All of lore.kernel.org
 help / color / mirror / Atom feed
From: Niklas Cassel <cassel@kernel.org>
To: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>,
	"Martin K. Petersen" <mkp@kernel.org>
Cc: linux-scsi@vger.kernel.org, Damien Le Moal <dlemoal@kernel.org>,
	John Garry <john.garry@linux.dev>,
	Niklas Cassel <cassel@kernel.org>
Subject: [PATCH v7 05/11] scsi: scsi_debug: Report the residual of a write
Date: Fri, 25 Sep 2026 09:17:32 +0200	[thread overview]
Message-ID: <20260925071726.140915-18-cassel@kernel.org> (raw)
In-Reply-To: <20260925071726.140915-13-cassel@kernel.org>

A write whose data buffer is larger than its transfer length leaves a
residual that the initiator is entitled to be told about. resp_read_dt0()
and resp_write_tape() report it, but the write paths for a disk do not,
so such a write completes with GOOD status and a residual of zero while
a READ of the same length into the same buffer reports it correctly.

Report it in resp_write_dt0(), resp_write_scat() and resp_atomic_write().

resp_write_scat() differs twice over. Its data-out buffer also holds the
parameter list header and the LBA range descriptors, so what the command
consumed is sg_off rather than the bytes that the last range
transferred; and sg_off counts what was asked for rather than what was
transferred, so it can exceed the buffer and needs a guard. Two of its
exits also bypass the end of the loop: a command with no LBA range
descriptors or a buffer transfer length of zero transfers nothing, so
the whole buffer is residual, and an injected error ends the command
after a range has been written, where resp_write_dt0() reports the
residual before injecting it.

Assisted-by: LLM
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Niklas Cassel <cassel@kernel.org>
---
Tested with:

  modprobe scsi_debug zbc=managed sector_size=512 physblk_exp=3 \
      zone_size_mb=8 dev_size_mb=128 zone_nr_conv=2

An eight logical block WRITE(16) with a 5000 byte buffer reports
resid=904, having transferred 4096. A READ(16) of the same length into
the same buffer reported that before this patch and the WRITE reported
0. A buffer of exactly 4096 bytes reports 0, as does a 512 byte buffer,
which is consumed in its entirety. WRITE ATOMIC (16) behaves like the
WRITE, on a device that is not zoned.

WRITE SCATTERED (16) with one LBA range descriptor of eight blocks and a
5632 byte buffer reports resid=1024, having consumed 512 bytes of
parameter list and 4096 bytes of data. The same command with a 1024 byte
buffer completes and reports resid=0: sg_off reaches 4608, past the end
of the buffer, and without the guard the residual would have been
reported as 4294963712.

Changes since v6: a WRITE SCATTERED (16) with no LBA range descriptors,
or with a buffer transfer length of zero, and a 1024 byte buffer
reports resid=1024, as does the command above with a RECOVERED ERROR
injected (opts=8, every_nth=1), on a device that is not zoned. All
three reported 0 in v6.
---
 drivers/scsi/scsi_debug.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c
index b64ae3ad300d..9220758bb801 100644
--- a/drivers/scsi/scsi_debug.c
+++ b/drivers/scsi/scsi_debug.c
@@ -5162,6 +5162,8 @@ static int resp_write_dt0(struct scsi_cmnd *scp, struct sdebug_dev_info *devip)
 			    "%s: write: cdb indicated=%u, IO sent=%d bytes\n",
 			    my_name, num * sdebug_sector_size, ret);
 
+	scsi_set_resid(scp, scsi_bufflen(scp) - ret);
+
 	if (unlikely((sdebug_opts & SDEBUG_OPT_RECOV_DIF_DIX) &&
 		     atomic_read(&sdeb_inject_pending))) {
 		if (sdebug_opts & SDEBUG_OPT_RECOVERED_ERR) {
@@ -5233,8 +5235,10 @@ static int resp_write_scat(struct scsi_cmnd *scp,
 					    "Unprotected WR to DIF device\n");
 		}
 	}
-	if ((num_lrd == 0) || (bt_len == 0))
+	if (num_lrd == 0 || bt_len == 0) {
+		scsi_set_resid(scp, scsi_bufflen(scp));
 		return 0;       /* T10 says these do-nothings are not errors */
+	}
 	if (lbdof == 0) {
 		if (sdebug_verbose)
 			sdev_printk(KERN_INFO, scp->device,
@@ -5327,6 +5331,9 @@ static int resp_write_scat(struct scsi_cmnd *scp,
 
