| 00:04.01 | ds2 | is that the 68K based one? |
| 00:18.56 | flyback- | <CosmicPenguin> In their defense, broadcom is good about letting people write drivers under NDA and then open sourcing them |
| 00:31.33 | *** join/#edev g1powermac (n=g1powerm@adsl-18-70-57.sdf.bellsouth.net) |
| 00:38.04 | *** join/#edev FireEgl (i=FireEgl@4.0.0.0.1.0.0.0.c.d.4.8.0.c.5.0.1.0.0.2.ip6.arpa) |
| 00:41.31 | Exiles | nah toss the card with it, get a 133 card |
| 00:42.27 | flyback- | no way |
| 00:55.33 | flyback- | http://video.google.com/videoplay?docid=614877293652284697&q=cat+5+cable&total=57&start=0&num=10&so=0&type=search&plindex=9 |
| 01:06.09 | flyback- | mostly worthless k6-2 boxes with buggy boards |
| 01:06.25 | flyback- | 120 volts for the 4th should be nice |
| 01:27.35 | *** join/#edev CosmicPenguin (n=nobody@www.crouse-house.com) |
| 01:27.35 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 01:49.35 | *** part/#edev joe_bleau (n=joe_blea@adsl-68-90-143-106.dsl.fyvlar.swbell.net) |
| 02:45.02 | *** join/#edev joe_bleau (n=joe_blea@adsl-69-152-222-9.dsl.fyvlar.swbell.net) |
| 03:37.46 | *** join/#edev DrZimmerman (n=theo@2001:5c0:8fff:fffe:0:0:0:6af1) |
| 05:20.22 | *** join/#edev pierrelux (n=pierre-l@144-125.sh.cgocable.ca) |
| 05:27.35 | *** join/#edev redblue (i=star@ppp027.108-253-207.mtl.mt.videotron.ca) |
| 06:33.13 | *** part/#edev LyosNorezel (n=LyosNore@unaffiliated/lyosnorezel) |
| 08:26.27 | *** join/#edev SpaceMonkey (n=andrew@ovation1.gotadsl.co.uk) |
| 08:28.46 | *** join/#edev steliosk (n=Stelios@ipa226.211.tellas.gr) |
| 11:20.24 | *** join/#edev redblue (i=star@ppp137.108-253-207.mtl.mt.videotron.ca) |
| 12:09.04 | *** join/#edev dijenerate (n=dijenera@72.22.153.43) |
| 12:16.20 | *** join/#edev GPSFan (n=kenm@12.10.255.248) |
| 12:16.20 | *** mode/#edev [+o GPSFan] by ChanServ |
| 13:32.37 | *** join/#edev T0mW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 13:59.13 | *** join/#edev g1powermac (n=g1powerm@adsl-18-70-57.sdf.bellsouth.net) |
| 14:00.07 | *** join/#edev prpplague (n=billybob@mail.americanmicrosystems.com) |
| 14:02.46 | prpplague | davidc__: ping |
| 14:07.30 | *** mode/#edev [+o prpplague] by ChanServ |
| 14:07.44 | *** mode/#edev [+b *!*n=nope@c-24-3-31-106.hsd1.pa.comcast.net] by prpplague |
| 14:07.44 | *** kick/#edev [flyback-!n=billybob@mail.americanmicrosystems.com] by prpplague (prpplague) |
| 14:16.18 | wmat | prpplague: heh |
| 14:17.37 | prpplague | hmm |
| 14:17.57 | prpplague | any kernel people awake? |
| 14:20.12 | wmat | prpplague: you still having compile problems? |
| 14:21.07 | prpplague | wmat: naw got that taken care of |
| 14:21.25 | prpplague | wmat: run across a function that i can't see to locate info on |
| 14:21.35 | prpplague | wmat: xchg(ptr,int); |
| 14:27.04 | g1powermac | prpplague, I'm so tempted to do that on ##microcontrollers, as he's idling there too. . . |
| 14:27.19 | g1powermac | but don't think the other ops would like that without him doing something bad first. . . |
| 14:27.46 | prpplague | g1powermac: hehe, yea, well he's had his chances |
| 14:33.03 | wmat | prpplague: what do you need to know about it? |
| 14:34.02 | prpplague | wmat: what it does |
| 14:34.24 | prpplague | wmat: i was unable to find it in my ldd book or a man page online |
| 14:34.58 | wmat | prpplague: i'd guess it exchanges 2 parameters, probably (dest,src) |
| 14:35.55 | prpplague | wmat: that would have been my guess, but the parameters don't match |
| 14:35.57 | wmat | like the assembly version |
| 14:36.24 | prpplague | wmat: xchg(&test->value, 0); |
| 14:43.04 | T0mW | prpplague: isn't it defined in linux/include/asm-{ARCH}/system.h ? |
| 14:44.43 | *** join/#edev CosmicPe1guin (n=nobody@crank.laptop.org) |
| 14:44.49 | prpplague | T0mW: yea, but it doesn't tell me much |
| 14:44.50 | prpplague | #define xchg(ptr,x) \ |
| 14:44.50 | prpplague | <PROTECTED> |
| 14:45.04 | T0mW | yeah, that one |
| 14:45.39 | T0mW | heh: find . -type f | xargs fgrep -w __xchg | grep -v = |
