Re: [9fans] more little hardware
On Tue, Mar 16, 2010 at 8:03 PM, Tim Newsham wrote: > Hmm.. There's the OK-labs android stuff which virtualizes > android on top of L4. If only p9 was running on top of L4 :) Get cracking Tim! how hard can it be? :-) Actually I still think igep + sparkfun phone module might be a path. ron
[9fans] recreational programming of an evening
I'm an admirer of acid but never found that I like the truss functions
all that much. I've used acid on just enough semi-working platforms,
where breakpoints don't do what is expected, that truss is not
frequently operational and hence not that useful. Also, it's not that
great for fork. And, I'm not that happy about the need to change the
process' memory image just to do tracing. truss() just feels
unfinished.
I pretty much prefer strace on Unix to what acid can do on Plan 9. It
follows forks nicely, and it shows what's going on. And, it works even
if debugging doesn't. And, and on a lightweight node, I don't have to
drag all the acid bits along.
Now, on Linux, strace is a beast: 256KB binary, needs shared
libraries, and so on. (come to think of it that's small for a GNU
program nowadays ...)
What I want is something that works like this:
echo stop > proc/75/ctl
while (echo startsyscall>/proc/75/ctl) {
cat /proc/75/truss
}
(note the undocumented startsyscall verb). You need to see the source,
not the man page :-)
Well, turns out to be trivial to do in Plan 9. I did the work against
9vx, since it's so easy to prototype this stuff. The diff is so small
I just attached it. Russ already had nice pretty-print examples for
tracing system calls so I cribbed them.
Here's the output:
75 [cat] Pread 0x13d7 0fffdf30 2000 2000h= 0x9
75 [cat] Pwrite 0x12ed 0001 0fffdf30 0009 0009h= 0x9
75 [cat] Pread 0x13d7 0fffdf30 2000 2000h= 0xc
75 [cat] Pwrite 0x12ed 0001 0fffdf30 000c 000ch= 0xc
75 [cat] Pread 0x13d7 0fffdf30 2000 2000h= 0
75 [cat] Open 0x1410 6990 0014 0014h= 0x3
75 [cat] Pread 0x13d7 0003 0f00 0014 0014h= 0xc
75 [cat] Close 0x143e 0003 0f00 0014 0014h= 0
75 [cat] Exits 0x128b hecho: write
error: process exited
So, voila, we have truss, it took 15 minutes to add this in, in fact
almost as long to write this message as to add it in. And a day to
look at the code and ruminate in the odd free moment how to do it. But
that's why I still like this kernel: I can image how much fun this
would have been in just about any Unix, since I've been through a
similar exercise of this sort frequently on Linux.
I may extend it just a bit to dump stuff like file names, read and
write data, and so on. Not that hard in my view.
Anyway, I'll be doing the same thing in the 9k kernel, but the patch I
just attached should give you a hint if you need this. I think it
would help people trying to find problems in 9vx. And, I bet somebody
out there can do a better job than I did.
Obviously, you need to do a bit more work to catch fork but that's
left to the student.
