![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
ext3 bug...
Sep. 22nd, 2004 12:21 amBah, this is what I get for being on the bleeding edge...
root@hevesy-int /dev/shm # dd if=/dev/urandom of=/tmp/testfile bs=1k count=4 4+0 records in 4+0 records out root@hevesy-int /dev/shm # file /tmp/testfile /tmp/testfile: empty root@hevesy-int /dev/shm # dd if=/dev/urandom of=/tmp/testfile bs=1k count=5 5+0 records in 5+0 records out root@hevesy-int /dev/shm # file /tmp/testfile /tmp/testfile: data
As at 2.6.9-rc1-mm1, ext3 fails with files that are exactly 4kb in size.
Due Dates - Reminder To Self
Sep. 13th, 2004 12:37 pmJust my personal list of everything still due that I need to do this module.
INTD401.2 - Sept 19 - Project Brainstorming
INTD401.4 - Oct 3 - Project Brief
ITEC401.1 - Sept 17 10am - Use and Requirements of Safety Systems
ITEC401.3 - Sept 24 10am - Implementation of Safety-Critical Systems
ITEC401.5 - Oct 8 - TechWeek Assignment
ITEC404.5 - Oct 8 - Project
ITEC413.2 - Sept 18 - Project plan and progress report
ITEC413.2 - Sept 19 - Muddiest Point
ITEC413.2 - Sept 21 - Conferencing Questions
ITEC413.3 - Sept 26 - Muddiest Point
ITEC413.3 - Sept 24 - Conferencing Questions
ITEC413.4 - Oct 3 - Muddiest Point
ITEC413.4 - Oct 6 - Conferencing Questions
ITEC413.5 - Oct 6 - Project
ITEC419.1 - Sept 16 - Assignment
ITEC419.2 - Sept 23 - Assignment
ITEC419.3 - Sept 30 - Assignment
MTEC401.2 - Sept 19 - Conferencing Questions
MTEC401.3 - Sept 26 - Conferencing Questions
MTEC401.4 - Oct 3 - Conferencing Questions
INTD401.2 - Sept 19 - Project Brainstorming
INTD401.4 - Oct 3 - Project Brief
ITEC401.1 - Sept 17 10am - Use and Requirements of Safety Systems
ITEC401.3 - Sept 24 10am - Implementation of Safety-Critical Systems
ITEC401.5 - Oct 8 - TechWeek Assignment
ITEC404.5 - Oct 8 - Project
ITEC413.2 - Sept 18 - Project plan and progress report
ITEC413.2 - Sept 19 - Muddiest Point
ITEC413.2 - Sept 21 - Conferencing Questions
ITEC413.3 - Sept 26 - Muddiest Point
ITEC413.3 - Sept 24 - Conferencing Questions
ITEC413.4 - Oct 3 - Muddiest Point
ITEC413.4 - Oct 6 - Conferencing Questions
ITEC413.5 - Oct 6 - Project
ITEC419.1 - Sept 16 - Assignment
ITEC419.2 - Sept 23 - Assignment
ITEC419.3 - Sept 30 - Assignment
MTEC401.2 - Sept 19 - Conferencing Questions
MTEC401.3 - Sept 26 - Conferencing Questions
MTEC401.4 - Oct 3 - Conferencing Questions
Computer stuff good
Sep. 12th, 2004 10:20 pmAfter waking at 1pm, and working on my large loads of classwork for a bit, I saw an email on the VANLUG list about some computer goodies being sold cheaply.
I wanted 5 of the items on the list, but two had already gone (a Creative DXR2 hardware MPEG decoder card and a BT848-based video capture card). But I did get a PCMCIA (type III) hard drive 340Mb in size - good for embedded systems, a Miro DC30Pro MotionJPEG card (with the breakout panel, all the manuals and an original of Premiere5.0 with licenses) and finally a APC SmartUPS 900 - good for 630W, needs some new batteries first. All of this cost me $75 :-).
