These are not really questions about uClinux, they’re questions about your
development board. So you should ask the vendor of that development board.
But the debug monitor has to be in FLASH. You can’t persistently save things
in SRAM.
From: uclinux-dev-boun...@uclinux.org
Quoth Larry Baker:
Yes, this is indeed what can happen on a system like uClinux that does
not have an MMU (memory management unit). The tiny controller chips
that uClinux targets do not protect any part of the memory from being
accidentally corrupted. As long as the memory being accessed
Quoth Anna Fischer:
I was wondering if anyone had any experiences they can share about using
XIP on ucLinux? As far as I can see it is supported for the bootloader,
the kernel and also for applications, if the toolchain supports this. My
question though is if anyone has done any studies on how
Quoth angelo:
Seems that for my case (m68k-nommu) it is not defined, so none of the
get_unmapped_area will be called, in any place.
It doesn't need to be defined. fbmem.c implements get_unmapped_area
regardless (just slightly differently in each case). If you had actually
tried just setting
Quoth Daniel Glöckner:
Are we talking about the same kernel tree?
Mainline had in 2.6.36.2 and to date still has
No, I was looking at my local kernel tree for an old project, which is
considerably older. But looking at the official tree now, it looks like I
might have patched the
Quoth angelo:
thanks, well, i implemented the handler:
static int amcorefb_mmap(struct fb_info *info, struct vm_area_struct
*vma) {
No, that's not the right one. You don't need to implement that.
Seems i miss the get_unmapped_area/BDI_CAP_MAP_DIRECT) capability.
fbmem.c implements the
Quoth Steve deRosier:
Can anyone tell me the proper or accepted way to enable
CONFIG_USB_ARCH_HAS_HCD
for my embedded uClinux project?
[...]
Obviously this is a board-specific thing, not a generic linux thing, so
what's the
better way of doing this?
You should have a board-specific
Quoth Muthuselvan Sivam:
We're having a function call fork_dispatcher inside a function loop() in
the avahi-autoip main.c file, which will create a child process. This
child will create a pipe to read the IPv4LL address from the parent,
which writes the address, interface name etc, into
Quoth Lennart Sorensen:
I wonder where you could get cycle counts for each instruction.
I remember seeing them years ago for 8085 and such, but no idea of
freescale provides them.
I still have something that uses the 68332 (the original CPU32, I believe);
for that there are detailed
Quoth Greg Ungerer:
On 24/05/11 18:06, Andreas Schwab wrote:
Geert Uytterhoevenge...@linux-m68k.org writes:
What exactly do you mean by does not support anything less? It
seems it does restrict instruction generation to 68000 if you
ask for it.
The point is that Linux/m68k
Quoth Greg Ungerer:
Impact for m68knommu:
- The table is now stored in .data instead of .text,
Do you mean .rodata ?
[...]
Yes, I suspect that was the original thinking. It has been that way
for a very long time (it is in .text in the 2.4 kernel patches for
this too).
I will test
Quoth Greg Ungerer:
Doesn't that have XIP consequences? .text (and presumably .rodata)
can be stored and executed from ROM, since they can't be changed at
runtime. .data has to be in RAM, since it can be.
Yes, but even in the kernel XIP case the very early startup code
moves the kernels
Quoth Gilles Ganault:
Thanks Ulisses. I'm not much of a developer so don't have enough
technical knowledge to tell, but others documents also say that
Asterisk checks a specific directory several times per second, hence
the need to use mv instead of cp, since mv is atomic.
This is true
Quoth Jate Sujjavanich:
Should the bit fields in ptrace.h? So then it would be
#ifdef CONFIG_COLDFIRE
unsigned format : 4; /* frame format specifier */
+ unsigned fault_status_H : 2;
- unsigned vector : 8; /* vector offset */
+ unsigned vector : 12; /* vector offset */
+ unsigned
Quoth Jerry:
This is precisely why I am asking the question :). given that the
application has a single main thread i do not see why this
expectation is unreasonable or irrational!
