**** BEGIN LOGGING AT Sun May 15 23:59:57 2005 May 16 12:00:30 [g2]: Perhaps we should discuss the layout changes here? May 16 12:01:04 <[g2]> sure here is fine with me May 16 12:01:15 <[g2]> I'm sure at some point the openslug guys will be interested May 16 12:01:23 Indeed. May 16 12:01:25 <[g2]> but the crew is all here May 16 12:01:33 <[g2]> you, me, jacques and dyoung-zzzz May 16 12:01:40 I just wanted to get an idea of what you need without the cross-talk. May 16 12:01:48 <[g2]> and possibly ka6sox-away May 16 12:02:05 <[g2]> nod. It's much quieter in here May 16 12:02:19 <[g2]> actually I was going to change the layout of firmware image May 16 12:02:43 Flash layout is really a config issue and shouldn't be more than a custom config. May 16 12:03:13 <[g2]> exactly there are no changes to APEX, maybe just another custom config May 16 12:03:21 <[g2]> or maybe that is the default config May 16 12:03:32 BTW, did you figure out why your patch wasn't propagating? May 16 12:03:51 <[g2]> no I think that's an open issue May 16 12:04:03 <[g2]> I can test some more if you like May 16 12:04:08 <[g2]> it real easy May 16 12:04:15 * [g2] types too fast May 16 12:04:38 <[g2]> the work-around is simple and quick enough, make clean && make May 16 12:05:25 <[g2]> beewoolie-afk, do you recommend having a config block for the APEX parameters ? May 16 12:05:45 what is a config block? May 16 12:05:54 <[g2]> a configuration block May 16 12:05:58 <[g2]> in flash May 16 12:06:05 APEX has an environment. May 16 12:06:12 VAR=VALUE May 16 12:06:50 <[g2]> and the real question is making changes to that environment persistent across cold-boots May 16 12:07:11 Right. It has an environment that it writes to flash. May 16 12:07:39 There are default values compiled into the code. These values can be overridden. May 16 12:07:53 <[g2]> Really ? does the writing ? But that's only if things are changed correct ? May 16 12:08:03 Right. May 16 12:08:36 I'm not using it at the moment on the slug because I don't want to clobber the second flash block. May 16 12:08:39 <[g2]> That'll need some testing and we''ll need to back up the last 144 bytes from the Redboot second block May 16 12:08:56 <[g2]> Ok.. that's good for now May 16 12:09:33 At the moment, the environment is stored in the second flash block. It could be put in the first block, after the loader. Trouble is that there is no logic that allows for erasing the environment in a block without erasing the whole thing. May 16 12:09:45 That was part of the issue. May 16 12:10:09 We can certainly add some logic that pulls the MAC address from the second block. May 16 12:10:30 <[g2]> that's why we just have the user save off the last 144 bytes because the first 250,000 are all the same for everyone May 16 12:11:23 We could write a piece of code that installs APEX on a runnin system s.t. it pulls the mac address automatically. and writes the first block with everything we need. May 16 12:11:27 <[g2]> it's a simple dd if=/dev/mdtblock0 of=redboot_save bs=1K skip=250 I think May 16 12:11:50 I'm suggesting something a little more elegant, but yeah. I get that it's easy. May 16 12:12:13 I like having automated processes so that fumble fingers joe doesn't end up with a brick. May 16 12:12:18 <[g2]> Ok.. are you saying we reserve the last 144 bytes and and just uses the first 128K-144 bytes ? May 16 12:12:43 <[g2]> agreed on the deployablility issue May 16 12:12:49 <[g2]> automated is much better May 16 12:13:22 <[g2]> actually that could be part of a script May 16 12:13:37 <[g2]> an init script May 16 12:13:39 It depends on whether or not you want to reorg flash. If we plan to replace the bootloader, we can put everything APEX needs in one block and give the other block to the kernel...which appears to need it. May 16 12:14:03 Sure. The init script could be smart about the upgrade and only do it when necessary. May 16 12:14:21 The presently isn't a MAGIC number in APEX, but we can add one easily enough. May 16 12:14:54 <[g2]> Since APEX is so good with space utilization, I'd prefer to use the second block May 16 12:15:02 <[g2]> as the environemnt May 16 12:15:12 Fair enough. May 16 12:15:22 <[g2]> Then we still gain back 3 blocks May 16 12:15:30 <[g2]> 1 for the sysconfig May 16 12:15:31 There may come a time when APEX needs the extra space for NPE or whatever. May 16 12:15:35 Right. May 16 12:15:42 Two others from??? May 16 12:15:44 <[g2]> 1 for the ramdisk start May 16 12:15:50 <[g2]> and 1 from the trailer May 16 12:16:03 They waste a whole block for the ramdisk start marker? May 16 12:16:23 <[g2]> they only waste 16 bytes, but jffs2 needs to start on a block boundary :( May 16 12:16:32 <[g2]> which makes total sense to me May 16 12:16:34 That 768K. Kindofa lot. May 16 12:16:44 <[g2]> 386K May 16 12:16:50 <[g2]> 128 * 3 May 16 12:17:01 Oh. Yeah. I forgot that blocks are only 128k on this baby. May 16 12:17:02 <[g2]> 382 May 16 12:17:12 Still, that will help. May 16 12:17:20 <[g2]> absolutely May 16 12:17:28 Now, we're talking about Openslug, right? May 16 12:17:33 <[g2]> Right May 16 12:17:45 <[g2]> Do you want to support Unslung ? May 16 12:18:07 So, this can be the upgrade plan. Boot to existing firmware, run this script, reboot and you've got openslug. May 16 12:18:22 No, I don't care about unslung. I just wanted to make sure we're talking about the same thing. May 16 12:18:54 <[g2]> Ok.. I need to think about OpenSlug compatibility for rwhitby May 16 12:19:02 <[g2]> but of the top of my head... May 16 12:19:31 <[g2]> The current default config runs for both Unslung and Openslug, but Unslung untested right now May 16 12:19:48 <[g2]> and we need to add some minor config changes like the 1 second autoboot May 16 12:20:09 <[g2]> well they'll both run after the minor changes... May 16 12:20:44 <[g2]> the second custom config is for the camp #5 guys that cut the cord with the linksyss compatibility May 16 12:20:45 I still don't get why you're so hot to lengthen the boot. May 16 12:21:09 <[g2]> it's the autoboot that's important May 16 12:21:16 <[g2]> at 1 second May 16 12:21:36 Making it longer inconveniences everyone. Leaving it shorter just means that people who need to break into the loader need to press the key when the power the unit on. May 16 12:21:54 <[g2]> 1 second is too long ? May 16 12:22:07 I thought that you wanted it longer. May 16 12:22:10 <[g2]> no... May 16 12:22:19 <[g2]> I just want it *AUTO* May 16 12:22:20 Oh. The default config is already one second to autoboot. May 16 12:22:27 <[g2]> mine didn't May 16 12:22:33 You must have overridden it in your config. May 16 12:22:47 The default for nslu2 has autoboot enabled in 1.2.11. May 16 12:22:50 <[g2]> I'll double check but I don't think that's in 1.2.11 May 16 12:23:09 # May 16 12:23:09 # Environment May 16 12:23:09 # May 16 12:23:09 CONFIG_NOR_BOOT=y May 16 12:23:09 # CONFIG_NAND_BOOT is not set May 16 12:23:11 CONFIG_AUTOBOOT=y May 16 12:23:13 CONFIG_AUTOBOOT_DELAY=10 May 16 12:23:15 That's the default. May 16 12:23:31 <[g2]> yes, but you must select the ixp4xx profile and that may change it May 16 12:23:32 10 as in 10 10ths. May 16 12:23:39 Huh? May 16 12:23:47 # make nslu2_config May 16 12:23:52 That will set the