Nice to see that the new cvs-server is up and running
again. I will try the new release you made now and build NSLU2 packages too.
Christian
I havent thought much about it before, but isnt
it wrong to install the include-files in the install-target?
Who cares about the content in ow_1820.h for example
(after owfs is installed).
I fell into some problem when I compiled owfs for
NSLU2 before and all include-files where
The directory structure worked in 2.3p5 when starting owfs/owserver like
this:
owserver -p 3001 -u
owfs -s 3001 /tmp/1wire
[EMAIL PROTECTED]:~# ls -l /var/1wire/
drwxr-xr-x1 root root1 May 16 11:37 10.061847000800
drwxr-xr-x1 root root1 May 16 11:37
Damn Paul's 1-wire sensors... He got too many temperature sensors now
and he should donate some of them to the poor people who don't know
anything about owfs.
/Christian
On Tue, 2006-05-09 at 23:46 -0700, Vadim Tkachenko wrote:
Hello all,
Had to approve two messages today because of the
I made a quick compilation of temploggerd to NSLU2, and it seems to
work It depends on rrdtool, and there already exist rrdtool-1.2.12
in the repository... It's much bigger than rrdtool-1.0.50 and won't fit
in the flash, so it's not useful in the openslug distribution. Perhaps I
will compile
root 3076 May 9 06:24
time_links.html
Makes me wonder about permissions on cgi-bin
Paul
On 5/9/06, Christian Magnusson [EMAIL PROTECTED] wrote:
I made a quick compilation of temploggerd to NSLU2, and it
seems to
work It depends
-rw-rw-rw-1 root
root210
May9 06:24 meta_tags.html
drwxrwxrwx2 root
root 4096 May9
06:22 rrd
-rw-rw-rw-1 root
root 2735 May9 06:24
style.css
-rw-rw-rw-1 root
root 3076 May9
06:24
time_links.html
Makes me wonder about permissions on cgi-bin
Paul
On 5/9/06, Christian Magnusson [EMAIL
I have finally made some ipk-packages for Unslung
6.8. They might work for other distributions, but I dont know yet.
http://home.mag.cx/nslu2/packages/_README
http://home.mag.cx/nslu2/packages/
/Christian
I have finally made some ipk-packages for Unslung
6.8. They might work for other distributions, but I dont know yet.
http://home.mag.cx/nslu2/packages/_README
http://home.mag.cx/nslu2/packages/
/Christian
Seem to be a bug in module/owfs/src/c/owfs.c
#ifndef OW_MT
Fuse_add(-s , fuse_options) ; // single threaded
#endif /* OW_MT */
#if FUSE_VERSION = 26 // requires extra parameter
c = fuse_main(fuse_options.argc, fuse_options.argv, owfs_oper) ;
//printf(fuse_main=%d\n,c);
#else
I finally managed to compile a working fuse-2.6.0-pre2
and owfs-2.3p3 version for the NSLU2.
Its compiled natively right now, so I dont
have any ipkg package yet.
Ill try to make a cross-compilation just to
create a ipkg package If I fail you could always get a tar-file from me
with
owfs and owserver have been stable for almost 20 hours on WRT54GS now,
so I guess you found the dead-lock I had problem with.
/Christian
On Tue, 2006-05-02 at 15:58 +0200, Christian Magnusson wrote:
Great... I think the dead-lock problem on the WRT54G router disappeared
now. I'll tell you
oops..
My latest build might be compiled without libusb... I'll fix that..
/Christian
On Wed, 2006-05-03 at 11:06 +0200, Kurt André Selbach wrote:
Hi again,
Never got the LINK adapter to work on my router,
however i have my DS9490R now, and trying to get it to work.
but owserver/owfs
Great... I think the dead-lock problem on the WRT54G router disappeared
now. I'll tell you more in a couple of hours...
/Christian
On Mon, 2006-05-01 at 23:00 -0400, Paul Alfille wrote:
I found a locking error with multiple clients against the same device.
New release 2.3p2
One other
I fixed this bug in the
CVS When you compile owfs without support for usb-adapters you get into
this problem. It was a bug in ow_opt.c which tries to call the function DS9490_enumerate()
which doesnt exist if USB is disabled.