If you look at the documentation for signal handlers, you'll see that they
get called asynchronous to normal
Quoth John Master:
I don’t think he is referring to an application that runs on both
platforms, but code that can be compiled with different compilers.
The short answer is yes, you can. However, there may be slight changes
in memory management code.
Also in anything that accesses
Quoth David Howells:
Stub out vm_get_page_prot() if there's no MMU.
[...]
in the fbmem video driver, but the function doesn't exist on NOMMU,
resulting in an undefined symbol at link time.
[...]
+#ifdef CONFIG_MMU
pgprot_t vm_get_page_prot(unsigned long vm_flags);
+#else
+static inline
Quoth Andrew Kohlsmith:
I would hate to try and do it from software with an interrupt.
Yuck.
It's not that bad. :-) IRQ on transmit holding or shift register
empty (ideally the latter). If you have only a THRE type interrupt
then you have to set a timer for 1 word time (usually 9 bits)
Quoth 李 軍:
Does uClinux support some kind of architecture with MMU?
I mean are there some boards with mmu, if we disable the mmu option (or
undef CONFIG_MMU), the uClinux can run on it.
At this stage, uClinux is mostly just the embedded build environment, and isn't
really tied to MMU/!MMU.
Quoth Slava Apter:
Yes. I am using Coldfire arch 5271. So may I assume, that redefinition
of Flash Memory Size should be done in Bootloader?
Sorry, I think I misinterpreted your question. The chipselect is used to
determine the size of the RAM, not Flash.
(Normally, uClinux neither knows nor
Quoth Slava Apter:
I am working with 4MB flash memory. But at this stage it is
not enough.
Where do I have to insert changes in uclinux-dist in order
to redefine Flash memory?
Or maybe I have to change Bootloader???
It depends on the arch, but AFAIK the startup code for the Coldfires at
Quoth Erwin Authried:
you are right, if you call ld directly, -elf2flt can't be used. I don't
think that this is really a problem, because you'll hardly link a BFLT
application with ld directly. I haven't ever seen that since I got one
of the first uCsimms. If you still want to do that, you
Quoth Niklas Molin:
Problem is that the code is using fork and each child process is running
simultaneously (plus the parent process needs to be in charge of handle
all information each child processes are sending/receiving).
If I understood vfork() right, it will halt the parent process
Quoth Prasad:
I am running uclinux 2.4.31 in microblaze. Whenever an application
gets killed (by running killall or kill -9), does all the dynamic
memory (allocated through malloc by that application) gets freed ? Is
there any limitation to that. I am chasing a possible memory leak in
some of
Quoth Jamie Lokier:
Some arches have a page which is per-thread, called the vsyscall/vdso
kernel page. User-space calls routines in that page. In that page, a
pointer to the TLS area might be stored, swapped on context switch by
the kernel, and retrieved by calling a routine at a fixed
Quoth Courousse, Damien:
The following lines (see the complete init log below) look strange to me
because I did not setup ROMfs :
uclinux[mtd]: RAM probe address=0x18c90c size=0xdf000
Creating 1 MTD partitions on RAM:
0x-0x000df000 : ROMfs
uclinux[mtd]: set ROMfs to be root
Quoth Bob Furber:
Currently defined functions:
[, basename, busybox, cat, chgrp, chmod, chown, chroot, clear,
cmp, cp, date, dd, df, dirname, du, echo, env, expr, false, find,
free, ftpget, ftpput, getopt, grep, halt, head, hostname, hwclock,
id, ifconfig,
Quoth Courousse, Damien:
My problem is that I cannot get a full gdb setup working or even compiling
from the uClinux source tree.
Please could you indicate me what are the options that would lead me to get a
fully working setup,
including gdb / gdbserver?
While I haven't done it myself
Quoth Michael Schnell:
I did not know about this wording. With exec*() the process stays the
same (same pid), and it's still a child of the same parent. It just gets
new memory and some initialization is done (man execve=).
