[00:11:40] it sounds a lot "nicer" than ours, but ours is louder / more noticable :P [00:29:06] TheSeven: do you have a complete rockbox install (zip) for 6g with your patches? [00:50:46] TheSeven: using the rockbox.ipod file from your 7z archive made windows think that the filesystem is corrupted, then it said that there's no disk in the drive. ejecting usb crashed rockbox (panic: mount: 0, pc:0804a6c4, sp:080d2e08) [00:51:16] it was launched from rolo, if this matters [01:15:56] http://paste.pm/i3g.pl [01:16:01] ...and I'm sure there are even more [02:42:58] *** Quits: bcobco (~bcobco@77.225.204.119) () [03:20:25] user890104: still awake? [03:20:34] I'm observing some really weird behavior here [03:24:30] Hit abort (05) at 08034838 (accessing 00ffc3ea)! [03:24:30] r00: 08b30820, r04: 00ffc3ea, r08: 0000000f, r12: ea00ffc3 [03:24:30] r01: 200000d3, r05: 00000001, r09: 39a00094, r13: 08b30820 [03:24:30] r02: 000000c0, r06: 089aed2c, r10: 00008000, r14: 08034838 [03:24:30] r03: 02000000, r07: 0812cc38, r11: 00000020, r15: 0803483c [03:24:30] spsr: 200000d3, cpsr: 200000d7, fsr: 00000005, far: 00ffc3ea [03:24:32] Enabled IRQs: 0 1 2 3 8 16 17 18 21 23 24 29 33 (IRQs globally disabled!) [03:24:47] 08B30820: C0 00 00 00 EA C3 FF 00 18 08 AC 08 E0 4C 03 08 |........ .....L..| [03:24:47] 08B30830: F0 00 00 00 44 2F 00 22 B4 00 A0 39 00 0F E0 38 |....D/." ...9...8| [03:24:47] 08B30840: 01 00 00 00 08 20 E0 38 E0 C1 E5 0B 50 82 54 09 |..... .8 ....P.T.| [03:24:48] 08B30850: 1F 99 99 99 00 00 00 00 00 00 00 00 18 4E 00 00 |........ .....N..| [03:26:08] 4e18 is the global IRQ handler, right after it jumps into the actual handler for that IRQ [03:42:37] ok, the abort is in the IRQ1 handler [03:46:35] that handler is at 080b6f38 [03:46:42] it writes 5 to r0 [03:46:49] then jumps to 080166e0 [03:47:00] that's a veneer for 22002eb8 [03:47:17] no stack activity up to that point [03:48:50] 22002eb8 then saves r3-r11 and lr to the stack (lr=00004e18, r11=0, r10=0, r9=9999991f, r8=09548250, r7=0befc1e0, r6=38e02008, r5=1, r4=38e00f00, r3=39a000b4) [03:49:25] the next instruction copys r0 to r6 [03:50:04] then, assuming the usual procedure call convention, nothing should mess with the stack or r6 until the final (crashing) call to 22003598 [03:50:45] that's a veneer for 080a50b4, which dereferences r0 (which was loaded from an address that depends on r6), and jumps to 08034ccc [03:51:50] that one saves r4 and lr (lr=22002f44, r4=000000f0), doesn't touch r6, and calls atomic_add (08034824) [03:52:36] that then saves r3-r5 and lr (lr=08034ce0, r5=08ac0818, r4=00ffc3ea, r3=000000c0), and doesn't touch r6 either before crashing [03:52:52] after all of that, r6 is 089aed2c... why!? [03:53:35] r0/r6 being corrupted at the call to 22002eb8 would explain the crash [03:54:01] and I don't see anything afterwards messing with it, so it seems likely that we see the actual r6 value back from that time [03:54:22] however the r0 value is set right before calling that function! how can that possibly get corrupted!? [03:55:10] and why does all of that happen ~15 seconds into the boot process (for the first time)? [03:57:41] the irq handler gets called much earlier already and seems to work without problems at that time [03:57:52] (to make matters worse, we don't even know what that IRQ is) [04:01:35] I've even checked for runtime code corruption