Menu
▾
▴
njbfs-devel — Development List - (High Traffic)
You can subscribe to this list here.
| 2001 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(5) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2002 |
Jan
(2) |
Feb
(60) |
Mar
(9) |
Apr
(3) |
May
|
Jun
(4) |
Jul
|
Aug
|
Sep
(24) |
Oct
(2) |
Nov
|
Dec
(1) |
| 2003 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(1) |
Nov
|
Dec
|
| 2007 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Betty C. <bos...@a2...> - 2007-02-26 00:23:47
|
Hi, Chaepest CALIS and VAGRA online. http://parkerokay+now.com Remove "+" from the link. Ha, ha, ha, said Hermione sarcastically. Goblins dont need protection. Havent you been listening to what Professor Binns has been telling us about goblin rebellions? |
|
From: SourceForge.net <no...@so...> - 2003-10-27 21:05:19
|
Support Requests item #831317, was opened at 2003-10-27 13:04 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=422360&aid=831317&group_id=38403 Category: Install Problem (example) Group: None Status: Open Priority: 5 Submitted By: Nobody/Anonymous (nobody) Assigned to: Nobody/Anonymous (nobody) Summary: Compiling Error Initial Comment: Hi i'm tryin to compile the njbfs Filesystem. I used make install but i got the following error message gcc -O2 -fomit-frame-pointer -fno-strict-aliasing -pipe -mpreferred-stack-boundary=2 -Wall -I- -I. -I/usr/src/linux/include -DMODULE -D__KERNEL__ -DLINUX -DEXPORT_SYMTAB -DDEBUG -c nomad.c -o nomad.o In file included from nomad.c:42: /usr/src/linux/include/linux/malloc.h:4:2: #error linux/malloc.h is deprecated, use linux/slab.h instead. make: *** [nomad.o] Fehler 1 What can i do ? mail: pag...@we... ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=422360&aid=831317&group_id=38403 |
|
From: Ken M. <km...@ma...> - 2002-12-19 15:38:47
|
I just got a 10GB Nomad "Jukebox 2" which I'm trying to use with the 0.2.1 njbfs driver. I'm running on Redhat 8.0 with a 2.4.18 kernel and hotplug for usb. When I plug in the drive the log shows Dec 19 10:11:41 kmlinux kernel: hub.c: USB new device connect on bus1/2, assigned device number 2 Dec 19 10:11:41 kmlinux kernel: usb.c: USB device 2 (vend/prod 0x41e/0x4100) is not claimed by any active driver. Dec 19 10:11:45 kmlinux /etc/hotplug/usb.agent: Setup njbfs for USB product 41e/4100/1 Dec 19 10:11:45 kmlinux kernel: usb.c: registered new driver nomad Dec 19 10:11:45 kmlinux kernel: nomad.c: USB Creative Nomad II/Jukebox support registered. Dec 19 10:11:45 kmlinux kernel: driver.c: /proc/njb created Dec 19 10:11:45 kmlinux kernel: driver.c: /proc/njb/driver created and cat /proc/njb/driver shows Nomad Jukebox Driver Status --------------------------- NJB File System Version : 0.1 Connected : No A mount attempt yields Dec 19 10:12:41 kmlinux kernel: njb_usb.c: check: jukebox not connected Dec 19 10:12:41 kmlinux kernel: inode.c: could not connect Is the problem that this is a "Jukebox 2" and not the older "1", or am I running into a Linux issue with Redhat? Any suggestions? Ken Mandelberg | km...@ma... Emory University | Dept of Math and CS | Phone: Voice (404) 727-7963 Atlanta, GA 30322 | FAX 727-5611 |
|
From: <joe...@ne...> - 2002-10-09 22:40:07
|
> I tried njbfs a while ago and thought it had great promise. > > But for months...nothing. > > Is this project dead ? Sorry for the suspension of the project. When I pick up a NJB3 with the=20 IEEEE1394 interface, I will add some code. It will be around end of the = year=20 when the price becomes affordable. Cheers, J=F6rg |
|
From: Lon J. <lo...@di...> - 2002-10-08 05:39:53
|
I tried njbfs a while ago and thought it had great promise. But for months...nothing. Is this project dead ? Thanks LJ |
|
From: Ben O. <be...@gi...> - 2002-09-19 21:30:25
|
> > Hint: insmod errors can be caused by incorrect module parameters, > > including invalid IO or IRQ parameters. > > You may find more information in syslog or the output from dmesg Something that seems to commonly trigger this error for me is forgetting to load a usb module. |
|
From: <joe...@ne...> - 2002-09-19 21:28:59
|
> That leads me to a question: > What about me as a developer at least for documentation of njbfs? Frank, thank you for your support. If you submit something valuable here = on=20 this list, and people like it, it will get included into the project asap= =2E Best regards,=20 J=F6rg |
|