Right, but it's the new memory that does the trick. After calling
Quoth David Wu:
CACHE_ENABLE is defined to nop if CONFIG_UCBOOTLOADER is defined. The cache
setting is defined in the
bootloader in this case.
Also I would like to point out that at least the cache setting is not quite
correct if RAM size is not
16M for CONFIG_M527x.
The settings made by
Quoth Jan Ringoš:
1) If properly synchronized I can access data of other processes.
True only for !MMU, but it's bad style. You should use the normal shared
memory facilities, just as you would on
Windows.
3) There is no practical distinction between process and thread in uClinux
as there
Quoth Wolfgang Mües:
The linking of crtbegin and crtend is disabled with --no-stdlibs. In the old
uClinux distributions, this switch is embedded in ucfront.
Wait, ucfront is old now? I'm still using a pre-ucfront distribution ;)
___
uClinux-dev
Quoth loody:
At beginning, I use the combination you suggest, but uclinux makes me
cry when I find that the only executable file format that can run on
uclinux is FLAT.
This is because the other file formats rely on the VMM being able to make
code/data appear at specific addresses.
Without a
Quoth Greg Ungerer:
Use readelf. FLAT format files are generated by a trivial
conversion from ELF. (That is they are compiled as ELF, and
a final conversion step is done to convert them to FLAT).
The ELF format file is the myapp.gdb file created next to the myapp (FLAT) file.
Quoth RAM size autodetection:
We plan to double up the SDRAM size of our 5272-based board. It would be
more convenient for us that the kernel automagically detects the size of
the installed RAM instead of managing two different kernels or
bootloaders; it would also be better than placing some
Quoth Lin KJ:
You mentioned that the uClinux binaries do not necessarily have to be
XIP and can be run on non-XIP mode.
How do i switch to run non-XIP uClinux?
Bypass the -fpic -msingle-pic-base CFLAGS and use some
specific elf2flt options?
If uClinux could be non-XIP, why the -fpic
Quoth Michael Schnell:
I am using uCLinux for non-MMU NIOS2 processor and I was told that the
gdbserver in the uCllinux-distr for same does not support threads (gdb
does not show the threads when a threaded application is monitored). As
I will need this some day soon, we need to fix this
Quoth David Wu:
I have created attched patch using Gavin's original patch plus my
changes
-- adding fdwatch in cgi_interpose_output(). The timeout for read
is 5000ms
-- terminating the child process if the timeout expires.
again this patch applies to uClinux-dist-20080808.
Is the
Quoth David Wu:
OK. For normally CGI program(I mean exit normally) it works fine. But If
I try to call cat kmsg in a CGI program then it will block further
request from browser. It seems there is one missing close on the socket.
That's just vfork at work, isn't it? vfork blocks the
Quoth David Wu:
Here is a patch for thttpd against
uClinux-dist-200808011(uClinux-dist-20080808 + 20080811 patch)
This patch fixes CGI calls for uClinux(vfork). Here is an example to
test:
I've previously sent (larger) patches for thttpd in that make it use vfork.
Been running for ages
Quoth David Howells:
Once it's loaded into ramfs, then XIP is unnecessary.
As far as I was aware, that's not true. Regardless of filesystem (ramfs or
other), if XIP isn't enabled then an *extra* copy of the executable will get
loaded into RAM when it's run.
Quoth Robin Getz:
This just takes longer - since you either need to copy things by hand,
or run a 'make romfs' every time you edit something in vendors
How long does a 'make romfs' take on your system? It's usually only a second
on mine, so it doesn't bother me.
If I'm just tweaking
Quoth rwarner:
Are the patches supplied only for 20080808 or will they also work with
the 2007 version?
