Date: Tue, 13 Mar 2018 17:38:16 +0000 From: bugzilla-noreply@freebsd.org To: freebsd-bugs@FreeBSD.org Subject: [Bug 226578] panic: _mtx_lock_sleep: recursed on non-recursive mutex CAM device lock @ /usr/home/trasz/svn-ssh/head/sys/cam/scsi/scsi_da.c:2042 Message-ID: <bug-226578-8-B8wn0RYVLb@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-226578-8@https.bugs.freebsd.org/bugzilla/>
index | next in thread | previous in thread | raw e-mail
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=226578 Warner Losh <imp@FreeBSD.org> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |imp@FreeBSD.org --- Comment #1 from Warner Losh <imp@FreeBSD.org> --- OK. XPT is locking the periph lock in xpt_async_process_dev sometimes, but not other times. The following only takes out the lock when we're likely to change the flags. It's an OK workaround for the moment, if it works, so we can validate the funky locking done in xpt_async_process_dev() and stylize the test a bit better... diff --git a/sys/cam/scsi/scsi_da.c b/sys/cam/scsi/scsi_da.c index f5be34f96e0..90045ca3e76 100644 --- a/sys/cam/scsi/scsi_da.c +++ b/sys/cam/scsi/scsi_da.c @@ -2039,10 +2039,10 @@ daasync(void *callback_arg, u_int32_t code, * Handle all UNIT ATTENTIONs except our own, * as they will be handled by daerror(). */ - cam_periph_lock(periph); if (xpt_path_periph(ccb->ccb_h.path) != periph && scsi_extract_sense_ccb(ccb, &error_code, &sense_key, &asc, &ascq)) { + cam_periph_lock(periph); if (asc == 0x2A && ascq == 0x09) { xpt_print(ccb->ccb_h.path, "Capacity data has changed\n"); @@ -2057,8 +2057,8 @@ daasync(void *callback_arg, u_int32_t code, softc->flags &= ~DA_FLAG_PROBED; dareprobe(periph); } + cam_periph_unlock(periph); } - cam_periph_unlock(periph); break; } case AC_SCSI_AEN: -- You are receiving this mail because: You are the assignee for the bug.help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-226578-8-B8wn0RYVLb>