| 19:37:10 | | BZFlag was last seen on #tuxscreen 3 days, 15 hours, 17 minutes and 0 seconds ago, saying: goes to box tonights pending orders. [Fri Aug 30 05:20:10 2002] |
| 19:37:10 | andersee | ibot: Seen BZFlag? |
| 19:49:16 | aaronl | andersee! |
| 19:49:26 | aaronl | i know i've asked you this before, but refresh my memory |
| 19:49:38 | aaronl | what's the best IDE i can run on my server? |
| 19:49:41 | aaronl | i'm running |
| 19:49:44 | aaronl | 2.4.19-pre5-ac3 |
| 19:49:58 | aaronl | and bugs in the oom killer keep bringing it down |
| 19:50:02 | aaronl | so i NEED to switch kernels |
| 19:51:39 | andersee | IDE? |
| 19:51:54 | andersee | You mean like VS Visual Studio .NET? |
| 19:52:20 | aaronl | uh no |
| 19:52:22 | andersee | Or like an ATA compliant hard drive? |
| 19:52:31 | aaronl | i mean the ATA thing that makes everyone go insane |
| 19:52:53 | aaronl | i would upgrade to the latest -ac but i've heard reports of disk corruption |
| 19:53:15 | andersee | So by best IDE, do you mean best kernel, or best hardware? |
| 19:54:26 | andersee | I am running 2.4.20-pre5 with a few (~30 or so) patches. I would studiously avoid the recent -ac kernels -- the new IDE code is pretty scary and (at elast for my system) misbehaves |
| 19:54:51 | andersee | s/elast/least/ |
| 20:04:07 | aaronl | best kernel |
| 20:04:16 | aaronl | or patch combination |
| 20:04:21 | aaronl | best ide subsystem after patches... |
| 20:06:38 | andersee | I suggest sticking with 2.4.20-pre5... |
| 20:07:27 | andersee | The newer -ac kernels have quite a lot of experimental IDE code... Really not 100% safe IMHO |
| 20:08:06 | aaronl | ah |
| 20:08:28 | aaronl | i have 200gb of ide in this box and i really want it to be safe |
| 20:08:33 | aaronl | 2.4.20-pre sounds like a good bet |
| 20:08:34 | aaronl | thanks |
| 20:08:39 | andersee | no probe |
| 20:08:51 | andersee | BTW, did you ever settle on a MP lib? |
| 20:09:13 | aaronl | we're almost set on BNlib |
| 20:09:21 | aaronl | apt-cache show bnlib1 |
| 20:09:27 | aaronl | but we haven't tested it yet |
| 20:09:29 | andersee | Looked at the one in libtomcrypt? http://libtomcrypt.sunsite.dk/ |
| 20:09:36 | andersee | Its freeware... |
| 20:09:45 | aaronl | no, is that the link you gave me a few days ago? |
| 20:10:00 | andersee | nope |
| 20:10:04 | andersee | A different one |
| 20:10:32 | andersee | I was looking at this (for the blowfish code) and saw it had the mp code as well. Looks nice and small. |
| 20:11:01 | aaronl | i'm almost considering architecting it to be mp-lib neutral |
| 20:11:15 | aaronl | there are so many libraries out there and different ones are better on different platforms |
| 20:11:28 | aaronl | there are very few MP-related functions that my code needs |
| 20:11:43 | aaronl | (creation, deletion, addition, multiplication, modular exponentiation...) |
| 20:11:43 | andersee | http://libtomcrypt.sunsite.dk/src/mpi.h |
| 20:11:52 | andersee | http://libtomcrypt.sunsite.dk/src/mpi.c |
| 20:12:03 | andersee | That header pretty much covers what it does |
| 20:14:06 | aaronl | running 'make oldconfig' on a marcelo kernel with a .config that comes from -ac shouldn't hurt, should it? |
| 20:14:28 | andersee | thats fine. |
| 20:14:42 | andersee | It'll just ask you a few extra questions. |
| 20:14:51 | aaronl | i ate lunch with a debian developer yesterday |
| 20:14:59 | aaronl | he was hacking on radeonfb |
| 20:15:02 | aaronl | i told him to get in touch with you |
| 20:15:07 | aaronl | he was very bitter at ATI |
| 20:15:14 | aaronl | the driver doesn't survive suspend/resume properly |
| 20:15:20 | aaronl | and they haven't been good about giving him specs |
| 20:15:36 | aaronl | they wanted him to sign an NDA, and when he mentioned that he needed to be able to GPL the code, they did not respond |
| 20:16:13 | andersee | grrr |
| 20:16:47 | andersee | They are also making a binary only module apparently for their upcoming radeon card... |
| 20:16:53 | andersee | grr |
| 20:17:01 | aaronl | argh |
| 20:17:09 | aaronl | binary only modules are a joke |
| 20:17:11 | andersee | Time to pull out 'objdump -d' |
| 20:17:13 | aaronl | even vmware has the source out |
| 20:17:27 | aaronl | who doesn't compile their own kernel? |
| 20:17:58 | andersee | Step one after an install == rebuild kerenel |
| 20:18:01 | aaronl | of course |
| 20:18:39 | aaronl | he mentioned something interesting too |
| 20:18:56 | aaronl | he'll be working on an 802.11a driver, for which the company is giving him specs |
| 20:19:12 | aaronl | but the card is completely configurable from software |
| 20:19:31 | aaronl | including the frequencies |
| 20:19:46 | aaronl | he may have to obfuscate some of the code (wtf?) to avoid the wrath of the FCC |
| 20:22:27 | andersee | hehe |
| 20:22:28 | andersee | That sounds pretty dang cool |
| 20:22:28 | andersee | run 801.11b at, say, 4 Ghz |
| 20:23:14 | aaronl | it's 802.11a |
| 20:23:24 | aaronl | wonders whether to chose eepro100 or e100 |
| 20:24:31 | andersee | dunno on that one |
| 20:26:28 | aaronl | bah |
| 20:26:37 | aaronl | they STILL havent fixed the "Fusion MPT" menu? |
| 20:26:46 | aaronl | nothing's happened when i selected that since, like, 2.4.17 |