My contribution to the patchset (#2) was originally developed with 20041215;
I don't think it needed much (if any) changes to work with the latest
version, so you shouldn't have too many
Quoth rwarner:
Problem with a ML is you get every single mail that may or may not
pertain to something you are working on with respect to the uclinux
dist and or kernel. Note other ML try to setup multiple ML, but
this becomes unwieldy with any break down into upper level subject
Quoth Markus Franke:
I am wondering whether it would be possible to have a huge FIFO of
like 16 MBytes allocated in kernel memory and to map this memory
region to the user space process. I am not really sure if this would
be possible with uClinux because of the lack of MMU support.
While it's
Quoth rwarner:
[42949379.66] uclinux[mtd]: RAM probe address=0x204cffd0
size=0x8e000
[42949379.67] Creating 1 MTD partitions on %s:
[42949379.68] 0x%8x-0x%x : NUEL
[42949379.70] uclinux[mtd]: set ROMfs to be root filesystem
There's something really weird going on there. Why
Quoth Paul Romero:
I made great progress in isolating the problem.
If you modify the driver to ignore interrupts that
are very close together rather than automatically send
SIGUSR2 to to the user space application process right
away, the problem does not occur. (i.e. Crude debouncing.)
This
Quoth Sebastian Siewior:
+ /*
+ * If the address is either in the text segment of the
kernel,
+ * then it *may* be the address of a calling routine; if
so,
+ * print it so that someone tracing down the cause of the
crash
+ *
Quoth JL SYNCHRO:
I try to compile my application with zlib support :
[...]
m68k-elf-gcc -m5307 -msep-data libz.c -L/opt/uClinux-
dist/uClibc/ -Wl,-elf2flt -lc -lgcc -lz
[...]
.elf2flt: In function `zcalloc':
/opt/uClinux-dist/lib/zlib/zutil.c:306: undefined reference to `malloc'
.elf2flt:
Quoth Sebastian Siewior:
+ /*
+ * If the address is in the text segment of the kernel or
in
+ * it *may* be the address of a calling routine; if so,
print
+ * it so that someone tracing down the cause of the crash
will
+ * be
Title: RE: [uClinux-dev] socket binding error
Quoth kailash toshniwal:
syslog, Errorlog, Accesslog all these types of files will not be
created since filesystem used is ROMFS. That's why I had commented
out all the commands that creates these files\xA0in boa.conf file.\xA0
This is
Quoth kailash toshniwal:
syslog, Errorlog, Accesslog all these types of files will not be
created since filesystem used is ROMFS. That's why I had commented
out all the commands that creates these files in boa.conf file.
This is what I understood,I might be wrong in reasoning here.
You can
Quoth kailash toshniwal:
Do I have to run 'boa' binary file in uClinux to start boa
web server? Well I thought so, so I gave it a try. When I
try to run it,I get error message like
/bin./boa
'pid:18 failed 1536'.
[...]
If I run it again then pid number increments by 1 each time.
What
Quoth Michael Schnell:
I don't have one. OTOH, I don't understand. uCLinux does not handle the
MMU and a user land program can't do that on itself. So it should not
matter it any MMU hardware is present.
Without an MMU, userland constructs such as fork()s and certain kinds of
mmap() won't
Quoth Michael Schnell:
sash does not support scripts ? That does not seem very useful.
Do you know a document that explains what shells provide which
capabilities ?
Looking at the docs for each shell is how I did it.
Note that what I said is slightly misleading -- you can write scripts and
Quoth kailash toshniwal:
Can someone please tell me the general instructions on setting
up the network in uClinux for any platform?
Much like any other linux system: you run ifconfig and/or a DHCP client.
Getting DHCP working isn't trivial, since most common DHCP clients tend to
use scripts,
Quoth Greg Ungerer:
Well in this sense there is no libraries. You lost any notion of
them at the link stage. With no shared libraries all you have is
the final binaries. (Yes each will contain some of code that was in
the build library archives).
That's my point though. If linking to the
[Resending because I got a listserver bounce; apologies if some people
already received this.]
