| 00:18.50 | *** join/#edev gandhijee_ (~akp@50.12.169.99) |
| 00:44.32 | *** join/#edev prpplague (~prpplague@ppp-70-249-149-70.dsl.rcsntx.swbell.net) |
| 01:26.17 | *** join/#edev dijenerate (~dijenerat@173.225.251.175) |
| 02:09.32 | *** join/#edev djerome (~djerome@ip68-2-20-108.ph.ph.cox.net) |
| 03:59.22 | *** join/#edev risca (~risca@wnpgmb0903w-ds01-249-233.dynamic.mtsallstream.net) |
| 04:37.50 | *** join/#edev mbuf (user@nat/redhat/x-edjvndufssfmartx) |
| 05:11.04 | *** join/#edev nullpuppy (~nullpuppy@freematrix/staff/nullpuppy) |
| 06:19.07 | *** join/#edev dijenerate (~dijenerat@173.225.251.175) |
| 07:09.42 | *** join/#edev calculu5 (~calculus@gentoo/user/calculus) |
| 07:20.11 | *** join/#edev amarsman (~marsman@90-145-17-249.wxdsl.nl) |
| 07:23.25 | *** join/#edev JoeLlama (~snork@unaffiliated/joellama) |
| 07:24.51 | *** join/#edev sjhill (~sjhill@home.bethel-hill.org) |
| 07:24.51 | *** mode/#edev [+o sjhill] by ChanServ |
| 07:53.40 | *** join/#edev amarsman (~marsman@90-145-17-249.wxdsl.nl) |
| 08:00.53 | *** join/#edev clio (~andrej@85.159.109.222) |
| 09:47.13 | *** join/#edev florian (~fuchs@port-217-146-132-69.static.qsc.de) |
| 09:47.13 | *** join/#edev florian (~fuchs@Maemo/community/contributor/florian) |
| 09:52.46 | *** join/#edev g1powermac (~g1powerma@rrcs-173-197-150-24.west.biz.rr.com) |
| 09:52.52 | *** join/#edev g1powermac (~g1powerma@unaffiliated/g1powermac) |
| 11:41.18 | *** join/#edev sledges (~sledges@unaffiliated/sledges) |
| 11:41.26 | sledges | hello |
| 11:41.32 | sledges | which Freescale kernel for imx5X boards would you find most stable and most covered in working peripheral support? |
| 11:42.18 | sledges | (including add-on LCD+Touchscreen) |
| 11:42.34 | sledges | got mad among numerous half-patched trees aroudn |
| 12:43.37 | av500 | sledges: |
| 12:43.40 | av500 | [13:37:57] <KotH> av500: tell him to use the LTIB kernel |
| 12:43.42 | av500 | [13:38:04] <KotH> av500: it has all the patches he needs |
| 12:43.43 | av500 | [13:38:18] <KotH> av500: if he's luck, then a few of those patches were cleaned up and send upstream |
| 12:43.45 | av500 | [13:38:25] <KotH> av500: but he shouldnt count on it |
| 13:36.59 | sledges | av500, thank you, but I used ltib 0911 with latest 1211 patches and the kernel wouldn't even boot |
| 13:37.24 | av500 | well, I am just the relay |
| 13:37.35 | av500 | and not I am happy not having to deal with FSC :) |
| 13:37.38 | av500 | and nwo* |
| 13:37.40 | av500 | now* |
| 13:47.11 | *** join/#edev Guest6973 (simoirc@109.73.162.119) |
| 14:09.36 | *** join/#edev GPSFan (~kenm@64.92.145.112) |
| 14:09.36 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:11.04 | *** join/#edev roxfan (dunno@91.180.76.196) |
| 14:12.45 | *** join/#edev dijenerate (~dijenerat@173.225.251.175) |
| 14:31.50 | *** join/#edev prpplague^2 (~danders@192.91.66.186) |
| 14:37.47 | *** join/#edev prp^2 (~prpplague@ppp-70-249-149-70.dsl.rcsntx.swbell.net) |
| 15:22.12 | *** join/#edev bgamari (~ben@physicsnat56.physics.umass.edu) |
| 15:31.34 | *** join/#edev risca (~risca@wi-secure-6007.cc.umanitoba.ca) |
| 16:14.02 | *** join/#edev sndcrb (~sndcrb@dsl-hkibrasgw2-fef8de00-126.dhcp.inet.fi) |
| 16:41.09 | *** join/#edev landley (~landley@140.242.26.2) |
| 16:41.09 | *** mode/#edev [+o landley] by ChanServ |
| 16:41.24 | landley | Fedora 16 is kind of appalling. |
| 16:41.34 | landley | I've been trying to install it under qemu since yesterday. |
| 16:41.42 | landley | The number of ways it's found to fail is kind of ingenious. |
| 17:03.21 | sledges | av500, apparently I was wrong. I was using another uboot (also for imx5) but from other (numerous) trees. apparently you have to use uboot which comes together with the given kernel, otherwise badness ensues |
| 17:03.58 | sledges | so latest ltib looks like it is the way to go (touchscreen works finally, only framebuffer issues appear out of nowhere, but hope to deal with them soon) |
| 17:07.32 | landley | sledges: uboot sometimes does some of the early hardware setup, and Linux leaves it that way. |
| 17:07.44 | landley | DRAM controller init, for example... |
| 17:07.56 | sledges | landley, that's what I reckoned |
| 17:08.24 | sledges | but that other uboot I had worked perfectly for that other kernel, which supported all things needed |