I wanted 5 of the items on the list, but two had already gone (a Creative DXR2 hardware MPEG decoder card and a BT848-based video capture card). But I did get a PCMCIA (type III) hard drive 340Mb in size - good for embedded systems, a Miro DC30Pro MotionJPEG card (with the breakout panel, all the manuals and an original of Premiere5.0 with licenses) and finally a APC SmartUPS 900 - good for 630W, needs some new batteries first. All of this cost me $75 :-).
XXS1500 Porting Status Report #5 - delays
Sep. 11th, 2004 01:26 amVanilla 2.6.8.1 appears to have PCMCIA glitches in it, and 2.6.9-rc1-* appears to have NFS problems amongst other things.
On the plus side, my XXS1500 is now mostly stable enough to be left building packages, but it still borks out occasionally :-(.
A new patch is up on my Gentoo dev space, against Linux-MIPS CVS + 2.6.9-rc1-bk12. It's a bit bloated, and should be consider dangerous to your PCMCIA [expect it to hard-hang the box in some cases]. See the latest console logs for examples of the problems.
An ALSA driver for the Au1x00 has floated around on the Linux-MIPS mailing list, so that should probably be added in here at some point.
I'm busy with school since the semester just started, and waiting for 2.6.9-rc2 before I continue work on the XXS1500.
Todo tracking:
PCMCIA - should work, just debugging it now
PCI - stubs in place
USB - needs to be revisited entirely, looks like non-PCI USB was dropped between 2.4 and 2.6
Audio - test outside code
NFSRoot Env - still a mismatch hybrid between Gentoo and Debian packages, need to build everything as -mips2 or -mips32 (I'm in favour of the former at the moment, as it can then be re-used on most mipsel boxes).
I've had some questions about what cross-platform toolchain I use, and I'm using Gentoo's crossdev which provides me with:
gcc-3.3.4-r1
binutils-2.14.90.0.8-r1
I've only got it in kernel mode, no glibc provided.
On the plus side, my XXS1500 is now mostly stable enough to be left building packages, but it still borks out occasionally :-(.
A new patch is up on my Gentoo dev space, against Linux-MIPS CVS + 2.6.9-rc1-bk12. It's a bit bloated, and should be consider dangerous to your PCMCIA [expect it to hard-hang the box in some cases]. See the latest console logs for examples of the problems.
An ALSA driver for the Au1x00 has floated around on the Linux-MIPS mailing list, so that should probably be added in here at some point.
I'm busy with school since the semester just started, and waiting for 2.6.9-rc2 before I continue work on the XXS1500.
Todo tracking:
PCMCIA - should work, just debugging it now
PCI - stubs in place
USB - needs to be revisited entirely, looks like non-PCI USB was dropped between 2.4 and 2.6
Audio - test outside code
NFSRoot Env - still a mismatch hybrid between Gentoo and Debian packages, need to build everything as -mips2 or -mips32 (I'm in favour of the former at the moment, as it can then be re-used on most mipsel boxes).
I've had some questions about what cross-platform toolchain I use, and I'm using Gentoo's crossdev which provides me with:
gcc-3.3.4-r1
binutils-2.14.90.0.8-r1
I've only got it in kernel mode, no glibc provided.
Scottish Webpages
Sep. 10th, 2004 12:02 amI had no idea that the scots wrote their webpages with accents...
http://www.scottish.parliament.uk/vli/language/scots/index.htm
http://www.scottish.parliament.uk/vli/language/scots/index.htm
Rambling HTTP error messages
Sep. 2nd, 2004 02:27 amSeen when BigLumber wasn't behaving:
500 - Something Went Boom
Well, now you've gone and done it. You had to mess around and cause the mysterious 500 error to appear. You probably caused a core dump and whose going to have to cleanup that mess? Me, that's who! Dang kids. Why, when I was your age I was writing programs with punch cards and vacuum tubes. In the dark. Uphill! You and your fancy "mod_perl" and "postgres." Ha! If we wanted to store things in a database, we'd punch some cards and arrange them all neat-like in a box. At least I think it was a box. It might have been my pants. Where's my medicine anyway?....
