Created attachment 1339 [details] 10 times beagleboard boot up log Poky: master:3dee534f1e25109e0bdb681de0746c336f4b8840 Test steps: 1. bootup beagleboard, modify /etc/inittab and set default runlevel to 3 ######## id:3:initdefault ######## 2. reboot it, and edit "kernel" line and add "psplash=false text" at the end, then boot it. 3. check the bootup info, sometimes, it will show Internal error, then use dmesg| grep error could find some error info during the booting. I did this test 10 times, and totally match 7 times "Internal error", all of them are Oops 5 and 817: Internal error: Oops: 5 Internal error: Oops: 817 Detail logs please check attachment. ===============some log============================= Internal error: Oops: 5 [#1] ARM Modules linked in: CPU: 0 Tainted: G W (3.4.52-yocto-standard #1) PC is at acct_process+0x3c/0x94 LR is at acct_process+0x28/0x94 pc : [<c0082fc4>] lr : [<c0082fb0>] psr: a0000013 sp : de61bf18 ip : de61bf18 fp : de61bf2c r10: 00000000 r9 : de5bb100 r8 : c000e468 r7 : 000000f8 r6 : 4c003b8c r5 : c0049780 r4 : 00000100 r3 : fffffffa r2 : de61bf18 r1 : 00000000 r0 : e50bc044 Flags: NzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user Control: 10c5387d Table: 9e620019 DAC: 00000015 Process modprobe (pid: -565289040, stack limit = 0xde61a2f0) Stack: (0xde61bf18 to 0xde61c000) bf00: 00000100 00000001 bf20: de61bf7c de61bf30 c003ba78 c0082f94 de61bf6c 00000001 de61a030 c0141d58 bf40: 00000000 00000000 de61bf94 00000000 00000000 00000100 00000001 4c003b8c bf60: 000000f8 c000e468 de61a000 00000000 de61bf94 de61bf80 c003bcf8 c003b518 bf80: 00000001 00000001 de61bfa4 de61bf98 c003bd78 c003bcb4 00000000 de61bfa8 bfa0: c000e200 c003bd64 00000001 00000001 00000001 b6fda6d0 00000008 00000001 bfc0: 00000001 00000001 4c003b8c 000000f8 be896ca8 00000000 00000000 000203c8 bfe0: 000974ec be896a50 00097508 4bf6dfe0 60000010 00000001 00000000 00000000 [<c0082fc4>] (acct_process+0x3c/0x94) from [<c003ba78>] (do_exit+0x56c/0x750) [<c003ba78>] (do_exit+0x56c/0x750) from [<c003bcf8>] (do_group_exit+0x50/0xb0) [<c003bcf8>] (do_group_exit+0x50/0xb0) from [<c003bd78>] (__wake_up_parent+0x0/0x30) [<c003bd78>] (__wake_up_parent+0x0/0x30) from [<c000e200>] (ret_fast_syscall+0x0/0x30) Code: 089da830 e5950024 e3500000 0a000010 (e5903008) ---[ end trace 1b75b31a2719ed1e ]--- Fixing recursive fault but reboot is needed! generic-usb 0003:413C:2003.0002: input: USB HID v1.10 Keyboard [Dell Dell USB Keyboard] on usb-ehci-omap.0-2.4.5/input0 Unable to handle kernel paging request at virtual address ffffffff pgd = de58c000 [ffffffff] *pgd=9f0fe821, *pte=00000000, *ppte=00000000 Internal error: Oops: 817 [#2] ARM Modules linked in: CPU: 0 Tainted: G D W (3.4.52-yocto-standard #1) PC is at __d_lookup_rcu+0x114/0x164 LR is at do_lookup+0x50/0x2d8 pc : [<c011eac0>] lr : [<c0114b28>] psr: 600f0013 sp : de585df4 ip : 00000035 fp : de585e1c r10: de513005 r9 : 00000005 r8 : de585ee8 r7 : de801280 r6 : 00000004 r5 : 131018d9 r4 : de919688 r3 : ffffffff r2 : dee80d90 r1 : de9196a8 r0 : de919680 Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user Control: 10c5387d Table: 9e58c019 DAC: 00000015 Process udevd (pid: 82, stack limit = 0xde5842f0) Stack: (0xde585df4 to 0xde586000) 5de0: de585e80 de585e88 00000051 5e00: de801280 de584020 de400e50 de585ee0 de585e5c de585e20 c0114b28 c011e9b8 5e20: de585e5c de585ee8 00000005 00000004 de585e54 de585e88 00000000 00000051 5e40: de585e80 de584020 de585ee0 00000000 de585ebc de585e60 c01157b8 c0114ae4 5e60: de585e78 de585e80 de585f20 c01217dc 51e904d5 de585ee8 00000000 00000041 5e80: dee80d90 c01129a4 dec9c178 de585e98 c08ab7d0 00000001 de585ee0 de513000 