| 17:08.42 | sledges | so probably it boils down on how you set the hardware up, and then later how the kernel uses it (or just ignores) |
| 17:08.59 | av500 | landley: dram is mostly in uboot, but all the pinmux seems to be split between bootloader and kernel in random ways |
| 17:09.19 | av500 | as in everybody does it differently |
| 17:09.40 | sledges | it was just unbelievable, that one kernel did not work, and instead just spat out UART boot console with garbled characters half-way into boot, and then would refuse to boot from any media (whereas the other kernel would perform fine) |
| 17:10.05 | sledges | av500, correct you are I sense |
| 17:10.20 | landley | sledges: kernel command line may have speed setting in console= or not. It's optional. |
| 17:10.43 | landley | If the kernels had built-in command lines (instead of getting them from u-boot), then one could have it and the other accept whatever speed the uart was currently set to... |
| 17:10.57 | landley | Or one uboot could send one command line and the other could send another... |
| 17:11.06 | landley | Again: you have to tell the system as a whole to do the right thing. |
| 17:11.20 | sledges | landley, I always had that set, and it would be picked up from uboot (i double check one of the first kernel boot lines) |
| 17:11.39 | sledges | landley, it was more weird, as the kernel would log boot console normally |
| 17:12.16 | sledges | and then would change to garbled text mid-way through _kernel's_ boot (probably not that the baud rate changes, cause I poked around in minicom through different baudrates), but some additional UART init would screw things |
| 17:12.55 | sledges | all delicate stuff, definitely, hence lessons learned |
| 17:13.01 | landley | IRQ initialization. |
| 17:13.20 | landley | Early printk is polling based. IRQ init happens just before launching PID 1. |
| 17:13.55 | landley | So something was horking your serial interrupt, which could be a GPIO depending on how your board was set up... |
| 17:15.09 | av500 | nod |
| 17:15.10 | sledges | but there ya go, I chucked the came-along uboot, and problems solved! |
| 17:15.26 | av500 | way too much stuff done in boot loaders |
| 17:15.42 | av500 | they should just setup ram and start the kernel |
| 17:26.32 | florian | ack |
| 17:27.16 | landley | There was an ELC video about teaching the kernel to do DRAM init... |
| 17:28.39 | landley | http://www.embeddedlinuxconference.com/elc_2010/sessions.html#Ungerer |
| 17:28.56 | landley | The video and slides are at http://free-electrons.com/blog/elc-2010-videos/ |
| 17:29.27 | av500 | landley: with e.g. TI you can add configuration headers |
| 17:29.43 | av500 | the rom code parses then and setus up sdram, then loads the kernel and starts it |
| 17:29.52 | av500 | that way you can boot a kernel directly |
| 17:30.09 | av500 | it just asks for the "raw" mode instead of "fat" mode |
| 17:31.03 | av500 | I mean, it only works in raw mode like that |
| 18:03.53 | *** join/#edev florian (~fuchs@sign-4d09191c.pool.mediaWays.net) |
| 18:03.53 | *** join/#edev florian (~fuchs@Maemo/community/contributor/florian) |
| 18:07.09 | *** join/#edev djerome (~djerome@ip68-2-20-108.ph.ph.cox.net) |
| 18:20.00 | *** join/#edev risca (~risca@wi-secure-5971.cc.umanitoba.ca) |
| 19:29.47 | *** join/#edev risca (~risca@wi-secure-5971.cc.umanitoba.ca) |
| 19:29.54 | *** join/#edev litejk (~litejk@80.202.230.199) |
| 19:29.58 | *** part/#edev litejk (~litejk@80.202.230.199) |
| 20:18.38 | *** join/#edev JoeLlama (~snork@unaffiliated/joellama) |
| 21:10.43 | *** join/#edev mrcan (~mrcan@unaffiliated/mrcan) |
| 21:17.16 | *** join/#edev JoeMoose (~snork@cpe-75-85-32-65.socal.res.rr.com) |
| 21:17.18 | *** join/#edev JoeMoose (~snork@unaffiliated/joellama) |
| 21:26.33 | *** join/#edev JoeMoose (~snork@cpe-75-85-45-219.socal.res.rr.com) |
| 21:26.35 | *** join/#edev JoeMoose (~snork@unaffiliated/joellama) |
| 21:48.42 | *** join/#edev rsFF (~rsFF@bl19-20-130.dsl.telepac.pt) |
| 22:36.21 | *** join/#edev bgamari (~ben@72.19.94.241) |
| 22:56.28 | *** join/#edev djerome (~djerome@ip68-2-20-108.ph.ph.cox.net) |
| 23:02.40 | *** join/#edev bgamari (~ben@physicsnat56.physics.umass.edu) |
| 23:58.14 | *** join/#edev bgamari (~ben@physicsnat56.physics.umass.edu) |