From: <joe...@ne...> - 2002-09-19 21:26:19
|
> Using /lib/modules/2.4.19/kernel/fs/njbfs/njbfs.o > Warning: loading /lib/modules/2.4.19/kernel/fs/njbfs/njbfs.o will taint > the kernel: forced load Why is that?=20 > See http://www.tux.org/lkml/#export-tainted for information about > tainted modules > /lib/modules/2.4.19/kernel/fs/njbfs/njbfs.o: init_module: Operation not > permitted Your kernel refuses loading because of a strict license policy. This shou= ldn't=20 happen, njbfs is GPL. > Hint: insmod errors can be caused by incorrect module parameters, > including invalid IO or IRQ parameters. > You may find more information in syslog or the output from dmesg Can you please have a look in /var/log/messages?=20 Best regards, J=F6rg |
|
From: Mental <Me...@Ne...> - 2002-09-18 23:56:22
|
sorry. here. oops. -- Mental (Me...@Ne...) This body. This body holding me. Be my reminder here that I am not alone in This body, this body holding me, feeling eternal All this pain is an illusion. --Tool "Parabola" CARPE NOCTEM, QUAM MINIMUM CREDULA POSTERO. GPG public key: http://www.neverlight.com/pas/Mental.asc |
|
From: Mental <Me...@Ne...> - 2002-09-18 23:55:22
|
Ok, I tried an older kernel. It panicked on me when usbmgr kicked on. I dropped the idea. I booted back into 2.4.19, and recompiled everything with modversions turned off. That fixed my unresolved symbols. I was still unable to inmod the module. I peppered the code with prink's to see what was going on. As near as I can tell, the struct for usb_drivers has changed. At this point I can insmod the module, and list mp3's in the directories as usual. My new issue of the moment is copying files. So far I've only tried copying from the jukebox. This is what I get: cp velvet\ acid\ christ\ -\ glority\ satan\ and\ suicide.mp3 ~ cp: skipping file `velvet acid christ - glority satan and suicide.mp3', as it was replaced while being copied I'll dive back into the code in a bit and see what I can figure out, but a nudge in the right direction would be cool. What're fops anyways, and where'd they go? Were they used to lock files for copying perchance? I've attached a unified diff (diff -urN njbfs njbfs.mental). Its probably over kill for what was essentially a 3 line change (I replaced a deprecated kernel header in an include statement as well). PS If I'm being a pest, say something and I'll spam the list a little less often. I'm just having a lot of fun with this at the moment :) -- Mental (Me...@Ne...) This body. This body holding me. Be my reminder here that I am not alone in This body, this body holding me, feeling eternal All this pain is an illusion. --Tool "Parabola" CARPE NOCTEM, QUAM MINIMUM CREDULA POSTERO. GPG public key: http://www.neverlight.com/pas/Mental.asc |
|
From: <joe...@ne...> - 2002-09-18 17:22:31
|
> depmod -ae > depmod: *** Unresolved symbols in > /lib/modules/2.4.19/kernel/fs/njbfs/njbfs.o > depmod: _mmx_memcpy > depmod: schedule_timeout > depmod: __generic_copy_from_user > depmod: usb_bulk_msg > depmod: generic_file_write > depmod: unlock_page This shouldn't happen. Maybe depmod is confused. E.g. you build a kernel = from=20 source but you have different headers in /usr/include/linux. Or the kerne= l=20 build wasn't clean, disturbed by older kernel object files, maybe a missi= ng=20 'make mrproper'. > Tomorrow I'm heading off to the bookstore to pick up the linux drivers > o'reilly book. I tried tracking down what was causing the unresolved > symbols to gain insight into why modules_init was failing. I didnt get = too > far with that, but I did learn a bit about how modules work. The O'Reilly book is a little bit outdated, it relies more on the 2.2 ker= nel. The missing symbols are generated by the tools of the modutils package by= =20 Keith Owens, and I think it is a build problem. There can be changed symb= ol=20 exports between 2.4.18 and 2.4.19 - which I do not expect - or a version=20 mismatch of kernel modules/kernel object files. Modutils has a sophistica= ted=20 method to recognize different kernel versions. > Anyways, I figured I'd just pass on the info in case anyone notices any > glaring errors in what I'm doing or had any suggestions as to things to > try. Thank you, it helps a lot, if I get the same messages. J=F6rg |
|
From: <joe...@ne...> - 2002-09-15 09:53:48
|