| 14:45.46 | prpplague | CosmicPenguin: morning |
| 14:45.48 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 14:45.56 | prpplague | T0mW: way ahead of you |
| 14:45.57 | CosmicPenguin | greets |
| 14:46.02 | *** join/#edev CP|too (n=nobody@www.crouse-house.com) |
| 14:46.02 | *** mode/#edev [+o CP|too] by ChanServ |
| 14:46.03 | prpplague | CosmicPenguin: maybe you know |
| 14:46.06 | T0mW | :-D heh |
| 14:46.16 | CosmicPenguin | famous last words |
| 14:46.17 | prpplague | CosmicPenguin: xchg(ptr,int); |
| 14:46.25 | prpplague | CosmicPenguin: what the heck does that do? |
| 14:46.27 | T0mW | prpplague: then again, maybe he doesn't |
| 14:46.36 | prpplague | hehe |
| 14:47.02 | CosmicPenguin | best guess it emulates the x86 XCHG opcode |
| 14:47.08 | wmat | heh |
| 14:47.21 | CosmicPenguin | puts the int into *ptr and whatever is in *ptr into int |
| 14:47.40 | prpplague | CosmicPenguin: ok, problem is , i don't see it being used like that |
| 14:47.47 | T0mW | sounds logical to me too |
| 14:47.59 | prpplague | CosmicPenguin: its being use like this: xchg(&test->value,0); |
| 14:48.19 | CosmicPenguin | gimme some context |
| 14:49.02 | prpplague | CosmicPenguin: its being used in a pxa sdio driver to mark the controller as busy |
| 14:49.17 | prpplague | CosmicPenguin: xchg(&ctrllr->busy,0); |
| 14:49.22 | prpplague | or |
| 14:49.25 | prpplague | CosmicPenguin: xchg(&ctrllr->busy,1); |
| 14:49.27 | CosmicPenguin | no yeah - definately |
| 14:49.29 | CosmicPenguin | its an atomic set |
| 14:50.34 | CosmicPenguin | http://lxr.linux.no/source/include/asm-arm/system.h#L271 |
| 14:50.48 | prpplague | CosmicPenguin: pretend i'm wmat and give me a 30 second explanation |
| 14:52.10 | CosmicPenguin | okay - well lets look at the x86 behavior, because thats a good example |
| 14:52.15 | CosmicPenguin | and thats where this came from |
| 14:52.21 | T0mW | prpplague: __xchg is defined in linux/include/asm-{arch}/system.h, the second parm is used in a switch() statement |
| 14:52.37 | CosmicPenguin | xchg basically exchanges a register or memory with another register |
| 14:52.50 | prpplague | T0mW: ahh thats what was confusing me |
| 14:52.51 | CosmicPenguin | so say, xchg ax,bx would be an atomic exchange of registers |
| 14:53.07 | T0mW | prpplague: yeah, it is not the src/dst, merely an action modifier |
| 14:53.11 | CosmicPenguin | But if you happen to use a memory operand, something ineteresting happens |
| 14:53.37 | CosmicPenguin | the bus getes locked while the exchange happens |
| 14:53.45 | CosmicPenguin | regardless of the state of the machine |
| 14:53.47 | prpplague | gotcha |
| 14:54.00 | CosmicPenguin | so it turns out its an excellent way to atomically set a bit in memory |
| 14:54.03 | prpplague | so the contents don't get changed while the swap is occuring |
| 14:54.06 | CosmicPenguin | thats what you're doing |
| 14:54.08 | T0mW | from what I read, if you pass it a size of 0, it will return the value of the pointer |
| 14:54.09 | *** join/#edev SpaceMonkey (n=andrew@ovation1.gotadsl.co.uk) |
| 14:55.02 | CosmicPenguin | the switch in __xchg is on the third param, not the second |
| 14:56.04 | T0mW | nope, I am wrong |
| 14:56.10 | T0mW | CosmicPenguin: yes |
| 14:56.42 | prpplague | <PROTECTED> |
| 14:56.49 | prpplague | #define xchg(ptr,x) \ |
| 14:56.52 | prpplague | <PROTECTED> |
| 14:57.50 | T0mW | CosmicPenguin: oh. So that would be for a set & test type approach? |
| 14:58.03 | CosmicPenguin | I think just for a set |
| 14:58.11 | CosmicPenguin | set and test on the x86 is cmpxchg |
| 14:58.11 | T0mW | "set and see if was set" |
| 14:58.22 | T0mW | ok |
| 14:58.39 | CosmicPenguin | the arm version of __xchg is somewhat more complicated |
| 14:58.50 | CosmicPenguin | leading me to believe that probably this was an x86ism that was later ported |
| 14:59.11 | T0mW | interesting, there is something to be learned on #edev, other than where the State Department has Donna this week. heh |
