Re: kern/54275: nvme_poll() called too early?

[Thomas Klausner <wiz%netbsd.org@localhost>]

The following reply was made to PR kern/54275; it has been noted by GNATS.

From: Thomas Klausner <wiz%netbsd.org@localhost>
To: matthew green <mrg%eterna.com.au@localhost>
Cc: NetBSD bugtracking <gnats-bugs%NetBSD.org@localhost>
Subject: Re: kern/54275: nvme_poll() called too early?
Date: Thu, 6 Jun 2019 07:53:41 +0200

 Thanks for the patch!
 
 It makes the machine panic differently:
 
 (handcopied)
 
 nvme0: for admin queue interrupting at msix9 vec 0
 nvme0: Samsung SSD 970 EVO Plus 1TB, firmware 1B2QEXM7, serial SOMETHING
 ahcisata0 port 5: device present, speed: 6.GB/s
 ahcisata0 port 5: clearing WDCIL_RSI failed for drive 15 (THIS LINE DOES NOT HAVE A TIMESTAMP)
 ahcisata0 port 5: drive 15 reset failed, sdisabling PMP (THIS LINE DOES NOT HAVE A TIMESTAMP)
 ahcisata0 port 5: device present, speed: 6.GB/s
 uhidev...
 ukbd0...
 wskbd...
 uhidev...
 uhid...
 wd0...
 nvme0: for io queue 1 interrupting at msix9 vec 1 affinity to cpu0
 (repeat with increasing queue, vec, and cpuX)
 nvme0: for io queue 31 interrupting at msix9 vec 31 affinity to cpu30
 fatal page fault in supervisor mode
 trap type 6 code 0x10 rip 0 cs 0x8 rflags 0x10202 cr2 0 ilevel 0x3 rsp 0xffffac090732b028
 curlwp 0xfffffa619cdec980 pid 0.22 lowest kstack 0xffffac09073272c0
 kernel: page fault trap, code=0
 Stopped in pid 0.22 (system) at 0:fatal page fault in supervisor mode
 trap type 6 code 0x10 rip 0xffffffff8021e646 cs 0x8 rflags 0x10216 cr2 0xffff964b00000000 ilevel 0x8 rsp 0xffffac090732ac10
 curlwp 0xfffffa619cdec980 pid 0.22 lowest kstrack 0xffffac09073272c0
 kernel: page fault trap, code=0
 Stopped in pid 0.22 (system) at netbsd:db_disasm+0xec: movq 0(%rdx,%rcx,8),%
 rcx
 db{1}> bt
 db_disasm()
 db_trap()
 kdb_trap()
 trap()
 --- trap (number 6) ---
 ?() at 0
 softint_dispatch()
 DDB lost frame for netbsd:Xsoftintr
 Xsoftintr()
 --- interrupt ---
 1bce138d178e1389:
 db{1}>
 
 Cheers,
  Thomas