Am Sonntag, 15. September 2002 04:37 schrieb Mental: > At any rate, I've noticed that njbfs has issues with 2.4.19, or at leas= t > it does on my box at home. > Under 2.4.18-preemptable it would load, info would show up in /proc abo= ut > the jukebox, but I was unable to list files. I'll see if I have an olde= r > kernel I can try. Thank you for your report! I will include njbfs into my jp15 patch set wh= ich=20 is based on 2.4.20pre and will check what is going on. jp15 is to appear = in a=20 week or two, so please be patient. I think it's the Linux VFS change in=20 2.4.19pre which might confuse njbfs when listing files, giving "NFS stale= =20 handle" messages. I had to fix all related "remote" file systems, ftpfs f= or=20 example, or supermount. The URL: http://infolinux.de > I miss it working. It was very convenient :) Thanks! > Hopefully the project isnt dead. No, it's not dead. There had been some requests lately, so I wonder the n= jbfs=20 project has been many friends :-) It's only suspended for myself, still o= pen=20 for Ben and all other code committers, and if you feel comfortable writin= g C=20 code, as you say, I would appreciate you joining the project.=20 For myself, I suspended my activities until I have a new NJB 3 with IEEE = 1394=20 interface, which will be a new challenge for writing a modular njbfs driv= er,=20 serving both USB and IEEE 1394. But that will take time to pay a reasonab= le=20 price, maybe end of 2002. And, I will need some extra skills of programmi= ng=20 IEEE 1394 devices, so it will be fun. Best regards, J=F6rg |
|
From: Mental <Me...@Ne...> - 2002-09-15 02:41:28
|
Hello. A while back, I had njbfs up and working. I think you have a perl script I wrote for moving files in cvs. At any rate, I've noticed that njbfs has issues with 2.4.19, or at least it does on my box at home. The module compiles fine, but refuses to load. Under 2.4.18-preemptable it would load, info would show up in /proc about the jukebox, but I was unable to list files. I'll see if I have an older kernel I can try. At any rate, I was wondering about the state of the project. I greatly prefered it over libnjb since I didnt feel like writing my own app to manage music. If anyone has any ideas or suggestions, I'd be happy to hear them. If this is a known issue or whatever, just let me know. I could try to help. I've been doing a little C and I feel less intimidated by the code now. I miss it working. It was very convenient :) Hopefully the project isnt dead. -- Mental (Me...@Ne...) This body. This body holding me. Be my reminder here that I am not alone in This body, this body holding me, feeling eternal All this pain is an illusion. --Tool "Parabola" CARPE NOCTEM, QUAM MINIMUM CREDULA POSTERO. GPG public key: http://www.neverlight.com/pas/Mental.asc |
|
From: <joe...@ne...> - 2002-09-10 21:35:30
|
Am Dienstag, 10. September 2002 11:15 schrieb Steinhardt, Ingolf: > Hi J=F6rg, > > ich habe mir Dein/Euer Projekt mal angesehen - wede den Treiber demn=E4= chst > mal ausprobieren.... > > wie sieht den die "Roadmap" aus - gibt=B4s Neuigkeiten?? > > Gruss Ingolf > Berlin Hi Ingo, my "roadmap" is to buy an NJB 3 when it's cheaper, and extend the njbfs f= or=20 IEEE 1394 :-) but no special schedules yet. I'm happy if I get my fingers= on=20 it before Christmas... Cheers, J=F6rg |
|
From: Steinhardt, I. <Ing...@in...> - 2002-09-10 09:19:02
|
Hi J=F6rg, ich habe mir Dein/Euer Projekt mal angesehen - wede den Treiber = demn=E4chst mal ausprobieren.... wie sieht den die "Roadmap" aus - gibt=B4s Neuigkeiten?? Gruss Ingolf Berlin |
|
From: Ben O. <be...@gi...> - 2002-09-05 09:42:26
|
On Thu, 5 Sep 2002 10:54:22 +0200 Jörg Prante <joe...@ne...> wrote: Hey Jörg, how's tricks? > Right now, for simplicity, > the USB data transfer follows the kernel block size. Anyway, the USB data can > be transferred at *once*, i.e. some megabytes at a single command, without > interuption, called burst mode, and this is much faster. Easy from userspace, > easy from W$ndows, but it needs some extra work in Linux kernel. But this > extra work is preferable. It'd be nice, but just try grabbing 2mb of kernel memory at once section... it doesn't seem that the kernel allocates it like that. (am I missing something)? However, I do believe that bumping it up to 64k or so would bring us nearly on par with playcenter etc. I'm currently doing this with uploading, (see the "block_cache" code in proc.c in my branch) and the njb can't keep up. =) It'd be fairly easy to apply the same principal to downloads, although perhaps a bit trickier. > It would fit to data preprocessing. I started > months ago with ID3 tag processing preparing a more intelligent upload, to > get rid of special filenames for uploading proper MP3 files. This bit is pretty much done in my branch. check it out. |
