irclog2html for blob on 2002.03.09

00:01:34Russ|werkSA-1110.h lays it all out pretty clear for a quick ref
00:02:13prpplaguelooks
00:08:36prpplagueRuss|werk: i don't have a SA-1110.h in my kernel, is that part of a patch?
00:09:11Russ|werker, try SA-1100.h
00:09:21prpplagueohh
00:09:50Russ|werkwriting an email filter is fun
00:10:06Russ|werk"there's a CR in the header field, outlook express chokes, mother it"
00:10:41Russ|werkalso, if a \r\n..\r\n crosses packet boundries between the .'s, outlook express will think the pop server signaled end of email
00:10:50Russ|werkand then you can't get your email
00:11:20Russ|werkboth, along with dozens of other DoS and buffer overflow exploits are out there with no fix
00:11:23prpplaguedoh
00:15:47prpplagueRuss|werk: can you explain something about the SA-1100.h , the only gpio's associated with irq's listed there are 0-10, but erikm said 11-27
00:17:03BZFlagerikm: yep. thanx. /me feels like a hardware designer.
00:18:04prpplagueBZFlag: mind if i ask what spawned you question?
00:18:09BZFlag11-27 all share one irq bit.
00:18:29BZFlagvercel http://www.vercel.com/ hardware folk asking me questions.
00:18:46BZFlagright now they have power buttons mapped to gpio26. this is not a good idea.
00:19:00erikmBZFlag: definatively not
00:19:07BZFlagpower button(s) should map to some non-shared irq.
00:19:32BZFlagso I told em so. ;-)
00:19:42prpplagueBZFlag: else events on the other gpios could be interepted as a power down?
00:19:52BZFlagerikm: want to review the schematics under a verbal NDA for me?
00:20:01BZFlagprpplague: exactly.
00:20:22BZFlagprpplague: well... actually it's the power up I'm worried about.
00:20:30prpplagueBZFlag: mind if i write up a note on that?
00:20:43BZFlagknock yourself out.
00:20:55BZFlaggrins
00:21:28erikmBZFlag: actually... I want to go to sleep...
00:21:31prpplagueBZFlag: you'd be surprised how simple questions like that can be hard to find docs on
00:22:13BZFlagerikm: I could mail em to ya to review when you get a chance.
00:23:25Russ|werkBZFlag: I'll do a quick review if you like
00:24:00BZFlagRuss|werk: cool. I copy you.
00:24:01erikmBZFlag: ok, send them by email. I'll have a look at them on sunday
00:24:21BZFlagemail will have some known issues that are being changed in the next rev.
00:24:31BZFlagI expect a new rev in less than a week.
00:25:15BZFlag"I copy you" wow. I guess I can't think and type. I meant "I'll copy you" which you probably guessed.
00:25:29Russ|werkprpplague: sorry, I meant arch-sa1100/irqs.h
00:26:08BZFlagRuss|werk, erikm: either of you have SD/MMC designs? If so how did you attach them?
00:27:06prpplagueRuss|werk: there is also info there in the SA-1100.h file
00:27:38BZFlagThey are thinking of using the SPI mode on the uart. hmm.. 0 or 1? checking...
00:28:17BZFlag1 iirc.
00:28:56BZFlagnope. I was wrong. it's 0 =(
00:30:03prpplagueRuss|werk: from SA-1100.h - #define IC_GPIO11_270x00000800/* GPIO [11:27]
00:30:29Russ|werkah, well, irqs.h is another good source for info anyway
00:30:43Russ|werkBZFlag: what verbal NDA do you want?
00:30:50erikmBZFlag: what do you mean with SD design?
00:31:30BZFlagRuss|werk: you need to say "I won't pass this on to anyone without asking" ;-)
00:31:52Russ|werkok, I will not pass this on to anyone without asking
00:31:52BZFlagsa11xx systems with a SD/MMC slot.
00:32:03BZFlagRuss|werk: that will do. ;-)
00:32:33Russ|werkoops, evolution has a vunerability in common with outlook
00:33:04erikmBZFlag: ah, right. IIRC Steve Wiseman's balloon2 board has that
00:33:14Russ|werkoh, nm, my asu.edu email servers fixed the email for me
00:34:18Russ|werka incorrect fix, but fixed none the less
00:34:20erikmgoes zzz
00:34:22erikmbye
00:34:28prpplaguebye
00:46:47prpplaguehttp://www.abcsinc.com/small-linux/gpio-irqs.txt
00:46:55prpplagueBZFlag: check that out
00:47:06prpplagueRuss|werk: you too
00:47:22prpplaguesee if there is anything i need to add or is incorrect
00:50:41prpplagueRuss|werk: looks like everything is set on the camion, it should go out monday or tuesday
00:52:35Russ|werkcool
00:52:45Russ|werkhmm...something is eating my memory...
