>>>> check that /usr/xenomai/lib is in the ld.so.conf on your target (and
>>>>> that you have run ldconfig) or use LD_LIBRARY_PATH
>>>>>
>>>>> > > $ readelf --sections /usr/xenomai/bin/latency
>>>>>
>>>>> I was rather thinking about readelf -d /usr/xenomai/bin/latency
>>>>>
>>>>>
>>>>>
>>>> I actuall don't have ldd and readelf on the target. I did a ldconfig -p. The
>>>> Xenomai Libraries were listed there.
>>>>
>>> readelf can work on the host. But ldd on the target is the best way to
>>> know if the installation worked correctly. If you are looking for ldd,
>>> it is compiled as part of the glibc. You may even found it in the
>>> toolchain files.
>>>
>> Actually ldd is a script, it is not compiled.
>>
>
> All what it does is launching
>
> LD_TRACE_LOADED_OBJECTS=1 /usr/xenomai/bin/latency
>
> So, you do not even need the script.
>
>
> On Tue, May 20, 2008 at 10:22 AM, Gilles Chanteperdrix
> <gilles.chanteperdrix@xenomai.org> wrote:
>
>> On Tue, May 20, 2008 at 10:20 AM, Gilles Chanteperdrix
>> <gilles.chanteperdrix@xenomai.org> wrote:
>>
>>> On Tue, May 20, 2008 at 10:19 AM, Paradoxxa <paradoxxa2001@domain.hid> wrote:
>>>
>>>> Gilles Chanteperdrix schrieb:
>>>>
>>>>> paradoxxa wrote:
>>>>> > Gilles Chanteperdrix schrieb:
>>>>> > > On Mon, May 19, 2008 at 4:28 PM, paradoxxa <paradoxxa2001@domain.hid>
>>>>> wrote:
>>>>> > > > >> Gilles Chanteperdrix schrieb:
>>>>> > >> > >>> On Mon, May 19, 2008 at 3:04 PM, paradoxxa
>>>>> <paradoxxa2001@domain.hid> wrote:
>>>>> > >>>
>>>>> > >>> > >>>> Gilles Chanteperdrix schrieb:
>>>>> > >>>>
>>>>> > >>>> > >>>>> On Mon, May 19, 2008 at 2:17 PM, paradoxxa
>>>>> <paradoxxa2001@domain.hid> wrote:
>>>>> > >>>>>
>>>>> > >>>>>
>>>>> > >>>>> No, you are talking about CONFIG_DEBUG_LL, I am talking about
>>>>> > >>>>> CONFIG_DEBUG_INFO. These are two different options. So, this make
>>>>> at
>>>>> > >>>>> least two options that you changed. Could you check if a kernel
>>>>> with
>>>>> > >>>>> an unmodified defconfig boots ?
>>>>> > >>>>>
>>>>> > >>>>>
>>>>> > >>>>>
>>>>> > >>>>> > >>>> The CONFIG_DEBUG_INFO is already set in the
>>>>> defconfig. The
>>>>> > >>>> CONFIG_DEBUG_LL i
>>>>> > >>>> set manually.
>>>>> > >>>> As well I had to changed the AT91 Master Clock Frequency.
>>>>> > >>>>
>>>>> > >>>> Here are the logs of a unmodified defconfig boot:
>>>>> > >>>>
>>>>> > >>>> > >>> Ok. But this is a boot with your old compiler.
>>>>> Could you try a
>>>>> > >>> defconfig build (with I-pipe AT91 master clock frequency modified)
>>>>> > >>> with CodeSourcery compiler ?
>>>>> > >>> If you can not get CodeSourcery compiler to compile correctly a
>>>>> Linux
>>>>> > >>> kernel, I can send you the 2007q3 compiler, or you could try to
>>>>> > >>> compile the kernel with your old compiler, and to compile
>>>>> user-space
>>>>> > >>> with CodeSourcery compiler.
>>>>> > >>>
>>>>> > >>> > >> I tried it with the both compiler. When i execute
>>>>> /usr/xenomai/bin/latency i
>>>>> > >> get the message "sh: /usr/xenomai/bin/latency: not found"
>>>>> > >> > >
>>>>> > > Have you tried running file, ldd, or readelf on the binary to see
>>>>> what
>>>>> > > it is made of ?
>>>>> > >
>>>>> > > > >> By the way, how i can figure out if the binary or the
>>>>> compiler is a
>>>>> > >> oabi/eabi file/compiler?
>>>>> > >> > >
>>>>> > > codesourcery compiles for eabi by default. So, you have to enable
>>>>> eabi
>>>>> > > in kernel configuration to be able to run binaries generated by
>>>>> > > codesourcery compiler.
>>>>> > >
>>>>> > > > The eabi was disabled in the kernel configuration. But to enable
>>>>> it make > no changes.
>>>>> > > > $ file usr/xenomai/bin/latency
>>>>> > usr/xenomai/bin/latency: ELF 32-bit LSB executable, ARM, version 1 >
>>>>> (SYSV), for GNU/Linux 2.6.14, dynamically linked (uses shared libs), not >
>>>>> stripped
>>>>> > > > $ ldd usr/xenomai/bin/latency
>>>>> > not a dynamic executable
>>>>>
>>>>> This is wrong. ldd should be run on the target, not on the host. Also
>>>>> check that /usr/xenomai/lib is in the ld.so.conf on your target (and
>>>>> that you have run ldconfig) or use LD_LIBRARY_PATH
>>>>>
>>>>> > > $ readelf --sections /usr/xenomai/bin/latency
>>>>>
>>>>> I was rather thinking about readelf -d /usr/xenomai/bin/latency
>>>>>
>>>>>
>>>>>
>>>> I actuall don't have ldd and readelf on the target. I did a ldconfig -p. The
>>>> Xenomai Libraries were listed there.