This was my fix for it in
module/owlib/src/c/ow_opt.c
I noticed some time ago that some directories are
missing in the /bus.x libraries nowadays
Its possible to access them, but they are not
listed in owhttpd or owfs when they are connected to a remote owserver.
# owserver p 3000 u
# owhttpd p 3001 s 3000
uncached, bus.0, settings,
I noticed the
path_busless never were created as a new string WITHOUT the /bus.x
part in Parse_Bus()
I think I fixed the
problem and left some #if 0 in the code. Havent tested it too
much yet.
/Christian
Från: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] För Christian
I recompiled the latest cvs now, and I noticed you changed the timeout from
0.5 to 1.0 second.
Problem with include i2c-dev.h and I2C_SLAVE in ow_ds2482.c, but it have
comment it out... Will you checkin that new include file?
I have actually noticed some dead-locks on my wrt54gs router
I'll make a new build from this latest cvs again...
/Christian
-Ursprungligt meddelande-
Från: [EMAIL PROTECTED] [mailto:owfs-developers-
[EMAIL PROTECTED] För Paul Alfille
Skickat: den 25 april 2006 01:30
Till: owfs-developers@lists.sourceforge.net
Ämne: Re: [Owfs-developers]
@lists.sourceforge.net
Ämne: Re: [Owfs-developers] OWFS on a ASUS wl500g Deluxe
Thanks!
Both of you - just noted some new 03:XXam files on the server going to
try them know.
On 4/25/06, Christian Magnusson [EMAIL PROTECTED] wrote:
I'll make a new build from this latest cvs again
for
openwrt. It appears the current version works fine. It also appears that
we
can't blame the serial, link, or usb components.
Christian Magnusson knows how to build for embedded platforms. I don't
know
how, and don't see an obvious target in Makefile.am
Paul
On Friday 21 April 2006 07:01
@lists.sourceforge.net
Kopia: [EMAIL PROTECTED]
Ämne: openwrt (kamikaze version 3695, Linux OpenWrt 2.6.16.1) owfs problem
- Transport endpoint is not connected
I have the owfs running (thanks Christian Magnusson ) under an older
openwrt Linux OpenWrt 2.6.12.5 on a Netgear wgt634u using the following
ipk-package to you.
/Christian
-Ursprungligt meddelande-
Från: [EMAIL PROTECTED] [mailto:owfs-developers-
[EMAIL PROTECTED] För Christian Magnusson
Skickat: den 30 mars 2006 22:59
Till: owfs-developers@lists.sourceforge.net
Ämne: SV: [Owfs-developers] owfs and nslu
I have
Paul,
Sorry for the lazyness, but did you fix the problem
with crashing owfs when starting with -u parameter AND no
usb-adapter connected?
I was trying my latest binaries for NSLU2 and (2.2p3)
and it crashed at once (as expected) since I didnt have any adapter
connected.
(Havent
I have managed to compile everything except kmod-fuse now, and I'm working
on it as much as I can... I have a busy life here at home right now with two
twin-daughters who cries for attention all the time.
Owfs, owlib, libfuse, fuseutils are done so far...
/Christian
-Ursprungligt
-files?
/Christian
Från:
[EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] För Christian Magnusson
Skickat: den 23 mars 2006 19:44
Till: owfs-developers@lists.sourceforge.net
Ämne: SV: [Owfs-developers] owfs
and nslu
I have ordered a NSLU2,
but I have started to do a full cross
I'm not very used to svn at all... but I will of course try to learn
something more about svn if you are changing to it.
I have no clue which benefits we could use from svn that makes it important
to change now... The cvs-statistics which be lost... that's bad... so are
there any similar utility
I just noticed that a bug was introduced lately when
owserver doesnt find any usb adapter at startup. This could eventually
happen when adapter was found and later was reconnected due to some error.
BUS_selectection_error() is called, and reconnect is
initiated.
