selftests: ublk: add generic_06 for covering fault inject

Add one simple fault inject target, and verify if an application using ublk
device sees an I/O error quickly after the ublk server dies.

Signed-off-by: Uday Shankar <ushankar@purestorage.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20250416035444.99569-9-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

tools/testing/selftests/ublk/Makefile

@@ -8,6 +8,7 @@ TEST_PROGS += test_generic_02.sh
 TEST_PROGS += test_generic_03.sh
 TEST_PROGS += test_generic_04.sh
 TEST_PROGS += test_generic_05.sh
+TEST_PROGS += test_generic_06.sh
 
 TEST_PROGS += test_null_01.sh
 TEST_PROGS += test_null_02.sh
@@ -31,7 +32,8 @@ TEST_GEN_PROGS_EXTENDED = kublk
 
 include ../lib.mk
 
-$(TEST_GEN_PROGS_EXTENDED): kublk.c null.c file_backed.c common.c stripe.c
+$(TEST_GEN_PROGS_EXTENDED): kublk.c null.c file_backed.c common.c stripe.c \
+	fault_inject.c
 
 check:
 	shellcheck -x -f gcc *.sh

tools/testing/selftests/ublk/fault_inject.c

@@ -0,0 +1,98 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/*
+ * Fault injection ublk target. Hack this up however you like for
+ * testing specific behaviors of ublk_drv. Currently is a null target
+ * with a configurable delay before completing each I/O. This delay can
+ * be used to test ublk_drv's handling of I/O outstanding to the ublk
+ * server when it dies.
+ */
+
+#include "kublk.h"
+
+static int ublk_fault_inject_tgt_init(const struct dev_ctx *ctx,
+				      struct ublk_dev *dev)
+{
+	const struct ublksrv_ctrl_dev_info *info = &dev->dev_info;
+	unsigned long dev_size = 250UL << 30;
+
+	dev->tgt.dev_size = dev_size;
+	dev->tgt.params = (struct ublk_params) {
+		.types = UBLK_PARAM_TYPE_BASIC,
+		.basic = {
+			.logical_bs_shift	= 9,
+			.physical_bs_shift	= 12,
+			.io_opt_shift		= 12,
+			.io_min_shift		= 9,
+			.max_sectors		= info->max_io_buf_bytes >> 9,
+			.dev_sectors		= dev_size >> 9,
+		},
+	};
+
+	dev->private_data = (void *)(unsigned long)(ctx->fault_inject.delay_us * 1000);
+	return 0;
+}
+
+static int ublk_fault_inject_queue_io(struct ublk_queue *q, int tag)
+{
+	const struct ublksrv_io_desc *iod = ublk_get_iod(q, tag);
+	struct io_uring_sqe *sqe;
+	struct __kernel_timespec ts = {
+		.tv_nsec = (long long)q->dev->private_data,
+	};
+
+	ublk_queue_alloc_sqes(q, &sqe, 1);
+	io_uring_prep_timeout(sqe, &ts, 1, 0);
+	sqe->user_data = build_user_data(tag, ublksrv_get_op(iod), 0, 1);
+
+	ublk_queued_tgt_io(q, tag, 1);
+
+	return 0;
+}
+
+static void ublk_fault_inject_tgt_io_done(struct ublk_queue *q, int tag,
+					  const struct io_uring_cqe *cqe)
+{
+	const struct ublksrv_io_desc *iod = ublk_get_iod(q, tag);
+
+	if (cqe->res != -ETIME)
+		ublk_err("%s: unexpected cqe res %d\n", __func__, cqe->res);
+
+	if (ublk_completed_tgt_io(q, tag))
+		ublk_complete_io(q, tag, iod->nr_sectors << 9);
+	else
+		ublk_err("%s: io not complete after 1 cqe\n", __func__);
+}
+
+static void ublk_fault_inject_cmd_line(struct dev_ctx *ctx, int argc, char *argv[])
+{
+	static const struct option longopts[] = {
+		{ "delay_us", 	1,	NULL,  0  },
+		{ 0, 0, 0, 0 }
+	};
+	int option_idx, opt;
+
+	ctx->fault_inject.delay_us = 0;
+	while ((opt = getopt_long(argc, argv, "",
+				  longopts, &option_idx)) != -1) {
+		switch (opt) {
+		case 0:
+			if (!strcmp(longopts[option_idx].name, "delay_us"))
+				ctx->fault_inject.delay_us = strtoll(optarg, NULL, 10);
+		}
+	}
+}
+
+static void ublk_fault_inject_usage(const struct ublk_tgt_ops *ops)
+{
+	printf("\tfault_inject: [--delay_us us (default 0)]\n");
+}
+
+const struct ublk_tgt_ops fault_inject_tgt_ops = {
+	.name = "fault_inject",
+	.init_tgt = ublk_fault_inject_tgt_init,
+	.queue_io = ublk_fault_inject_queue_io,
+	.tgt_io_done = ublk_fault_inject_tgt_io_done,
+	.parse_cmd_line = ublk_fault_inject_cmd_line,
+	.usage = ublk_fault_inject_usage,
+};