00:54:54prpplagueRuss|werk: parana app?
00:55:23prpplagueRuss|werk: you look over the gpio doc?
00:56:59Russ|werkit was galeon, one sec
00:59:44Russ|werkyou might mention something briefly about GPLR, GPDR, GPSR, GPCR, and GAFR
01:00:06prpplaguewill do
01:00:11Russ|werkand how for each GPIO there is a bit mask, a number, and an IRQ number associated with it
01:01:03prpplaguehmm, your right, i didn't explain that well
01:19:44Russ|werkalso, gives gpio 11-27 an chained irq number
01:19:53Russ|werks/ gives/linux gives/
01:20:03Russ|werk#defineIRQ_GPIO1132
01:20:03Russ|werk#defineIRQ_GPIO1233
01:20:03Russ|werk...
01:20:27prpplagueyes i saw that, but i'm not sure how that is used
01:20:34Russ|werkwhen you register it
01:20:46Russ|werkthere are a lot of steps you go though
01:21:12Russ|werkGFAR, GPDR, set the edge, register the IRQ (I think thats right)
01:21:58prpplagueGFER and GPDR, ok
01:22:00Russ|werkGFAR and GPDR take the mask, setting the edge might take the gpio or the irq, I forget which, and registering it takes the irq number
01:22:36prpplagueok i'm with ya
01:23:19Russ|werkrequest_irq and set_GPIO_IRQ_edge
01:23:57Russ|werk(see arch/arm/mach-sa1100/irq.c)
01:28:04prpplagueRuss|werk: oh i see now, each GPIO has its own irq defined however, irq_gpio11 through irq_gpio27 only triger irq_gpio11_27
01:28:48Russ|werkright, but it gets propogated to the right irq handler
01:29:13prpplagueRuss|werk: gotcha
01:29:34Russ|werksa1100_GPIO11_27_demux()
01:33:04prpplagueso after resceiving a irq_gpio11_27 the demux() polls each of the 11-27 gpios?
01:40:48prpplaguewb
01:40:58prpplaguedid you get my last question?
01:44:19Russ|werktake a quick look at arch/arm/mach-sa1100/irq.c
01:45:04prpplagueRuss|werk: ya thats where i'm looking
01:46:08prpplaguefor (i = 11; i <= 27; ++i) {
01:46:08prpplagueif (irq & (1<<i)) {
01:46:08prpplaguedo_IRQ(IRQ_GPIO11 + i - 11, regs);
01:46:08prpplague}
01:46:08prpplague}
01:52:17russ_just goes through, and sees what irq's need servicing
01:53:08prpplaguethats what i thought
02:02:23Test Failed!
02:02:23BZFlagibot test
02:02:26Test Successful!
02:02:26BZFlagibot test
02:02:29BZFlagheh
02:03:27Russ|werk: excuse me?
02:03:27Russ|werkibot: go test yourself
02:03:40Russ|werkbrb
02:06:11prpplagueRuss|werk: http://www.abcsinc.com/small-linux/gpio-irqs.txt
02:06:30prpplagueRuss|werk: i'll work on the control registers info a little more tomorrow
02:06:48prpplagueRuss|werk: but i think i got things strait
02:07:12prpplagueBZFlag: you welcome to add any additional info too
02:09:22prpplagueruss_: having problems?
02:11:52russ_testing stuff
02:11:59prpplagueohh
02:12:24prpplagueruss_: well thanks for the 101 clas
02:12:27prpplagueclas
02:12:29russ_brb again
02:12:48prpplagueBZFlag: and thanks for the question
02:15:22prpplagueruss__: holy cow its russ, long time no see
02:17:45russ_last time
02:18:43prpplagueround n round he goes, where he'll no body knows
02:18:57prpplagueargh too much coffee
02:20:47russ_or not...going to just disconnect for a bit to avoid bugging people
02:21:12prpplagueohh well
02:23:06prpplagueBZFlag: i'm out of here, thanks again, i know a guru like you don't need them much, but i've start collecting notes here - http://www.abcsinc.com/small-linux/notes.htm
02:26:32BZFlagnice, I'll browse sometime.
02:31:29prpplaguelater
06:23:32RussBZFlag: can you put that pdf up again?
06:37:33Russer, nm
06:45:01prpplagueRuss: should you be out at some college party or something?
07:14:32Russwatching movie with girl
07:15:29prpplaguecat attention > /dev/gf
07:36:28prpplaguewb BZFlag

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with infobot logs, split per channel and by date, etc.