From: Jens Axboe <axboe@kernel.dk>
To: fio@vger.kernel.org
Subject: Recent changes (master)
Date: Sat, 30 Jan 2021 06:00:01 -0700 (MST) [thread overview]
Message-ID: <20210130130001.95E371BC0134@kernel.dk> (raw)
The following changes since commit 119bb82143bd6c4a577c135ece4ed6b702443f50:
Merge branch 'fio-fix-detecting-libpmem' of https://github.com/ldorau/fio (2021-01-27 09:51:01 -0700)
are available in the Git repository at:
git://git.kernel.dk/fio.git master
for you to fetch changes up to 2ef3c1b02473a14bf7b8b52e28d0cdded9c5cc9a:
zbd: relocate Coverity annotation (2021-01-29 22:06:49 -0700)
----------------------------------------------------------------
Aravind Ramesh (1):
zbd: initialize sectors with data at start time
Dmitry Fomichev (26):
zbd: return ENOMEM if zone buffer allocation fails
zbd: use zbd_zone_nr() more actively in the code
zbd: add get_zone() helper function
zbd: introduce zone_unlock()
zbd: engines/libzbc: don't fail on assert for offline zones
zbd: remove dependency on zone type during i/o
zbd: skip offline zones in zbd_convert_to_open_zone()
zbd: avoid zone buffer overrun
zbd: don't unlock zone mutex after verify replay
zbd: use zone_lock() in zbd_process_swd()
zbd: don't log "zone nnnn is not open" message
zbd: handle conventional start zone in zbd_convert_to_open_zone()
zbd: improve replay range validation
engines/libzbc: enable block backend
zbd: avoid failing assertion in zbd_convert_to_open_zone()
zbd: set thread errors in zbd_adjust_block()
t/zbd: check for error in test #2
t/zbd: add run-tests-against-nullb script
t/zbd: add an option to bail on a failed test
t/zbd: prevent test #31 from looping
t/zbd: add checks for offline zone condition
t/zbd: add test #54 to exercise ZBD verification
t/zbd: show elapsed time in test-zbd-support
t/zbd: increase timeout in test #48
t/zbd: avoid looping on invalid command line options
zbd: relocate Coverity annotation
Jens Axboe (1):
zbd: fix 32-bit compile warnings for logging
Shin'ichiro Kawasaki (12):
zbd: do not lock conventional zones on I/O adjustment
zbd: do not set zbd handlers for conventional zones
zbd: count sectors with data for write pointer zones
zbd: initialize min_zone and max_zone for all zone types
zbd: disable crossing from conventional to sequential zones
t/zbd: add -t option to run-tests-against-nullb
t/zbd: skip tests when test prerequisites are not met
t/zbd: skip tests that need too many sequential zones
t/zbd: test that conventional zones are not locked during random i/o
t/zbd: test that zone_reset_threshold calculation is correct
t/zbd: test random I/O direction in all-conventional case
t/zbd: fix wrong units in test case #37
Makefile | 5 +-
engines/libzbc.c | 5 +-
oslib/linux-blkzoned.c | 2 +-
t/run-fio-tests.py | 8 +-
t/zbd/functions | 56 +++++-
t/zbd/run-tests-against-nullb | 354 +++++++++++++++++++++++++++++++++
t/zbd/run-tests-against-regular-nullb | 27 ---
t/zbd/run-tests-against-zoned-nullb | 53 -----
t/zbd/test-zbd-support | 299 ++++++++++++++++++++++++----
zbd.c | 357 +++++++++++++++++++++-------------
zbd.h | 5 +
11 files changed, 911 insertions(+), 260 deletions(-)
create mode 100755 t/zbd/run-tests-against-nullb
delete mode 100755 t/zbd/run-tests-against-regular-nullb
delete mode 100755 t/zbd/run-tests-against-zoned-nullb
---
Diff of recent changes:
diff --git a/Makefile b/Makefile
index f74e59e1..612344d1 100644
--- a/Makefile
+++ b/Makefile
@@ -626,9 +626,10 @@ fulltest:
make -j && \
sudo make install) \
fi && \
- sudo t/zbd/run-tests-against-regular-nullb && \
+ sudo t/zbd/run-tests-against-nullb -s 1 && \
if [ -e /sys/module/null_blk/parameters/zoned ]; then \
- sudo t/zbd/run-tests-against-zoned-nullb; \
+ sudo t/zbd/run-tests-against-nullb -s 2; \
+ sudo t/zbd/run-tests-against-nullb -s 4; \
fi
install: $(PROGS) $(SCRIPTS) $(ENGS_OBJS) tools/plot/fio2gnuplot.1 FORCE
diff --git a/engines/libzbc.c b/engines/libzbc.c
index 4b900233..2aacf7bb 100644
--- a/engines/libzbc.c
+++ b/engines/libzbc.c
@@ -86,7 +86,8 @@ static int libzbc_open_dev(struct thread_data *td, struct fio_file *f,
return -ENOMEM;
ret = zbc_open(f->file_name,
- flags | ZBC_O_DRV_SCSI | ZBC_O_DRV_ATA, &ld->zdev);
+ flags | ZBC_O_DRV_BLOCK | ZBC_O_DRV_SCSI | ZBC_O_DRV_ATA,
+ &ld->zdev);
if (ret) {
log_err("%s: zbc_open() failed, err=%d\n",
f->file_name, ret);
@@ -283,7 +284,7 @@ static int libzbc_report_zones(struct thread_data *td, struct fio_file *f,
default:
/* Treat all these conditions as offline (don't use!) */
zbdz->cond = ZBD_ZONE_COND_OFFLINE;
- break;
+ zbdz->wp = zbdz->start;
}
}
diff --git a/oslib/linux-blkzoned.c b/oslib/linux-blkzoned.c
index 0a8a577a..f37c67fc 100644
--- a/oslib/linux-blkzoned.c
+++ b/oslib/linux-blkzoned.c
@@ -203,7 +203,7 @@ int blkzoned_report_zones(struct thread_data *td, struct fio_file *f,
default:
/* Treat all these conditions as offline (don't use!) */
z->cond = ZBD_ZONE_COND_OFFLINE;
- break;
+ z->wp = z->start;
}
}
diff --git a/t/run-fio-tests.py b/t/run-fio-tests.py
index e5c2f17c..a59cdfe0 100755
--- a/t/run-fio-tests.py
+++ b/t/run-fio-tests.py
@@ -879,8 +879,8 @@ TEST_LIST = [
{
'test_id': 1007,
'test_class': FioExeTest,
- 'exe': 't/zbd/run-tests-against-regular-nullb',
- 'parameters': None,
+ 'exe': 't/zbd/run-tests-against-nullb',
+ 'parameters': ['-s', '1'],
'success': SUCCESS_DEFAULT,
'requirements': [Requirements.linux, Requirements.zbd,
Requirements.root],
@@ -888,8 +888,8 @@ TEST_LIST = [
{
'test_id': 1008,
'test_class': FioExeTest,
- 'exe': 't/zbd/run-tests-against-zoned-nullb',
- 'parameters': None,
+ 'exe': 't/zbd/run-tests-against-nullb',
+ 'parameters': ['-s', '2'],
'success': SUCCESS_DEFAULT,
'requirements': [Requirements.linux, Requirements.zbd,
Requirements.root, Requirements.zoned_nullb],
diff --git a/t/zbd/functions b/t/zbd/functions
index 1a64a215..40ffe1de 100644
--- a/t/zbd/functions
+++ b/t/zbd/functions
@@ -71,7 +71,7 @@ first_sequential_zone() {
if [ -n "${blkzone}" ] && [ ! -n "${use_libzbc}" ]; then
${blkzone} report "$dev" |
- sed -n 's/^[[:blank:]]*start:[[:blank:]]\([0-9a-zA-Z]*\),[[:blank:]]len[[:blank:]]\([0-9a-zA-Z]*\),.*type:[[:blank:]]2(.*/\1 \2/p' |
+ sed -n 's/^[[:blank:]]*start:[[:blank:]]\([0-9a-zA-Z]*\),[[:blank:]]len[[:blank:]]\([0-9a-zA-Z]*\),.*zcond:\(14\|[[:blank:]][0-4]\)(.*type:[[:blank:]]\([2]\)(.*/\1 \2/p' |
{
read -r starting_sector length &&
# Convert from hex to decimal
@@ -79,7 +79,7 @@ first_sequential_zone() {
}
else
${zbc_report_zones} "$dev" |
- sed -n 's/^Zone [0-9]*: type 0x2 .*, sector \([0-9]*\), \([0-9]*\) sectors,.*$/\1 \2/p' |
+ sed -n 's/^Zone [0-9]*: type 0x2 .*,[[:blank:]]cond[[:blank:]]0x[0-4e][[:blank:]].*, sector \([0-9]*\), \([0-9]*\) sectors.*$/\1 \2/p' |
head -n1
fi
}
@@ -121,6 +121,58 @@ total_zone_capacity() {
echo $((capacity * 512))
}
+# Reports the starting sector and length of the first zone of device $1
+# that is not in offline (or similar) condition.