 		if (unlikely((sdebug_opts & SDEBUG_OPT_RECOV_DIF_DIX) &&
 			     atomic_read(&sdeb_inject_pending))) {
+			if (scsi_bufflen(scp) > sg_off + num_by)
+				scsi_set_resid(scp, scsi_bufflen(scp) -
+					       (sg_off + num_by));
 			if (sdebug_opts & SDEBUG_OPT_RECOVERED_ERR) {
 				mk_sense_buffer(scp, RECOVERED_ERROR,
 					FAILURE_PREDICTION_THRESHOLD_EXCEEDED);
@@ -5350,6 +5357,13 @@ static int resp_write_scat(struct scsi_cmnd *scp,
 		sg_off += num_by;
 		cum_lb += num;
 	}
+	/*
+	 * sg_off counts what the command asked for, which can exceed the
+	 * buffer: lbdof is not validated against it, and a range is counted
+	 * in full even if do_device_access() copied less.
+	 */
+	if (scsi_bufflen(scp) > sg_off)
+		scsi_set_resid(scp, scsi_bufflen(scp) - sg_off);
 	ret = 0;
 err_out_unlock:
 	sdeb_meta_write_unlock(sip);
@@ -6206,6 +6220,8 @@ static int resp_atomic_write(struct scsi_cmnd *scp,
 		return DID_ERROR << 16;
 	if (unlikely(ret != len * sdebug_sector_size))
 		return DID_ERROR << 16;
+
+	scsi_set_resid(scp, scsi_bufflen(scp) - ret);
 	return 0;
 }
 
-- 
2.55.0


  parent reply	other threads:[~2026-09-25  7:17 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-25  7:17 [PATCH v7 00/11] scsi: scsi_debug: fix zoned write validation Niklas Cassel
2026-09-25  7:17 ` [PATCH v7 01/11] scsi: scsi_debug: Refuse a zoned device with a non-zero lowest aligned LBA Niklas Cassel
2026-09-25  7:17 ` [PATCH v7 02/11] scsi: scsi_debug: Make atomic writes and ZBC emulation mutually exclusive Niklas Cassel
2026-09-25  7:17 ` [PATCH v7 03/11] scsi: scsi_debug: Take the zone metadata lock before the data lock Niklas Cassel
2026-09-25  7:17 ` [PATCH v7 04/11] scsi: scsi_debug: Evaluate scsi_debug_lbp() only once Niklas Cassel
2026-09-25  7:17 ` Niklas Cassel [this message]
2026-09-25  7:17 ` [PATCH v7 06/11] scsi: scsi_debug: Enforce physical block alignment of zoned writes Niklas Cassel
2026-09-25  7:17 ` [PATCH v7 07/11] scsi: scsi_debug: Do not write a partial physical block to a zoned device Niklas Cassel
2026-09-25  7:17 ` [PATCH v7 08/11] scsi: scsi_debug: Advance the write pointer over the data written Niklas Cassel
2026-09-25  7:17 ` [PATCH v7 09/11] scsi: scsi_debug: Refuse a short WRITE ATOMIC (16) before writing it Niklas Cassel
2026-09-25  9:30   ` John Garry
2026-09-26 18:03     ` Niklas Cassel
2026-09-28  8:52       ` John Garry
2026-09-25  7:17 ` [PATCH v7 10/11] scsi: scsi_debug: Map the region written by WRITE ATOMIC (16) Niklas Cassel
2026-09-25  9:33   ` John Garry
2026-09-25  7:17 ` [PATCH v7 11/11] scsi: scsi_debug: Validate the access parameters of " Niklas Cassel

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260925071726.140915-18-cassel@kernel.org \
    --to=cassel@kernel.org \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=dlemoal@kernel.org \
    --cc=john.garry@linux.dev \
    --cc=linux-scsi@vger.kernel.org \
    --cc=mkp@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.