defaults. May 16 12:23:56 <[g2]> Ah... May 16 12:24:02 Just like the kernel. May 16 12:24:04 <[g2]> here's the difference... May 16 12:24:21 <[g2]> I was going into make xconfig and just picking the ixp4xxx profile May 16 12:24:50 <[g2]> I like the make xxxx_config better May 16 12:24:59 Best to start with the nslu2_defaults and then make your changes. May 16 12:25:24 <[g2]> well I've got to change things right out of the box because my toolchain is in a different place May 16 12:25:38 ah. May 16 12:25:57 <[g2]> can I make nslu2_default CROSS_COMPILE=.....linux- May 16 12:25:59 I think that you can use an environment variable for that. May 16 12:26:08 At least, it is suppose to work that way. May 16 12:26:18 <[g2]> or CROSS_COMILE=....linux- make nslu2_default May 16 12:26:26 <[g2]> and it gets picked up ? May 16 12:26:36 Hmm. No, I think it needs to be defined for the build. May 16 12:26:44 #CROSS_COMPILE=... make May 16 12:27:16 <[g2]> well it's easier to just sed the nslu2_defaults file and make nslu2_defaults May 16 12:28:08 alternatively, we can make it possible to put the compiler in the user's path. May 16 12:28:26 cross_compile=arm-linux- ... May 16 12:28:32 Perhaps that should be the default. May 16 12:28:51 <[g2]> we're splitting hairs here I think May 16 12:29:06 <[g2]> we just need to make it a little clearer on the setup May 16 12:29:11 <[g2]> I was doing it wrong May 16 12:29:11 Sort of, except the cleanliness of the build is the first thing that most people see. May 16 12:29:20 <[g2]> sure... May 16 12:29:20 Like you. May 16 12:29:45 I'll add some notes about that in the readme. May 16 12:36:42 <[g2]> beewoolie-afk, I've got an env.c patch and a new default config for openslug May 16 12:36:56 <[g2]> then we'll have nslu2_config and openslug_config May 16 12:36:57 email? May 16 12:37:06 That sounds good. May 16 12:37:08 <[g2]> sure I'll e-mail in a couple minutes May 16 12:37:16 <[g2]> I was going to test them first :) May 16 12:37:36 test schmest May 16 12:37:46 <[g2]> Right not the only change in the Opeslug is the path for me :) May 16 12:38:22 <[g2]> Does the memuconfig know which default profile was used ? May 16 12:39:14 I don't know what that would mean. May 16 12:39:20 What is a profile? May 16 12:39:36 <[g2]> sorry which default configuration May 16 12:40:16 <[g2]> you are correct about the 1 second autoboot being there when one makes the nslu2/openslug_config May 16 12:40:40 It works the same way as the kernel. May 16 12:40:45 # make foo_config May 16 12:40:52 Will set the configuration from a template. May 16 12:41:11 Successive calls to either menuconfig for xconfig will use those defaults instead of an arch default. May 16 12:41:39 I don't have arch defaults, so I think the defaults, given no other information, are completely empty. May 16 12:41:40 <[g2]> nod and what I'm saying is when I go into menuconfig or xconfig it would be nice to have a little variable about the default config May 16 12:42:30 <[g2]> I mean in case I'm playing with three different default configs May 16 12:42:43 <[g2]> Avila, Openslug, and nslu2 May 16 12:43:53 Well, again it's like the kernel. I used to symlink configs so that changes would be saved to the original file. I don't do that anymore. May 16 12:43:59 <[g2]> what would be "nice" is that the IX{42x and it should really be IXP4xx would say "Linksys NSLU2", "Avila", "OpenSlug" May 16 12:44:30 if you make foo_config, then edit the config, build, and do make bar_config, the foo changes are lost unless you copy .config to another file. May 16 12:44:48 Huh? May 16 12:44:48 <[g2]> exactly and that's totally fine May 16 12:45:49 Are you saying that you want the implementations to list openslug/unslung? May 16 12:46:05 <[g2]> What I'm saying is just like in the kernel there are processor selections in the processor family having a similar concept or a different heading could reminded the APEX config users which profile is currently being used May 16 12:46:13 The avilla should be added to the implementation list, just in case there are some hardware changes. May 16 12:46:34 When you use xconfig, you can see the implementation name May 16 12:46:43 <[g2]> I'll do the avila port and submit the patches and files to you May 16 12:46:54 Menucofig shows it, too. May 16 12:47:14 I suppose that the default memory config should change for the avilla. May 16 12:47:25 I don't know if there is anything else, though. May 16 12:47:51 <[g2]> Ok... are you saying the MACH_NSLU2 ? May 16 12:47:55 BTW, do you use quilt? It makes it really easy to develop patches. May 16 12:48:08 <[g2]> beewoolie-afk, not yet on the quilt thing May 16 12:48:23 there should be a MACH_AVILLA, I suppose. May 16 12:48:34 <[g2]> but it would be nice to say I go quilting with my friends :) May 16 12:48:39 The MACH's are implementations of the CPU or CPU family. May 16 12:48:57 quilting on-line with your pals. May 16 12:49:58 Now that I think of it, it is important to have an AVILLA target because there is some NSLU2 specific GPIO code in APEX already. May 16 12:50:41 <[g2]> right May 16 12:51:02 <[g2]> althought they don't effect the avila platform right now May 16 12:51:13 <[g2]> it's GPIO 8 right ? May 16 12:51:41 <[g2]> The default IXP425 platform used 8-11 for INTA-D iirc May 16 12:52:06 <[g2]> The slug uses 8-10, and avila uses 8-11 but it's split between two pci buses May 16 12:52:09 I think so. The NSLU2 doesn't. It uses 8 for power-down, IIRC. May 16 12:52:24 <[g2]> ok... maybe I'm off by 1 May 16 12:52:28 <[g2]> maybe it's 9-12 May 16 12:52:34 14 and 15 are clocks May 16 12:52:38 <[g2]> and 9-11 on the slug May 16 12:52:39 13 is PCI reset May 16 12:52:48 12, 11, 10, 9 are ints, IIRC. May 16 12:52:54 something like that. May 16 12:52:57 <[g2]> nod. May 16 12:53:18 <[g2]> Oh... so 13 is a hard PCI reset ? May 16 12:53:28 Should this be AVILA_XXXX or just generic AVILA May 16 12:53:53 <[g2]> I haven't looked closely at the boards and the does only the 2347 May 16 12:53:54 Well, there is a pin that performs PCI reset. I think, though, that it's a PCI peripheral reset that goes to the USB controller. May 16 12:54:19 <[g2]> so that resets all the target devices May 16 12:54:29 <[g2]> PCI target devices that is May 16 12:55:31 Is there an ixp425 on that board? May 16 12:55:50 <[g2]> it's a 422 but they do 420, 422 and 425 May 16 12:55:57 <[g2]> I've got two 425s here May 16 12:56:03 On the same model number? May 16 12:56:06 <[g2]> a BB and BD May 16 12:56:15 <[g2]> the RV042 and RV082 May 16 12:56:28 <[g2]> 266 @ 425 and 533 @ 425 May 16 12:56:54 <[g2]> Those are the devices that are getting pimped out tomorrow or the next day May 16 12:57:03 There are a couple of ways to handle this. I can add a sub-menu for avila so that all of the options are selectable. May 16 12:57:13 Or I can add top-level entries for each. May 16 12:57:24 It depends on how many choices there are May 16 12:57:27 <[g2]> Let's leave the avila stuff alone for now May 16 12:57:42 I just added one. May 16 12:58:04 <[g2]> At the top level ? May 16 12:58:10 Yeah. Below nslu2 May 16 12:58:37 config MACH_AVILA2347 May 16 12:58:37 bool "Gateworks Avila 2347" May 16 12:58:37 select ARCH_IXP422 May 16 12:58:37 select MACH_AVILA May 16 