+first_online_zone() {
+ local dev=$1
+
+ if [ -z "$is_zbd" ]; then
+ echo 0
+ return
+ fi
+
+ if [ -n "${blkzone}" ] && [ ! -n "${use_libzbc}" ]; then
+ ${blkzone} report "$dev" |
+ sed -n 's/^[[:blank:]]*start:[[:blank:]]\([0-9a-zA-Z]*\),[[:blank:]]len[[:blank:]]\([0-9a-zA-Z]*\),.*zcond:\(14\|[[:blank:]][0-4]\)(.*type:[[:blank:]][12](.*/\1/p' |
+ head -n1 |
+ {
+ read -r starting_sector &&
+ # Convert from hex to decimal
+ echo $((starting_sector))
+ }
+ else
+ ${zbc_report_zones} "$dev" |
+ sed -n 's/^Zone[[:blank:]][0-9]*:[[:blank:]]type[[:blank:]]0x[12][[:blank:]].*,[[:blank:]]cond[[:blank:]]0x[0-4e][[:blank:]].*,[[:blank:]]sector[[:blank:]]\([0-9]*\),.*$/\1/p' |
+ head -n1
+ fi
+}
+
+# Reports the starting sector and length of the last zone of device $1
+# that is not in offline (or similar) condition.
+last_online_zone() {
+ local dev=$1
+
+ if [ -z "$is_zbd" ]; then
+ echo 0
+ return
+ fi
+
+ if [ -n "${blkzone}" ] && [ ! -n "${use_libzbc}" ]; then
+ ${blkzone} report "$dev" |
+ sed -n 's/^[[:blank:]]*start:[[:blank:]]\([0-9a-zA-Z]*\),[[:blank:]]len[[:blank:]]\([0-9a-zA-Z]*\),.*zcond:\(14\|[[:blank:]][0-4]\)(.*type:[[:blank:]][12](.*/\1/p' |
+ tail -1 |
+ {
+ read -r starting_sector &&
+ # Convert from hex to decimal
+ echo $((starting_sector))
+ }
+ else
+ ${zbc_report_zones} "$dev" |
+ sed -n 's/^Zone[[:blank:]][0-9]*:[[:blank:]]type[[:blank:]]0x[12][[:blank:]].*,[[:blank:]]cond[[:blank:]]0x[0-4e][[:blank:]].*,[[:blank:]]sector[[:blank:]]\([0-9]*\),.*$/\1/p' |
+ tail -1
+ fi
+}
+
max_open_zones() {
local dev=$1
diff --git a/t/zbd/run-tests-against-nullb b/t/zbd/run-tests-against-nullb
new file mode 100755
index 00000000..db901179
--- /dev/null
+++ b/t/zbd/run-tests-against-nullb
@@ -0,0 +1,354 @@
+#!/bin/bash
+#
+# Copyright (C) 2020 Western Digital Corporation or its affiliates.
+#
+# This file is released under the GPL.
+#
+# Run t/zbd/test-zbd-support script against a variety of conventional,
+# zoned and mixed zone configurations.
+#
+
+usage()
+{
+ echo "This script runs the tests from t/zbd/test-zbd-support script"
+ echo "against a nullb device in a variety of conventional and zoned"
+ echo "configurations."
+ echo "Usage: ${0} [OPTIONS]"
+ echo "Options:"
+ echo -e "\t-h Show this message."
+ echo -e "\t-L List the device layouts for every section without running"
+ echo -e "\t tests."
+ echo -e "\t-s <#section> Only run the section with the given number."
+ echo -e "\t-l Use libzbc ioengine to run the tests."
+ echo -e "\t-t <#test> Only run the test with the given number in every section."
+ echo -e "\t-o <max_open_zones> Specify MaxOpen value, (${set_max_open} by default)."
+ echo -e "\t-n <#number of runs> Set the number of times to run the entire suite "
+ echo -e "\t or an individual section/test."
+ echo -e "\t-q Quit t/zbd/test-zbd-support run after any failed test."
+ echo -e "\t-r Remove the /dev/nullb0 device that may still exist after"
+ echo -e "\t running this script."
+ exit 1
+}
+
+cleanup_nullb()
+{
+ for d in /sys/kernel/config/nullb/*; do [ -d "$d" ] && rmdir "$d"; done
+ modprobe -r null_blk
+ modprobe null_blk nr_devices=0 || exit $?
+ for d in /sys/kernel/config/nullb/*; do
+ [ -d "$d" ] && rmdir "$d"
+ done
+ modprobe -r null_blk
+ [ -e /sys/module/null_blk ] && exit $?
+}
+
+create_nullb()
+{
+ modprobe null_blk nr_devices=0 &&
+ cd /sys/kernel/config/nullb &&
+ mkdir nullb0 &&
+ cd nullb0 || return $?
+}
+
+configure_nullb()
+{
+ echo 0 > completion_nsec &&
+ echo ${dev_blocksize} > blocksize &&
+ echo ${dev_size} > size &&
+ echo 1 > memory_backed || return $?
+
+ if ((conv_pcnt < 100)); then
+ echo 1 > zoned &&
+ echo "${zone_size}" > zone_size || return $?
+
+ if ((zone_capacity < zone_size)); then
+ if ((!zcap_supported)); then
+ echo "null_blk does not support zone capacity"
+ return 2
+ fi
+ echo "${zone_capacity}" > zone_capacity
+ fi
+ if ((conv_pcnt)); then
+ if ((!conv_supported)); then
+ echo "null_blk does not support conventional zones"
+ return 2
+ fi
+ nr_conv=$((dev_size/zone_size*conv_pcnt/100))
+ echo "${nr_conv}" > zone_nr_conv
+ fi
+ fi
+
+ echo 1 > power || return $?
+ return 0
+}
+
+show_nullb_config()
+{
+ if ((conv_pcnt < 100)); then
+ echo " $(printf "Zoned Device, %d%% Conventional Zones (%d)" \
+ ${conv_pcnt} ${nr_conv})"
+ echo " $(printf "Zone Size: %d MB" ${zone_size})"
+ echo " $(printf "Zone Capacity: %d MB" ${zone_capacity})"
+ if ((max_open)); then
+ echo " $(printf "Max Open: %d Zones" ${max_open})"
+ else
+ echo " Max Open: Unlimited Zones"
+ fi
+ else
+ echo " Non-zoned Device"
+ fi
+}
+
+#
+# Test sections.