ron
x
Description: Binary data
Re: [9fans] Question on /net with respect to anonymity
However, the absence of ipfilters / iptables means, the gateways really have no history information and no control over who is having access to either of its interfaces. If I mounted my gateway's /net onto my machine, I would expect that the gateway would do no extra filtering on my connections than it did on any of its own connections. That's a feature, not a bug. Am I misunderstanding what you're saying? Rahul Murmuria Tim Newsham | www.thenewsh.com/~newsham | thenewsh.blogspot.com
Re: [9fans] more little hardware
9phone? Just an idea to run Plan 9 on phone hardware, "son of bitsy". Looks like there's a lot of very cheap Android devices coming from China in the near future, so we may be in luck there. Hmm.. There's the OK-labs android stuff which virtualizes android on top of L4. If only p9 was running on top of L4 :) John Tim Newsham | www.thenewsh.com/~newsham | thenewsh.blogspot.com
Re: [9fans] more little hardware
On 16/03/2010, at 5:02 PM, Patrick Kelly wrote: > Any thought as to using the OpenMoko as a phone platform? Unfortunately, I don't think either the Neo1973 or the FreeRunner are useful as a daily-use phone due to various hardware bugs or limitations. d
Re: [9fans] interesting qemu problem
On Tue, Mar 16, 2010 at 4:04 PM, Eric Van Hensbergen wrote: > Yeah, but Plan 9's a cluster environment, nothing wrong with the venti server > being elsewhere (in fact, thats kind of expected) -- unless of course you are > debugging the venti server. I'm using qemu to debug a problem I'm having on a real machine. So, yeah, I need to do this. I don't see why venti has gotten so memory hungry, this seems new behavior. I realize I can twist the knobs myself but geez, this is a 4 GB disk -- why does it think it needs nearly 400 MB RSS to deal with it? ron
[9fans] Question on /net with respect to anonymity
I have been working on a couple of networking related projects, one of them being the /net filesystem on Linux kernel for the Glendix[1] project. I was however unclear on certain networking aspects of /net on pure Plan 9 / 9P networks. The question is, is there a proper document explaining and analyzing the networking aspects of using /net? [1] http://www.glendix.org When a user mounts the /net of another computer system, (and I suppose the protocol used there also is 9P), the client essentially gains the ability to send packets out of the server's ethernet port. Hence a gateway or/and a VPN is implemented in the Plan 9 world. I understand that the main advantage is that we get to use the same userspace tools to work with files in the /net directory, without needing to know whether it is on the local system, or is a remote filesystem. It's just a brilliant and clean way of performing the same task we have done for decades using more complex sockets-based implementations. However, the absence of ipfilters / iptables means, the gateways really have no history information and no control over who is having access to either of its interfaces. That can be taken care of, if we run an extra service to provide these features, but it is no more a requirement to run a gateway. There is a mention of anonymous traffic here, which I have not been able to locate in literature. So how about if we created a, say MANET, where we used /net, and hence used 5 different nodes one after the other, each mounting the /net of the next node, and ultimately send the packet out to the Internet from the last node. Is there anything different we will see, from the networking side of things, than what we know of based on widely used implementations? Regards, -- Rahul Murmuria
Re: [9fans] interesting qemu problem
On Mar 16, 2010, at 6:52 PM, ron minnich wrote: > On Tue, Mar 16, 2010 at 3:47 PM, Eric Van Hensbergen wrote: > >> You could configure venti to be less aggressive with its use of memory, but >> that would likely hurt performance. >> >> Running venti inside qemu is silly. If you really want venti for your vm, >> run venti on the host and target your in-vm fossil at it. > > > I tend to disagree. If I'm running qemu it is because I want to > simulate a whole-machine environment. If I don't need that simulation, > I'll go back to 9vx. > > but YMMV. > Yeah, but Plan 9's a cluster environment, nothing wrong with the venti server being elsewhere (in fact, thats kind of expected) -- unless of course you are debugging the venti server. -eric