12:58:39 select USES_NOR_CFI May 16 12:58:43 select USES_NOR_BOOTFLASH May 16 12:58:45 select BIGENDIAN May 16 12:59:20 <[g2]> Ok, I'll send you the default config file and any other patches required May 16 12:59:37 ko May 16 13:00:33 <[g2]> So let me test the OpenSlug changes first and we'll have that fixed and finished, then we can have the default template for the avila stuff May 16 13:01:09 <[g2]> I might float a message on the avila board and let them know it all works for 2.6 and I could possibly test their default config too May 16 13:54:44 <[g2]> jacques I've got an APEX question for you ? May 16 13:54:57 for me? May 16 13:55:02 <[g2]> yeah for you May 16 13:55:17 <[g2]> you're interested in APEX right ? May 16 13:55:33 yes May 16 13:55:58 <[g2]> Ok.. after I give beewoolie-afk some minor tweaks I think APEX is ready to go on the slug May 16 13:56:50 <[g2]> basically, you'll download the latest, tweak for your compile path, make openslug_config; make and have an apex.bin for flashing May 16 13:57:29 <[g2]> jacques, your running OpenSlug right ? May 16 13:57:33 <[g2]> you are May 16 13:57:39 yes May 16 13:57:52 <[g2]> ok... so the question then becomes...... May 16 13:58:18 <[g2]> That solves the normal (current Redboot configuration). May 16 13:58:41 <[g2]> With a different APEX config file we can have a custom flash layout May 16 13:58:56 <[g2]> where we lose all the legacy stuff May 16 13:59:05 <[g2]> and gain back 3 128K blocks May 16 13:59:26 <[g2]> plus a APEX config area for persistent changes May 16 14:01:00 how does one currently get the firmware image to the slug using APEX? May 16 14:01:23 jacques: fedex May 16 14:01:38 <[g2]> Initially, it would be Block 0 APEX, Block 1 APEX environ, Block 2-10 (1.128M) Kernel, and the rest all JFFS2 May 16 14:01:44 <[g2]> jacques, it's totally cake May 16 14:01:58 <[g2]> from the openslug command line: May 16 14:02:22 <[g2]> 1) Backup the current redboot on the JFFS2 partition "dd if=/dev/mtdblock0 of=redboot.bin" May 16 14:03:13 <[g2]> 2) Copy over apex.bin "scp path..../apex.bin root@host:/home/root/apex_1.2.11_openslug.bin" May 16 14:03:41 <[g2]> 3) Reflash the bootloard, "dd if=apex_1.2.11_openslug.bin of=/dev/mtdblock0" May 16 14:04:04 <[g2]> 4) make sure /initrd isn't mounted (umount /initrd) May 16 14:04:13 <[g2]> 5) reboot "shutdown -r now" May 16 14:04:24 <[g2]> 6) do the happy dance as it all comes up May 16 14:04:55 ok, now I want to flash a new openslug image, how do I do it? May 16 14:05:23 since AFAIK NIC isn't working under APEX May 16 14:05:28 so no tftp May 16 14:05:59 <[g2]> jacques, The NIC and USB are 100% with the lastest APEX May 16 14:06:19 <[g2]> I'm booting via APEX to SDA1 May 16 14:06:23 you mean once linux is booted May 16 14:06:34 <[g2]> and the IXP automatically comes up properly May 16 14:06:41 <[g2]> sure May 16 14:06:49 I mean, what is the equivalent of load -r -v -b 0x01000000 /nslu2/rootfs.jffs2 May 16 14:07:02 <[g2]> well for now... May 16 14:07:13 xmodem? :-\ May 16 14:07:17 <[g2]> boot into linux and dd if=reboot.bin of=/dev/mtdblock0 May 16 14:07:28 <[g2]> and restore redboot or xmodem stuff down. May 16 14:07:38 ipkg upgrade openslug-kernel May 16 14:07:47 should work. May 16 14:07:56 <[g2]> dyoung-zzzz, welcome. May 16 14:08:01 <[g2]> there are minor issues with that May 16 14:08:14 let's say I can't boot into linux May 16 14:08:18 you guys made quite the backlog. May 16 14:08:22 <[g2]> :) May 16 14:08:26 still catching it up. May 16 14:08:28 I have to JTAG redboot back on ? May 16 14:08:38 <[g2]> ok... May 16 14:08:46 <[g2]> You've got serial May 16 14:08:52 <[g2]> and the kernel is still good right ? May 16 14:09:04 sounds like I need a slug specifically for testing APEX since I can't do dual duty as an openslug tester with it May 16 14:09:14 <[g2]> I think we can just change the kernel cmd line from APEX and NFS boot May 16 14:09:14 no, let's assume the kernel is non-bootable May 16 14:09:39 only thing that workd is APEX May 16 14:10:11 <[g2]> Ok... You X or ZMODEM the kernel in (untested) then NFS boot May 16 14:10:32 <[g2]> well'll have to add a small initrd for the modules networking modules to be loaded from the ramdisk May 16 14:10:32 OK so APEX *does* have [xz]modem that's good, untested is bad May 16 14:11:17 <[g2]> Ummmm... would you like me to wipe your butt next time you go to the bathroom too ? May 16 14:11:17 <[g2]> :) May 16 14:11:34 your point? May 16 14:12:16 <[g2]> My point I thought you were interested in APEX. May 16 14:12:26 I think I need at least one more slug, then the JTAG'ed one can be specifically for testing APEX May 16 14:12:37 Good Plan. May 16 14:12:59 Just as I should serial-enable my openslug test bed so I can use the jtag one for testing funner stuff. May 16 14:13:01 <[g2]> very good plan May 16 14:13:03 sure, I'm just trying to see where it is feature-wise May 16 14:13:22 <[g2]> that makes great sense May 16 14:13:39 yeah, I have two serial kits left - one usb and one not May 16 14:14:05 I'll probably use the USB one - wish I could have found more May 16 14:14:22 <[g2]> So we'll test the serial and make sure it works May 16 14:14:45 <[g2]> My real question to you and dyoung-zzzz is about the flash layout May 16 14:15:10 where will the partition table be stored? passed in on the kernel command line only? May 16 14:15:16 <[g2]> would you stay with the current layout or want to gain the 3 flash block back and move to a non-standard layout May 16 14:16:05 <[g2]> well APEX doesn't need the partition table right now May 16 14:16:13 only thing I can think of that's necessary in the current layout is partition table, and if we can somehow store that in the APEX config area, we can regain all that space May 16 14:16:39 <[g2]> Right I'd move it to the APEX environement area May 16 14:16:46 how are the partitions communicated to the kernel? May 16 14:17:04 I'm in favor or swiping the mac address from the end of redboot and putting it someplace intelligent. May 16 14:17:40 <[g2]> dyoung-zzzz, nod. we are going to backup the last 144 bytes of Redboot which are the most of the unique ones May 16 14:18:34 <[g2]> although if you're running OpenSlug the mac addr is already in /etc/network/interfaces May 16 14:18:51 yeah, but it would need a way to get there in the first place right? May 16 14:19:13 <[g2]> right you have to boot OpenSlug first right ? May 16 14:19:21 you do? May 16 14:19:58 seems on first boot openslug extracts MAC from redboot and puts it in /etc/network/interfaces May 16 14:20:05 in that case, its critical to have that information someplace smart because when you reflash then turnup the thing, the turnup wont know what its mac is anymore May 16 14:20:41 <[g2]> sure turnup copies /etc/network/interfaces after its been built May 16 14:20:47 anyways, thats one of my peeves about redboot. May 16 14:20:55 jacques: BTW, xmodem receive in APEX *is* tested. I use it frequently. May 16 14:21:20 <[g2]> beewoolie-afk, I was just speaking from personal experience May 16 14:21:27 dyoung-zzzz: I believe we can teach APEX to program the MAC address. May 16 14:21:59 <[g2]> I see near zero need to go there May 16 14:22:32 You're fussing about MAC addresses. I'm saying there are options. May 