tools/testing/selftests/ublk/kublk.c

@@ -12,6 +12,7 @@ static const struct ublk_tgt_ops *tgt_ops_list[] = {
 	&null_tgt_ops,
 	&loop_tgt_ops,
 	&stripe_tgt_ops,
+	&fault_inject_tgt_ops,
 };
 
 static const struct ublk_tgt_ops *ublk_find_tgt(const char *name)
@@ -1234,7 +1235,7 @@ static void __cmd_create_help(char *exe, bool recovery)
 {
 	int i;
 
-	printf("%s %s -t [null|loop|stripe] [-q nr_queues] [-d depth] [-n dev_id]\n",
+	printf("%s %s -t [null|loop|stripe|fault_inject] [-q nr_queues] [-d depth] [-n dev_id]\n",
 			exe, recovery ? "recover" : "add");
 	printf("\t[--foreground] [--quiet] [-z] [--debug_mask mask] [-r 0|1 ] [-g 0|1]\n");
 	printf("\t[-e 0|1 ] [-i 0|1]\n");

tools/testing/selftests/ublk/kublk.h

@@ -68,6 +68,11 @@ struct stripe_ctx {
 	unsigned int    chunk_size;
 };
 
+struct fault_inject_ctx {
+	/* fault_inject */
+	unsigned long   delay_us;
+};
+
 struct dev_ctx {
 	char tgt_type[16];
 	unsigned long flags;
@@ -81,6 +86,9 @@ struct dev_ctx {
 	unsigned int	fg:1;
 	unsigned int	recovery:1;
 
+	/* fault_inject */
+	long long	delay_us;
+
 	int _evtfd;
 	int _shmid;
 
@@ -89,6 +97,7 @@ struct dev_ctx {
 
 	union {
 		struct stripe_ctx 	stripe;
+		struct fault_inject_ctx fault_inject;
 	};
 };
 
@@ -384,6 +393,7 @@ static inline int ublk_queue_use_zc(const struct ublk_queue *q)
 extern const struct ublk_tgt_ops null_tgt_ops;
 extern const struct ublk_tgt_ops loop_tgt_ops;
 extern const struct ublk_tgt_ops stripe_tgt_ops;
+extern const struct ublk_tgt_ops fault_inject_tgt_ops;
 
 void backing_file_tgt_deinit(struct ublk_dev *dev);
 int backing_file_tgt_init(struct ublk_dev *dev);

tools/testing/selftests/ublk/test_generic_06.sh

@@ -0,0 +1,41 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+. "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
+
+TID="generic_06"
+ERR_CODE=0
+
+_prep_test "fault_inject" "fast cleanup when all I/Os of one hctx are in server"
+
+# configure ublk server to sleep 2s before completing each I/O
+dev_id=$(_add_ublk_dev -t fault_inject -q 2 -d 1 --delay_us 2000000)
+_check_add_dev $TID $?
+
+STARTTIME=${SECONDS}
+
+dd if=/dev/urandom of=/dev/ublkb${dev_id} oflag=direct bs=4k count=1 status=none > /dev/null 2>&1 &
+dd_pid=$!
+
+__ublk_kill_daemon ${dev_id} "DEAD"
+
+wait $dd_pid
+dd_exitcode=$?
+
+ENDTIME=${SECONDS}
+ELAPSED=$(($ENDTIME - $STARTTIME))
+
+# assert that dd sees an error and exits quickly after ublk server is
+# killed. previously this relied on seeing an I/O timeout and so would
+# take ~30s
+if [ $dd_exitcode -eq 0 ]; then
+        echo "dd unexpectedly exited successfully!"
+        ERR_CODE=255
+fi
+if [ $ELAPSED -ge 5 ]; then
+        echo "dd took $ELAPSED seconds to exit (>= 5s tolerance)!"
+        ERR_CODE=255
+fi
+
+_cleanup_test "fault_inject"
+_show_result $TID $ERR_CODE