Re: [9fans] interesting qemu problem
> I tend to disagree. If I'm running qemu it is because I want to > simulate a whole-machine environment. If I don't need that simulation, > I'll go back to 9vx. seems that keeping up with qemu is at least as hard as keeping up with real hardware. - erik
Re: [9fans] interesting qemu problem
On Tue, Mar 16, 2010 at 3:47 PM, Eric Van Hensbergen wrote: > You could configure venti to be less aggressive with its use of memory, but > that would likely hurt performance. > > Running venti inside qemu is silly. If you really want venti for your vm, > run venti on the host and target your in-vm fossil at it. I tend to disagree. If I'm running qemu it is because I want to simulate a whole-machine environment. If I don't need that simulation, I'll go back to 9vx. but YMMV. ron
Re: [9fans] more little hardware
On Tue, 2010-03-16 at 15:38 -0800, ron minnich wrote: > On Tue, Mar 16, 2010 at 1:02 PM, Patrick Kelly wrote: > > > Any thought as to using the OpenMoko as a phone platform? > > vapor. That thing was pure vapor from start to end. Figures sickening how many potentially nice, GPL'd, things turn into that. > > ron >
Re: [9fans] interesting qemu problem
On Mar 16, 2010, at 6:04 PM, ron minnich wrote: > I was wrong. I built a new kernel from sources and performance is > still very bad, with a load of 2500 minimum. > > Also, venti, on this little machine, is a bit hungry for memory. > venti...2010/0316 20:31:06 venti: conf.../boot/venti: mem 1,048,576 > bcmem 140,753,578 icmem 211,130,368...httpd > tcp!127.1!8000...init...icache 211,130,368 bytes = 3,298,912 entries; > 16 scache > sync...2010/0316 20:31:12 arenas01: indexing 430 clumps... > announce tcp!127.1!17034...serving. > fossil(#S/sdC0/fossil)...fsys: dialing venti at tcp!127.1!17034 > > > This image formerly ran in 256M, now requires 512M, because venti > footprint is 140+211+211 ... wait, how does it ever fit in 512 anyway. > > I think qemu changed too. An old kernel has the same problems. Not > sure what is up here. but for now qemu on unongo 9.10 is useless for > plan 9 from what I can see. > You could configure venti to be less aggressive with its use of memory, but that would likely hurt performance. Running venti inside qemu is silly. If you really want venti for your vm, run venti on the host and target your in-vm fossil at it. -eric
Re: [9fans] more little hardware
On Tue, Mar 16, 2010 at 1:02 PM, Patrick Kelly wrote: > Any thought as to using the OpenMoko as a phone platform? vapor. That thing was pure vapor from start to end. ron
Re: [9fans] interesting qemu problem
On Tue, Mar 16, 2010 at 3:07 PM, erik quanstrom wrote: >> This image formerly ran in 256M, now requires 512M, because venti >> footprint is 140+211+211 ... wait, how does it ever fit in 512 anyway. > > swap? this would answer two questions. I should mention that another person here tried qemu recently and commented that it was dog slow as well. Something changed in qemu I think and it's affecting plan 9. That was a very old qemu image and it was peppy in the old days. ron
Re: [9fans] interesting qemu problem
> This image formerly ran in 256M, now requires 512M, because venti > footprint is 140+211+211 ... wait, how does it ever fit in 512 anyway. swap? this would answer two questions. - erik
Re: [9fans] interesting qemu problem
I was wrong. I built a new kernel from sources and performance is still very bad, with a load of 2500 minimum. Also, venti, on this little machine, is a bit hungry for memory. venti...2010/0316 20:31:06 venti: conf.../boot/venti: mem 1,048,576 bcmem 140,753,578 icmem 211,130,368...httpd tcp!127.1!8000...init...icache 211,130,368 bytes = 3,298,912 entries; 16 scache sync...2010/0316 20:31:12 arenas01: indexing 430 clumps... announce tcp!127.1!17034...serving. fossil(#S/sdC0/fossil)...fsys: dialing venti at tcp!127.1!17034 This image formerly ran in 256M, now requires 512M, because venti footprint is 140+211+211 ... wait, how does it ever fit in 512 anyway. I think qemu changed too. An old kernel has the same problems. Not sure what is up here. but for now qemu on unongo 9.10 is useless for plan 9 from what I can see. ron
