Rock64 snapshot main-n260538-6f52f8504814 gets mx2510 related "lock order reversal: (sleepable after non-sleepable)" during boot
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Sun, 05 Feb 2023 22:24:25 UTC
I just booted the Rock64 via today's main snapshot put on a microsd card. It reported: . . . Building /boot/kernel/linker.hints Autoloading module: mx25l mx25l0: <M25Pxx Flash Family> at cs 0 mode 0 on spibus0 lock order reversal: (sleepable after non-sleepable) 1st 0xffffa00000dfa320 spi0 (spi0, sleep mutex) @ /usr/src/sys/arm64/rockchip/rk_spi.c:431 2nd 0xffff000000cda970 Clock topology lock (Clock topology lock, sx) @ /usr/src/sys/dev/extres/clk/clk.c:1200 lock order spi0 -> Clock topology lock attempted at: #0 0xffff0000004cc540 at witness_checkorder+0xadc #1 0xffff000000467660 at _sx_xlock+0x7c #2 0xffff00000018037c at clk_set_freq+0x4c #3 0xffff000000877eec at rk_spi_transfer+0x15c #4 0xffff0000e24117a0 at mx25l_attach+0xd8 #5 0xffff000000497f28 at device_attach+0x3f8 #6 0xffff000000497a98 at device_probe_and_attach+0x7c #7 0xffff000000499ce4 at bus_generic_driver_added+0x88 #8 0xffff00000049510c at devclass_driver_added+0x44 #9 0xffff00000049d934 at device_do_deferred_actions+0x70 #10 0xffff00000049d4ec at devctl2_ioctl+0x3f0 #11 0xffff000000300c3c at devfs_ioctl+0xd4 #12 0xffff00000055e0e4 at vn_ioctl+0xc0 #13 0xffff00000030132c at devfs_ioctl_f+0x20 #14 0xffff0000004d2474 at kern_ioctl+0x2dc #15 0xffff0000004d2118 at sys_ioctl+0x118 #16 0xffff00000078cfe0 at do_el0_sync+0x524 #17 0xffff00000076910c at handle_el0_sync+0x48 mx25l0: device type gd25q128, size 16384K in 256 sectors of 64K, erase size 4K . . . (I'm guessing that this LOR is not an expected one.) A separate note is that, at least for this combination, the Rock64 boot was not blocked by the U-Boot 2023.01 lmb code's lack of initialization of memory.max and reserved.max (bounds on memory.cnt and reserved.cnt for how many address ranges are allowed). (2022.10 contained the same lack initializations as it turns out. But a boot failure for 2023.01 has been observed in another context.) === Mark Millard marklmi at yahoo.com