500 - Something Went Boom
Well, now you've gone and done it. You had to mess around and cause the mysterious 500 error to appear. You probably caused a core dump and whose going to have to cleanup that mess? Me, that's who! Dang kids. Why, when I was your age I was writing programs with punch cards and vacuum tubes. In the dark. Uphill! You and your fancy "mod_perl" and "postgres." Ha! If we wanted to store things in a database, we'd punch some cards and arrange them all neat-like in a box. At least I think it was a box. It might have been my pants. Where's my medicine anyway?....
So while trying to debug my kernel, I found a bug in another peice of software... it turns out ksymoops doesn't behavior correctly on cases where nm/objdump decide a file format is ambiguous. Ksymoops patch: http://dev.gentoo.org/~robbat2/xxs1500/ksymoops-2.4.9-target-fix.patch.gz
Latest kernel patch: http://dev.gentoo.org/~robbat2/xxs1500/linux-xxs1500-20040825.patch.gz
Had markedly more success with the XXS1500 today. After digging in the old config file, I found out that I'd totally missed HW_HAS_PCI, so the kernel was acting as if PCI didn't exist. Sorted it out in the Kconfigs, and a bunch more things fell into place. PCMCIA and PCI _nearly_ work. See my kernel patch above, and look for the two XXX markers. I need to see about the actual wiring schematic of the system to solve those two.
Cleaned up the include tree a bit, properly creating the asm/mach-xxs1500 instead of having the header directly in asm.
Cleaned up the au1000 handling of the boot cmdline, so that it actually uses CONFIG_CMDLINE, and makes life easier (editing your params straight into the file is nasty).
I suspect there is a bug somewhere in the NoOp I/O scheduler, as if I use it, I get a lot of panics. Switched back to the deadline scheduler, works much better :-).
Got my initial NFSRoot mostly up and working, a few slight glitches left, as the initial NFSRoot I got my hands on is for 2.4 only, I'll have to ask `Kumba if he has 2.6 mipsel stage available, or could build be a few packages I could use to convert the stage as needed. Packages needed: pcmcia-cs*, module-init-tools, nfs-utils, portmap
Once I've got the initial NFSRoot up (it uses glibc), I can build some uclibc stages, and nfsroot into them, ready to create some embedded disk images. I need to see about adding mtd-utils into the Gentoo tree first, as I'll need it for doing this.
Console logs in the usual place.
Latest kernel patch: http://dev.gentoo.org/~robbat2/xxs1500/linux-xxs1500-20040825.patch.gz
Had markedly more success with the XXS1500 today. After digging in the old config file, I found out that I'd totally missed HW_HAS_PCI, so the kernel was acting as if PCI didn't exist. Sorted it out in the Kconfigs, and a bunch more things fell into place. PCMCIA and PCI _nearly_ work. See my kernel patch above, and look for the two XXX markers. I need to see about the actual wiring schematic of the system to solve those two.
Cleaned up the include tree a bit, properly creating the asm/mach-xxs1500 instead of having the header directly in asm.
Cleaned up the au1000 handling of the boot cmdline, so that it actually uses CONFIG_CMDLINE, and makes life easier (editing your params straight into the file is nasty).
I suspect there is a bug somewhere in the NoOp I/O scheduler, as if I use it, I get a lot of panics. Switched back to the deadline scheduler, works much better :-).
Got my initial NFSRoot mostly up and working, a few slight glitches left, as the initial NFSRoot I got my hands on is for 2.4 only, I'll have to ask `Kumba if he has 2.6 mipsel stage available, or could build be a few packages I could use to convert the stage as needed. Packages needed: pcmcia-cs*, module-init-tools, nfs-utils, portmap
Once I've got the initial NFSRoot up (it uses glibc), I can build some uclibc stages, and nfsroot into them, ready to create some embedded disk images. I need to see about adding mtd-utils into the Gentoo tree first, as I'll need it for doing this.