|
From: <joe...@ne...> - 2002-09-05 08:58:20
|
On Wednesday 04 September 2002 21:39, Frank Meurer wrote: > BTW - libnjb seems to work faster. Can you give some time measurements? Here, njbfs is fast enough, only 15-20% slower than say PlayCenter. njbfs can be tuned for speed. There is some extra delay in the code right= now=20 because the USB data is transferred in 4KB blocks (which is the page size= of=20 the virtual file system in the Linux kernel), disconnecting from USB,=20 reconnecting to the USB again etc. for each block! Right now, for simplic= ity,=20 the USB data transfer follows the kernel block size. Anyway, the USB data= can=20 be transferred at *once*, i.e. some megabytes at a single command, withou= t=20 interuption, called burst mode, and this is much faster. Easy from usersp= ace,=20 easy from W$ndows, but it needs some extra work in Linux kernel. But this= =20 extra work is preferable. It would fit to data preprocessing. I started=20 months ago with ID3 tag processing preparing a more intelligent upload, t= o=20 get rid of special filenames for uploading proper MP3 files. Cheers, J=F6rg |
|
From: Ben O. <be...@gi...> - 2002-09-04 19:54:11
|
> > > > BTW - libnjb seems to work faster. > > > > ooooooooooooooooooooooooohhhhhhhhhhhhh. > > > > But er, um, njbfs is a um, cooler idea. Yeah. > > Yep! And it would be even more kewl with "rm" support at which I struggled > and fell into the deeps of your code. :-) Ah, the deeps of my code. How poetic. The "ben" branch actually does have "rm" support... check it out. |
|
From: Ben O. <be...@gi...> - 2002-09-04 19:11:05
|
Hi Daniel, Yup! The project is mostly dead, or at least has been stagnant for quite some time. There's near-release code sitting around on my hard drive (and in the "ben" branch of cvs, if you want to check it out), but there's still some show-stoppers (deadlocks) that need to be fixed. Maybe I'll lose my job soon and have the time again. Alternatives? Well, there's "Gnomad", which relies on libusb. There's also a windows plug-in that makes the Nomad show up in the explorer window. -ben On 04 Sep 2002 19:46:57 +0200 Daniel Demus <da...@de...> wrote: > Hello anyonoe who's out there! > > There hasn't been any activity in CVS for a long time. Is the project > dead, and if so what alternatives are there? > > Daniel Demus > > > > > ------------------------------------------------------- > This sf.net email is sponsored by: OSDN - Tired of that same old > cell phone? Get a new here for FREE! > https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390 > _______________________________________________ > njbfs-devel mailing list > njb...@li... > https://lists.sourceforge.net/lists/listinfo/njbfs-devel |
|
From: <joe...@ne...> - 2002-09-04 19:04:19
|
On Wednesday 04 September 2002 19:46, Daniel Demus wrote: > Hello anyonoe who's out there! > > There hasn't been any activity in CVS for a long time. Is the project > dead, and if so what alternatives are there? > No, the project is not really dead, it's suspended. I want to buy an NJB 3 and add extra IEEE 1394 capabilities one time, but now, it's way too expensive. If you feel you can go with userspace tools, there is the fabulous http://libnjb.sf.net project. Cheers Jörg |
|
From: Daniel D. <da...@de...> - 2002-09-04 17:41:43
|
Hello anyonoe who's out there! There hasn't been any activity in CVS for a long time. Is the project dead, and if so what alternatives are there? Daniel Demus |
|
From: <joe...@ne...> - 2002-06-23 09:53:06
|