A segmentation fault
Alfille
Skickat: den 19 februari 2006 07:44
Till: owfs-developers@lists.sourceforge.net
Ämne: Re: [Owfs-developers] Timeout again
On Monday 13 February 2006 05:22 pm, Christian Magnusson wrote:
Nope... 2 seconds isn't enough... Can't tell you why, but it's not
working for me at least..
/Christian
I have looked a bit on your new updates, and I'm trying to verify that it
works for the linksys routers at least. My first try seemed successful, and
I will continue to run the latest cvs this night and see if it crash or
hangs.
It seems to be very slow response-time now, but I'll tell you more
Paul Alfille
Skickat: den 19 februari 2006 20:25
Till: owfs-developers@lists.sourceforge.net
Ämne: Re: SV: SV: [Owfs-developers] Timeout again
On Sunday 19 February 2006 02:07 pm, Christian Magnusson wrote:
It seems to be very slow response-time now, but I'll tell you more about
my
experience when
I have updated the version number in configure.ac to 2.2p1cvs.
I hope this is ok for you so people could refer to which a more specific
version. When its time to release a new version, cvs
should be released before creating the tar and rpm packages.
I hope this is ok for you
How is
That seems to be one solution to the problem.
We are really close to have children now (twins actually), so I doubt I
will have time to do this myself right now. If nothing happens until
Monday, we are going to the hospital to manually start it.
Until then I have lots of things to take care of
I found out that the recently added read-timeout doesnt
work good at all
A timeout of 1 second in ow_server.c is too low
Dont know exactly why, but I end up with EIO errors when reading
temperatures very often.
Host1: owserver
d /dev/tty/1 (Slow Linksys router with 1-wire adapter
.
I've always wanted to implement owfake, allowing fake devices for testing.
Paul Alfille
On Monday 13 February 2006 04:08 am, Christian Magnusson wrote:
I found out that the recently added read-timeout doesn't work good at all.
A timeout of 1 second in ow_server.c is too low. Don't know
Nope... 2 seconds isn't enough... Can't tell you why, but it's not
working for me at least..
/Christian
[EMAIL PROTECTED] owfs]# cat /var/1wire/uncached/1*/temperature
26.5625 20.4375 19.75 15.93759.25 22.0625
0.125cat:
will come up later.
-
Gregg C Levine [EMAIL PROTECTED]
---
Remember the Force will be with you. Always. Obi-Wan Kenobi
-Original Message-
From: [EMAIL PROTECTED]
[mailto:owfs-developers-
[EMAIL PROTECTED] On Behalf Of Christian Magnusson
Sent: Monday, February
Great,
We finally have a stable release which also could be used for the
embedded systems. I'll download it and try at at once and make sure it
works (without any missing files) for all my systems at home.
/Christian
On Fri, 2006-02-03 at 22:39 -0500, Paul Alfille wrote:
I've just created a
I don't know if you managed to fix the problem yourself, but the main
reason is that your new installation contains fuse-2.4 from beginning
+ you have fuse-2.5.x installed at /usr/local/include/fuse*
and /usr/local/lib/libfuse*.
You should either install the rpms fuse, fuse-libs and fuse-devel,
messages in English in the Moscow subway.
On 1/7/06, Christian Magnusson [EMAIL PROTECTED] wrote:
Has anyone else tried DS2436B devices?
I have no device which I could test, so could anyone else verify if the
code
is broken?
/Christian
-Ursprungligt meddelande-
Från: [EMAIL PROTECTED
New ipkg-packages are available at my repository now.
(http://home.mag.cx/openwrt/packages/)
/Christian
-Ursprungligt meddelande-
Från: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] För Christian Magnusson
Skickat: den 6 januari 2006 23:50
Till: owfs-developers@lists.sourceforge.net
-Ursprungligt meddelande-
Från: OpenWrt Mailer [mailto:[EMAIL PROTECTED]
Skickat: den 7 januari 2006 16:11
Till: [EMAIL PROTECTED]
Ämne: Reply to topic: 'OWFS and temploggerd'
leonardk has replied to the topic 'OWFS and temploggerd' to which you are
subscribed. There may be more new
I will build a new ipkg package tomorrow based on the latest cvs-version. I
have been very busy at work lately, so I haven't been able to do much for
owfs the last weeks.