Console logs in the usual place.
For the geeks and security-conscious amongst us, there will be a PGP keysigning event here in Vancouver. September 1st, 7pm. I will be attending, as will a notary on the Gossamer Spider Web of Trust (GSWoT) and an OpenCA Assurer.
Think Coffee Lounge & Bistro
4512 10th Avenue West,
Vancouver, BC V6R2J1
(604) 228-9510
Please see http://lunar.ca/keysigning.htm for further details and instructions of things that must be done before attending.
Please circulate this announcement amongst anybody you know that uses PGP.
Think Coffee Lounge & Bistro
4512 10th Avenue West,
Vancouver, BC V6R2J1
(604) 228-9510
Please see http://lunar.ca/keysigning.htm for further details and instructions of things that must be done before attending.
Please circulate this announcement amongst anybody you know that uses PGP.
Working on the PCMCIA driver today. Doing a fair bit of code refactoring to make life easier.
Have two snippets:
a) au_writel((au_readl(GPIO2_PINSTATE) & ~(1<<14))|(1<<30), GPIO2_OUTPUT);
b) au_writel((au_readl(GPIO2_PINSTATE) | (1<<14))|(1<<30), GPIO2_OUTPUT);
The 1<<14 indicates a specific location to set, and the 1<<30 says to enable output on that location.
In arch/mips/au1000/xxs1500/board_setup.c, snippet a is commented as 'turn off power'.
In drivers/pcmcia/au1000_xxs1500.c, snippet a is commented as 'turn on power', and snippet b is commented as 'turn off power'.
Which one do I trust? I don't have any wiring diagrams for the unit, so your guess is as good as mine. The CPU here uses a mix of negated and normal pins, so both options could work.
I guess it's just time for trial and error.
I'm still not sure which is right (If I can't solve it, I'll ask mycable), and I've got a few other PCMCIA glitches to deal with still, but it's mostly done. Patch up in the usual place, along with the latest console log.
Have two snippets:
a) au_writel((au_readl(GPIO2_PINSTATE) & ~(1<<14))|(1<<30), GPIO2_OUTPUT);
b) au_writel((au_readl(GPIO2_PINSTATE) | (1<<14))|(1<<30), GPIO2_OUTPUT);
The 1<<14 indicates a specific location to set, and the 1<<30 says to enable output on that location.
In arch/mips/au1000/xxs1500/board_setup.c, snippet a is commented as 'turn off power'.
In drivers/pcmcia/au1000_xxs1500.c, snippet a is commented as 'turn on power', and snippet b is commented as 'turn off power'.
Which one do I trust? I don't have any wiring diagrams for the unit, so your guess is as good as mine. The CPU here uses a mix of negated and normal pins, so both options could work.
I guess it's just time for trial and error.
I'm still not sure which is right (If I can't solve it, I'll ask mycable), and I've got a few other PCMCIA glitches to deal with still, but it's mostly done. Patch up in the usual place, along with the latest console log.
XXS1500 Gentoo Porting, Status Report #2
Aug. 24th, 2004 02:59 amSo I emailed mycable about the bitrot of 2.6 xxs1500, and I got a patch back, that contained the same compile fixes I had for 2.6 (just accounting for moved headers), and it fixed networking and MTD. Thanks guys!
I had to do a single tweak to their MTD driver to work with the latest CVS (buswidth changed to bankwidth). They also had a hack to force CONFIG_BCM5222_DUAL_PHY on for the network driver, which I converted to some correctly done Kconfig magic, which should be suitable to send to Linux-MIPS upstream. I also put in a bunch of missing .cvsignore files, that were annoying me :-).
I've started some work on the PCMCIA updates required to get the CF working, as I'm really out of my depth on the USB side. Later on I should look at converting the broken OSS driver to a working ALSA driver :-).
My Initrd seems to be lacking libwrap, but a bunch of things are compiled against it (like portmap - so no nfs mounting yet), so I can't test out my prototype nfsroot environment either yet, but it should work. I'll try directly booting the nfsroot tommorrow, instead of chrooting into it.
