[PATCH blktests] tests/block/031: Add a test for sharing a tag set across hardware queues

[PATCH blktests] tests/block/031: Add a test for sharing a tag set across hardware queues

[Bart Van Assche]

Support for sharing a tag set across hardware queues has been added
recently to the Linux kernel. See also the BLK_MQ_F_TAG_HCTX_SHARED flag,
Linux kernel commit 32bc15afed04 ("blk-mq: Facilitate a shared sbitmap per
tagset"; v5.10) and commit 0905053bdb5b ("null_blk: Support shared tag
bitmap"; v5.10). Add a test that triggers the shared tag set code in the
block layer core.

Cc: John Garry <john.garry@huawei.com>
Cc: Don Brace<don.brace@microsemi.com>
Cc: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 tests/block/031     | 43 +++++++++++++++++++++++++++++++++++++++++++
 tests/block/031.out |  1 +
 2 files changed, 44 insertions(+)
 create mode 100755 tests/block/031
 create mode 100644 tests/block/031.out

diff --git a/tests/block/031 b/tests/block/031
new file mode 100755
index 000000000000..7940f9fa1114
--- /dev/null
+++ b/tests/block/031
@@ -0,0 +1,43 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright 2021 Google LLC
+#
+# Test support for sharing a tag set across hardware queues ("host tag set").
+
+. tests/block/rc
+. common/null_blk
+
+DESCRIPTION="test host tagset support"
+TIMED=1
+
+requires() {
+	_have_null_blk && _have_module_param null_blk shared_tag_bitmap
+}
+
+test() {
+	local fio_status bs=512
+
+	: "${TIMEOUT:=30}"
+	if ! _init_null_blk nr_devices=0 queue_mode=2 shared_tag_bitmap=1; then
+		echo "Loading null_blk failed"
+		return 1
+	fi
+	if ! _configure_null_blk nullb0 completion_nsec=0 blocksize=$bs size=1\
+	     submit_queues="$(nproc)" memory_backed=1 power=1; then
+		echo "Configuring null_blk failed"
+		return 1
+	fi
+	fio --verify=md5 --rw=randwrite --bs=$bs --loops=$((10**6)) \
+		--iodepth=64 --group_reporting --sync=1 --direct=1 \
+		--ioengine=libaio --runtime="${TIMEOUT}" --thread \
+		--name=block-031 --filename=/dev/nullb0 \
+		--output="${RESULTS_DIR}/block/fio-output-031.txt" \
+		>>"$FULL"
+	fio_status=$?
+	rmdir /sys/kernel/config/nullb/nullb0
+	_exit_null_blk
+	case $fio_status in
+		0) echo "Passed";;
+		*) echo "Failed (fio status = $fio_status)";;
+	esac
+}
diff --git a/tests/block/031.out b/tests/block/031.out
new file mode 100644
index 000000000000..863339fb8ced
--- /dev/null
+++ b/tests/block/031.out
@@ -0,0 +1 @@
+Passed

[PATCH blktests v2] rdma: Use rdma link instead of /sys/class/infiniband/*/parent

[Bart Van Assche]

The approach of verifying whether or not an RDMA interface is associated
with the rdma_rxe interface by looking up its parent device is deprecated
and will be removed soon from the Linux kernel. Hence this patch that uses
the rdma link command instead.

Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: Yi Zhang <yi.zhang@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
v2: Added a _have_program check for 'rdma' as requested by Omar.
---
 common/multipath-over-rdma | 111 +++++++++++--------------------------
 tests/nvmeof-mp/rc         |   2 +-
 tests/srp/rc               |  12 +---
 3 files changed, 35 insertions(+), 90 deletions(-)

diff --git a/common/multipath-over-rdma b/common/multipath-over-rdma
index 9d9d2b27af83..9e68189b56fd 100644
--- a/common/multipath-over-rdma
+++ b/common/multipath-over-rdma
@@ -57,6 +57,9 @@ _multipathd_version_ge() {
 }
 
 get_ipv4_addr() {
+	if [ ! -e "/sys/class/net/$1" ]; then
+		echo "get_ipv4_addr(): $1 is not a network interface" 1>&2
+	fi
 	ip -4 -o addr show dev "$1" |
 		sed -n 's/.*[[:blank:]]inet[[:blank:]]*\([^[:blank:]/]*\).*/\1/p'
 }
@@ -76,41 +79,11 @@ is_number() {
 	[ "$1" -eq "0$1" ] 2>/dev/null
 }
 