Re: [9fans] more little hardware
On Tue, 2010-03-16 at 11:44 -0500, John Floren wrote: > On Tue, Mar 16, 2010 at 9:54 AM, wrote: > >> http://www.engadget.com/2010/03/15/qi-hardwares-tiny-hackable-ben-nanonote-now-shipping/ > > > > Okay, Maht. You just cost me $125 :) I just couldn't resist. > > Of course, it remains to be seen whether this will be another > > project that gets pushed onto the stack or we will see something > > come of it. What I've got in mind at the moment is a native > > Inferno port. We'll see what happens. > > > > BLS > > > > Shame it doesn't have a cell phone radio built in, or Ron and I might > have just what we needed for the 9phone. Any thought as to using the OpenMoko as a phone platform? > > At 32 MB of RAM, it's basically the very lower limit of what we could > use for a terminal. Does it have a touchscreen? I didn't see that > listed. > > > John
Re: [9fans] more little hardware
> What flavor MIPS is it? 64 bit (I doubt it)? Is it a version that the > compilers will like? Good point. I do know it's 32-bit, but so far that's all I know. Honestly, I had noticed it was MIPS and didn't really think any further about it. The good news is that both big-endian and little-endian compilers are there, and they're described as being for the M3000 family. I suspect (or hope anyway) that the processor on the device is a superset of the 3000, and given how old the 3000 is, that's probably not too unreasonable an expectation. BLS
Re: [9fans] more little hardware
On Tue, Mar 16, 2010 at 3:35 PM, Tim Newsham wrote: >> Shame it doesn't have a cell phone radio built in, or Ron and I might >> have just what we needed for the 9phone. > > 9phone? > Just an idea to run Plan 9 on phone hardware, "son of bitsy". Looks like there's a lot of very cheap Android devices coming from China in the near future, so we may be in luck there. John -- "Object-oriented design is the roman numerals of computing" -- Rob Pike
Re: [9fans] more little hardware
Shame it doesn't have a cell phone radio built in, or Ron and I might have just what we needed for the 9phone. 9phone? John Tim Newsham | www.thenewsh.com/~newsham | thenewsh.blogspot.com
Re: [9fans] more little hardware
What flavor MIPS is it? 64 bit (I doubt it)? Is it a version that the compilers will like? ron
Re: [9fans] omap3 notebook
On 14 Mar 2010, at 10:22, Richard Miller wrote: There's also the Nokia's NIT devices if you're looking for pocketable. Might be a relatively hostile environment for anything but Maemo, though. Hosted inferno works well on these. That's good to know. It may be my best choice for my Zaurus too. -- Simplicity does not precede complexity, but follows it. -- Alan Perlis
Re: [9fans] interesting qemu problem
On Tue, Mar 16, 2010 at 9:48 AM, erik quanstrom wrote: >> And have noticed that an old image I use for qemu is going astray. >> Same kernel as it has been for quite some time, but the load is pegged >> at about 2500 at all times. > > 2.5, i assume. what are the 2-3 processes running? Never mind, this fools me every time. It was so long since I booted this thing that it had to do a snap, and after a replica -s, well, that takes a long time. ron
Re: [9fans] more little hardware
> As I get it, it does not feauture a USB host controller, but acts > like an USB device that you can connect to your PC. Maybe it > will work anyhow... > I was wondering how you'd network one of those things: http://en.qi-hardware.com/wiki/Wi-Fi_in_Nanonote >> >> I thought that was terribly cute. The other option is talking >> PPP over the USB. You'd be tethered, but you could at least >> talk. I think it will. I've done PPP over USB before, and I'm pretty sure it was a case of the PC acting as host and the other end as a device. But it's been about a year, and I'm having trouble recalling the details. But as I'm thinking right now, that would just be an interim step before hopefully moving to the 802.11 device in the microSD card. BLS