Download location as always:
http://dev.gentoo.org/~robbat2/xxs1500/
Contains the new serial console log, and the patch as described above.
I had to do a single tweak to their MTD driver to work with the latest CVS (buswidth changed to bankwidth). They also had a hack to force CONFIG_BCM5222_DUAL_PHY on for the network driver, which I converted to some correctly done Kconfig magic, which should be suitable to send to Linux-MIPS upstream. I also put in a bunch of missing .cvsignore files, that were annoying me :-).
I've started some work on the PCMCIA updates required to get the CF working, as I'm really out of my depth on the USB side. Later on I should look at converting the broken OSS driver to a working ALSA driver :-).
My Initrd seems to be lacking libwrap, but a bunch of things are compiled against it (like portmap - so no nfs mounting yet), so I can't test out my prototype nfsroot environment either yet, but it should work. I'll try directly booting the nfsroot tommorrow, instead of chrooting into it.
Download location as always:
http://dev.gentoo.org/~robbat2/xxs1500/
Contains the new serial console log, and the patch as described above.
XXS1500 Gentoo Porting, Status Report #1
Aug. 23rd, 2004 01:26 amNote: Blogs are useful for many things, including keeping track of technical work. This geek-talk will baffle and confuse many people.
Ok, so I had a discussion with somebody at mycable.de about their XXS1500 device, and managed to buy a used one very cheaply for the express purpose of porting Gentoo Linux to it.
Although I recieved it more than a week ago, since I was on vacation last week, I didn't exactly do anything &em; the vacation was AWAY from technology.
Once I got around a problem of a powerbar (wallwarts take too much space!), I was good to start trying it out.
It shipped with 2.4.21 and a uclibc environment with busybox providing most functionality. It only has an extremely minimal thttpd, and no native way of getting data onto the machine (I'm ignoring the serial console here).
After a bit of hacking around, I set up a crosscompiler chain using Gentoo's crossdev:
crossdev -u -k --vgcc=3.3.4-r1 --arch=mipsel
Built up a kernel from the Linux-MIPS CVS sources, using the xxs1500_defconfig in the tree.
A number of things failed to build, as it seems the XXS1500 stuff has suffered some bitrot.
Mostly trivial things like #include locations changing, and a few defines changing. However there are several major breakages I couldn't fix myself:
PCMCIA fails to compile &en; looks like it needs an update to the latest 2.6 pcmcia system, a sample on how to do it is available.
Audio (OSS) fails to compile &en; I have no idea about it.
Turning those off, and fixing the other minor things, I got a kernel, that didn't want to boot. Dug some, found this is the first time I needed an SREC kernel :-). gmake vmlinux.srec to the rescue. I hadn't used YAMON before.
Stuck the kernel into my tftp system, tried to boot with an nfsroot. kernel panic somewhere in networking :-(. Turned off networking &em; need to fix it later as well. Grabbed a mipsel initrd image from `Kumba's work (thanks for the pointer to it Hardave) [http://dev.gentoo.org/~kumba/mips/cobalt/netboot/]. Kernel boots up and gives me an initial environment now, but no networking as noted.
Issues to work on: (mainly my own todo list)
- Kernel fixin': Network, PCMCIA, Audio [in that order]
- NFSRoot environment
- Device speed: this is supposed to be a 500mhz CPU, but it only says 396Mhz, and there is an undocumented YAMON command 'ss' I don't quite get yet.
- produce kernel patch to send back upstream
- uclibc
- catalyst target
I'm going to keep a collection of downloads for this work on my Gentoo dev-space here:
http://dev.gentoo.org/~robbat2/xxs1500/
The first installment today has my serial console log and my initial working netboot kernel+initrd.
(no subject)
Aug. 1st, 2004 02:31 amPay what you think music is worth, or have it as MP3 for free - legitimately.
http://magnatune.com/
a wide collection of music.
I esp. recommend the group 'Cargo Cult'.