> My goal was downloading of the classic music MP3s on the NJB which are > delivered with it. I got a kernel hang (or panic). Many lines flowed > through the console (if you need the output I will have to set up a serial > console). Currrently, njbfs only works with self-uploaded files. The "ls" command gives weird (unusuable) file sizes for the pre-installed files, zeroes or negative numbers, but not for the self-uploaded ones. The problem is the "verify" command fails for the pre-installed music. I learned the NJB comes in Europe/Germany with a different set of pre-installed music as in the US. Since I don't know how the NJB handles pre-installed files, I am unable to present a solution. Maybe the files are "protected", but it is only a question of how to deal with the header information of such files. The header format is not fully understood, at least for me. There is an exciting NJB reverse engineering activity at http://www.metalhead.ws/mike/ and this work is very useful to understand njbfs "bugs and features" because it gives insights into the NJB. > A: Yes, I'm interested in helping/working at your code, but unfortunately > my skills doesn't fit. Anyway: I own a Saehan (Germany: Schneider) MPMan > F20 with software for linux (some guy at UK wrote it) and I'm also very > interested in using it in a fs way. I will take a closer look at the > sources. I turned away from njbfs to build bigger patch collections for the Linux kernel. But if I got the new IEEE1394 "Nomad 3" device in my hands, I am sure I will pick up the njbfs work and extend the capabilities. Jörg |
|
From: Ben O. <be...@gi...> - 2002-06-23 02:59:40
|
On Sat, 22 Jun 2002 12:37:44 +0200 (CEST), "Frank Meurer" <me...@ma...> dithered away: > My goal was downloading of the classic music MP3s on the NJB which are > > delivered with it. I got a kernel hang (or panic). Many lines flowed > through the console (if you need the output I will have to set up a > serial console). As near as I can remember, the silly protection that Creative tried to impose on the mp3s that are shipped with the njb was located in Playcenter, not the firmware. So you may be seeing a more general problem... can you successfully download other tracks off the njb? > Q: Also deleting files doesn't work. Did I miss something? There's code sitting on my hard drive right now with a lot of new features. Including deleting tracks. I hope to release it fairly soon now. > A: Yes, I'm interested in helping/working at your code, but > unfortunately my skills doesn't fit. Anyway: I own a Saehan (Germany: > Schneider) MPMan F20 with software for linux (some guy at UK wrote it) > and I'm also very interested in using it in a fs way. I will take a > closer look at the sources. Skillset shmillest. I knew nothing about linux fs/drivers coming into this, now I know lots. -ben |
|
From: Frank M. <me...@ma...> - 2002-06-22 10:34:19
|
Hello,
last night I read about njbfs when looking for libnjb.
I got v 0.2.1 compiled it without problems (only already known warnings
about deprecated malloc-includes etc.).
First I compiled it with gcc-3.1 and linux-2.4.18 and then run it on RH
kernel linux-2.4.8-5.
As an automunt-user I used your make-target, but the the script-restart
failed ("/etc/rc.d/init.d/auto.... restart" more compatible?).
I set it up manually and it works, too.
My goal was downloading of the classic music MP3s on the NJB which are
delivered with it. I got a kernel hang (or panic). Many lines flowed
through the console (if you need the output I will have to set up a serial
console).
Then I tried it on another system (Mdk 7.2, gcc 2.95.2, linux-2.4.9 with
Alan Cox patches). It also worked fine except for downloading the classic
music MP3s. The attempt to download these caused an immediately reboot.
Q: Are the delivered MP3s protected in any way? I don't read about it in
your docs.
I got the latest version of njbfs from CVS but it shows the same
behaviour.
Q: Also deleting files doesn't work. Did I miss something?
A: Yes, I'm interested in helping/working at your code, but unfortunately
my skills doesn't fit. Anyway: I own a Saehan (Germany: Schneider) MPMan
F20 with software for linux (some guy at UK wrote it) and I'm also very
interested in using it in a fs way. I will take a closer look at the
sources.
Thanx for great work!
regards,
Frank
--
Sending unsolicited commercial email to this address may be a violation
of the Washington State Consumer Protection Act, chapter 19.86 RCW.
Frank Meurer <me...@ST...>, PGP ID: 0x5E756DA8
Key fingerprint = 169A 1138 8DB4 528F 2F01 20A6 EDD8 49C3 5E75 6DA8
|
|
From: Ben O. <be...@gi...> - 2002-04-21 07:30:57
|
sorry, just switched from sendmail to exim and had things all f'ed up. Should be receiving mail now... -ben |
9 messages has been excluded from this view by a project administrator.