16 14:22:32 beewoolie-afk, great :-) May 16 14:22:49 <[g2]> right there are lots of options May 16 14:22:54 There isn't really a need to worry about losing the MAC address as it's printed on the box. May 16 14:23:34 <[g2]> I'm just trying figure out if you guys want to regain the 3 blocks and lose the legacy stuff May 16 14:23:54 jacques: FYI, the only important feature missing from APEX which Redboot has is NPE support. It's kinda hairy, so I've elected to avoid it for now. May 16 14:24:14 * beewoolie-afk raises his hand in approval of getting 3 blocks May 16 14:24:49 beewoolie-afk, yeah, I can understand how that would be a difficult painful feature to add (NPE) May 16 14:24:50 <[g2]> beewoolie-afk, you saw the note about moving the partition table to the APEX environment area right ? May 16 14:24:53 dont we really only gain back 1 block? May 16 14:25:32 <[g2]> dyoung-zzzz, sysconf area 1, Ramdisk fragmentation 1, Trailer 1, 1+1+1=3 May 16 14:25:37 or are you counting the last block too, in which case thats 2 May 16 14:25:57 <[g2]> Plus the sysconfig block May 16 14:26:02 Ok, so the suggestion is that apex will always fit in 1 block? May 16 14:26:25 <[g2]> 1 block APEX code, 1 Block APEX environment May 16 14:26:37 <[g2]> same size as Redboot which is 2 blocks right ? May 16 14:27:01 right, I saw talk of "future exansion" above someplace with the suggestion of reserving 2blocks for apex code. May 16 14:28:56 <[g2]> Ok so the 144 bytes (or less) from the end of Redboot and the partition table from the trailer get moved to the APEX environment block May 16 14:29:08 <[g2]> say the last 1 K or 2K May 16 14:29:55 I think max partition table size is 4kB (not positive) May 16 14:30:13 256 bytes x 16 entries May 16 14:30:14 * beewoolie-afk is on the phone. May 16 14:30:27 <[g2]> Ok... it's a 128K block with hardly anything in it May 16 14:30:40 <[g2]> I'm sure we've got plenty of room May 16 14:31:04 yes, I don't think that will be an issue May 16 14:32:15 <[g2]> so for discussion sake... Let's gzip the sysconf info, put the last 144 bytes, and have room for a 4K partition table at the end of the environment block May 16 14:32:38 <[g2]> that's the last 144 bytes of Redboot May 16 14:32:53 <[g2]> and fill the partition table with the proper values May 16 14:33:38 <[g2]> OK... This makes total sense to me.... All these things can be done at runtime with APEX May 16 14:34:12 <[g2]> So by default, APEX is like it currently is, meaning it doesn't initialize the environment block May 16 14:34:40 <[g2]> The procedure is flash OpenSlug May 16 14:34:52 <[g2]> and one can ipkg install APEX May 16 14:37:17 <[g2]> which saves off a copy of Redboot and reflashes APEX as the bootloader May 16 14:38:38 <[g2]> then there's another routine that can enable the environment block and move stuff around on the jffs2 partition to setup the extended functionality May 16 14:39:22 * beewoolie-afk hangs up phone and reads the logs May 16 14:39:41 Can I get a clarification here? Is the 'partition table' one of those redboot partition table thingys? May 16 14:39:55 <[g2]> it's a kernel partition table May 16 14:39:58 I ask, because there isn't a standard for it. May 16 14:40:09 there is May 16 14:40:13 Flash partitions are configured with the command line. May 16 14:40:31 <[g2]> actually there and ordered dependency May 16 14:40:37 <[g2]> the commandline is just one option May 16 14:40:37 we used existing kernel code to read the partition table May 16 14:40:46 and kernel by default looks at start of last block for it May 16 14:41:03 I think you're talking about a redboot partition table. May 16 14:41:22 static const char *part_probes[] __initdata = {"cmdlinepart", "RedBoot", NULL}; May 16 14:41:35 That's how the physmap driver handles partition table parsing. May 16 14:41:46 It looks to the command line and then looks for a RedBoot partition table. May 16 14:42:26 IOW, we don't really need the redboot legacy stuff. May 16 14:42:39 It's better IMHO to use the command line. It's more clear and easier to change. May 16 14:43:26 I don't see a problem with that May 16 14:43:34 <[g2]> nslu2_2.6.11.patch:+static const char *probes[] = { "cmdlinepart", "RedBoot", "NSLU2", NULL }; May 16 14:43:52 <[g2]> that's the patch applied by OE May 16 14:43:57 I was just saying before that the partition info had to be stored somewhere and passed to the kernel somehow May 16 14:43:59 Ah. The it's a non-standard format. May 16 14:44:05 All the more reason to go to the command line. May 16 14:44:19 non-standard how? May 16 14:44:33 NSLU2 is not a partition 'standard'. No other platform uses it. May 16 14:44:33 it's a standard RedBoot partition table May 16 14:44:44 Then why call it something different? May 16 14:44:56 oh, I dunno what that is, I was referring to the "RedBoot" part above May 16 14:45:21 <[g2]> well we could easily change the patch to be "cmdlinepart", "RedBoot", NULL May 16 14:45:34 when I wrote the code in slugimage to build tha partition table, I referred to the kernel code for RedBoot partition table May 16 14:45:43 In the end,I think it's better to go to the command line. We don't even have to change the kernel. May 16 14:45:52 so I dunno what NSLU2 format means May 16 14:46:18 yes, we are ascillating between two different things here May 16 14:46:22 oscillating May 16 14:46:53 I have already agreed command line is fine, but we are also arguing about the current partition table format name May 16 14:47:01 I'm trying to pull it back. The more that we do that is simple, command-line or ATAG based, the more conforming our solution will be. May 16 14:47:04 Right. May 16 14:47:34 <[g2]> beewoolie-afk, nod. May 16 14:47:50 <[g2]> jacques, As far as I'm aware it's the RedBoot format May 16 14:48:04 [g2], me too May 16 14:48:31 and I wrote the slugimage code by looking at the kernel redboot format code so I think I know what I'm talking about May 16 14:48:38 NSLU2 from that kernel patch is the stock factory non-standard static partition scheme that linksys uses. We're pretty much ignoring it for everything made post-slugimage. May 16 14:48:38 (could be wrong tho) May 16 14:48:43 jacques: Just to clarify one other point. APEX is really small. It will remain so unless we decide we have to have NPE. I've been plannng to add a feature where it is possible to put the environment in the same block as the loader. It means we could reduce the memory usage by another block. May 16 14:49:48 beewoolie-afk, I think that would be for specialized application tho - because I would like to be able to change the env data without worrying about corrupting bootloader May 16 14:50:02 personally for slug I don't think it's worth it for that one block May 16 14:50:03 That isn't really the problem. May 16 14:50:59 The environment code is smart about writing data. The trouble is that the allocated space can fill with old settings. Erasing the whole block requires erasing the same block that the loader is in and rewriting it. May 16 14:51:12 It just needs a special command to do it. May 16 14:51:12 ah nice May 16 14:51:38 so you have a section of 0xFF and just write new config data until it's full May 16 14:51:39 As I said, it's just a small clarification. May 16 14:51:48 