Quoth [EMAIL PROTECTED]:
I'm compiling uclinux with 20080305-patch on net+arm, and i
get the following error from mtd-utils:
make[3]: Entering directory `/home/u011514/uClinux-dist/user/mtd-utils
Quoth [EMAIL PROTECTED]:
PAGE_OFFSET question:
Looked through archives and grepped through code. It's still unclear
to me how PAGE_OFFSET is getting set to 0x2000. When i view
vmlinux.lds the replaced value for PAGE_OFFSET is 0x2000, however
the SDRAM configuration from make
Quoth Andrey Kamchatnikov:
I have the next kernel message during booting:
kernel BUG at sched.c:561!
What is wrong with the kernel?
Perhaps you should look at that line of code and find out. Without even
knowing what kernel version and configuration you're talking about, there's
no way
Quoth Mike Frysinger:
On Wednesday 20 February 2008, Mike Frysinger wrote:
On Tuesday 12 September 2006, Nathan Sidwell wrote:
static void fatal (const char *, ...); static void fatal_perror
(const char *, ...);
are there really no libiberty funcs that do this ?
looks like the
Quoth [EMAIL PROTECTED]:
I have made an hello world application in c++ that when run on the
board returns the message : BINFMT_FLAT: reloc outside program
0xb658 (0 - 0x19504/0xdde0), killing hello!
[...]
ucfront-g++ m68k-elf-g++ -m5307 -DCONFIG_COLDFIRE -Wl,--fatal-warnings
-Wl,-elf2flt
Quoth Sean McGranaghan:
I could not get the mcf_qspi driver to leave the chip select asserted
after the 16th word in the transfer. (I tried all manner of
configurations using the QCRn[CONT] bit (bit 15 in the Command
Register) and QWR[CSIV] bit (bit 12 of the Wrap Register). The short
of it
Quoth António Silva:
From 0xC000 to 0xC01F there is one flash chip
From 0xC020 to 0xC03F nothing is mapped
From 0xC040 to 0xC060 there is another flash chip (exact same
model as the previous one)
[...]
However, I was expecting that by specifying 0x80 as the address
Quoth Christian Gieseler:
I have to use a named semaphore in my application. When i tried to
compile i received an error message undefined reference to sem_open.
I could not understand why because i included all the necessary headers.
Are semaphores enabled in your libc configuration?
Quoth Michael Schnell:
Is it true that 2.6 supports an I²C Subsystem that allows for a
driver for the I²C transport hardware and - independent of same -
multiple drivers for the chips to be accesses via I²C. That would
be a major advantage and rule out earlier versions in many cases.
Yes,
Quoth Robert Warner:
I'm looking into the prospect of preloading all of the necessary tables
for applications and kernel such that when the device is turned on it
doesn't boot it just starts executing as if it had already gone through
all the booting and application(s) initialization(s). Has
Quoth Robert Warner:
I may be misunderstanding something here. Do i apply the patch to a
Linux 2.6.24 kernel tree from kernel.org? Or, do i apply the patch at
the uCLinux-dist/linux-2.6.x directory? From the FAQ sounds like the
first choice. Then after i apply the patch should i
Quoth Mike Frysinger:
diff --git a/Makefile b/Makefile
[...]
@@ -109,25 +109,17 @@ Kconfig:
[...]
+config menuconfig qconfig gconfig xconfig: Kconfig conf
+ $(SCRIPTSDIR)/$(SCRIPTS_BINARY_$@) Kconfig
@if [ ! -f .config ]; then \
echo; \
echo You have
Quoth Claude:
On the other hand /dev/mtdblock1 is used for mount:
mount -t jffs2 /dev/mtdblock1 /mnt
Actually that's not true. The block device is used for mounting any
filesystem *other than* JFFS2, however JFFS2 actually prefers being given a
character device. (It will still work ok
Quoth Jamie Lokier:
A small note about licensing. uClibC is copyright licensed under the
LGPL.
If you have shared libraries on the target, it's easy to meet the LGPL
requirements. If you don't have shared libraries (e.g. ARM without
MMU), and some of your programs aren't open source, it's
Quoth Alessandro Rubini:
Not really: it's more than that for static linking. You must allow the
recipient to replace the library with a different version (rationale:
the library is free, so you should be able to fix a bug by upgrading).