-# Check whether a device is an RDMA device. An example argument:
-# /sys/devices/pci0000:00/0000:00:03.0/0000:04:00.0
-is_rdma_device() {
-	local d i inode1 inode2
-
-	inode1=$(stat -c %i "$1")
-	# echo "inode1 = $inode1"
-	for i in /sys/class/infiniband/*; do
-		d=/sys/class/infiniband/"$(readlink "$i")"
-		d=$(dirname "$(dirname "$d")")
-		inode2=$(stat -c %i "$d")
-		# echo "inode2 = $inode2"
-		if [ "$inode1" = "$inode2" ]; then
-			return
-		fi
-	done
-	false
-}
-
 # Lists RDMA capable network interface names, e.g. ib0 ib1.
 rdma_network_interfaces() {
-	(
-		cd /sys/class/net &&
-			for i in *; do
-				[ -e "$i" ] || continue
-				# Skip IPoIB (ARPHRD_INFINIBAND) network
-				# interfaces.
-				[ "$(<"$i"/type)" = 32 ] && continue
-				[ -L "$i/device" ] || continue
-				d=$(readlink "$i/device" 2>/dev/null)
-				if [ -n "$d" ] && is_rdma_device "$i/$d"; then
-					echo "$i"
-				fi
-			done
-	)
+	rdma link show |
+		sed -n 's/^.*[[:blank:]]netdev[[:blank:]]\+\([^[:blank:]]*\)[[:blank:]]*/\1/p' |
+		sort -u
 }
 
 # Check whether any stacked block device holds block device $1. If so, echo
@@ -411,47 +384,36 @@ all_primary_gids() {
 	done
 }
 
-# Check whether or not an rdma_rxe instance has been associated with network
-# interface $1.
-has_rdma_rxe() {
-	local f
-
-	for f in /sys/class/infiniband/*/parent; do
-		if [ -e "$f" ] && [ "$(<"$f")" = "$1" ]; then
-			return 0
-		fi
-	done
-
-	return 1
+# Check whether or not an rdma_rxe or siw instance has been associated with
+# network interface $1.
+has_soft_rdma() {
+	rdma link | grep -q " netdev $1[[:blank:]]*\$"
 }
 
 # Load the rdma_rxe or siw kernel module and associate it with all network
 # interfaces.
 start_soft_rdma() {
+	local type
+
 	{
 	if [ -n "$use_siw" ]; then
 		modprobe siw || return $?
-		(
-			cd /sys/class/net &&
-				for i in *; do
-					[ -e "$i" ] || continue
-					[ -e "/sys/class/infiniband/${i}_siw" ] && continue
-					rdma link add "${i}_siw" type siw netdev "$i" ||
-						echo "Failed to bind the siw driver to $i"
-				done
-		)
+		type=siw
 	else
 		modprobe rdma_rxe || return $?
-		(
-			cd /sys/class/net &&
-				for i in *; do
-					if [ -e "$i" ] && ! has_rdma_rxe "$i"; then
-						echo "$i" > /sys/module/rdma_rxe/parameters/add ||
-							echo "Failed to bind the rdma_rxe driver to $i"
-					fi
-				done
-		)
+		type=rxe
 	fi
+	(
+		cd /sys/class/net &&
+			for i in *; do
+				[ -e "$i" ] || continue
+				[ "$i" = "lo" ] && continue
+				[ "$(<"$i/addr_len")" = 6 ] || continue
+				has_soft_rdma "$i" && continue
+				rdma link add "${i}_$type" type $type netdev "$i" ||
+				echo "Failed to bind the $type driver to $i"
+			done
+	)
 	} >>"$FULL"
 }
 