clever May 16 14:51:51 <[g2]> well you are risking the boot block when you do that and it isn't worth it for that 1 block May 16 14:52:27 yeah. to both of you. That's why it hasn't come up. it is more important on systems with larger erase block sizes. May 16 14:52:39 <[g2]> agreeed May 16 14:53:29 <[g2]> so do you want the partition table copied or not ? May 16 14:53:54 maybe not copied so much as translated into a command line format May 16 14:54:10 <[g2]-dinner> that's totally possible too May 16 14:54:22 in practice, we never got close to the 16 entry limit May 16 14:54:34 seems a waste of space 4kB just for partition table May 16 14:54:55 <[g2]-dinner> well it's empty anyway May 16 14:55:00 <[g2]-dinner> off to dinner May 16 14:55:24 You guys are the experts. May 16 14:55:31 * dyoung-zzzz is a slacker May 16 15:02:51 dang I wish slugs would go on sale somewhere May 16 15:11:10 Is there a clearing house of tivo hacking info someplace? Googling returns too many hits to be helpful. May 16 15:11:54 Basically I inheireted a Series2 which I just slapped a 120G into. its software version was updated to 7.1; but I want to be able to telnet to it and screw around wiht it, etc... May 16 15:12:11 do I need to downgrade to 4.0.1 or something to do that? May 16 15:12:20 * dyoung-zzzz wonders aloud.... May 16 15:15:21 tivo underground May 16 15:15:50 * dyoung-zzzz googles more May 16 15:16:10 http://www.tivocommunity.com/tivo-vb/forumdisplay.php?f=24 May 16 15:19:01 <[g2]-dinner> jacques, last I bought they were $79US at newegg May 16 15:19:21 [g2], yeah, I know, I would prefer to pay less May 16 15:19:45 I check newegg almost every day and they are holding steady May 16 15:20:25 dyoung-zzzz, if you need info about things which aren't allowed on tivocommunity.com, there are other forums May 16 15:20:49 Ok May 16 15:21:07 I just wanna be able to telnet to the thinig for now. May 16 15:21:16 I'll get to "those other topics" a little later. May 16 15:37:04 <[g2]> jbowler, welcome May 16 15:37:13 <[g2]> beewoolie-afk, around ? May 16 15:37:41 <[g2]> jbowler, beewoolie is the author of APEX May 16 15:38:02 Ah, ok. Let me state the problem... May 16 15:38:06 <[g2]> I've been working with him to get stuff fully running on the slug and other IXP OEM boards May 16 15:38:30 With RedBoot, RedBoot SysConf partition->openslug jffs2 /etc/default/sysconf (say) and this supplies the eth0 MAC address May 16 15:38:34 <[g2]> the problems is if we move sysconf you lose where you are getting your MAC frome May 16 15:38:46 With APEX something needs to read and generate the same file with equivalent information. May 16 15:39:17 That's easy - we just need some 1130 bytes somewhere to store the useful part of the original RedBoot SysConf May 16 15:39:27 <[g2]> 144 May 16 15:39:58 <[g2]> for the Redboot, and we've been talking about saving the whole sysconf gzipped or something like that May 16 15:40:01 It's not just 144. The user may have run LinkSys config and stored useful data there. May 16 15:40:21 <[g2]> 144 for Redboot, more for sysconf May 16 15:40:42 There's no need to gzip it - actually that's bad. Putting gunzip into /linuxrc is bad ;-) May 16 15:41:13 <[g2]> that makes a bunch of sense if you are pulling it from there May 16 15:41:36 <[g2]> actually, we can add something to /proc like macaddr May 16 15:41:46 [g2] right. May 16 15:41:53 that was the latest plan, but I slacked on it. May 16 15:41:58 <[g2]> and it can be pulled from flash or passed on the cmd line May 16 15:42:01 and the patch never got written. May 16 15:42:15 [g2]: if it can be added to /proc then it can be added to the driver. Much better, no change for user land mistakes. May 16 15:42:20 <[g2]> well APEX passes cmdline options fine May 16 15:42:24 s/change/chance/ May 16 15:42:50 * dyoung-zzzz hides May 16 15:42:54 <[g2]> yeah.. except for those licensing issues it makes total sense May 16 15:43:08 My feeling is that the MAC should not need to be set by ifconfig - it should default to the allocated MAC if at all possible. May 16 15:43:13 <[g2]> that's in the 425 section which is GPL right ? May 16 15:43:22 [g2]: correct May 16 15:43:26 <[g2]> jbowler, totally agreed May 16 15:43:59 Anyway, so what, if the NSLU2 kernel just has the two numbers stored somewhere that's fine (I assume there are two, but maybe LinkSys only bought one...) May 16 15:44:12 <[g2]> OK.. SO.. that begs the question of what do we do with the current OpenSlug for setting the MAC addr ? May 16 15:44:37 isn't that going to be bootloader specific? May 16 15:44:54 Why do we need to do anything? If I can't read hw_addr from SysConf I won't set it ;-) May 16 15:45:54 <[g2]> jbowler, how do you know it's sysconf ? are you checking the partition name, are you checking the lenght at the beginning are you parsing the tag ? May 16 15:46:08 [g2]: yes, to all of the above May 16 15:46:30 <[g2]> Ok. so if there's no sysconf partition, you could look elsewhere May 16 15:46:31 If I can find something called "SysConf" in /proc/mtd I can find it. May 16 15:47:23 [g2]: yes, and if I can't find SysConf in /proc/mtd I know to look elsewhere. May 16 15:47:45 <[g2]> jbowler, is *such* a reasonable guy :) May 16 15:48:25 Good Plan. May 16 15:48:28 <[g2]> longer term, I'd like to see just 3 partitions in the simplest case May 16 15:48:38 <[g2]> 1) APEX code, May 16 15:48:43 <[g2]> 2) APEX environ May 16 15:48:59 <[g2]> 3) jffs2 rootfs where the kernel is loaded from the rootfs May 16 15:49:05 yeah May 16 15:49:18 <[g2]> we need to teach APEX about JFFS2 first May 16 15:49:23 I'd really like that feature May 16 15:49:41 * dyoung-zzzz me toos May 16 15:49:43 here we go again. May 16 15:49:50 :-) May 16 15:49:52 <[g2]> we are just some driver tweaks away from doing that with ext2 May 16 15:50:04 I feel like a yes-man today. May 16 15:50:15 <[g2]> but usb console was deemed important May 16 15:50:24 I feel like implementing jffs2 in APEX just to shut you guys up. May 16 15:50:29 :-D May 16 15:50:37 <[g2]> well I'm trying to make sure beewoolie-afk and I are working on the features you guys care about right ? May 16 15:50:44 "I'll buy that" May 16 15:50:50 "Whos baby is that" May 16 15:50:58 <[g2]> the jffs2 driver ? May 16 15:51:00 "I'll buy that baby" May 16 15:51:11 So how about adding 'Kernel 1' and 'Kernel 2', 'Kernel 2' gets used unless it's bogus, I make it bogus on boot ;-) May 16 15:51:29 <[g2]> that's a possibility too. May 16 15:51:39 <[g2]> I was thinking more rootfs1 and rootfs2 May 16 15:51:39 "I'd buy that for a dollar!" May 16 15:51:46 lol May 16 15:52:07 Can anyone name the reference? May 16 15:52:09 * jacques goes for a ride in his 6000SUX May 16 15:52:18 Tee hee May 16 15:52:21 rootfs much bigger than kernel May 16 15:52:33 <[g2]> not our kernels :) May 16 15:52:41 both too damn big May 16 15:52:48 Robocop May 16 15:53:02 Nice. May 16 15:53:09 <[g2]> jbowler, APEX can x|zmodem over serial also May 16 15:53:12 ah, I'm slow today... must be the binary release hangover May 16 15:53:34 [g2]: No z modem. I'd like to support it, but the code is big. May 16 15:53:47 <[g2]> X and Y then ? May 16 15:54:01 Just x. May 16 15:54:06 <[g2]> ok. just X May 16 15:54:15 apex> xreceive ADDRESS May 16 15:54:27 