+#
+# Fully conventional device.
+section1()
+{
+ conv_pcnt=100
+ max_open=0
+}
+
+# Zoned device with no conventional zones, ZCAP == ZSIZE, unlimited MaxOpen.
+section2()
+{
+ conv_pcnt=0
+ zone_size=1
+ zone_capacity=1
+ max_open=0
+}
+
+# Zoned device with no conventional zones, ZCAP < ZSIZE, unlimited MaxOpen.
+section3()
+{
+ conv_pcnt=0
+ zone_size=4
+ zone_capacity=3
+ max_open=0
+}
+
+# Zoned device with mostly sequential zones, ZCAP == ZSIZE, unlimited MaxOpen.
+section4()
+{
+ conv_pcnt=10
+ zone_size=1
+ zone_capacity=1
+ max_open=0
+}
+
+# Zoned device with mostly sequential zones, ZCAP < ZSIZE, unlimited MaxOpen.
+section5()
+{
+ conv_pcnt=10
+ zone_size=4
+ zone_capacity=3
+ max_open=0
+}
+
+# Zoned device with mostly conventional zones, ZCAP == ZSIZE, unlimited MaxOpen.
+section6()
+{
+ conv_pcnt=66
+ zone_size=1
+ zone_capacity=1
+ max_open=0
+}
+
+# Zoned device with mostly conventional zones, ZCAP < ZSIZE, unlimited MaxOpen.
+section7()
+{
+ dev_size=2048
+ conv_pcnt=66
+ zone_size=4
+ zone_capacity=3
+ max_open=0
+}
+
+# Zoned device with no conventional zones, ZCAP == ZSIZE, limited MaxOpen.
+section8()
+{
+ dev_size=1024
+ conv_pcnt=0
+ zone_size=1
+ zone_capacity=1
+ max_open=${set_max_open}
+ zbd_test_opts+=("-o ${max_open}")
+}
+
+# Zoned device with no conventional zones, ZCAP < ZSIZE, limited MaxOpen.
+section9()
+{
+ conv_pcnt=0
+ zone_size=4
+ zone_capacity=3
+ max_open=${set_max_open}
+ zbd_test_opts+=("-o ${max_open}")
+}
+
+# Zoned device with mostly sequential zones, ZCAP == ZSIZE, limited MaxOpen.
+section10()
+{
+ conv_pcnt=10
+ zone_size=1
+ zone_capacity=1
+ max_open=${set_max_open}
+ zbd_test_opts+=("-o ${max_open}")
+}
+
+# Zoned device with mostly sequential zones, ZCAP < ZSIZE, limited MaxOpen.
+section11()
+{
+ conv_pcnt=10
+ zone_size=4
+ zone_capacity=3
+ max_open=${set_max_open}
+ zbd_test_opts+=("-o ${max_open}")
+}
+
+# Zoned device with mostly conventional zones, ZCAP == ZSIZE, limited MaxOpen.
+section12()
+{
+ conv_pcnt=66
+ zone_size=1
+ zone_capacity=1
+ max_open=${set_max_open}
+ zbd_test_opts+=("-o ${max_open}")
+}
+
+# Zoned device with mostly conventional zones, ZCAP < ZSIZE, limited MaxOpen.
+section13()
+{
+ dev_size=2048
+ conv_pcnt=66
+ zone_size=4
+ zone_capacity=3
+ max_open=${set_max_open}
+ zbd_test_opts+=("-o ${max_open}")
+}
+
+#
+# Entry point.
+#
+SECONDS=0
+scriptdir="$(cd "$(dirname "$0")" && pwd)"
+sections=()
+zcap_supported=1
+conv_supported=1
+list_only=0
+dev_size=1024
+dev_blocksize=4096
+set_max_open=8
+zbd_test_opts=()
+libzbc=0
+num_of_runs=1
+test_case=0
+quit_on_err=0
+
+while (($#)); do
+ case "$1" in
+ -s) sections+=("$2"); shift; shift;;
+ -o) set_max_open="${2}"; shift; shift;;
+ -L) list_only=1; shift;;
+ -r) cleanup_nullb; exit 0;;
+ -l) libzbc=1; shift;;
+ -n) num_of_runs="${2}"; shift; shift;;
+ -t) test_case="${2}"; shift; shift;;
+ -q) quit_on_err=1; shift;;
+ -h) usage; break;;
+ --) shift; break;;
+ *) usage; exit 1;;
+ esac
+done
+
+if [ "${#sections[@]}" = 0 ]; then
+ readarray -t sections < <(declare -F | grep "section[0-9]*" | tr -c -d "[:digit:]\n" | sort -n)
+fi
+
+cleanup_nullb
+
+#
+# Test creating null_blk device and check if newer features are supported
+#
+if ! eval "create_nullb"; then
+ echo "can't create nullb"
+ exit 1
+fi
+if ! cat /sys/kernel/config/nullb/features | grep -q zone_capacity; then
+ zcap_supported=0
+fi
+if ! cat /sys/kernel/config/nullb/features | grep -q zone_nr_conv; then
+ conv_supported=0
+fi
+
+rc=0
+test_rc=0
+intr=0
+run_nr=1
+trap 'kill ${zbd_test_pid}; intr=1' SIGINT
+
+while ((run_nr <= $num_of_runs)); do
+ echo -e "\nRun #$run_nr:"
+ for section_number in "${sections[@]}"; do
+ cleanup_nullb
+ echo "---------- Section $(printf "%02d" $section_number) ----------"
+ if ! eval "create_nullb"; then
+ echo "error creating nullb"
+ exit 1
+ fi
+ zbd_test_opts=()
+ if ((test_case)); then
+ zbd_test_opts+=("-t" "${test_case}")
+ fi
+ if ((quit_on_err)); then
+ zbd_test_opts+=("-q")
+ fi
+ section$section_number
+ configure_nullb
+ rc=$?
+ ((rc == 2)) && continue
+ if ((rc)); then
+ echo "can't set up nullb for section $(printf "%02d" $section_number)"
+ exit 1
+ fi
+ show_nullb_config
+ if ((libzbc)); then
+ if ((zone_capacity < zone_size)); then
+ echo "libzbc doesn't support zone capacity, skipping section $(printf "%02d" $section_number)"
+ continue
+ fi
+ if ((conv_pcnt == 100)); then
+ echo "libzbc only supports zoned devices, skipping section $(printf "%02d" $section_number)"
+ continue
+ fi
+ zbd_test_opts+=("-l")
+ fi
+ cd "${scriptdir}"
+ ((intr)) && exit 1
+ ((list_only)) && continue
+
+ ./test-zbd-support ${zbd_test_opts[@]} /dev/nullb0 &
+ zbd_test_pid=$!
+ if kill -0 "${zbd_test_pid}"; then
+ wait "${zbd_test_pid}"
+ test_rc=$?
+ else
+ echo "can't run ZBD tests"
+ exit 1
+ fi
+ ((intr)) && exit 1
+ if (($test_rc)); then
+ rc=1
+ ((quit_on_err)) && break
+ fi
+ done
+
+ ((rc && quit_on_err)) && break
+ run_nr=$((run_nr + 1))
+done
+
+if ((!list_only)); then
+ echo "--------------------------------"
+ echo "Total run time: $(TZ=UTC0 printf "%(%H:%M:%S)T\n" $(( SECONDS )) )"
+fi
+
+exit $rc
diff --git a/t/zbd/run-tests-against-regular-nullb b/t/zbd/run-tests-against-regular-nullb
deleted file mode 100755
index 5b7b4009..00000000
--- a/t/zbd/run-tests-against-regular-nullb
+++ /dev/null
@@ -1,27 +0,0 @@
-#!/bin/bash
-#
-# Copyright (C) 2018 Western Digital Corporation or its affiliates.
