[src/trunk]: src/sys/dev/ic Note added for a strange value of residual counts...

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

details:   https://anonhg.NetBSD.org/src/rev/4d8f24bd150e
branches:  trunk
changeset: 366497:4d8f24bd150e
user:      rin <rin%NetBSD.org@localhost>
date:      Wed Aug 29 16:51:51 2018 +0000

description:
Note added for a strange value of residual counts in successful read/write.
Seems like a H/W bug. The best thing we can do would be to leave the code as is.

No binary changes.

diffstat:

 sys/dev/ic/adv.c |  12 ++++++++++--
 1 files changed, 10 insertions(+), 2 deletions(-)

diffs (33 lines):

diff -r 49fe801fa0ba -r 4d8f24bd150e sys/dev/ic/adv.c
--- a/sys/dev/ic/adv.c  Wed Aug 29 16:26:25 2018 +0000
+++ b/sys/dev/ic/adv.c  Wed Aug 29 16:51:51 2018 +0000
@@ -1,4 +1,4 @@
-/*     $NetBSD: adv.c,v 1.47 2016/07/14 04:19:26 msaitoh Exp $ */
+/*     $NetBSD: adv.c,v 1.48 2018/08/29 16:51:51 rin Exp $     */
 
 /*
  * Generic driver for the Advanced Systems Inc. Narrow SCSI controllers
@@ -38,7 +38,7 @@
  */
 
 #include <sys/cdefs.h>
-__KERNEL_RCSID(0, "$NetBSD: adv.c,v 1.47 2016/07/14 04:19:26 msaitoh Exp $");
+__KERNEL_RCSID(0, "$NetBSD: adv.c,v 1.48 2018/08/29 16:51:51 rin Exp $");
 
 #include <sys/param.h>
 #include <sys/systm.h>
@@ -899,6 +899,14 @@
                switch (qdonep->d3.host_stat) {
                case ASC_QHSTA_NO_ERROR:
                        xs->error = XS_NOERROR;
+                       /*
+                        * XXX
+                        * According to the original Linux driver, xs->resid
+                        * should be qdonep->remain_bytes. However, its value
+                        * is bogus, which seems like a H/W bug. The best thing
+                        * we can do would be to ignore it, assuming that all
+                        * data has been successfully transferred...
+                        */
                        xs->resid = 0;
                        break;