/Christian
-Ursprungligt meddelande-
Från: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] För Paul Hilton
Skickat: den
I really think a new version (2.2) should be released from the cvs as
soon as possible. There are no major bugs in the current cvs-version,
and there are tons of new features and fixes compared to the old 2.1
version. That old version is really outdated nowadays.
Most compilation errors
Set ip-number with -p parameter if you want.
[EMAIL PROTECTED] ~]# owhttpd -p 127.0.0.1:5678 -s 3001
[EMAIL PROTECTED] ~]# netstat -an | grep 5678
tcp0 0 127.0.0.1:5678 0.0.0.0:* LISTEN
On Tue, 2005-12-20 at 21:39, Krzysztof wrote:
Hello
Is it possible to
Thanks for the e-mail... I'm a bit in a hurry, and have to leave in a few
minutes... I'll write a full reply later when I'm back.
3. Christian, it appears you have moved away from the complete firmware
solution for ewrt/owfs and gone to openwrt and packages, what were your
reasons for doing so
the embedded
platform.
Christian Magnusson has created the embedded distributions, but I'm not sure
he included owtcl in that.
It sounds like you are specifically interested in TCL for the WRT. That
means
the tcl library and binary (I can't see the utility of Tk without a display
device, unless you
RPATH is printed by objdump on i386 platforms... It's not an environment
variable. I didn't know otool was used on MacOS X.
The latest CVS would NOT compile,
ld: unknown flag: --rpath
make[4]: *** [owhttpd] Error 1
I have removed this patch for MacOS X again...
A quick look at
On Wed, 2005-11-09 at 06:44 -0600, Ross Cathriner wrote:
Can you try this...
mkdir $HOME/temp
mkdir $HOME/temp/lib
mkdir $HOME/temp/bin
cp owhttpd $HOME/temp/bin
cp libow.0.so $HOME/temp/lib
LD_LIBRARY_PATH=$HOME/temp/lib
export LD_LIBRARY_PATH
otool -L $HOME/temp/bin/owhttpd
If you look closer to the binary with objdump, you get some more
information about required libraries and paths.
objdump -x /opt/owfs/bin/owfs
Dynamic Section:
NEEDED libow.so.0
NEEDED libfuse.so.2
NEEDED libpthread.so.0
NEEDED libusb-0.1.so.4
NEEDED
I have updated the owfs code to work with the latest released
fuse-2.4.1, but it should also compile with all versions between
fuse-1.4 - fuse-2.3.
I don't know which version to recommend since all versions seem to
work when you compile them together from scratch.
Just to be sure about the
/owfs/bin/owfs -u /mnt/1wire/ again, and this worked too
even if owfs use an older fuse-library.
I guess this tells us that it's possible to make rpm's of owfs that
depends on fuse-2.2.
/Christian
On Tue, 2005-10-18 at 11:00 +0200, Christian Magnusson wrote:
I have updated the owfs code to work
On Sat, 2005-10-15 at 05:00, Ross Cathriner wrote:
Still no joy. The parsing of the address does look better, don't
know if its right or not.
INFO: Set DS9490 [818DE1240063] unique id to adapter
[004/007-04fa-2490-ff-ff]
Things we know: I changed to the 004/007 usb port , Vendor
On Sun, 2005-10-16 at 17:57, Ross Cathriner wrote:
HAPPY, HAPPY, JOY, JOY
httpd how fully operational on MacOSX. Put another feather in your
hat.
The full transscript can be viewed
here: http://homepage.mac.com/rnc/startup_log.txt
The key factor is the --foreground, and terminal
I have fixed the bug now. I hope you could try it now again. I re-built
the daily snapshot too, so download it tell me if it works.
/Christian
On Fri, 2005-10-14 at 07:24 +0200, Christian Magnusson wrote:
That looks good to me... But it's a VERY scary USB-device name... A big
buffer overflow
I made some updates to owfs to support fuse-2.4. Have tried to use
some older fuse-versions too, and I think it should work as it
should.
