Closed steeve closed 9 years ago
I should add that I've checked for doubles in case they were already there under a different name with shasum
.
$ shasum * | sort
052a1caa49ae698c009e16d529ff24d5792e3046 meson8b_m202_512M.dtd
09d8697e68f4486418c676ea43c5260eebc29a2d meson8_k200a_2G.dtd
0ece1569c94fa091226a5591e10e2d25cc07387d meson8_k200a_1G.dtd
1f9ff125c14ffc1295292b7603facbb6a7fa13cd Makefile
259a2a9464f5c0a18b6b1c62092dbdcf836b9a90 meson6tvc_h15.dtd
25a144a0fd63fc4e8b126c3a33264bd9c65283c9 meson8m2_pxp.dtd
2a75d06a4c5db7a0303e620d953ad5b05238c04f meson6_g33_1G.dtd
306804631b35f1c947b1e1277557064eb99e5557 meson6tvd_ref_512m.dtd
35d8f6b5e30abf1c8b4d5f2b7755d3af8be2cc55 meson8_k200b_1G_emmc_sdio.dtd
36d8a4074b76278e9353857520ceab76b3ef041c meson8_k101_512M.dtd
36f16d2cb64ac07799bca7de99928726fd0bb2ba aml_top.dtd
38cadb3dd691f38e33e9a9b4dd4c0546188341aa meson8m2_n100_1G.dtd
39ecba199f7a7d8a821641824ee3beaed19f5926 meson6tvc_h15_512m.dtd
3c1dc532ed3497dec2521d675bbb514b30420968 meson8_m82_1G.dtd <==================
3d0c363fa06f47dedb0d9d0f819f4fb4510e5cff meson6_g18_512M.dtd
4090f09a998aedbcf6a8190207a4e1f3540fb31d meson8_k200b_2G.dtd
46a71061767081c46e5f6af765fdb313dbe5e589 meson8_k200a_2G_4k2k.dtd
4cb1a12d3551c7906bce7d9c08e10584b9e0b8b7 meson8_k160_v1.dtd
5136fb42f4e72000a15e4b42a6f9760ceef84b93 meson6tv_ref.dtd
592d88d4bf7897cba7635fa3065f899fa57cbc26 meson6tvd_ref.dtd
5c36a43dd1e2ed3747c7fdd9e4c51198b305a862 meson8_k200b_1G_emmc_sdhc.dtd
5e4175919884ed07c2454f24b3c45dcc55a02910 meson8_k100_1G.dtd
5fb6c0be4b62edc4949af00d51af7a7ec04d30e7 meson8b_m200_1G.dtd
5fd731f511cac7b5a0bfe5b31b404180e7748899 meson8_k100_2G.dtd
6dfd0a721ebaf8de13403f9b5d16530d193ccffb meson8_k150_v1.dtd
6e48e9e808846c692d071558d97efb171db40b27 meson6_g33_1212.dtd
6fccea9207622e31e14563e6fdbfaa84e1dfa9f4 meson8b_pxp.dtd
70d8467629759272037d86ed1d55db024bf60e6b meson8_k101_1G.dtd
74a97f6a386c3ad78896747175ce472a47f95a1f meson8_ZeBu.dtd
75f7fa05efcfc6e6990f1384e2db20f5dcc23f44 meson8m2_skt.dtd
76f5ff87c5d20be132f2376b660a836080628d7b meson6_g18.dtd
7d178b62a31b53b04d6ef5b58ef5bbacaaece4c0 meson8b_m201_512M.dtd
840f165f40bc7cafa37ea7642273a4fe8971eecc meson8m2_n200_2G.dtd
8f613c10d392326ee3de88fa0f4cfcd656237ccc meson6_g33_512M.dtd
a2c9f06672a02ba99a674f1f853888eb61ca0751 meson8_k200a_2G_hdmiin.dtd
a8ca106ca9583681d4f4599c47d2e640f8b83d6d meson8_k200b_2G_hdmiin.dtd
ac9fc0c79e2c72d9eaccca7aa8684680357ad961 meson8_k200b_2G_4k2k.dtd
b5acb0b5359e7d8b0898e0dbdb0a66b2eeebb0b9 meson8_skt.dtd
b7b555e9403004707e4a8d68363514e531ffab87 meson8b_skt.dtd
be5ddc7f610f1491dac728d952c602d4dcbfa0fa meson8b_m201_1G.dtd
c3ce3fc98690de3c6539279c7b87699dfdeaa108 meson8m2_n101_512M.dtd
caeac17ec9d57ecf8f66ac0da8d0633b352eb112 meson8b_m100_1G.dtd
dc09a260b5dfb6bb263590f93008d33030050ba1 meson8b_m102_1G.dtd