ADDRESS can be direct to flash. May 16 15:54:38 changing a kernel is a user-land operation (i.e. it's not valid to assume a mod'ed NSLU2), so it's fine if APEX can do telnet+x/y-modem May 16 15:54:45 even blob in 2001 could fo jffs2 so how hard could it be? :-D :-D May 16 15:54:51 <[g2]> apex>xrecieve nor:0x0+0x2000 ? May 16 15:54:52 OK OK OK! May 16 15:55:03 lol May 16 15:55:04 You don't need to tell it how long the destination is. May 16 15:55:07 lol! May 16 15:55:12 <[g2]> beewoolie-afk, hey these guys haven't done anything with APEX May 16 15:55:16 Let me get the blob code. May 16 15:55:25 <[g2]> beside yapp about features they want May 16 15:55:36 * beewoolie-afk starts digging though pile of debris, looking for old listings May 16 15:55:55 <[g2]> and badger us hard-working under-paid APEX dev's May 16 15:55:59 If JFFS2 is as easy as that, it isn't worth not doing it. May 16 15:56:07 <[g2]> :) May 16 15:56:32 I know because we used blob with kernel in jffs2 on tuxscreen May 16 15:56:55 The blob code that I have doesn't have jffs. It says it is planned. May 16 15:57:03 There may be a patch for it. May 16 15:57:40 hmm, maybe the patch is somewhere in the buildroot-tux or on the tuxscreen.net wiki May 16 15:57:50 let me see if anyone's awake in #elinux May 16 15:58:14 Robocop? May 16 15:58:21 Nah all that stuff is from the Flintstones. May 16 15:58:43 Oh, "I'll buy that for a dollar", i didnt see that one. :-) May 16 15:59:07 jacques: looks like a patch alright. May 16 16:01:11 let me check buildroot-tux May 16 16:01:59 blob-20020405.patch May 16 16:02:09 hmm too small May 16 16:02:35 I found one on the bootldr site. May 16 16:02:57 youre too fast. I was just gonna paste that link. May 16 16:03:08 Heck, I've already read the patch. May 16 16:04:34 beewoolie-afk, it might already be in the CVS blob May 16 16:04:36 It's actually a lot of code. May 16 16:04:48 looks like buildroot-tux gets blob from CVS May 16 16:04:56 jacques: I didn't know there was a CVS repo for blob. May 16 16:05:11 and builds it with --enable-jffs2 May 16 16:05:27 CVS_PASSFILE=$(CVS_PASSFILE) \ May 16 16:05:27 cvs -d:pserver:anonymous@cvs.sourceforge.net:/cvsroot/blob \ May 16 16:05:27 co -D 20020621 blob) May 16 16:05:37 I see. It looks like the trick is performing a log replay so that the metadata can be recovered. May 16 16:06:28 Is this read-only or read-write? May 16 16:06:33 jacques: Why use the cvs date? May 16 16:07:35 beewoolie-afk, that's in the blob.mk for tuxscreen buildroot - they probably wanted to make sure they got a known good version (good thing with a bootloader) May 16 16:07:45 beewoolie-afk, I'm not saying you need to use that date May 16 16:07:50 Fair enough. May 16 16:08:17 I'm trying to look at the blob page on sf.net but it's being really fscking slow May 16 16:08:35 more fsking slow than usual May 16 16:08:41 This code looks more reasonable than the bootldr stuff. May 16 16:11:43 <[g2]> beewoolie-afk, currently you are not using the partition tables at all you are just using the address/offset/lenght from APEX environment vars May 16 16:12:24 <[g2]> when we move to the driver model are you planning to use those ? or would you rather not support that ? May 16 16:14:17 Presently, regions are specified literally. May 16 16:14:38 Once I add some of the features I've talked about for chaining driver, we can use aliases to refer to flash partitions. May 16 16:14:56 alias part0 nor:0+128k May 16 16:15:43 [g2]: BTW, I got distracted. you don't need to specify the length of the receive region for an xmodem receive. May 16 16:16:04 apex> xr nor:0 # download to start of flash May 16 16:30:21 * ka6sox-office got an edgekation. May 16 17:08:36 jacques: BTW, one of the problems with this code is that it's not endian-safe. It isn't complex, fortunately. Only two routines really. But it needs some help. May 16 17:09:05 beewoolie-afk, ah yeah, that does not surprise me actually :-\ May 16 17:09:50 Unfortunately, I haven't found a description of the JFFS2 format which would make this easier since I'd know if everything was kosher. May 16 17:11:15 hmm, I thought I had found one someplace when I was adding the "create empty jffs2 partition" feature to slugimage May 16 17:12:20 I just want something that explains the meanings of all of the bits and fields in the structures. There's a little out there, but not much. May 16 17:12:36 I assume you've already found this: http://sources.redhat.com/jffs2/ May 16 17:13:01 Just found it. thx May 16 17:14:14 http://mhonarc.axis.se/jffs-dev/maillist.html May 16 17:14:23 David Woodhouse still posts to that list May 16 17:15:21 and he's dwmw2 on #handhelds.org May 16 17:17:21 maybe some stuff here too: http://www.linux-mtd.infradead.org/index.html May 16 19:57:54 <[g2]> dyoung-zzzz, around ? May 16 20:05:08 <[g2]> dyoung-web, around ? May 16 20:05:09 * dyoung-web ear perk up May 16 20:05:25 <[g2]> Ok I've got a hw question for you May 16 20:05:31 Okay. May 16 20:05:53 <[g2]> Currently the RV042 and 082 had 32MB memory as in 2 8Mx16 chips May 16 20:06:27 <[g2]> tomorrow I'll drop them off at a re-work shop and they'll switch those chips to be 16Mx16 May 16 20:06:37 Okay. May 16 20:06:37 <[g2]> and add an extra set on the 082 May 16 20:06:43 Woo! May 16 20:06:58 <[g2]> I'm guessing that things should still boot with the old firmware May 16 20:07:08 Yes, they should. May 16 20:07:16 the upper 32M should be ignored. May 16 20:07:19 <[g2]> as the avila which had the same situtaion booted when it was only configured for 32M May 16 20:07:23 <[g2]> but had 64M May 16 20:07:42 <[g2]> until I enabled that extra 32M :) May 16 20:08:17 <[g2]> Okay.... we'll see in a couple days May 16 20:08:43 <[g2]> with some luck I'll have two units with one with 64MB one with 128MB and both the serial and JTAG May 16 20:08:52 <[g2]> maybe even double flash May 16 20:08:53 neato. May 16 20:09:05 * [g2] thinks so May 16 20:09:23 I wish I had as many toys as [g2] May 16 20:09:46 * [g2] is trying to start a hw business :) May 16 20:10:05 me too. May 16 20:10:15 <[g2]> what are you gonna build ? May 16 20:10:20 more like a "stuff integration" thing I guess. May 16 20:10:26 did you see that guys site? May 16 20:10:34 what was it.. May 16 20:10:48 anyways, he bought 300 nslu2's to use for that. May 16 20:10:58 <[g2]> NO I missed that one May 16 20:11:20 lemme poke around; its really neat stuff. May 16 20:11:25 huh? May 16 20:11:28 i wanna see that too May 16 20:11:28 <[g2]> it wasn't the German guy was it ? May 16 20:11:37 <[g2]> for the libraries ? May 16 20:11:49 no, this dudes in cali. May 16 20:12:12 <[g2]> I wish I had known May 16 20:12:24 <[g2]> that coulda been an avila purchase May 16 20:12:54 I tried to sell him that. May 16 20:13:01 but he said he already bought 300 nslu2's May 16 20:13:18 <[g2]> is there a follow-on buy ? May 16 20:13:25 <[g2]> does he need support ? :) May 16 20:13:34 hehe I think hes pretty self supported. May 16 20:13:52 the nslu2's are to be a gateway device from a network of wireless sensors. May 16 20:14:02 lemme dig it up, still looking. May 16 20:14:21 the 2 main guys are from berkeley. thats actually their "real job". May 16 20:14:58 <[g2]> oh the wireless motes thing May 16 20:15:17 