I have also made some test-packages of owfs-2.2p0RC and fuse-2.4.0 to
the WRT54G router and openwrt at my repository:
http://home.mag.cx/openwrt/packages/
I have updated the makefiles to make it possible to make dist and
rpm again. Library extensions for libow.so and libowcapi.so should be
fixed too.
The created tar-ball owfs-2.2p0RC.tar.gz seem to be complete after
make dist now, so it should be possible to create a new release
with this tar-ball
I have seen this too when cross-compiling rrdtools-1.0.50 to the WRT54G
router. I had to set an environment variable before building the
libraries just to add .so as the correct library-extension.
for bash:
shrext_cmds=.so
export shrext_cmds
./configure
make
for tcsh:
setenv shrext_cmds
That subject almost sounded ironic to me... :) Are you tired of
new changes?
Anyway, nice that this capi finally is developed. I will be back
with some additions to this later. I have written some c-applications
using a mounted fuse filesystem, and I have always been missing lots of
functions
On Sun, 2005-10-02 at 19:37 -0500, Ross Cathriner wrote:
I got myself confused and therefore gave you bad information.
Originally I found two versions of time.h, the first at
usr/include/time.h has
struct tm {
inttm_sec;/* seconds after the minute [0-60] */
in the code.
Try the latest CVS-version instead... If you don't have cvs on your
computer, download the daily snapshot instead at:
http://owfs.sourceforge.net/daily/owfs-latest.tar.gz
I have recently added a parameter called --disable-swig to the configure
script just to make it easier to remove the swig
BTW, the use of FF or any other opcode as done by some already IS
illegal and opens you and your companies to possible legal action from
Dallas. And the use of all the serial numbers is controlled by Dallas
(as it must be). And while Dallas is willing to sell serial number
blocks it is
For example from the combination:
DS18S20 + DS2405 + DS18S20 + TAI-8570 (AAG Pressure Sensor)
I can read only: DS18S20 + DS2406 (Pressure Sensor)
welcome to the bug-ground... try reading the adresses with digitemp -a
- -w -s serial ... you should be able to acess the directory using
Actually no... Since owhttpd doesn't work for you, the problem is
probably not located in fuse (or wrong installed fuse)
/Christian
On Tue, 2005-09-13 at 14:54 -0400, Phil Strong wrote:
OK so I heard recently that fuse is in the new kernel standard now.
Do you think upgrading to that new
I'm not sure why it crash, and I doubt I could fix it without really
logging in to your server and find out by myself.
The backtrace from owhttpd you sent, is either missing debug-information
or perhaps the stack got corrupted.
As far as I know you are running Fedora Core 3 which should have a
On Fri, 2005-09-09 at 07:46 +0200, Daniel Hiepler wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Vom: Fri, 09 Sep 2005 07:35:32 +0200
I presume all the data pins are connected together and all the
ground pins (and VDD for DS1820) are connected together. Except
for
I guess you have to debug this with gdb. Could you try to turn
on the core dumps in your shell:
tcsh: limit coredumpsize unlimited
bash: ulimit -c unlimited
Start owfs and do you crashing ls.
/opt/owfs/bin/owfs --error_level=9 -u /mnt/1wire/
ls /mnt/1wire/
Check if any core-file is generated
-1_mipsel.ipk
temploggerd_1.1-1_mipsel.ipk
When I have tested it some more I will upload the files to the
file-area.
--
Christian Magnusson [EMAIL PROTECTED]
---
SF.Net email is Sponsored by the Better Software Conference EXPO
September 19-22
On Sat, 2005-08-06 at 19:20, Sven Geggus wrote:
Hi there,
I try to use owfs on a alder Pentium 233 Machine which does not seem to
work:
[EMAIL PROTECTED]:./sven owserver -p 4711 -u --foreground --error_level 3
INFO: PARSENAME path=
ERR: Opened USB DS9490 adapter at 001/005.
: No data
I would like to keep it in the option to have shared libraries in the
configure script at least... Perhaps --enable-static should be
default, but it should be possible to configure with --enable-shared if
you want...