-#
-# This file is released under the GPL.
-
-scriptdir="$(cd "$(dirname "$0")" && pwd)"
-
-for d in /sys/kernel/config/nullb/*; do [ -d "$d" ] && rmdir "$d"; done
-modprobe -r null_blk
-modprobe null_blk nr_devices=0 || exit $?
-for d in /sys/kernel/config/nullb/*; do
- [ -d "$d" ] && rmdir "$d"
-done
-modprobe -r null_blk
-[ -e /sys/module/null_blk ] && exit $?
-modprobe null_blk nr_devices=0 &&
- cd /sys/kernel/config/nullb &&
- mkdir nullb0 &&
- cd nullb0 &&
- echo 0 > completion_nsec &&
- echo 4096 > blocksize &&
- echo 1024 > size &&
- echo 1 > memory_backed &&
- echo 1 > power || exit $?
-
-"${scriptdir}"/test-zbd-support "$@" /dev/nullb0
diff --git a/t/zbd/run-tests-against-zoned-nullb b/t/zbd/run-tests-against-zoned-nullb
deleted file mode 100755
index f9c9530c..00000000
--- a/t/zbd/run-tests-against-zoned-nullb
+++ /dev/null
@@ -1,53 +0,0 @@
-#!/bin/bash
-#
-# Copyright (C) 2018 Western Digital Corporation or its affiliates.
-#
-# This file is released under the GPL.
-
-scriptdir="$(cd "$(dirname "$0")" && pwd)"
-
-zone_size=1
-zone_capacity=1
-if [[ ${1} == "-h" ]]; then
- echo "Usage: ${0} [OPTIONS]"
- echo "Options:"
- echo -e "\t-h Show this message."
- echo -e "\t-zone-cap Use null blk with zone capacity less than zone size."
- echo -e "\tany option supported by test-zbd-support script."
- exit 1
-elif [[ ${1} == "-zone-cap" ]]; then
- zone_size=4
- zone_capacity=3
- shift
-fi
-
-for d in /sys/kernel/config/nullb/*; do [ -d "$d" ] && rmdir "$d"; done
-modprobe -r null_blk
-modprobe null_blk nr_devices=0 || exit $?
-for d in /sys/kernel/config/nullb/*; do
- [ -d "$d" ] && rmdir "$d"
-done
-modprobe -r null_blk
-[ -e /sys/module/null_blk ] && exit $?
-modprobe null_blk nr_devices=0 &&
- cd /sys/kernel/config/nullb &&
- mkdir nullb0 &&
- cd nullb0 || exit $?
-
-if ((zone_capacity < zone_size)); then
- if [[ ! -w zone_capacity ]]; then
- echo "null blk does not support zone capacity"
- exit 1
- fi
- echo "${zone_capacity}" > zone_capacity
-fi
-
-echo 1 > zoned &&
- echo "${zone_size}" > zone_size &&
- echo 0 > completion_nsec &&
- echo 4096 > blocksize &&
- echo 1024 > size &&
- echo 1 > memory_backed &&
- echo 1 > power || exit $?
-
-"${scriptdir}"/test-zbd-support "$@" /dev/nullb0
diff --git a/t/zbd/test-zbd-support b/t/zbd/test-zbd-support
index acde3b3a..1658dc25 100755
--- a/t/zbd/test-zbd-support
+++ b/t/zbd/test-zbd-support
@@ -14,6 +14,7 @@ usage() {
echo -e "\t-r Reset all zones before test start"
echo -e "\t-o <max_open_zones> Run fio with max_open_zones limit"
echo -e "\t-t <test #> Run only a single test case with specified number"
+ echo -e "\t-q Quit the test run after any failed test"
echo -e "\t-z Run fio with debug=zbd option"
}
@@ -190,6 +191,64 @@ prep_write() {
reset_zone "${dev}" -1
}
+SKIP_TESTCASE=255
+
+require_scsi_dev() {
+ if ! is_scsi_device "$dev"; then
+ SKIP_REASON="$dev is not a SCSI device"
+ return 1
+ fi
+ return 0
+}
+
+require_conv_zone_bytes() {
+ local req_bytes=${1}
+
+ if ((req_bytes > first_sequential_zone_sector * 512)); then
+ SKIP_REASON="$dev does not have enough conventional zones"
+ return 1
+ fi
+ return 0
+}
+
+require_zbd() {
+ if [[ -z ${is_zbd} ]]; then
+ SKIP_REASON="$dev is not a zoned block device"
+ return 1
+ fi
+ return 0
+}
+
+require_regular_block_dev() {
+ if [[ -n ${is_zbd} ]]; then
+ SKIP_REASON="$dev is not a regular block device"
+ return 1
+ fi
+ return 0
+}
+
+require_seq_zones() {
+ local req_seq_zones=${1}
+ local seq_bytes=$((disk_size - first_sequential_zone_sector * 512))
+
+ if ((req_seq_zones > seq_bytes / zone_size)); then
+ SKIP_REASON="$dev does not have $req_seq_zones sequential zones"
+ return 1
+ fi
+ return 0
+}
+
+require_conv_zones() {
+ local req_c_zones=${1}
+ local conv_bytes=$((first_sequential_zone_sector * 512))
+
+ if ((req_c_zones > conv_bytes / zone_size)); then
+ SKIP_REASON="$dev does not have $req_c_zones conventional zones"
+ return 1
+ fi
+ return 0
+}
+
# Check whether buffered writes are refused.
test1() {
run_fio --name=job1 --filename="$dev" --rw=write --direct=0 --bs=4K \
@@ -221,14 +280,15 @@ test2() {
if [ -z "$is_zbd" ]; then
opts+=("--zonesize=${zone_size}")
fi
- run_fio "${opts[@]}" >> "${logfile}.${test_number}" 2>&1 || return $?
- ! grep -q 'WRITE:' "${logfile}.${test_number}"
+ run_fio "${opts[@]}" >> "${logfile}.${test_number}" 2>&1 && return 1
+ grep -q 'buflen exceeds zone size' "${logfile}.${test_number}"
}
# Run fio against an empty zone. This causes fio to report "No I/O performed".
test3() {
local off opts=() rc
+ require_seq_zones 129 || return $SKIP_TESTCASE
off=$((first_sequential_zone_sector * 512 + 128 * zone_size))
size=$((zone_size))
[ -n "$is_zbd" ] && reset_zone "$dev" $((off / 512))
@@ -246,6 +306,7 @@ test3() {
test4() {
local off opts=()
+ require_seq_zones 130 || return $SKIP_TESTCASE
off=$((first_sequential_zone_sector * 512 + 129 * zone_size))
size=$((zone_size))
[ -n "$is_zbd" ] && reset_zone "$dev" $((off / 512))
@@ -327,10 +388,7 @@ test8() {
test9() {
local size
- if ! is_scsi_device "$dev"; then
- echo "$dev is not a SCSI device" >>"${logfile}.${test_number}"
- return 0
- fi
+ require_scsi_dev || return $SKIP_TESTCASE
prep_write
size=$((4 * zone_size))
@@ -346,10 +404,7 @@ test9() {
test10() {
local size
- if ! is_scsi_device "$dev"; then
- echo "$dev is not a SCSI device" >>"${logfile}.${test_number}"
- return 0
- fi
+ require_scsi_dev || return $SKIP_TESTCASE
prep_write
size=$((4 * zone_size))
@@ -409,18 +464,20 @@ test13() {
# Random write to conventional zones.
test14() {
- local size
+ local off size
+ if ! result=($(first_online_zone "$dev")); then
+ echo "Failed to determine first online zone"
+ exit 1
+ fi
+ off=${result[0]}
prep_write
size=$((16 * 2**20)) # 20 MB
- if [ $size -gt $((first_sequential_zone_sector * 512)) ]; then
- echo "$dev does not have enough sequential zones" \
- >>"${logfile}.${test_number}"
- return 0
- fi
+ require_conv_zone_bytes "${size}" || return $SKIP_TESTCASE
+
run_one_fio_job "$(ioengine "libaio")" --iodepth=64 --rw=randwrite --bs=16K \
--zonemode=zbd --zonesize="${zone_size}" --do_verify=1 \
- --verify=md5 --size=$size \
+ --verify=md5 --offset=$off --size=$size\
>>"${logfile}.${test_number}" 2>&1 || return $?
check_written $((size)) || return $?
check_read $((size)) || return $?