yeah, thos guys. May 16 20:15:47 i still would like a link because this is first I've heard of this May 16 20:16:19 sigh, sparkfun had to double their board making cost from $2.50/in to $5.00/in May 16 20:16:37 too good to last May 16 20:16:50 oh yeah, moteiv.com May 16 20:17:48 300 nslu2's is almost m ind boggling. May 16 20:18:02 dammit well they can't have my slugarmy domains! May 16 20:18:14 hehe you have slugarmy? May 16 20:18:19 yep May 16 20:18:32 thats petty cool! May 16 20:19:43 <[g2]> moteiv.com is that the site for the 300 slugs ? May 16 20:20:09 jacques, you know about barebonespcb.com and 4pcb.com right? May 16 20:20:44 been there...done that...didn't have good luck with either of them. May 16 20:20:47 the killer i guess is the $35 setup fee. May 16 20:20:57 Same here; I've had very good luck with them. May 16 20:21:12 really? May 16 20:21:14 they are the same company. May 16 20:21:29 ah..thats why... May 16 20:21:58 jacques, basically thats what sparkfun is doing. they are aggregating orders to place with 4pcb.com . May 16 20:22:15 [g2]: yes; thats the 300-slug guy. May 16 20:22:49 <[g2]> that's his company ? project ? ??? May 16 20:23:30 Oh excuse me, they are aggregating through Phoenix Tech. May 16 20:23:36 thats their company. May 16 20:23:43 they do similar stuff at work (at berk) May 16 20:24:44 http://bwrc.eecs.berkeley.edu/ May 16 20:25:52 now if I were a little smarter, I would setup a 501c3 research corporation so I/We could apply for research grants. May 16 20:28:31 [g2] anyways, thats what I want to do when I grow up. "Device consulting". May 16 20:28:58 me too May 16 20:29:10 maybe May 16 20:29:13 :-) May 16 20:33:16 <[g2]> Hey check this out :) http://www.eeproductcenter.com/micro/brief/showArticle.jhtml?articleID=163103857 May 16 20:33:45 Does it run windows xp? May 16 20:34:20 Bah, dont lead me on like that! May 16 20:34:23 Theyre sold out! May 16 20:34:40 argh May 16 20:34:48 <[g2]> sold-out :( May 16 20:35:03 I was gonna buy one right now too. May 16 20:36:51 [g2] are you a Avila reseller? May 16 20:36:58 <[g2]> nope May 16 20:37:26 <[g2]> Although if we had a big order I'm sure they would talk to us :) May 16 20:37:57 <[g2]> I don't think they really understand the premise May 16 20:38:12 <[g2]> that'll probably change in the next couple months May 16 20:38:29 <[g2]> hmmm.... May 16 20:39:48 <[g2]> well I'm off to bed May 16 20:39:55 <[g2]> sweet dreams gus May 16 20:39:57 <[g2]> guys May 16 20:39:59 ok nighty! May 16 20:40:20 <[g2]> thx for hw answer. cross your fingers :) May 16 20:40:46 ok, crossing! May 16 20:50:25 i love how hollywood has no clue how a glock works May 16 20:50:49 they're always making clicking noises as if you can cock and decock them May 16 20:50:53 what movie are you watching? May 16 20:51:14 pick any movie with a glock in it May 16 20:51:25 but right now some old eddie murphy thing called metro May 16 20:51:31 true... May 16 20:51:42 I'm still waiting to see a movie with a HK P90 in it. May 16 20:51:59 I dig the looks of that thing. I almost bought the airsoft toy of it. May 16 20:52:16 let's not even get into the number of clicks on the guns which can be cocked/decocked May 16 20:52:42 not familiar with the HK - they were always too rich for my blood May 16 20:54:06 I have a love affair with it because of a anime, Gunslinger Girl. May 16 20:54:11 I think it just came out here. May 16 20:54:15 last night I was watching "open range" which is a damn good western, but at one point kevin costner fired nine shots from his revolver in one volley May 16 20:55:16 Oop, my bad, its a FN not Hk. May 16 20:55:23 http://world.guns.ru/smg/smg13-e.htm May 16 20:55:34 its so "cute" in that anime. May 16 20:55:46 ah bp May 16 20:55:58 I was watchin some War movie and they fired 10 shots out of a M1911A1 May 16 20:56:01 http://www.gunslingergirl.com/ May 16 20:56:02 bullpups do have an interesting look May 16 20:56:19 the feed on this thing is messed up though. May 16 20:56:23 ka6sox-away, heh May 16 20:56:28 7+1 baby May 16 20:56:41 in those days, these days my shooting star mags get me 8+1 May 16 20:56:59 the mag is sideways, and theres a little spinny thing that makes then go in the right way. May 16 20:57:02 Ya...got one in the Ready Service Mag. May 16 20:57:38 dyoung-web, ah sounds like that French thing... forget the name May 16 20:57:45 tho I thought it was caseless May 16 20:57:45 Fal? May 16 20:57:47 yeah May 16 20:58:24 I should buy the P90 toy and violin case. May 16 20:58:40 "play ya a tune on my violin? May 16 20:58:59 that anime was kind of messed up too come to think of it.... May 16 20:59:21 the more I look at that the cooler it looks May 16 20:59:53 * ka6sox-away prefers the M1921 May 16 21:00:09 what's that? tommygun? :-) May 16 21:00:20 it was meant to be used as a PDW. the round is kinda smallish, but is perfect for cqb. May 16 21:00:36 or rather its sucessor the M1928A1 (tommy gun) May 16 21:01:01 that's what 10 .45's per second? sweeet May 16 21:01:04 that was one of my requirements for a possible "consulting" offer I had in the "middle east" a few years ago. May 16 21:01:15 but only with a Straight Stick Magazine..the Drums were "unreliable" May 16 21:01:53 they used to tape 2 sticks upside down together.... May 16 21:02:16 mine is "open bolt" May 16 21:02:38 Can ya tell I like .45ACP? May 16 21:02:55 it's my favorite handgun caliber by far May 16 21:03:05 Hmm, I wonder if I can make you cringe.... May 16 21:03:08 "Sten" May 16 21:04:24 "grease May 16 21:04:27 " ? May 16 21:07:46 mac 10 May 16 21:08:11 all of them Cheap, Stamped...not Tooled May 16 21:11:11 the Thompson is a very heavy piece. May 16 21:14:01 * ka6sox-away wonders if ByronT can tell what the subject is.... May 16 21:14:22 Thompson sounds like a gum May 16 21:14:24 gun May 16 21:14:30 yes...it is. May 16 21:14:38 Enfield. May 16 21:14:49 sorry, i just have to cringe whenever someone says sten. May 16 21:14:58 sten. May 16 21:17:49 <[g2]> jacques, around ? May 16 21:17:58 yeah May 16 21:18:11 <[g2]> can you look at my ID on AIM ? May 16 21:18:34 <[g2]> giantshoulderinc May 16 21:18:39 look at how? I see you're signed in May 16 21:18:57 <[g2]> I've got no away message ? May 16 21:19:29 it doesn't think you;re idle May 16 21:19:36 not sure how gaim shows away message May 16 21:19:54 ah I see it now May 16 21:20:13 it talks about nslu2-linux May 16 21:20:13 what is your AIM id, jacques? May 16 21:20:37 I just msged you ByronT May 16 21:21:05 [g2], that answer your question? May 16 21:21:25 you guys all use aim too? May 16 21:21:42 iChat uses aim contacts. May 16 21:22:08 dyoung-web, yeah, and it's sometimes useful as an OOB channel May 16 21:22:15 oob? May 16 21:22:32 * jacques waits for ka6sox-away to answer that one :-) May 16 21:22:41 (out of band) May 16 21:22:52 Ooh. May 16 21:23:08 <[g2]> I hardly ever use it. I'm trying to hunt down a RV042/082 hacker May 16 21:23:47 <[g2]> And being that I'm so AIM savvy I needed help from my tech buddie jacques May 16 21:24:04 my last job was all about IM May 16 21:24:11 AIM, MSN, Y! May 16 21:24:19 Jabber to a smaller extent May 16 21:24:25 <[g2]> you worked at zipit ? :) May 16 21:24:40 nope, activebuddy now called conversagent May 