5ea0: ffffff9c ffffff9c de584000 00000000 de585edc de585ec0 c0115dc8 c011569c 5ec0: de513000 00000001 de585ee0 de585f80 de585f64 de585ee0 c0117554 c0115da8 5ee0: de400e50 de801280 131018d9 00000005 de513005 de400d90 de9b2400 dec9c178 5f00: 00000041 00000002 00000000 00000000 de585f74 de585f20 c0106f8c c0122df4 5f20: 00000041 c00021b0 de584000 c0123b1c de400e50 de585f78 000021b0 00047070 5f40: de585f5c 00000215 00000005 00000000 000000d4 c000e468 de585f7c de585f68 5f60: c01175ac c01174fc 00000000 de585f78 de585fa4 de585f80 c0108274 c0117594 5f80: 00000215 00000000 de585fa4 00000215 00000000 00047070 00000000 de585fa8 5fa0: c000e200 c010824c 00000215 00000000 00047038 00000000 00000005 00000000 5fc0: 00000215 00000000 00047070 000000d4 00047038 00033008 000021b0 00000000 5fe0: 000322a8 beb958e4 00011ab4 4bf92cdc 000f0010 00047038 00000000 00000000 [<c011eac0>] (__d_lookup_rcu+0x114/0x164) from [<c0114b28>] (do_lookup+0x50/0x2d8) [<c0114b28>] (do_lookup+0x50/0x2d8) from [<c01157b8>] (path_lookupat+0x128/0x70c) [<c01157b8>] (path_lookupat+0x128/0x70c) from [<c0115dc8>] (do_path_lookup+0x2c/0x6c) [<c0115dc8>] (do_path_lookup+0x2c/0x6c) from [<c0117554>] (user_path_at_empty+0x64/0x98) [<c0117554>] (user_path_at_empty+0x64/0x98) from [<c01175ac>] (user_path_at+0x24/0x2c) [<c01175ac>] (user_path_at+0x24/0x2c) from [<c0108274>] (sys_chown+0x34/0x8c) [<c0108274>] (sys_chown+0x34/0x8c) from [<c000e200>] (ret_fast_syscall+0x0/0x30) Code: e51b203c e51b0038 e5836000 e51b3030 (e5832000) ---[ end trace 1b75b31a2719ed1f ]---
Simar: can you grab this. Note: I'm able to boot the board fine with the latest, so I suspect a h/w issue. If you can confirm a 3.4 and then 3.8 boot, this can be closed.
Hi, I reused a 3.4 kernel image(20130615, kernel: 3.4.36) image to do this test, i reboot around 10 times, did not occur any "Internal error" issue. So i think this is not a hardware issue. Thanks.
Created attachment 1366 [details] retest it on 3.4.36 kernel, no this error
In the same vein, I can boot the 3.4.52 kernel as many times as I like on my board, and never see the issue. So clearly something is different with your board than mine, we can't fix it, if we can see the problem.
Just to keep everyone level, I have two BeagleBoards to test with right now. So far I've tested 3.4 on both of them and there haven't been any issues whatsoever. I've repeated this multiple times but haven't been able to reproduce this at all on 3.4 yet. I will further test out 3.8 on both of them today to see how they behave.
Update: I've built and booted various 3.8.x so far. Although there are other issues with bootup, this bug in specific doesn't manifest itself on 3.8.x. Still trying to see if I can reproduce this with the latest kernel.
Update: So it seems that I was able to reproduce this issue on 3.8.13 and also fix it. I now just need to narrow it down from a bunch of changes I made to the specific one that's causing this. Should have a fix by EOD tomorrow if everything goes well.
Update: Issue exists in ehci-omap driver. The defconfig for omap2 has EHCI disabled by default so issue doesn't show up there. Currently looking at both testing 3.10 and upstream 3.4 driver.
Update: It seems that this issue will get fixed in the 3.10 update when it comes to linux-yocto. Tested it out with 3.10 on linux-yocto-dev and works just fine on both omap2plus_defconfig and stock yocto configuration.
The changes for beagleboard should be in 3.10 kernel that currently exists. Can someone test this again as I don't have access to a beagleboard anymore.
This issue doesn't happen on 3.10 kernel. Tested with 1.5 M5 rc4.
Fixed with the 3.10 uprev.