[9fans] 9vx on MacOS X: problem with contrib
Hi there, i'm newbie in Plan 9 so may be this is common mistake (but i still can't google it). I have Intel mac with 10.6 OSX and 9vx from official site. Also I have plan9.tar.bz from this thread: http://www.mail-archive.com/9fans@9fans.net/msg02125.html. Plan9 user is "glenda", OSX user is "semka", but if i "touch test; ls -l test" it have "semka" as owner. Every time I trying to install contrib it falls with "Permission denied" errors. There is log: term% /n/sources/contrib/fgb/root/rc/bin/contrib/install fgb/contrib ... error: copying /386/bin/contrib/cat: '/386/bin/contrib/cat' Permission denied error: copying /386/bin/contrib/create: '/386/bin/contrib/create' Permission denied error: copying /386/bin/contrib/gui: '/386/bin/contrib/gui' Permission denied error: copying /386/bin/contrib/install: '/386/bin/contrib/install' Permission denied error: copying /386/bin/contrib/list: '/386/bin/contrib/list' Permission denied error: copying /386/bin/contrib/local: '/386/bin/contrib/local' Permission denied ... And so on, I can send full contrib-install log if needed. Sorry for, may be, too noob question (: -- take care of the brain
Re: [9fans] more little hardware
As I get it, it does not feauture a USB host controller, but acts like an USB device that you can connect to your PC. Maybe it will work anyhow... >>> I was wondering how you'd network one of those things: >>> >>> http://en.qi-hardware.com/wiki/Wi-Fi_in_Nanonote > > I thought that was terribly cute. The other option is talking > PPP over the USB. You'd be tethered, but you could at least > talk. ___Köp bläckpatron, toner eller fotopapper för 1 kr hos Inkclub. Beställ före kl. 19 för leverans till din hemadress redan nästa vardag. Det är bekvämt, snabbt och enkelt. Beställ på http://secure-dk.imrworldwide.com/cgi-bin/b?ci=aller-kampanj-se&cg=spray&tu=http%3A%2F%2Fwww.inkclub.com%2Fspraylinkfeb10"; target="_blank">www.inkclub.com.
Re: [9fans] interesting qemu problem
> And have noticed that an old image I use for qemu is going astray. > Same kernel as it has been for quite some time, but the load is pegged > at about 2500 at all times. 2.5, i assume. what are the 2-3 processes running? - erik
[9fans] interesting qemu problem
I'm now running an upgraded qemu: QEMU PC emulator version 0.11.0 (qemu-kvm-0.11.0), Copyright (c) 2003-2008 Fabrice Bellard And have noticed that an old image I use for qemu is going astray. Same kernel as it has been for quite some time, but the load is pegged at about 2500 at all times. I'm not sure where to start; any suggestions here? I'm using qemu in this case because I don't quite have a working tiny arm board and I want to beat on the kernel source proper for something. thanks ron
Re: [9fans] more little hardware
>> I was wondering how you'd network one of those things: >> >> http://en.qi-hardware.com/wiki/Wi-Fi_in_Nanonote I thought that was terribly cute. The other option is talking PPP over the USB. You'd be tethered, but you could at least talk. > Off-topic-ish, that 320x240 screen is probably the biggest challenge, > trying to find some usable UI in that space. I think the idea of a > native Inferno port is great. Anyone doing anything fun on the UI side > with the Nintendo DS port? Yeah, that's what I was figuring too. Even at a 5x7 font, you can't do better than about a 50x30 character screen. If all the graphical elements in wm are shrunk down to somewhere around 10 pixels, it might just be possible to put a shell window that's almost usable. Acme in a single column might even work. Of course, I'd really like to have more resolution, but this is a start and who knows what we might see in a second generation. BLS
Re: [9fans] more little hardware
On Tue, Mar 16, 2010 at 8:10 AM, Jack Johnson wrote: > Off-topic-ish, that 320x240 screen is probably the biggest challenge, > trying to find some usable UI in that space. I think the idea of a > native Inferno port is great. Sorry, last of the blather. It also seems ideal for Octopus: http://plan9.escet.urjc.es/ls/octopus.html -Jack
Re: [9fans] more little hardware