For static linking, you need to provide the object
Quoth Xin Xie:
Sorry if this is question is too trivial. Currently I am using
m68k-uclinux-20061214 tool chain, and the uClinux-dist-20070130 tree. I
need to debug some user applications through the network gdbserver ,
but I cannot find the m68k-elf-gdb in the tool chain?
I recall having
Quoth Greg Ungerer:
I am sure some people will stand up and say threads work great
for them on specific architectures.
I've used both C++ and threads on Coldfire (5272) successfully with an old
(20041215) build of uClinux.
But what Greg said is true -- while the threading itself works fine
Quoth Bob Furber:
...YMMV. -- Whazzat?
http://www.google.com/search?q=define%3AYMMV
___
uClinux-dev mailing list
uClinux-dev@uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev@uclinux.org
To
Quoth Greg Ungerer:
In this example it is better to malloc(1000) than 1000 malloc(1)'s.
Lots of small allocations is slow and generally not good for
fragmentation.
Small in this case is a relative thing. Large allocations (especially
say larger than about 64k) can be difficult for the
Quoth Greg Ungerer:
This is due to the almost universal way that images are put
together on ColdFire. The single image file (which is a kernel
and filesystem concatenated) means that the fs needs to be copied
at kernel start to a more suitable place. And this is the
code you are seeing above,
Quoth Karthik Balaguru:
Is it good to boot from RAM or from a Flash that is good for XIP ?
Technically, you're always booting from Flash somewhere along the line :)
Starting the kernel from RAM is easier to get working, so that's probably
the best way to start out using uClinux. Obviously this
Quoth Harry Gunnarsson:
1) For you vendor, edit vendor/whatever/board/config.arch and change
to this
CROSS_COMPILE=m68k-uclinux-
2) For your vendor, edit vendor/whatever/board/Makefile and change
all explicit references to m68k-elf-objcopy, gcc etc to
m68k-uclinux-objcopy, gcc etc
If
Quoth Tiberiu Motoc:
test_program: test_program.o httpslib.o
$(CC) $(CFLAGS) $(LDFLAGS) -o $@ test_program.o httpslib.o
/home/motoct/uClinux-dist/lib/libcurl/compiled/lib/libcurl.a $(LDLIBS)
The linker error that I'm getting is:
Quoth David van Geest:
1. What are the options for including custom user software on the board?
Is it necessary to re-compile the kernel with your app included every time
you make a change to your program(s)?
Depends on how you've written the program :)
For compiled-in-kernel changes,
Quoth Mei Xu:
I'm developing a uClinux system on MicroBlaze on Spartan3E 1600 Dev
Board. The reference design runs a romfs. Can anyone tell me how to
set up a ramfs, since I need to run some user application on it?
Many of the vendor configs have an rc file or fstab file that mounts a ramfs
Quoth Mike Frysinger:
while some processors lack an MMU, then do not lack an MPU (memory
protection unit). for example, the Blackfin processor has an MPU which
means the Blackfin Linux port can (and does) prevent user space from
accessing peripherals. we can (and have an
Quoth Harry Gunnarsson:
So here's the problem, if I run this download-uncompress routine
back-to-back, it always works fine the first time, but it could
bail on the second/third/fourth time due to allocation failure
on one of the big buffers I need. I never figured out why this
is and I
Quoth Harry Gunnarsson:
Hm, you are saying one thing that gets me thinking. I assumed,
perhaps incorrectly, that the 'chunks' of memory in the
power-of-two scheme were pre-allocated at kernel start.
That is, n * 128k, m * 256k, y * 512k etc And I thought that
the big chunks of 2M and 4M
Quoth Felipe Uderman:
I am having some trouble finding uClinux documentation. I guess for
now that most of it is this list and the ucdot.org and uClinux
websites. Is there a document that explains how does uClinux works
on providing the drivers and abstracts the hardwares? Maybe it is
Moments ago, I said:
oom-killer: gfp_mask=0x40d0, order=9
[...]