@@ -477,17 +534,26 @@ test16() {
# Random reads and writes in the last zone.
test17() {
- local io off read size written
+ local io off last read size written
off=$(((disk_size / zone_size - 1) * zone_size))
size=$((disk_size - off))
+ if ! last=($(last_online_zone "$dev")); then
+ echo "Failed to determine last online zone"
+ exit 1
+ fi
+ if [[ "$((last * 512))" -lt "$off" ]]; then
+ off=$((last * 512))
+ size=$zone_size
+ fi
if [ -n "$is_zbd" ]; then
reset_zone "$dev" $((off / 512)) || return $?
fi
prep_write
run_one_fio_job "$(ioengine "libaio")" --iodepth=8 --rw=randrw --bs=4K \
--zonemode=zbd --zonesize="${zone_size}" \
- --offset=$off --loops=2 --norandommap=1\
+ --offset=$off --loops=2 --norandommap=1 \
+ --size="$size"\
>>"${logfile}.${test_number}" 2>&1 || return $?
written=$(fio_written <"${logfile}.${test_number}")
read=$(fio_read <"${logfile}.${test_number}")
@@ -604,6 +670,7 @@ test27() {
test28() {
local i jobs=16 off opts
+ require_seq_zones 65 || return $SKIP_TESTCASE
off=$((first_sequential_zone_sector * 512 + 64 * zone_size))
[ -n "$is_zbd" ] && reset_zone "$dev" $((off / 512))
prep_write
@@ -628,6 +695,7 @@ test28() {
test29() {
local i jobs=16 off opts=()
+ require_seq_zones 80 || return $SKIP_TESTCASE
off=$((first_sequential_zone_sector * 512 + 64 * zone_size))
size=$((16*zone_size))
prep_write
@@ -664,12 +732,18 @@ test31() {
local bs inc nz off opts size
prep_write
- # Start with writing 128 KB to 128 sequential zones.
+ # Start with writing 128 KB to max_open_zones sequential zones.
bs=128K
- nz=128
+ nz=$((max_open_zones))
+ if [[ $nz -eq 0 ]]; then
+ nz=128
+ fi
# shellcheck disable=SC2017
inc=$(((disk_size - (first_sequential_zone_sector * 512)) / (nz * zone_size)
* zone_size))
+ if [ "$inc" -eq 0 ]; then
+ require_seq_zones $nz || return $SKIP_TESTCASE
+ fi
opts=()
for ((off = first_sequential_zone_sector * 512; off < disk_size;
off += inc)); do
@@ -696,6 +770,8 @@ test31() {
test32() {
local off opts=() size
+ require_zbd || return $SKIP_TESTCASE
+
prep_write
off=$((first_sequential_zone_sector * 512))
size=$((disk_size - off))
@@ -773,7 +849,7 @@ test37() {
local bs off size capacity
prep_write
- capacity=$(total_zone_capacity 1 $first_sequential_zone_sector $dev)
+ capacity=$(total_zone_capacity 1 $((first_sequential_zone_sector*512)) $dev)
if [ "$first_sequential_zone_sector" = 0 ]; then
off=0
else
@@ -805,16 +881,23 @@ test38() {
# Read one block from a block device.
read_one_block() {
+ local off
local bs
+ if ! result=($(first_online_zone "$dev")); then
+ echo "Failed to determine first online zone"
+ exit 1
+ fi
+ off=${result[0]}
bs=$((logical_block_size))
- run_one_fio_job --rw=read "$(ioengine "psync")" --bs=$bs --size=$bs "$@" 2>&1 |
+ run_one_fio_job --rw=read "$(ioengine "psync")" --offset=$off --bs=$bs \
+ --size=$bs "$@" 2>&1 |
tee -a "${logfile}.${test_number}"
}
# Check whether fio accepts --zonemode=none for zoned block devices.
test39() {
- [ -n "$is_zbd" ] || return 0
+ require_zbd || return $SKIP_TESTCASE
read_one_block --zonemode=none >/dev/null || return $?
check_read $((logical_block_size)) || return $?
}
@@ -824,7 +907,7 @@ test40() {
local bs
bs=$((logical_block_size))
- [ -n "$is_zbd" ] || return 0
+ require_zbd || return $SKIP_TESTCASE
read_one_block --zonemode=strided |
grep -q 'fio: --zonesize must be specified when using --zonemode=strided' ||
return $?
@@ -834,21 +917,21 @@ test40() {
# Check whether fio checks the zone size for zoned block devices.
test41() {
- [ -n "$is_zbd" ] || return 0
+ require_zbd || return $SKIP_TESTCASE
read_one_block --zonemode=zbd --zonesize=$((2 * zone_size)) |
grep -q 'job parameter zonesize.*does not match disk zone size'
}
# Check whether fio handles --zonesize=0 correctly for regular block devices.
test42() {
- [ -n "$is_zbd" ] && return 0
+ require_regular_block_dev || return $SKIP_TESTCASE
read_one_block --zonemode=zbd --zonesize=0 |
grep -q 'Specifying the zone size is mandatory for regular block devices with --zonemode=zbd'
}
# Check whether fio handles --zonesize=1 correctly for regular block devices.
test43() {
- [ -n "$is_zbd" ] && return 0
+ require_regular_block_dev || return $SKIP_TESTCASE
read_one_block --zonemode=zbd --zonesize=1 |
grep -q 'zone size must be at least 512 bytes for --zonemode=zbd'
}
@@ -862,7 +945,7 @@ test44() {
test45() {
local bs i
- [ -z "$is_zbd" ] && return 0
+ require_zbd || return $SKIP_TESTCASE
prep_write
bs=$((logical_block_size))
run_one_fio_job "$(ioengine "psync")" --iodepth=1 --rw=randwrite --bs=$bs\
@@ -901,6 +984,9 @@ test47() {
test48() {
local i jobs=16 off opts=()
+ require_zbd || return $SKIP_TESTCASE
+ require_seq_zones 80 || return $SKIP_TESTCASE
+
off=$((first_sequential_zone_sector * 512 + 64 * zone_size))
size=$((16*zone_size))
prep_write
@@ -922,7 +1008,7 @@ test48() {
{ echo; echo "fio ${opts[*]}"; echo; } >>"${logfile}.${test_number}"
- timeout -v -s KILL 45s \
+ timeout -v -s KILL 180s \
"${dynamic_analyzer[@]}" "$fio" "${opts[@]}" \
>> "${logfile}.${test_number}" 2>&1 || return $?
}
@@ -930,11 +1016,7 @@ test48() {
# Check if fio handles --zonecapacity on a normal block device correctly
test49() {
- if [ -n "$is_zbd" ]; then
- echo "$dev is not a regular block device" \
- >>"${logfile}.${test_number}"
- return 0
- fi
+ require_regular_block_dev || return $SKIP_TESTCASE
size=$((2 * zone_size))
capacity=$((zone_size * 3 / 4))
@@ -948,12 +1030,137 @@ test49() {
check_read $((capacity * 2)) || return $?