| 14:59.32 | prpplague | CosmicPenguin: i'm a little confused though, for the example in the sdio code, they are passing a fixed value, not a location |
| 15:00.04 | prpplague | CosmicPenguin: how can you exchange a static value |
| 15:01.34 | prpplague | CosmicPenguin: nm, i get it now |
| 15:02.04 | prpplague | CosmicPenguin: the value gets put in a registers first, which is then swapped |
| 15:02.54 | chouimat | morning |
| 15:03.35 | prpplague | chouimat: greetings |
| 15:04.37 | prpplague | CosmicPenguin: so basically its a method of setting a value in the ptr with the bus locked |
| 15:41.14 | *** join/#edev pierrelux (n=pierre-l@144-125.sh.cgocable.ca) |
| 16:28.30 | *** join/#edev lyakh (n=lyakh@dynamic-unidsl-85-197-31-110.westend.de) |
| 16:51.11 | davidc__ | prpplague, pong |
| 16:59.33 | *** join/#edev g1powermac_ (n=g1powerm@adsl-147-129-72.sdf.bellsouth.net) |
| 17:12.55 | prpplague | davidc__: those nokia style lcd's |
| 17:13.07 | davidc__ | mmhmm? |
| 17:13.07 | prpplague | davidc__: you guys have a pretty stable/large source of those? |
| 17:13.15 | davidc__ | somewhat6 |
| 17:13.26 | davidc__ | the model we end up with tends to keep changing |
| 17:13.32 | davidc__ | as they're just sold as bulk nokia replacements |
| 17:14.01 | prpplague | davidc__: but the physcial form factor is the same? |
| 17:14.08 | davidc__ | yup |
| 17:14.21 | davidc__ | prpplague, but sometimes you need to tweak contrast / etc values |
| 17:14.56 | prpplague | davidc__: ahh ok, well, the boss is going to do some plastics using that lcd |
| 17:15.12 | prpplague | just wanted to make sure they'd be around for a little while |
| 17:15.33 | davidc__ | prpplague, we plan to continue stocking them, but its all sourced in china, and those guys do things their own way ;) |
| 17:15.44 | prpplague | hehe yea |
| 17:15.57 | prpplague | davidc__: hows that psp display? |
| 17:16.08 | davidc__ | prpplague, its gorgeous - seen the pix of it running? |
| 17:16.14 | prpplague | davidc__: yea |
| 17:16.23 | prpplague | davidc__: what about the OLED one |
| 17:16.41 | prpplague | davidc__: that one going to be in stock for awhile? |
| 17:17.28 | davidc__ | we're having some stocking issues on it at the moment :P |
| 17:17.44 | prpplague | ahh |
| 17:18.30 | prpplague | davidc__: well that nokia one looks pretty sweet for a project we want to post |
| 17:18.44 | davidc__ | cool - fire me a link to it when its up |
| 17:18.53 | prpplague | davidc__: will do |
| 17:19.12 | prpplague | davidc__: the prototype plastics should be in next week |
| 17:19.28 | davidc__ | cool. |
| 17:20.16 | prpplague | ~seen mallum |
| 17:20.37 | purl | mallum <n=mallum@host81-156-120-160.range81-156.btcentralplus.com> was last seen on IRC in channel #oe, 62d 2h 2m 53s ago, saying: 'hey koen'. |
| 17:39.10 | *** join/#edev dfarnsworth (n=dfarnswo@65.200.49.156) |
| 17:40.31 | ds2 | morning |
| 18:09.19 | *** join/#edev g1powermac__ (n=g1powerm@adsl-147-129-138.sdf.bellsouth.net) |
| 18:46.18 | *** join/#edev g1powermac_ (n=g1powerm@adsl-147-129-138.sdf.bellsouth.net) |
| 18:59.42 | *** join/#edev redb1ue (i=star@ppp122.108-253-207.mtl.mt.videotron.ca) |
| 20:01.33 | *** join/#edev g1powermac_ (n=g1powerm@adsl-18-84-243.sdf.bellsouth.net) |
| 20:10.42 | *** join/#edev lyakh (n=lyakh@dynamic-unidsl-85-197-31-110.westend.de) |
| 21:20.28 | *** join/#edev Crofton__ (n=balister@jhr-476-5192.mprg.ee.vt.edu) [NETSPLIT VICTIM] |
| 21:20.28 | *** join/#edev MonMotha (i=monmotha@c-69-245-184-142.hsd1.in.comcast.net) [NETSPLIT VICTIM] |
| 21:20.28 | *** join/#edev davidc__ (n=davidc@sta-204-188-103-98.rockynet.com) [NETSPLIT VICTIM] |
| 21:20.28 | *** join/#edev ccjoe (n=ccjoe@c-67-177-42-240.hsd1.ut.comcast.net) [NETSPLIT VICTIM] |
| 22:36.44 | *** join/#edev dijenerate (n=dijenera@72.22.153.43) |
| 23:18.03 | ds2 | Ob complaint - wtf can't does an assignment of 2 to a single bit variable result in the variable being set to a 1...@#$!@#!@#@ |
| 23:25.35 | ds2 | s/can't/ |
| 23:25.40 | ds2 | <PROTECTED> |