16 21:24:53 talk to smarterchild on aim and you'll see May 16 21:25:52 SC is a cluster of linux boxes, about 20 dual processor machines May 16 21:27:28 all together I admined about 120 linux boxes in that job May 16 21:28:25 <[g2]> cool May 16 21:29:03 <[g2]> so If I'm away I can't get messages May 16 21:29:15 you should be able to May 16 21:29:29 <[g2]> ah.. they go through ? May 16 21:29:35 tho I admit I don't use that feature May 16 21:29:50 <[g2]> oh.. it has to be enabled ? May 16 21:30:29 i thought away was just a courtesy for ppl who might want to talk to you May 16 21:30:48 did you get my test message just now? May 16 21:30:50 <[g2]> can you send me a message ? May 16 21:30:55 I just did - "test" May 16 21:31:00 <[g2]> nod May 16 21:31:03 <[g2]> hi there May 16 21:31:06 <[g2]> cool May 16 21:31:39 <[g2]> THX jacques May 16 21:31:42 np May 16 21:34:50 my tivo knows how much I like "fargo" May 16 21:35:02 it recorded it and now I'm watching it May 16 21:35:27 jacques, I think I might not be smart enough to hack my Tivo. May 16 21:35:30 jacques, apofdoom May 16 21:36:23 dyoung-web, I know the series 2 are harder, I have series 1 May 16 21:36:49 <[g2]> ok good nite for real... May 16 21:36:50 anyone know a AIM screenname ending in monkey ? May 16 21:36:52 so what are some good PS2 games May 16 21:36:58 :) May 16 21:37:09 they just added me as a buddy and it wants to know if I want to add them May 16 21:37:10 starts with moss? May 16 21:37:13 yah May 16 21:37:18 and then an _ May 16 21:37:20 <[g2]> save your money XBOX 360 is coming out soon :) May 16 21:37:30 ok I added it May 16 21:37:31 :-) May 16 21:38:15 my wife gave me a PS2 for my birthday, :P May 16 21:38:19 dyoung-web, whereabouts are wou getting stuck? May 16 21:38:25 you May 16 21:38:27 PS2 is nice! May 16 21:38:50 it's small! May 16 21:39:03 I dunno, I think I spent 1hr searching for info on 7.1 and came up pretty empty. May 16 21:39:33 so does that mean I need to back it up, then downgrade to 4.0.1 to perform any of these hacsk? May 16 21:39:59 I successfully backed up the original 80 to a 120; but I had really hoped to use a 200G disk. May 16 21:40:16 I dunno, maybe I didnt put enough effort into poking around yet. May 16 21:40:30 that's a huge version jump to 7.1 - I had no idea it was that high on the series 2 - on series 1 it's stuck at 3.0.1 May 16 21:41:18 Well, I plugged it in and it updated itself from 4.0.1 to 7.1a. May 16 21:41:25 <- has a directivo... we were just told that my Series 2 will be upgraded to 6.2 May 16 21:41:33 you see your model on this page: http://9thtee.com/tivoupgrades.htm May 16 21:41:34 when I first turned it on it was 4.0.1. May 16 21:41:50 i guess 7.x has that new fast forward ad thingie May 16 21:42:16 it was all annoying. I didnt intend for it to upgrade itself. It just did it. May 16 21:42:31 I have a 240080. May 16 21:42:34 and the folders May 16 21:43:13 they have folders now? May 16 21:43:32 Folders for music and video (I thought) May 16 21:44:17 sorting of the Now Playing list by show names, and by Season Passes May 16 21:45:37 hopefully my series 1 will never be "upgraded" again May 16 21:45:54 last upgrade screwed it up and I'd rather it not get any more screwed up May 16 21:45:59 time for NWN. May 16 21:46:11 have fun ka6sox-away May 16 21:46:35 my rogue needs more practise. May 16 21:47:15 Oh well. I have the original hard drive safely stored away, I guess I'll screw around with the 200 until I bend it to my will. May 16 21:49:08 lol fargo reminds me how much I hate car salesmen May 16 21:51:21 dakota ppl talk funny May 16 21:53:37 Bastage! Tivo rates just go up recently? May 16 21:54:26 they did? May 16 21:54:30 how much? May 16 21:54:36 I have lifetime from when it was $200 May 16 21:55:23 L/T $300, Monthly $13. May 16 21:55:32 ouch May 16 21:55:44 i think it's been $13 for a while May 16 21:55:53 and LT has been $300 for a while too :-( May 16 21:55:53 with DirecTV, the Tivo service is only $4.99 per receiver May 16 21:56:03 yeah DT is a different deal May 16 21:56:13 Really? Last time I looked it was $10. May 16 21:56:19 completely different deal May 16 21:56:23 but that might have been a while ago May 16 21:57:00 but you can *blackmail* DirecTV into giving you an 80 hour DirecTivo unit May 16 21:57:10 by telling them that you might switch to cable May 16 21:57:32 jacques: so is your Tivo happily telnet enabled? May 16 21:57:54 I'm reading there a lot of hassles it making things work right on a series2 May 16 21:58:07 dyoung-web, yeah, I added a turbonet card to it - mostly so I could get rid of the land line requirement May 16 21:58:32 and I can telnet to it, but I haven't done much else - I wanted to setup the webserver etc May 16 21:59:09 thats what I want to do too... but.... May 16 21:59:12 sigh May 16 21:59:25 I need a brain transplant or something. May 16 21:59:45 hey dyoung-web...you have to stand in line behind me! May 16 22:00:28 dyoung-web, there's an iso image somewhere you can burn onto a cd, then boot a machine with that, hook up the drive, and go from there May 16 22:00:36 that's how series 1 is done these days May 16 22:00:42 pretty sure series 2 as well May 16 22:01:38 Well, I guess more fiddling is required. May 16 22:01:44 I should put it in the wiki. May 16 22:01:48 I hate searching forums. May 16 22:02:05 basically, the cdrom boots a linux kernel which has the tivo fs built in May 16 22:02:14 forums suck. May 16 22:02:18 so it can manipulate the tivo partitions May 16 22:02:23 they are a nightmare. May 16 22:02:37 I downloaded one such iso. May 16 22:02:42 it seems to work. May 16 22:02:52 also, endian-ness is different, but the new stuff should make that transparent May 16 22:02:56 but I think I might have had PEBKAC. May 16 22:03:49 it's been about three years since I upgraded the capacity in my series 1, and about 8 months since I added the turbonet, so I don't remember everything May 16 22:07:27 I hvae tivo on the brain now May 16 22:07:38 I think I need to ignore this CRC problem and go home and play with Tivo. May 16 22:08:21 lol! May 16 22:11:32 ka6sox-away: I forgot to ask earlier. Will changing the value of my pull-up from 100k (my frankenstien test board) TO 47k (your production board) significantly affect rise times and stuff? May 16 22:14:01 I'll find the answer in the log May 16 22:14:06 bbl May 16 23:32:32 yes May 16 23:32:37 it *should* help May 16 23:32:42 r May 16 23:32:44 rats May 16 23:34:29 should "help" in which direction? May 16 23:37:43 make it faster May 16 23:42:23 Aha, that makes sense then May 16 23:42:25 how much faster? May 16 23:44:37 2x? May 16 23:44:44 RC May 16 23:45:00 C is small but R is relatively large. May 16 23:47:13 2x (100/47) unless C varies with I, which only happens if I is too large (and it only happens once). May 16 23:48:54 HCT input May 16 23:49:08 that explains everything then. May 16 23:49:58 Opencollector driving a single HCT input. May 16 23:50:13 and the OC can sink 20ma May 16 23:50:29 (which from a 5v source is impossible. May 16 23:50:31 ) May 16 23:57:56 at least with a 100k pullup. May 16 23:58:01 :) May 16 23:59:54 You could try a 220ohm and see what happens ;-) **** ENDING LOGGING AT Mon May 16 23:59:56 2005