}
+# Verify that conv zones are not locked and only seq zones are locked during
+# random read on conv-seq mixed zones.
+test50() {
+ local off
+
+ require_zbd || return $SKIP_TESTCASE
+ require_conv_zones 8 || return $SKIP_TESTCASE
+ require_seq_zones 8 || return $SKIP_TESTCASE
+
+ reset_zone "${dev}" -1
+
+ off=$((first_sequential_zone_sector * 512 - 8 * zone_size))
+ run_fio --name=job --filename=${dev} --offset=${off} --bs=64K \
+ --size=$((16 * zone_size)) "$(ioengine "libaio")" --rw=randread\
+ --time_based --runtime=3 --zonemode=zbd --zonesize=${zone_size}\
+ --direct=1 --group_reporting=1 ${job_var_opts[@]} \
+ >> "${logfile}.${test_number}" 2>&1 || return $?
+}
+
+# Verify that conv zones are neither locked nor opened during random write on
+# conv-seq mixed zones. Zone lock and zone open shall happen only on seq zones.
+test51() {
+ local off jobs=16
+ local -a opts
+
+ require_zbd || return $SKIP_TESTCASE
+ require_conv_zones 8 || return $SKIP_TESTCASE
+ require_seq_zones 8 || return $SKIP_TESTCASE
+
+ prep_write
+
+ off=$((first_sequential_zone_sector * 512 - 8 * zone_size))
+ opts+=("--size=$((16 * zone_size))" "$(ioengine "libaio")")
+ opts+=("--zonemode=zbd" "--direct=1" "--zonesize=${zone_size}")
+ opts+=("--max_open_zones=2" "--offset=$off")
+ opts+=("--thread=1" "--group_reporting=1")
+ opts+=("--time_based" "--runtime=30" "--rw=randwrite")
+ for ((i=0;i<jobs;i++)); do
+ opts+=("--name=job${i}" "--filename=$dev")
+ opts+=("--bs=$(((i+1)*16))K")
+ opts+=($(job_var_opts_exclude "--max_open_zones"))
+ done
+ run_fio "${opts[@]}" >> "${logfile}.${test_number}" 2>&1 || return $?
+}
+
+# Verify that zone_reset_threshold only takes logical blocks from seq
+# zones into account, and logical blocks of conv zones are not counted.
+test52() {
+ local off io_size
+
+ require_zbd || return $SKIP_TESTCASE
+ require_conv_zones 8 || return $SKIP_TESTCASE
+ require_seq_zones 8 || return $SKIP_TESTCASE
+
+ reset_zone "${dev}" -1
+
+ # Total I/O size is 1/8 = 0.125 of the I/O range of cont + seq zones.
+ # Set zone_reset_threshold as 0.1. The threshold size is less than
+ # 0.125, then, reset count zero is expected.
+ # On the other hand, half of the I/O range is covered by conv zones.
+ # If fio would count the conv zones for zone_reset_threshold, the ratio
+ # were more than 0.5 and would trigger zone resets.
+
+ off=$((first_sequential_zone_sector * 512 - 8 * zone_size))
+ io_size=$((zone_size * 16 / 8))
+ run_fio --name=job --filename=$dev --rw=randwrite --bs=$((zone_size/16))\
+ --size=$((zone_size * 16)) --softrandommap=1 \
+ --io_size=$((io_size)) "$(ioengine "psync")" --offset=$off \
+ --zonemode=zbd --direct=1 --zonesize=${zone_size} \
+ --zone_reset_threshold=.1 --zone_reset_frequency=1.0 \
+ ${job_var_opts[@]} --debug=zbd \
+ >> "${logfile}.${test_number}" 2>&1 || return $?
+
+ check_written ${io_size} || return $?
+ check_reset_count -eq 0 || return $?
+}
+
+# Check both reads and writes are executed by random I/O to conventional zones.
+test53() {
+ local off capacity io read_b=0 written_b=0
+
+ require_zbd || return $SKIP_TESTCASE
+ require_conv_zones 4 || return $SKIP_TESTCASE
+
+ off=$((first_sequential_zone_sector * 512 - 4 * zone_size))
+ capacity=$(total_zone_capacity 4 $off $dev)
+ run_fio --name=job --filename=${dev} --rw=randrw --bs=64K \
+ --size=$((4 * zone_size)) "$(ioengine "psync")" --offset=${off}\
+ --zonemode=zbd --direct=1 --zonesize=${zone_size} \
+ ${job_var_opts[@]} \
+ >> "${logfile}.${test_number}" 2>&1 || return $?
+
+ written_b=$(fio_written <"${logfile}.${test_number}")
+ read_b=$(fio_read <"${logfile}.${test_number}")
+ io=$((written_b + read_b))
+ echo "Number of bytes read: $read_b" >>"${logfile}.${test_number}"
+ echo "Number of bytes written: $written_b" >>"${logfile}.${test_number}"
+ echo "Total number of bytes read and written: $io <> $capacity" \
+ >>"${logfile}.${test_number}"
+ if ((io==capacity && written_b != 0 && read_b != 0)); then
+ return 0
+ fi
+ return 1
+}
+
+# Test read/write mix with verify.
+test54() {
+ require_zbd || return $SKIP_TESTCASE
+ require_seq_zones 8 || return $SKIP_TESTCASE
+
+ run_fio --name=job --filename=${dev} "$(ioengine "libaio")" \
+ --time_based=1 --runtime=30s --continue_on_error=0 \
+ --offset=$((first_sequential_zone_sector * 512)) \
+ --size=$((8*zone_size)) --direct=1 --iodepth=1 \
+ --rw=randrw:2 --rwmixwrite=25 --bsrange=4k-${zone_size} \
+ --zonemode=zbd --zonesize=${zone_size} \
+ --verify=crc32c --do_verify=1 --verify_backlog=2 \
+ --experimental_verify=1 \
+ --alloc-size=65536 --random_generator=tausworthe64 \
+ ${job_var_opts[@]} --debug=zbd \
+ >> "${logfile}.${test_number}" 2>&1 || return $?
+}
+
+SECONDS=0
tests=()
dynamic_analyzer=()
reset_all_zones=
use_libzbc=
zbd_debug=
max_open_zones_opt=
+quit_on_err=
while [ "${1#-}" != "$1" ]; do
case "$1" in
@@ -968,8 +1175,10 @@ while [ "${1#-}" != "$1" ]; do
-o) max_open_zones_opt="${2}"; shift; shift;;
-v) dynamic_analyzer=(valgrind "--read-var-info=yes");
shift;;
+ -q) quit_on_err=1; shift;;
-z) zbd_debug=1; shift;;
--) shift; break;;
+ *) usage; exit 1;;
esac
done
@@ -1087,10 +1296,12 @@ fi
logfile=$0.log
passed=0
+skipped=0
failed=0
if [ -t 1 ]; then
red="\e[1;31m"
green="\e[1;32m"
+ cyan="\e[1;36m"
end="\e[m"
else
red=""
@@ -1101,14 +1312,23 @@ rc=0
intr=0
trap 'intr=1' SIGINT
+ret=0
for test_number in "${tests[@]}"; do
rm -f "${logfile}.${test_number}"
+ unset SKIP_REASON
echo -n "Running test $(printf "%02d" $test_number) ... "
- if eval "test$test_number" && check_log $test_number; then
+ eval "test$test_number"
+ ret=$?