@@ -459,27 +421,16 @@ start_soft_rdma() {
 # unload the rdma_rxe kernel module.
 stop_soft_rdma() {
 	{
-	(
-		cd /sys/class/net &&
-			for i in *; do
-				if [ -e "$i" ] && has_rdma_rxe "$i"; then
-					{ echo "$i" > /sys/module/rdma_rxe/parameters/remove; } \
-						2>/dev/null
-				fi
-			done
-	)
+	rdma link |
+		sed -n 's,^link[[:blank:]]*\([^/]*\)/.* netdev .*,\1,p' |
+		while read -r i; do
+		      echo "$i ..."
+		      rdma link del "${i}" || echo "Failed to remove ${i}"
+		done
 	if ! unload_module rdma_rxe 10; then
 		echo "Unloading rdma_rxe failed"
 		return 1
 	fi
-	(
-		cd /sys/class/net &&
-			for i in *_siw; do
-				[ -e "$i" ] || continue
-				rdma link del "${i}" ||
-					echo "Failed to unbind the siw driver from ${i%_siw}"
-			done
-	)
 	if ! unload_module siw 10; then
 		echo "Unloading siw failed"
 		return 1
diff --git a/tests/nvmeof-mp/rc b/tests/nvmeof-mp/rc
index ab7770f6aac0..0a12825e2c11 100755
--- a/tests/nvmeof-mp/rc
+++ b/tests/nvmeof-mp/rc
@@ -42,7 +42,7 @@ and multipathing has been enabled in the nvme_core kernel module"
 	)
 	_have_modules "${required_modules[@]}" || return
 
-	for p in mkfs.ext4 mkfs.xfs multipath multipathd pidof; do
+	for p in mkfs.ext4 mkfs.xfs multipath multipathd pidof rdma; do
 		_have_program "$p" || return
 	done
 
diff --git a/tests/srp/rc b/tests/srp/rc
index 700cd71ea155..2986bfd5439d 100755
--- a/tests/srp/rc
+++ b/tests/srp/rc
@@ -59,7 +59,8 @@ group_requires() {
 	)
 	_have_modules "${required_modules[@]}" || return
 
-	for p in mkfs.ext4 mkfs.xfs multipath multipathd pidof sg_reset; do
+	for p in mkfs.ext4 mkfs.xfs multipath multipathd pidof rdma \
+		 sg_reset; do
 		_have_program "$p" || return
 	done
 
@@ -142,14 +143,7 @@ do_ib_cm_login() {
 }
 
 rdma_dev_to_net_dev() {
-	local b d rdma_dev=$1
-
-	b=/sys/class/infiniband/$rdma_dev/parent
-	if [ -e "$b" ]; then
-		echo "$(<"$b")"
-	else
-		echo "${rdma_dev%_siw}"
-	fi
+	rdma link show "$1/1" | sed 's/.* netdev //;s/[[:blank:]]*$//'
 }
 
 # Tell the SRP initiator to log in to an SRP target using the RDMA/CM.

Re: [PATCH blktests v2] rdma: Use rdma link instead of /sys/class/infiniband/*/parent

[Bart Van Assche]

On 3/28/21 4:12 PM, Bart Van Assche wrote:
> The approach of verifying whether or not an RDMA interface is associated
> with the rdma_rxe interface by looking up its parent device is deprecated
> and will be removed soon from the Linux kernel. Hence this patch that uses
> the rdma link command instead.

Please ignore this patch - it has already been applied.

Bart.

Re: [PATCH blktests] tests/block/031: Add a test for sharing a tag set across hardware queues

[Omar Sandoval]

On Sun, Mar 28, 2021 at 04:12:09PM -0700, Bart Van Assche wrote:
> Support for sharing a tag set across hardware queues has been added
> recently to the Linux kernel. See also the BLK_MQ_F_TAG_HCTX_SHARED flag,
> Linux kernel commit 32bc15afed04 ("blk-mq: Facilitate a shared sbitmap per
> tagset"; v5.10) and commit 0905053bdb5b ("null_blk: Support shared tag
> bitmap"; v5.10). Add a test that triggers the shared tag set code in the
> block layer core.
> 
> Cc: John Garry <john.garry@huawei.com>
> Cc: Don Brace<don.brace@microsemi.com>
> Cc: Douglas Gilbert <dgilbert@interlog.com>
> Signed-off-by: Bart Van Assche <bvanassche@acm.org>
> ---
>  tests/block/031     | 43 +++++++++++++++++++++++++++++++++++++++++++
>  tests/block/031.out |  1 +
>  2 files changed, 44 insertions(+)
>  create mode 100755 tests/block/031
>  create mode 100644 tests/block/031.out

Thanks, Bart, applied.