dd7f843d8dfeb1d5528bf6d44b00f00b12a122b1 meson8_k102_v1.dtd
e1e82d9c05eb47fdc8fa5f670f12309604135227 meson8m2_n01_2G.dtd
e29142090423dd98dcef7ea926210c8728bd5060 meson6_g35.dtd
ea0c5ff4d86cdd26a57bc605a42806c5d3b0d7e4 meson6_g24.dtd
ead21d59933380bdd0ca40a30b97ec2b97501185 meson8_m82_2G.dtd <==================
f388e92a51c59355a6fdc08f516d79da9e349e14 meson6_skt.dtd
fd1e0207b86de7a61e7acdab23044cc3ebfd8d35 meson8_pxp.dtd
fe18afef47841b32fbb381d5ecfe0270b4ae33d2 meson8b_m101_512M.dtd
Hi Steeve, It seems that MXIII uses the same K200 board as the M8 box. So you can use existing Meson8_k200a_1G.dtd and Meson8_k200a_2g.dtd. They may look a liitle different from the ones that you extracted from MIII firmwares, but that's mainly because the extracted device trees don't have comments.
Indeed the update files has the k200
prefix, which indeed would point in that direction.
The weird thing though is that I've tried to use dtc
to do a dts
to dts
conversion, for instance:
$ dtc -I dts meson8_k200a_1G.dtd -O dts -o tmp.dtd
$ diff tmp.dtd meson8_m82_1G.dtd
And I'm finding some significant differences, such as:
- phys_offset = <0x200000>;
- linux,total-memory = <0x3fe00000>;
+ phys_offset = <0x0>;
+ linux,total-memory = <0x40000000>;
I guess I should really test building the firmware and see if it boots
Please try to build with meson8_k200a_1G.dtd and let me know if it works.
Will do, many thanks
Alright, I just tested it. The kernel boots, but no output to HDMI. Here is the console log: http://pastebin.com/pMDMcXCq
I've just unpacked/repacked/signed your OTA update with the new dtb
.
I'm going to test the extracted dtd
.
Weird, with the extracted dtd, the kernel doesn't boot...
Starting kernel ...
[ 28.070009@0] BUG: soft lockup - CPU#0 stuck for 23s! [swapper/0:1]
[ 28.072125@0] Kernel panic - not syncing: softlockup: hung tasks
[ 28.076615@0] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 3.10.57 #1
[ 28.082774@0] [<c0014bec>] (unwind_backtrace+0x0/0xf8) from [<c0011cd8>] (show_stack+0x10/0x14)
[ 28.091442@0] [<c0011cd8>] (show_stack+0x10/0x14) from [<c0769400>] (panic+0x8c/0x1e4)
[ 28.099330@0] [<c0769400>] (panic+0x8c/0x1e4) from [<c008dec0>] (watchdog_timer_fn+0x1f8/0x270)
[ 28.107999@0] [<c008dec0>] (watchdog_timer_fn+0x1f8/0x270) from [<c0050014>] (__run_hrtimer.isra.13+0x88/0x108)
[ 28.118053@0] [<c0050014>] (__run_hrtimer.isra.13+0x88/0x108) from [<c0050dcc>] (hrtimer_run_queues+0x160/0x188)
[ 28.128195@0] [<c0050dcc>] (hrtimer_run_queues+0x160/0x188) from [<c003ca5c>] (run_local_timers+0x8/0x14)
[ 28.137729@0] [<c003ca5c>] (run_local_timers+0x8/0x14) from [<c003ca98>] (update_process_times+0x30/0x68)
[ 28.147265@0] [<c003ca98>] (update_process_times+0x30/0x68) from [<c0074f70>] (tick_handle_periodic+0x18/0x7c)
[ 28.157233@0] [<c0074f70>] (tick_handle_periodic+0x18/0x7c) from [<c001c254>] (meson_timer_interrupt+0x64/0x94)
