irclog2html for #htc-blueangel on 20060228

00:11.02*** join/#htc-blueangel mindfunk (n=mindfunk@lala.ices.utexas.edu)
00:11.28mindfunkbhima: TimRiker told me you wanted a tuxscreen....
00:33.49*** join/#htc-blueangel dullard (n=jim@adsl-static-1-30.uklinux.net)
01:57.04mindfunk~botmail for bhima: msg me if you want a tuxscreen
01:57.28*** part/#htc-blueangel mindfunk (n=mindfunk@lala.ices.utexas.edu)
06:39.46*** join/#htc-blueangel jeanseb (i=jeanseb@ipv4.sabine-et-benoit.com)
06:39.48jeansebhello
06:49.24*** join/#htc-blueangel cr2 (n=konversa@crpl22.physik.uni-wuppertal.de)
08:23.49*** join/#htc-blueangel awelux__ (n=awelux__@dslb-084-058-168-134.pools.arcor-ip.net)
09:06.46*** join/#htc-blueangel rob_w (n=bob@p213.54.206.111.tisdip.tiscali.de)
09:06.46*** mode/#htc-blueangel [+o rob_w] by ChanServ
16:22.27*** join/#htc-blueangel Dat (n=datty@88.144.12.151)
18:13.06awelux__I still can't send privaate messages. cr2 are you here?
18:14.42cr2yes, but i need to buy some meals. till later :)
18:15.07awelux__hehe. cu
18:39.12*** join/#htc-blueangel dullard (n=jim@adsl-static-1-30.uklinux.net)
19:13.43cr2ok. i am back
19:15.23cr2the latest haret in files/haret dumps even more pxa27x info.
19:29.55*** join/#htc-blueangel cr2 (n=konversa@crpl22.physik.uni-wuppertal.de)
20:04.06awelux__I'm trying to add a memory check to the asmstuff part
20:06.54awelux__http://home.arcor.de/gu.de/haret/haret-gnu-awelux-200602271925.tar.gz
20:12.47cr2can't download.
20:13.04cr2can you clear the unnecessary bits in CKEN ?
20:13.15awelux__all in good time
20:13.39awelux__hmm the file should be readable
20:14.11awelux__but I can'T downlöoad either...
20:14.18cr2i have no connection at all.
20:15.11awelux__hmm. strange
20:15.35awelux__maybe  the page just jumped over the download limti
20:15.48cr2the directory haret is ok.
20:17.11cr2ok. downloaded
20:17.55awelux__I've no idea why but it took a few tries here..
20:23.49cr2btw. how big is vram ?
20:24.12awelux__240x480x2 at least. Maybe more.
20:24.18cr2256K
20:24.43cr2240x320x2 youe mean.
20:24.50awelux__uh yea
20:25.02cr2memPhysMap maps from 32K to 64K, imho.
20:25.31cr2maps guaranteed.
20:25.41cr2it _can_ be more.
20:26.47cr2your fb is 150K. not good ?
20:27.43awelux__I didn't check how much it is.
20:27.54awelux__it's enough
20:28.35cr2hmm. a proper ioremap with VirtualCopy should be cleaner...
20:31.18cr2CeSetThreadQuantum (GetCurrentThread (), 0);
20:31.23cr2can be enabled.
20:31.40cr2i commented it out because it did not work on wince2.11
20:31.48awelux__ok
20:36.48cr2i don't get the purpose of procb ?
20:36.58awelux__just for tests.
20:37.18awelux__I wanted to see if I can jump from proca to procb, pass parameters etc.
20:37.46awelux__I didn't remove anything so far
20:38.35cr2you read from the virtual address 0x1234. normally the lower 64K are not mapped. i don't think it is good.
20:39.29cr2ldr     r6, [sp, #+4]
20:39.33cr2looks funny :)
20:39.44awelux__what's the usual way for that ?
20:40.10awelux__I thought the ldr     r0, =0x00001234 is like r0 = 0x00001234
20:40.38awelux__ldr r0, 0x00001234 should be reading from that address
20:41.01awelux__maybe ldr     r0, =0x00001234 is a gnu asm shortcut
20:43.08cr2i always mix them. i will look at disasm.
20:44.37cr2yes. you are right.
20:45.05awelux__about it being a macro or what it does?
20:46.42cr2you load a number in a register. but this number is put behind the linux_preloader_end, that's why i was worried.
20:47.14awelux__ah. I don't use it from within the preloader
20:47.26cr2ldr     r0, [pc, #640]  ; 324 <linux_preloader_end+0xc>
20:47.39awelux__proca wass a _very_ basic test
20:47.51awelux__hu ?