I'm trying to patch the configure script just to make it possible to
configure
wrote:
Christian Magnusson [EMAIL PROTECTED] wrote:
You might need to bind owserver to listen on all addresses such as
owserver -u -p 0.0.0.0:5005
owhttpd -p 8080 -s 5005
Hm, I did not need to do this with the CVS Version from about one month
ago.
Anyway, if I use
On Sun, 2005-07-31 at 21:57, Paul Alfille wrote:
On Sunday 31 July 2005 03:27 pm, Paul Alfille wrote:
No change was intended. Sounds like a new bug.
Paul
On Sunday 31 July 2005 06:35 am, Sven Geggus wrote:
Christian Magnusson [EMAIL PROTECTED] wrote:
You might need to bind
Hi there... I found the problem for you,
On Wed, 2005-07-20 at 12:07 -0700, Miguel FC wrote:
Hi Christian, thanks for answering.
--- Christian Magnusson
[EMAIL PROTECTED] wrote:
Seems like the makefiles are trying to build the
owperl when you
don't have perl installed. Try disable
--
Christian Magnusson [EMAIL PROTECTED]
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need
On Tue, 2005-07-12 at 18:13 +0200, Christian Magnusson wrote:
I have checked in the changes now... I have tried it a couple of
times on both Solaris and Linux x86 and I don't think I have broken
anything at least.
Please check it and tell me if it doesn't compile for you.
/Christian
called.
INFO: MOUNT: MNT
(10 seconds delay)
INFO: MOUNT: MNT directory = /
INFO: Mount program called.
INFO: MOUNT: UMNT
--
Christian Magnusson [EMAIL PROTECTED]
---
This SF.Net email is sponsored by the 'Do More With Dual!' webinar
have any ideas of changes to make it
cleaner. Will check in all changes as soon as I can here.
/Christian
On Tue, 2005-07-12 at 09:57, Christian Magnusson wrote:
On Mon, 2005-07-11 at 19:24 +0200, Krzysztof wrote:
Christian Magnusson wrote:
There were no dependencies to libpthread when
://www.hp.com/go/dualwebinar
___
Owfs-developers mailing list
Owfs-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/owfs-developers
--
Christian Magnusson [EMAIL PROTECTED
There were no dependencies to libpthread when compiling libow.so,
so OW.so loaded libow.so and didn't understand to use libpthread.so
too.
I have updated some of the makefiles and configure.ac to fix the
problem (I hope). I guess this fix only was needed for older libc.
/Christian
On Sun,
Hi Jan,
I have been able to reproduce your error, and I have found the problem
now... Trying to solve it and a solution will soon be available. Ignore
the latest source file I sent you yesterday evening.
--
Christian Magnusson [EMAIL PROTECTED
] on adapter [001/009]
INFO: Set DS9490 [814657230051] unique id to adapter [001/009]
/Christian
On Fri, 2005-07-08 at 16:25, Krzysztof wrote:
Christian Magnusson wrote:
You should probably try the latest cvs-version of fuse. That version
works much better with mount-permissions.
cvs
usb_release_interface() is
called according to the output you sent.
If you for some reason have many file-descriptors open to the same
usb-device there are other things to fix for me.
BTW: which kernel do you use?
/Christian
On Thu, 2005-07-07 at 08:58 +0200, Jan Kandziora wrote:
Hallo Christian Magnusson, am
/dualwebinar
___
Owfs-developers mailing list
Owfs-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/owfs-developers
--
Christian Magnusson [EMAIL PROTECTED
.
--
Christian Magnusson [EMAIL PROTECTED]
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need
for WRT54G and Coldfire,
therefor some of the other strange changes. There are no leaks for my
Coldfire board now at least, but haven't tried the WRT54G router yet.