[ 28.167288@0] [<c001c254>] (meson_timer_interrupt+0x64/0x94) from [<c008e8d8>] (handle_irq_event_percpu+0x34/0x178)
[ 28.177690@0] [<c008e8d8>] (handle_irq_event_percpu+0x34/0x178) from [<c008ea58>] (handle_irq_event+0x3c/0x5c)
[ 28.187659@0] [<c008ea58>] (handle_irq_event+0x3c/0x5c) from [<c0091708>] (handle_fasteoi_irq+0xf8/0x13c)
[ 28.197194@0] [<c0091708>] (handle_fasteoi_irq+0xf8/0x13c) from [<c008e1f0>] (generic_handle_irq+0x24/0x38)
[ 28.206903@0] [<c008e1f0>] (generic_handle_irq+0x24/0x38) from [<c000f074>] (handle_IRQ+0x68/0x90)
[ 28.215829@0] [<c000f074>] (handle_IRQ+0x68/0x90) from [<c0008450>] (gic_handle_irq+0x3c/0x5c)
[ 28.224411@0] [<c0008450>] (gic_handle_irq+0x3c/0x5c) from [<c000dd00>] (__irq_svc+0x40/0x70)
[ 28.232902@0] Exception stack(0xe1a79eb0 to 0xe1a79ef8)
[ 28.238104@0] 9ea0: 00000001 00000000 000c03c2 fe040000
[ 28.246427@0] 9ec0: 00000001 ffff8ad8 c0afa0c0 e1a72ec0 00000000 00000000 c0b0c4ec 00000000
[ 28.254748@0] 9ee0: 00000001 e1a79ef8 c076759c c001bc68 20000113 ffffffff
[ 28.261512@0] [<c000dd00>] (__irq_svc+0x40/0x70) from [<c001bc68>] (meson_get_cpu_ctrl_addr+0x4/0x18)
[ 28.270700@0] [<c001bc68>] (meson_get_cpu_ctrl_addr+0x4/0x18) from [<c076759c>] (meson_boot_secondary+0x54/0x13c)
[ 28.280928@0] [<c076759c>] (meson_boot_secondary+0x54/0x13c) from [<c0766d88>] (boot_secondary+0x18/0x28)
[ 28.290463@0] [<c0766d88>] (boot_secondary+0x18/0x28) from [<c0766e08>] (__cpu_up+0x70/0x118)
[ 28.298957@0] [<c0766e08>] (__cpu_up+0x70/0x118) from [<c07677fc>] (_cpu_up+0xd4/0x154)
[ 28.306932@0] [<c07677fc>] (_cpu_up+0xd4/0x154) from [<c07678e0>] (cpu_up+0x64/0x84)
[ 28.314649@0] [<c07678e0>] (cpu_up+0x64/0x84) from [<c0ac4114>] (smp_init+0x9c/0xd4)
[ 28.322365@0] [<c0ac4114>] (smp_init+0x9c/0xd4) from [<c0abaaf4>] (kernel_init_freeable+0x78/0x1cc)
[ 28.331379@0] [<c0abaaf4>] (kernel_init_freeable+0x78/0x1cc) from [<c0764858>] (kernel_init+0x8/0xe4)
[ 28.340566@0] [<c0764858>] (kernel_init+0x8/0xe4) from [<c000e1d8>] (ret_from_fork+0x14/0x3c)
But you even use different kernel from this one. You have 3.10.57 version in your log, but this has 3.10.33 version number.
Thanks for your quick answer
Where do you see 3.10.33? I've used the same kernel/ramdisk when mkbootimg
I see 3.10.57:
[ 28.076615@0] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 3.10.57 #1
I'm using the kernel you provided in OpenELEC-CodeSnake-Meson8-K200-devel-320a148.zip
I shouldn't ?
You on IRC maybe?
No, you use different kernel from the one in this repository. I never used kernel 3.10.57 in my builds.
Ok I must have screwed up somewhere. Anyway, many thanks. I'll reopen the PR if need be
These S802 boards come in 2 flavours:
They were extracted from the firmwares at http://www.bee-link.com/en/article.php?id=134
Thank you for your great work.