in the relevant spots... everything's intact [04:02:23] the problem is that I can't easily breakpoint at the interesting spots because it isn't crashing when the code is run for the first time [06:36:19] *** Quits: TheSeven (~quassel@rockbox/developer/TheSeven) (Ping timeout: 272 seconds) [06:37:29] *** Joins: TheSeven (~quassel@rockbox/developer/TheSeven) [06:48:27] *** Quits: steffengy (~quassel@p57B49A5B.dip0.t-ipconnect.de) (Disconnected by services) [06:48:28] *** Joins: steffengy1 (~quassel@p57B49361.dip0.t-ipconnect.de) [09:50:03] *** Joins: slenselink__ (~STeeF@office.hostnetbv.nl) [09:50:19] *** slenselink__ is now known as STeeF [09:58:41] *** Joins: bcobco (~bcobco@77.225.204.119) [10:33:45] *** Quits: bcobco (~bcobco@77.225.204.119) (Remote host closed the connection) [10:34:11] *** Joins: bcobco (~bcobco@77.225.204.119) [11:27:13] *** Quits: bcobco (~bcobco@77.225.204.119) (Remote host closed the connection) [11:27:39] *** Joins: bcobco (~bcobco@77.225.204.119) [11:29:24] *** Quits: Basstard` (~sturgeon@host-95-195-153-67.mobileonline.telia.com) (Ping timeout: 255 seconds) [13:11:12] *** Quits: bcobco (~bcobco@77.225.204.119) (Remote host closed the connection) [13:11:38] *** Joins: bcobco (~bcobco@77.225.204.119) [13:40:27] *** Joins: steffengy (~quassel@p57B48556.dip0.t-ipconnect.de) [13:43:03] *** Quits: steffengy1 (~quassel@p57B49361.dip0.t-ipconnect.de) (Ping timeout: 240 seconds) [14:04:11] *** Quits: bcobco (~bcobco@77.225.204.119) (Remote host closed the connection) [14:04:33] *** Joins: bcobco (~bcobco@77.225.204.119) [15:28:06] *** Joins: jan_bee (~some@vpn.janbeerden.be) [15:30:59] *** Quits: jan_bee_ (~some@vpn.janbeerden.be) (Ping timeout: 245 seconds) [18:32:43] TheSeven: remeber the issue with chkdisk on windows? someone also reported it in #rockbox, and pasted the following message: [18:32:45] The type of File System is FAT32. Cannot Lock the Current Drive. An Unspecified Error Occured <666174766f6c2e63 d6> [18:33:12] so we need to offer some interface for locking the device? [18:33:51] this should be a mass storage instruction, where we can simply return true, right? [18:34:03] I don't really understand what is happening there [18:34:13] I think it works on older ipods, but fails on nano2g/classic [18:34:19] (when running rockbox) [18:34:40] those share the same USB MSC driver, so I doubt that it's a problem at that layer [18:35:11] also chkdsk should be able to read-only-scan an device that it cannot lock, and even that fails [18:35:23] the locking thing might also be some process still accessing it [18:35:25] did you pass /x? [18:36:07] I suspect that it's related to either the sector size, or some other detail about how we are formatting the drive [18:36:17] does it still fail after formatting the drive with windows? [18:36:29] also, did you read my rant from last night? [18:39:46] yes, i did [18:40:09] but didn't have time to look into [18:42:00] I must be overlooking something somewhere... [19:18:17] *** Quits: bcobco (~bcobco@77.225.204.119) (Remote host closed the connection) [19:18:39] *** Joins: bcobco (~bcobco@77.225.204.119) [20:51:17] *** Joins: Basstard` (~sturgeon@host-95-199-25-205.mobileonline.telia.com) [22:51:26] *** Quits: TheSeven (~quassel@rockbox/developer/TheSeven) (Disconnected by services) [22:51:41] *** Joins: [7] (~quassel@rockbox/developer/TheSeven)