+ if ((!ret)) && check_log $test_number; then
status="PASS"
cc_status="${green}${status}${end}"
((passed++))
+ elif ((ret==SKIP_TESTCASE)); then
+ status="SKIP"
+ echo "${SKIP_REASON}" >> "${logfile}.${test_number}"
+ cc_status="${cyan}${status}${end} ${SKIP_REASON}"
+ ((skipped++))
else
status="FAIL"
cc_status="${red}${status}${end}"
@@ -1118,10 +1338,15 @@ for test_number in "${tests[@]}"; do
echo -e "$cc_status"
echo "$status" >> "${logfile}.${test_number}"
[ $intr -ne 0 ] && exit 1
+ [ -n "$quit_on_err" -a "$rc" -ne 0 ] && exit 1
done
echo "$passed tests passed"
+if [ $skipped -gt 0 ]; then
+ echo " $skipped tests skipped"
+fi
if [ $failed -gt 0 ]; then
- echo " and $failed tests failed"
+ echo " $failed tests failed"
fi
+echo "Run time: $(TZ=UTC0 printf "%(%H:%M:%S)T\n" $(( SECONDS )) )"
exit $rc
diff --git a/zbd.c b/zbd.c
index f2599bd4..6a26fe10 100644
--- a/zbd.c
+++ b/zbd.c
@@ -131,15 +131,6 @@ static uint32_t zbd_zone_idx(const struct fio_file *f, uint64_t offset)
return min(zone_idx, f->zbd_info->nr_zones);
}
-/**
- * zbd_zone_swr - Test whether a zone requires sequential writes
- * @z: zone info pointer.
- */
-static inline bool zbd_zone_swr(struct fio_zone_info *z)
-{
- return z->type == ZBD_ZONE_TYPE_SWR;
-}
-
/**
* zbd_zone_end - Return zone end location
* @z: zone info pointer.
@@ -171,11 +162,12 @@ static bool zbd_zone_full(const struct fio_file *f, struct fio_zone_info *z,
{
assert((required & 511) == 0);
- return zbd_zone_swr(z) &&
+ return z->has_wp &&
z->wp + required > zbd_zone_capacity_end(z);
}
-static void zone_lock(struct thread_data *td, struct fio_file *f, struct fio_zone_info *z)
+static void zone_lock(struct thread_data *td, const struct fio_file *f,
+ struct fio_zone_info *z)
{
struct zoned_block_device_info *zbd = f->zbd_info;
uint32_t nz = z - zbd->zone_info;
@@ -183,6 +175,8 @@ static void zone_lock(struct thread_data *td, struct fio_file *f, struct fio_zon
/* A thread should never lock zones outside its working area. */
assert(f->min_zone <= nz && nz < f->max_zone);
+ assert(z->has_wp);
+
/*
* Lock the io_u target zone. The zone will be unlocked if io_u offset
* is changed or when io_u completes and zbd_put_io() executed.
@@ -199,11 +193,26 @@ static void zone_lock(struct thread_data *td, struct fio_file *f, struct fio_zon
}
}
+static inline void zone_unlock(struct fio_zone_info *z)
+{
+ int ret;
+
+ assert(z->has_wp);
+ ret = pthread_mutex_unlock(&z->mutex);
+ assert(!ret);
+}
+
static bool is_valid_offset(const struct fio_file *f, uint64_t offset)
{
return (uint64_t)(offset - f->file_offset) < f->io_size;
}
+static inline struct fio_zone_info *get_zone(const struct fio_file *f,
+ unsigned int zone_nr)
+{
+ return &f->zbd_info->zone_info[zone_nr];
+}
+
/* Verify whether direct I/O is used for all host-managed zoned drives. */
static bool zbd_using_direct_io(void)
{
@@ -235,7 +244,7 @@ static bool zbd_is_seq_job(struct fio_file *f)
zone_idx_b = zbd_zone_idx(f, f->file_offset);
zone_idx_e = zbd_zone_idx(f, f->file_offset + f->io_size - 1);
for (zone_idx = zone_idx_b; zone_idx <= zone_idx_e; zone_idx++)
- if (zbd_zone_swr(&f->zbd_info->zone_info[zone_idx]))
+ if (get_zone(f, zone_idx)->has_wp)
return true;
return false;
@@ -286,7 +295,7 @@ static bool zbd_verify_sizes(void)
}
zone_idx = zbd_zone_idx(f, f->file_offset);
- z = &f->zbd_info->zone_info[zone_idx];
+ z = get_zone(f, zone_idx);
if ((f->file_offset != z->start) &&
(td->o.td_ddir != TD_DDIR_READ)) {
new_offset = zbd_zone_end(z);
@@ -302,7 +311,7 @@ static bool zbd_verify_sizes(void)
f->file_offset = new_offset;
}
zone_idx = zbd_zone_idx(f, f->file_offset + f->io_size);
- z = &f->zbd_info->zone_info[zone_idx];
+ z = get_zone(f, zone_idx);
new_end = z->start;
if ((td->o.td_ddir != TD_DDIR_READ) &&
(f->file_offset + f->io_size != new_end)) {
@@ -316,10 +325,6 @@ static bool zbd_verify_sizes(void)
(unsigned long long) new_end - f->file_offset);
f->io_size = new_end - f->file_offset;
}
-
- f->min_zone = zbd_zone_idx(f, f->file_offset);
- f->max_zone = zbd_zone_idx(f, f->file_offset + f->io_size);
- assert(f->min_zone < f->max_zone);
}
}
@@ -415,6 +420,7 @@ static int init_zone_info(struct thread_data *td, struct fio_file *f)
p->type = ZBD_ZONE_TYPE_SWR;
p->cond = ZBD_ZONE_COND_EMPTY;
p->capacity = zone_capacity;
+ p->has_wp = 1;
}
/* a sentinel */
p->start = nr_zones * zone_size;
@@ -443,7 +449,7 @@ static int parse_zone_info(struct thread_data *td, struct fio_file *f)
struct fio_zone_info *p;
uint64_t zone_size, offset;
struct zoned_block_device_info *zbd_info = NULL;
- int i, j, ret = 0;
+ int i, j, ret = -ENOMEM;
zones = calloc(ZBD_REPORT_MAX_ZONES, sizeof(struct zbd_zone));
if (!zones)
@@ -475,7 +481,6 @@ static int parse_zone_info(struct thread_data *td, struct fio_file *f)
zbd_info = scalloc(1, sizeof(*zbd_info) +
(nr_zones + 1) * sizeof(zbd_info->zone_info[0]));
- ret = -ENOMEM;
if (!zbd_info)
goto out;
mutex_init_pshared(&zbd_info->mutex);
@@ -499,8 +504,17 @@ static int parse_zone_info(struct thread_data *td, struct fio_file *f)
p->wp = z->wp;
break;
}
+
+ switch (z->type) {
+ case ZBD_ZONE_TYPE_SWR:
+ p->has_wp = 1;
+ break;
+ default:
+ p->has_wp = 0;
+ }
p->type = z->type;
p->cond = z->cond;
+
if (j > 0 && p->start != p[-1].start + zone_size) {
log_info("%s: invalid zone data\n",
f->file_name);
@@ -512,8 +526,9 @@ static int parse_zone_info(struct thread_data *td, struct fio_file *f)
offset = z->start + z->len;
if (j >= nr_zones)
break;
- nrz = zbd_report_zones(td, f, offset,
- zones, ZBD_REPORT_MAX_ZONES);
+ nrz = zbd_report_zones(td, f, offset, zones,
+ min((uint32_t)(nr_zones - j),
+ ZBD_REPORT_MAX_ZONES));
if (nrz < 0) {
ret = nrz;
log_info("fio: report zones (offset %llu) failed for %s (%d).\n",
@@ -662,6 +677,18 @@ int zbd_setup_files(struct thread_data *td)
if (!zbd)
continue;
+ f->min_zone = zbd_zone_idx(f, f->file_offset);
+ f->max_zone = zbd_zone_idx(f, f->file_offset + f->io_size);
+
+ /*
+ * When all zones in the I/O range are conventional, io_size
+ * can be smaller than zone size, making min_zone the same
+ * as max_zone. This is why the assert below needs to be made
+ * conditional.