--
Christian Magnusson [EMAIL PROTECTED]
---
SF.Net email is sponsored by: Discover
___
Owfs-developers mailing list
Owfs-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/owfs-developers
--
Christian Magnusson [EMAIL PROTECTED]
---
SF.Net email is sponsored by: Discover
You might be helped by looking at the makefiles I use for the linksys
router at this address:
http://home.mag.cx/uclinux/owfs/patches/linksys/
/Christian
On Mon, 2005-07-04 at 10:46 +0200, Christian Magnusson wrote:
On Sun, 2005-07-03 at 21:44 -0400, Paul Alfille wrote:
On Sunday 03 July
-wire device is saved as an identifier.
/Christian
On Fri, 2005-07-01 at 17:50, Alfille, Paul H.,M.D. wrote:
This works for newer DS9490s, the older ones used an 01 device.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of
Christian Magnusson
Sent: Friday
it?
/Christian
On Fri, 2005-07-01 at 09:06 +0200, Jan Kandziora wrote:
Am Donnerstag, 30. Juni 2005 21:56 schrieb Christian Magnusson:
You are right... The kernel thinks the usb adapter is physically
removed and then inserted again and get a new device number.
This problem is very fatal
Forgot to say that kernel should be compiled with CONFIG_PROFILING=y
to get information about accumulated time in each function.
/Christian
On Wed, 2005-06-29 at 14:47 +0200, Christian Magnusson wrote:
I have optimized some functions which I noticed took very long time to
execute, and I have
Have you succeeded to use the latest cvs-version yet? Just wondering...
/Christian
On Tue, 2005-06-28 at 16:21 +0200, Jan Kandziora wrote:
Am Dienstag, 28. Juni 2005 13:54 schrieb Christian Magnusson:
When I removed one iButton and really tried to simulate bad connection
by moving
13:17 schrieb Christian Magnusson:
Try the latest cvs again... I have fixed a missing
usb_release_interface() and some other statistics from those errors.
/Christian
Sorry for the delay. I had my business partners here, which messed up
things a lot.
I've done some testing
PROTECTED] Behalf Of
Christian Magnusson
Sent: Tuesday, June 28, 2005 7:54 AM
To: owfs-developers@lists.sourceforge.net
Subject: RE: [Owfs-developers] More on Hanging Problems
I feel pretty comfortable with the USB reconnection now. I started
owhttpd and made a loop reading some files from
install
/opt/owfs/bin/owhttpd --foreground -p -u
(Stress-test it for a while)
Ctrl-C
gprof -b /opt/owfs/bin/owhttpd
--
Christian Magnusson [EMAIL PROTECTED]
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from
(at next read it reconnects with success)
ERR: USB DS9490 adapter reconnected
On Tue, 2005-06-28 at 07:31 +0200, Christian Magnusson wrote:
I agree... 3 attempts are perhaps not necessary. If it fails it should
only be necessary to try 1 attempt and then return read-error, since
it's
great that Jan has a setup where he can consistently trigger
the errors.
Paul
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of
Christian Magnusson
Sent: Sunday, June 26, 2005 4:31 PM
To: owfs-developers
Subject: Re: [Owfs-developers] More
offer some feedback on this.
Paul
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of
Christian Magnusson
Sent: Monday, June 27, 2005 7:18 AM
To: owfs-developers@lists.sourceforge.net
Cc: [EMAIL PROTECTED]
Subject: Re: [Owfs-developers] More on Hanging
= fread(tmp[pos], 1, 1, fp)) = 0) {
tmp[pos] = 0;
break;
}
pos++;
} while(1);
fclose(fp);
/Christian
On Wed, 2005-06-22 at 09:40 +0200, Christian Magnusson wrote:
I noticed some strange behavior with the direct_io flag in fuse. My
cat return an errno Cannot assign
-06-22 at 16:18, Christian Magnusson wrote:
I noticed I had to fix all those FS_r_all and the array-stuff too, just
to support offset for all files. I'm almost done and will checkin the
other changes later tonight or tomorrow morning.
/Christian
On Wed, 2005-06-22 at 14:23 +0200, Christian
the 1st process stopped thus after 10
seconds. Is this the right behaviour? If so why?
Data from the first read is cached, and immediately returned to the
second read, once the first was completed.
You can try '-odirect_io' to make each read() start a new request.
Miklos
--
Christian
201 - 300 of 319 matches
Mail list logo