Spread the word.
http://magnatune.com/
a wide collection of music.
I esp. recommend the group 'Cargo Cult'.
Spread the word.
New photos in my gallery
Jul. 31st, 2004 03:09 amStart here:
http://www.orbis-terrarum.net/?l=people.robbat2.gallery
Navigate to one of these:
Places/Naniamo River - August 26 2003
Misc/Amazing Clouds - 19 September 2003
TechBC/Navajo - Communications Box - 12 July 2004
Events/Grouse Mountain, The Observatory - Mom's Birthday Dinner - 15 July 2004
Mom&Dad/Capilano Suspension Bridge - 22 July 2004
People/Marissa/Titty Kitty July 2004
Events/Celebration of Light - China - 28 July 2004
The first two are a bunch of pictures that had never been posted previously.
The thumbnails are still generating, so give them a while.
I really need to get my hands on a decent tripod for the fireworks photos. However, some of them are quite nice already in a blurred kind of way.
http://www.orbis-terrarum.net/?l=people.robbat2.gallery
Navigate to one of these:
Places/Naniamo River - August 26 2003
Misc/Amazing Clouds - 19 September 2003
TechBC/Navajo - Communications Box - 12 July 2004
Events/Grouse Mountain, The Observatory - Mom's Birthday Dinner - 15 July 2004
Mom&Dad/Capilano Suspension Bridge - 22 July 2004
People/Marissa/Titty Kitty July 2004
Events/Celebration of Light - China - 28 July 2004
The first two are a bunch of pictures that had never been posted previously.
The thumbnails are still generating, so give them a while.
I really need to get my hands on a decent tripod for the fireworks photos. However, some of them are quite nice already in a blurred kind of way.
Microsoft Word Does Grammer Good
Jul. 26th, 2004 09:41 pm
Yeah, go ahead, change the ENTIRE meaning of the sentence.
I'm not a person that generally posts petitions or anything, but I think this is something that's going to have a serious influence.
(The IT) Developer Declaration of Independence
http://www.opengroup.org/declaration/declaration.htm
(The IT) Developer Declaration of Independence
http://www.opengroup.org/declaration/declaration.htm
(no subject)
Jul. 12th, 2004 04:01 amnew image categories posted on my photo gallery.
http://www.orbis-terrarum.net/?l=people.robbat2.gallery
Misc/Found_Objects/Found NJ Transit - 15 April 2004
Misc/Found_Objects/Found Vogue 87 - 27 March 2004
Misc/Found_Objects/Found Bag - 15 May 2004
Misc/Found_Objects/Genki - 10 May 2004
Trips/Ladysmith - 02 May 2004
People/Greg/San Francisco 19 June 2004 - 06 July 2004
People/Robin/Thai Dinner - 28 April 2004
People/Marissa/Making a Whip 18 June 2004
Places/Waterfront_Station_25_May_2004_Chemical_Spill
Places/Chinatown - 17 April 2004
Mom&Dad/Squirrel - 27 April 2004
Mom&Dad/Flowers + Gardens - April 2004
Computers/Crashed 7200rpm SCSI drive - 16 May 2004
http://www.orbis-terrarum.net/?l=people.robbat2.gallery
Misc/Found_Objects/Found NJ Transit - 15 April 2004
Misc/Found_Objects/Found Vogue 87 - 27 March 2004
Misc/Found_Objects/Found Bag - 15 May 2004
Misc/Found_Objects/Genki - 10 May 2004
Trips/Ladysmith - 02 May 2004
People/Greg/San Francisco 19 June 2004 - 06 July 2004
People/Robin/Thai Dinner - 28 April 2004
People/Marissa/Making a Whip 18 June 2004
Places/Waterfront_Station_25_May_2004_Chemical_Spill
Places/Chinatown - 17 April 2004
Mom&Dad/Squirrel - 27 April 2004
Mom&Dad/Flowers + Gardens - April 2004
Computers/Crashed 7200rpm SCSI drive - 16 May 2004