> Shame it doesn't have a cell phone radio built in, or Ron and I might > have just what we needed for the 9phone. That would be cool. Unfortunately, the cell phone people seem a lot less friendly about releasing the information necessary to program their chips. > At 32 MB of RAM, it's basically the very lower limit of what we could > use for a terminal. Does it have a touchscreen? I didn't see that > listed. As near as I can tell, it doesn't. I think the large square key is basically a mouse. As to buttons, I don't know yet. In the worst case, I can probably usurp the function keys as mouse buttons. BLS
Re: [9fans] more little hardware
On Tue, Mar 16, 2010 at 7:53 AM, Jack Johnson wrote: > On Tue, Mar 16, 2010 at 5:54 AM, wrote: >>> http://www.engadget.com/2010/03/15/qi-hardwares-tiny-hackable-ben-nanonote-now-shipping/ >> >> Okay, Maht. You just cost me $125 :) I just couldn't resist. > > I was wondering how you'd network one of those things: > > http://en.qi-hardware.com/wiki/Wi-Fi_in_Nanonote Off-topic-ish, that 320x240 screen is probably the biggest challenge, trying to find some usable UI in that space. I think the idea of a native Inferno port is great. Anyone doing anything fun on the UI side with the Nintendo DS port? It also looks like Android on this thing might be a possibility: http://www.laptopmag.net/3837-google-android-port-for-xburst-cpus-on-its-way.html ...so drawterm for Android might also be a worthwhile direction. -Jack
Re: [9fans] more little hardware
On Tue, Mar 16, 2010 at 5:54 AM, wrote: >> http://www.engadget.com/2010/03/15/qi-hardwares-tiny-hackable-ben-nanonote-now-shipping/ > > Okay, Maht. You just cost me $125 :) I just couldn't resist. I was wondering how you'd network one of those things: http://en.qi-hardware.com/wiki/Wi-Fi_in_Nanonote -Jack
Re: [9fans] more little hardware
On Tue, Mar 16, 2010 at 9:54 AM, wrote: >> http://www.engadget.com/2010/03/15/qi-hardwares-tiny-hackable-ben-nanonote-now-shipping/ > > Okay, Maht. You just cost me $125 :) I just couldn't resist. > Of course, it remains to be seen whether this will be another > project that gets pushed onto the stack or we will see something > come of it. What I've got in mind at the moment is a native > Inferno port. We'll see what happens. > > BLS > Shame it doesn't have a cell phone radio built in, or Ron and I might have just what we needed for the 9phone. At 32 MB of RAM, it's basically the very lower limit of what we could use for a terminal. Does it have a touchscreen? I didn't see that listed. John -- "Object-oriented design is the roman numerals of computing" -- Rob Pike
Re: [9fans] more little hardware
> http://www.engadget.com/2010/03/15/qi-hardwares-tiny-hackable-ben-nanonote-now-shipping/ Okay, Maht. You just cost me $125 :) I just couldn't resist. Of course, it remains to be seen whether this will be another project that gets pushed onto the stack or we will see something come of it. What I've got in mind at the moment is a native Inferno port. We'll see what happens. BLS
Re: [9fans] install filsystem hangs on bad block
Yes. As a note, I kept having write errors pop up, so I ran a low level disk format. So far it is running better, but have not tried to replicate various weird things I tripped over earlier. I'll try that again later. EBo -- erik quanstrom said: > > The second problem arose when I repartitioned the disk with partdisk and it > > did not appear to have been updated when I continued with the install. I > > wonder if the partition map had not been updated with plan9's partprobe > > equivelent. > > did you type 'w' then 'q' at the fdisk, the prep prompts, thus accepting > and writing the defaults? > > - erik > --
Re: [9fans] install filsystem hangs on bad block
> The second problem arose when I repartitioned the disk with partdisk and it > did not appear to have been updated when I continued with the install. I > wonder if the partition map had not been updated with plan9's partprobe > equivelent. did you type 'w' then 'q' at the fdisk, the prep prompts, thus accepting and writing the defaults? - erik
[9fans] more little hardware
http://www.engadget.com/2010/03/15/qi-hardwares-tiny-hackable-ben-nanonote-now-shipping/