Trying to allocate a further 1MB must therefore split the 2MB block in
half (since there's no VMM it's not possible to stitch smaller blocks
together).
In any case, allocating 1024KB from the 3040KB free would leave
Quoth Robin Getz:
Someone asked about why when their module was marked Proprietary,
they couldn't use GPL symbols, a suggestion was to:
3. create a GPL-licensed module that uses the GPL-only symbols and
make your proprietary driver talk through that module.
Although technically
Quoth Stanislav Meduna:
Actually what I need are the basic classes such as
vector or list - I don't need the exception support
as I found the exception handling too heavyweight
and error-prone in an embedded world - however, the
STL itself uses exceptions.
uClibc++ supports disabling
Quoth Eftimovski, Ivica:
I have also compiled and used the m68k-uclinux-tools-20061214
(binutils-2.16, uClibc-20060608 and gcc-4.1.1), but caveat emptor: the
cygwin shell parses text the DO$ way, and makefiles and .config will
fail because of CR/LF in the text. You will have to remove CR/LF
Quoth Aaron Zhong:
And df shows there is still aboe 2M free space in flash. I have seen
Gavin's reply on the previous letter. If it is caused by MTD driver,
how could I confirm raw MTD functions working correctly? Or there is
other methods to find the bug? Thanks in advance.
What I did was
Quoth [EMAIL PROTECTED]:
This strange problem isn't fixed yet. We've tried every trick we could
come up with, including the use of tools recommended by Matt Waddel,
remapping of the flash memory, the use of logs from JFFS2 and MTD and
nothing so far.
I had some problems with JFFS2 at first
Quoth Jamie Lokier:
But this part of the kernel will relocate incorrectly in border
cases with XIP:
if (r text_len) /* In text segment */
addr = r + start_code;
else/* In data segment */
Quoth Jamie Lokier:
It might be useful to see these files, and the exact version of
toolchain, and exact command line, and exact compiler messages (with
-v option, to show exactly which libraries are used), as there aren't
a lot of people reporting this problem.
I used to have it quite a lot
Quoth Bob Grimes:
I had tried this because of comments such as yours, Gavin (no, I am
most certainly NOT blaming you! :) ) in a last attempt to get C++ apps
to work, but that killed everything. (Well, okay, not everything, but
damn, I couldn't even change or list directories!).
Well, it'll
Quoth Bob Grimes:
Okay, I think I accomplished all this, with much gnashing of teeth on
this end. At one point, nearly everything was giving me the
BINFMT_FLT errors!!! But I've finally gotten it rebuilt, and things
seem somewhat better. For example, another program I wrote in C looks
like
Quoth Bob Grimes
BINFMT_FLAT: reloc outside program 0x73735f74 (0 -
0xffa52e54/0xd900), killing hello!
[...]
I've also seen some suggestions about disabling XIP, and I tried to
change that in vendors/Freescale/M5329EVB/config.arch, but I didn't
see any evidence that it had any
Quoth Claude:
I've got an stupid question, I think: Given a uclinux release, how
patches are applied?
I always have avoided patching code by downloading the latest
releases, but I think I should learn how to patch them.
This doesn't exactly answer your question, but probably the easiest thing
Quoth David McCullough:
Others have the embedded systems so well oiled it's quicker to
write/load/debug directly on the board.
Also, if you're testing user mode changes only (and your user code doesn't
require being in a fixed directory), then it's easy enough to tftp/ftp
something to a RAM
Quoth David Spain:
Anyone on the list know if its possible to make msh the default
shell at boot time rather than sash? And if so, how to go about
it? I am using uclinux-2.4.32-uc0.
Yes, it's easy -- there's a config option near where you turned on msh in
the first place. (You probably also
1 - 100 of 119 matches
Mail list logo
