On 2021-03-12 12:04, David Wolfskill wrote: > On Thu, Mar 11, 2021 at 07:57:36PM -0800, Alastair Hogge wrote: >> Turns out, EFI boot stopped working. The following boots but fails with >> ... >> umass0: SCSI over Bulk-Only; quirks = 0x4001 >> umass0:6:0: Attached to scbus6 >> Root mount waiting for: usbus5 CAM >> panic: malloc(M_WAITOK) with sleeping prohibited >> cpuid = 0 >> time = 4 >> KDB: stack backtrace: >> db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame >> 0xfffffe0081dcf5f0 >> vpanic() at vpanic+0x181/frame 0xfffffe0081dcf640 >> panic() at panic+0x43/frame 0xfffffe0081dcf6a0 >> malloc_dbg() at malloc_dbg+0xd4/frame 0xfffffe0081dcf6c0 >> malloc() at malloc+0x34/frame 0xfffffe0081dcf720 >> disk_alloc() at disk_alloc+0x1c/frame 0xfffffe0081dcf740 >> daregister() at daregister+0x3f4/frame 0xfffffe0081dcf9c0 >> cam_periph_alloc() at cam_periph_alloc+0x57b/frame 0xfffffe0081dcfa90 >> daasync() at daasync+0x2c2/frame 0xfffffe0081dcfb00 >> xpt_async_process_dev() at xpt_async_process_dev+0x152/frame >> 0xfffffe0081dcfb50 >> xpt_async_process() at xpt_async_process+0x334/frame 0xfffffe0081dcfc60 >> xpt_done_process() at xpt_done_process+0x3a3/frame 0xfffffe0081dcfca0 >> xpt_done_td() at xpt_done_td+0xf5/frame 0xfffffe0081dcfcf0 >> fork_exit() at fork_exit+0x80/frame 0xfffffe0081dcfd30 >> fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe0081dcfd30 >> --- trap 0, rip = 0, rsp = 0, rbp = 0 --- >> .... > > You may find the change that's in https://reviews.freebsd.org/D29210 > helps -- the trace looks similar to what I saw, and it has worked for > me. > > Note that I only saw the panic while running a kernel with INVARIANTS. Fantastic! Thanks for those pointers, I will try imp's fix after the current build finishes. To health, AlastairReceived on Fri Mar 12 2021 - 03:08:27 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:41:27 UTC