>>>>
>>> readelf can work on the host. But ldd on the target is the best way to
>>> know if the installation worked correctly. If you are looking for ldd,
>>> it is compiled as part of the glibc. You may even found it in the
>>> toolchain files.
>>>
>> Actually ldd is a script, it is not compiled.
>>
>
> All what it does is launching
>
> LD_TRACE_LOADED_OBJECTS=1 /usr/xenomai/bin/latency
>
> So, you do not even need the script.
>
>
I think that doesn't look good:
root@domain.hid:~# LD_TRACE_LOADED_OBJECTS=1 /usr/xenomai/bin/latency
-sh: /usr/xenomai/bin/latency: not found
root@domain.hid -d /usr/xenomai/bin/latency
Dynamic section at offset 0x48a0 contains 29 entries:
Tag Type Name/Value
0x00000001 (NEEDED) Shared library: [libnative.so.1]
0x00000001 (NEEDED) Shared library: [librtdm.so.1]
0x00000001 (NEEDED) Shared library: [libpthread.so.0]
0x00000001 (NEEDED) Shared library: [libm.so.6]
0x00000001 (NEEDED) Shared library: [libc.so.6]
0x0000000f (RPATH) Library rpath: [/usr/xenomai/lib]
0x0000000c (INIT) 0x8bbc
0x0000000d (FINI) 0xbdc8
0x00000019 (INIT_ARRAY) 0x14894
0x0000001b (INIT_ARRAYSZ) 4 (bytes)
0x0000001a (FINI_ARRAY) 0x14898
0x0000001c (FINI_ARRAYSZ) 4 (bytes)
0x00000004 (HASH) 0x8168
0x00000005 (STRTAB) 0x86b4
0x00000006 (SYMTAB) 0x82f4
0x0000000a (STRSZ) 703 (bytes)
0x0000000b (SYMENT) 16 (bytes)
0x00000015 (DEBUG) 0x0
0x00000003 (PLTGOT) 0x149b0
0x00000002 (PLTRELSZ) 336 (bytes)
0x00000014 (PLTREL) REL
0x00000017 (JMPREL) 0x8a6c
0x00000011 (REL) 0x8a4c
0x00000012 (RELSZ) 32 (bytes)
0x00000013 (RELENT) 8 (bytes)
0x6ffffffe (VERNEED) 0x89ec
0x6fffffff (VERNEEDNUM) 3
0x6ffffff0 (VERSYM) 0x8974
0x00000000 (NULL) 0x0
Thomas
next prev parent reply other threads:[~2008-05-20 12:43 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-05-19 6:26 [Xenomai-help] at91sam9263 and user-space support compile fails Paradoxxa2001
2008-05-19 9:39 ` Gilles Chanteperdrix
2008-05-19 11:20 ` paradoxxa
2008-05-19 12:06 ` Gilles Chanteperdrix
2008-05-19 12:17 ` paradoxxa
2008-05-19 12:36 ` Gilles Chanteperdrix
2008-05-19 13:04 ` paradoxxa
2008-05-19 13:30 ` Gilles Chanteperdrix
2008-05-19 14:24 ` paradoxxa
2008-05-19 14:28 ` paradoxxa
2008-05-19 14:34 ` Gilles Chanteperdrix
2008-05-20 6:43 ` paradoxxa
2008-05-20 6:52 ` Gilles Chanteperdrix
2008-05-20 8:19 ` Paradoxxa
2008-05-20 8:20 ` Gilles Chanteperdrix
2008-05-20 8:22 ` Gilles Chanteperdrix
2008-05-20 8:23 ` Gilles Chanteperdrix
2008-05-20 12:43 ` Paradoxxa [this message]
2008-05-20 15:16 ` Gilles Chanteperdrix
2008-05-21 6:43 ` Paradoxxa
2008-05-21 6:56 ` Gilles Chanteperdrix
-- strict thread matches above, loose matches on Subject: below --
2008-05-16 14:34 Paradoxxa2001
2008-05-16 15:26 ` Gilles Chanteperdrix
2008-05-16 13:53 Paradoxxa2001
2008-05-16 13:57 ` Gilles Chanteperdrix
2008-05-16 14:01 ` Gilles Chanteperdrix
2008-05-16 12:40 Paradoxxa2001
2008-05-16 12:42 ` Gilles Chanteperdrix
2008-05-16 11:07 Paradoxxa2001
2008-05-16 12:18 ` Gilles Chanteperdrix
2008-05-16 9:34 Paradoxxa2001
2008-05-16 9:45 ` Gilles Chanteperdrix
2008-05-16 9:57 ` Gilles Chanteperdrix
2008-05-16 9:24 Paradoxxa2001
2008-05-16 9:27 ` Gilles Chanteperdrix
2008-05-16 9:10 Paradoxxa2001
2008-05-16 9:18 ` Gilles Chanteperdrix
2008-05-16 8:15 Paradoxxa2001
2008-05-16 8:44 ` Gilles Chanteperdrix
2008-05-15 12:28 Paradoxxa2001
2008-05-15 8:36 Paradoxxa2001
2008-05-15 9:21 ` Gilles Chanteperdrix
2008-05-15 9:46 ` Gilles Chanteperdrix
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=4832C76D.3040803@domain.hid \
--to=paradoxxa2001@domain.hid \
--cc=xenomai@xenomai.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.