[src/trunk]: src/sys/arch/xen/xen do not retry when xen_shm_map() fails after...

[jdolecek <jdolecek%NetBSD.org@localhost>]

details:   https://anonhg.NetBSD.org/src/rev/ad6b270bc0d5
branches:  trunk
changeset: 931032:ad6b270bc0d5
user:      jdolecek <jdolecek%NetBSD.org@localhost>
date:      Mon Apr 20 14:11:04 2020 +0000

description:
do not retry when xen_shm_map() fails after having returned error for it,
it corrupts the ring; just go to next request

should fix the problem with ZFS causing failscade when req fails,
reported by Brian Buhrow on port-xen

diffstat:

 sys/arch/xen/xen/xbdback_xenbus.c |  7 ++++---
 1 files changed, 4 insertions(+), 3 deletions(-)

diffs (28 lines):

diff -r 87cb12bf8804 -r ad6b270bc0d5 sys/arch/xen/xen/xbdback_xenbus.c
--- a/sys/arch/xen/xen/xbdback_xenbus.c Mon Apr 20 13:56:07 2020 +0000
+++ b/sys/arch/xen/xen/xbdback_xenbus.c Mon Apr 20 14:11:04 2020 +0000
@@ -1,4 +1,4 @@
-/*      $NetBSD: xbdback_xenbus.c,v 1.81 2020/04/20 03:00:33 msaitoh Exp $      */
+/*      $NetBSD: xbdback_xenbus.c,v 1.82 2020/04/20 14:11:04 jdolecek Exp $      */
 
 /*
  * Copyright (c) 2006 Manuel Bouyer.
@@ -26,7 +26,7 @@
  */
 
 #include <sys/cdefs.h>
-__KERNEL_RCSID(0, "$NetBSD: xbdback_xenbus.c,v 1.81 2020/04/20 03:00:33 msaitoh Exp $");
+__KERNEL_RCSID(0, "$NetBSD: xbdback_xenbus.c,v 1.82 2020/04/20 14:11:04 jdolecek Exp $");
 
 #include <sys/atomic.h>
 #include <sys/buf.h>
@@ -1804,7 +1804,8 @@
                SLIST_INSERT_HEAD(&xbdi->xbdi_va_free, xbd_io->xio_xv, xv_next);
                xbd_io->xio_xv = NULL;
                xbdi->xbdi_io = NULL;
-               xbdi->xbdi_cont = xbdi->xbdi_cont_aux;
+               // do not retry
+               xbdi->xbdi_cont = xbdback_co_main_incr;
                return xbdi;
        }
 }