20:47.54awelux__where is that ?
20:48.09cr2objdump. in the machine code.
20:48.22awelux__oh
20:48.28cr2<PROTECTED>
20:48.39awelux__so the = stuff is like an unnamed variable?
20:48.46cr2yes.
20:49.05awelux__can't use this in the preloader then
20:49.05cr2you can name it, and will have much more control.
20:49.55cr2i did
20:49.58cr2ldr     r6, myfb
20:50.08cr2myfb:           .long   0xa3faa000
20:50.09awelux__it can be convenient to place a value directly into the code
20:50.17cr2linux_preloader_end:
20:50.36awelux__I saw that and added a few of my own ;)
20:51.01cr2yes. but this constant is above linux_preloader_end
20:51.06cr2unlike the =
20:51.26awelux__since I write directly to a gpio the code is very magician specific
20:53.00cr2looks good. by i can't test it...
20:53.16awelux__doesn't matter.
20:53.38awelux__It's just there because I wantet to identify the problem I had
20:54.11awelux__if all is done and we don't know what to do we can write a fancy graphical bootloader
20:54.25cr2:) ok.
21:06.18awelux__AllocPhysMem is used to guarantee to allocate continous memory ?
21:06.37awelux__<PROTECTED>
21:06.37awelux__<PROTECTED>
21:07.02awelux__this should only work if the memory was allocated continously.
21:08.04cr2yes. this is the purpose of AllocPhysMem
21:08.39cr2i could have used LockPages, but for 6MB it is an overkill.
21:08.44cr2hmm.
21:09.02cr2you should check that the pointer is high enough.
21:09.07awelux__It did allocate continous memory for me
21:09.20awelux__And pointer was high enough. no overwriting
21:09.20cr2but with 32MB ramdisk it is not an issue.
23:15.32awelux__cr2, where did you get this mapping of cken bits to clocks from ?
23:18.48awelux__ah found it
23:22.10awelux__or not. A lot less seams to be on compared to dump pxa27x
23:22.43cr2i don't understand this opie touchscreen stuff :(
23:22.54cr2clear these bits before booting.
23:24.10awelux__uint32 *cken = (uint32 *)memPhysMap (0x41300004);
23:24.16awelux__*cken = 0 ?
23:24.24cr2no.
23:24.36cr2only the bits you don't need.
23:24.55cr2you need lcd, memclk, pwm.
23:24.59cr2something else ?
23:25.18cr2what is off ?
23:25.24cr2s/off/on
23:25.25awelux__I've to check. the last output of cken was obviously bogus
23:28.34awelux__Clock Enable: 00515ee9
23:29.46cr2look at dump pxa27x
23:29.56cr2last version
23:30.16awelux__It's hard to build on that. you didn't upload the code. right?
23:31.01awelux__I'm looking at the mapping from the pxa27x manual
23:31.17awelux__*pxa27x reference
23:33.35cr2i can upload the code
23:34.32awelux__not necessary. It's working well with the bits
23:34.48awelux__almost the same stuff on as yesturday
23:40.21awelux__on are: MEMCLK, SRAM, LCD, I2C, MMC, USBC, USBH,OSTIM, BTUART, FFUART, STUART, SSP2
23:41.21awelux__btw. if I have to use a certain mem location for a driver, is there another way than to load the kernel somewhere else?
23:42.05cr2switch off usb* *uart ssp2
23:42.14cr2for a driver ?
23:42.26awelux__I always switch off the display.
23:43.46cr2ok. i think memclk is necessary.
23:46.02awelux__<PROTECTED>
23:46.02awelux__<PROTECTED>
23:46.19awelux__thought this would switch off bit 10, USB
23:46.39cr2yes.
23:47.14awelux__using this the lcd switches off after a while
23:47.49awelux__and I've to coldboot
23:48.08awelux__I will switch off both USB clks now
23:48.09cr2usbc or usbh ? i don't understand the purpose of usbh.
23:50.11awelux__strange
23:50.39awelux__just playing a little with these clocks and I've to coldboot
23:52.50cr2you are playing with haret ? or after disabling interrupts ?
23:53.35awelux__hmm, strange:
23:53.45awelux__<PROTECTED>
23:53.45awelux__<PROTECTED>
23:53.52awelux__and things fail later.
23:53.59awelux__Maybe the problem is something else
23:57.47awelux__lol. maybe I've a magician with a defective memory ;)

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