+ */
+ if (zbd_is_seq_job(f))
+ assert(f->min_zone < f->max_zone);
+
zbd->max_open_zones = zbd->max_open_zones ?: ZBD_MAX_OPEN_ZONES;
if (td->o.max_open_zones > 0 &&
@@ -695,10 +722,10 @@ int zbd_setup_files(struct thread_data *td)
return 0;
}
-static unsigned int zbd_zone_nr(struct zoned_block_device_info *zbd_info,
- struct fio_zone_info *zone)
+static inline unsigned int zbd_zone_nr(const struct fio_file *f,
+ struct fio_zone_info *zone)
{
- return zone - zbd_info->zone_info;
+ return zone - f->zbd_info->zone_info;
}
/**
@@ -716,15 +743,16 @@ static int zbd_reset_zone(struct thread_data *td, struct fio_file *f,
{
uint64_t offset = z->start;
uint64_t length = (z+1)->start - offset;
+ uint64_t data_in_zone = z->wp - z->start;
int ret = 0;
- if (z->wp == z->start)
+ if (!data_in_zone)
return 0;
assert(is_valid_offset(f, offset + length - 1));
dprint(FD_ZBD, "%s: resetting wp of zone %u.\n", f->file_name,
- zbd_zone_nr(f->zbd_info, z));
+ zbd_zone_nr(f, z));
switch (f->zbd_info->model) {
case ZBD_HOST_AWARE:
case ZBD_HOST_MANAGED:
@@ -737,7 +765,8 @@ static int zbd_reset_zone(struct thread_data *td, struct fio_file *f,
}
pthread_mutex_lock(&f->zbd_info->mutex);
- f->zbd_info->sectors_with_data -= z->wp - z->start;
+ f->zbd_info->sectors_with_data -= data_in_zone;
+ f->zbd_info->wp_sectors_with_data -= data_in_zone;
pthread_mutex_unlock(&f->zbd_info->mutex);
z->wp = z->start;
z->verify_block = 0;
@@ -757,11 +786,8 @@ static void zbd_close_zone(struct thread_data *td, const struct fio_file *f,
if (f->zbd_info->open_zones[open_zone_idx] == zone_idx)
break;
}
- if (open_zone_idx == f->zbd_info->num_open_zones) {
- dprint(FD_ZBD, "%s: zone %d is not open\n",
- f->file_name, zone_idx);
+ if (open_zone_idx == f->zbd_info->num_open_zones)
return;
- }
dprint(FD_ZBD, "%s: closing zone %d\n", f->file_name, zone_idx);
memmove(f->zbd_info->open_zones + open_zone_idx,
@@ -770,7 +796,7 @@ static void zbd_close_zone(struct thread_data *td, const struct fio_file *f,
sizeof(f->zbd_info->open_zones[0]));
f->zbd_info->num_open_zones--;
td->num_open_zones--;
- f->zbd_info->zone_info[zone_idx].open = 0;
+ get_zone(f, zone_idx)->open = 0;
}
/*
@@ -794,11 +820,11 @@ static int zbd_reset_zones(struct thread_data *td, struct fio_file *f,
assert(min_bs);
dprint(FD_ZBD, "%s: examining zones %u .. %u\n", f->file_name,
- zbd_zone_nr(f->zbd_info, zb), zbd_zone_nr(f->zbd_info, ze));
+ zbd_zone_nr(f, zb), zbd_zone_nr(f, ze));
for (z = zb; z < ze; z++) {
- uint32_t nz = z - f->zbd_info->zone_info;
+ uint32_t nz = zbd_zone_nr(f, z);
- if (!zbd_zone_swr(z))
+ if (!z->has_wp)
continue;
zone_lock(td, f, z);
if (all_zones) {
@@ -812,12 +838,11 @@ static int zbd_reset_zones(struct thread_data *td, struct fio_file *f,
}
if (reset_wp) {
dprint(FD_ZBD, "%s: resetting zone %u\n",
- f->file_name,
- zbd_zone_nr(f->zbd_info, z));
+ f->file_name, zbd_zone_nr(f, z));
if (zbd_reset_zone(td, f, z) < 0)
res = 1;
}
- pthread_mutex_unlock(&z->mutex);
+ zone_unlock(z);
}
return res;
@@ -866,29 +891,37 @@ enum swd_action {
};
/* Calculate the number of sectors with data (swd) and perform action 'a' */
-static uint64_t zbd_process_swd(const struct fio_file *f, enum swd_action a)
+static uint64_t zbd_process_swd(struct thread_data *td,
+ const struct fio_file *f, enum swd_action a)
{
struct fio_zone_info *zb, *ze, *z;
uint64_t swd = 0;
+ uint64_t wp_swd = 0;
- zb = &f->zbd_info->zone_info[f->min_zone];
- ze = &f->zbd_info->zone_info[f->max_zone];
+ zb = get_zone(f, f->min_zone);
+ ze = get_zone(f, f->max_zone);
for (z = zb; z < ze; z++) {
- pthread_mutex_lock(&z->mutex);
+ if (z->has_wp) {
+ zone_lock(td, f, z);
+ wp_swd += z->wp - z->start;
+ }
swd += z->wp - z->start;
}
pthread_mutex_lock(&f->zbd_info->mutex);
switch (a) {
case CHECK_SWD:
assert(f->zbd_info->sectors_with_data == swd);
+ assert(f->zbd_info->wp_sectors_with_data == wp_swd);
break;
case SET_SWD:
f->zbd_info->sectors_with_data = swd;
+ f->zbd_info->wp_sectors_with_data = wp_swd;
break;
}
pthread_mutex_unlock(&f->zbd_info->mutex);
for (z = zb; z < ze; z++)
- pthread_mutex_unlock(&z->mutex);
+ if (z->has_wp)
+ zone_unlock(z);
return swd;
}
@@ -899,37 +932,28 @@ static uint64_t zbd_process_swd(const struct fio_file *f, enum swd_action a)
*/
static const bool enable_check_swd = false;
-/* Check whether the value of zbd_info.sectors_with_data is correct. */
-static void zbd_check_swd(const struct fio_file *f)
-{
- if (!enable_check_swd)
- return;
-
- zbd_process_swd(f, CHECK_SWD);
-}
-
-static void zbd_init_swd(struct fio_file *f)
+/* Check whether the values of zbd_info.*sectors_with_data are correct. */
+static void zbd_check_swd(struct thread_data *td, const struct fio_file *f)
{
- uint64_t swd;
-
if (!enable_check_swd)
return;
- swd = zbd_process_swd(f, SET_SWD);
- dprint(FD_ZBD, "%s(%s): swd = %" PRIu64 "\n", __func__, f->file_name,
- swd);
+ zbd_process_swd(td, f, CHECK_SWD);
}
void zbd_file_reset(struct thread_data *td, struct fio_file *f)
{
struct fio_zone_info *zb, *ze;
+ uint64_t swd;
if (!f->zbd_info || !td_write(td))
return;
- zb = &f->zbd_info->zone_info[f->min_zone];
- ze = &f->zbd_info->zone_info[f->max_zone];
- zbd_init_swd(f);
+ zb = get_zone(f, f->min_zone);
+ ze = get_zone(f, f->max_zone);
+ swd = zbd_process_swd(td, f, SET_SWD);
+ dprint(FD_ZBD, "%s(%s): swd = %" PRIu64 "\n", __func__, f->file_name,
+ swd);
/*
* If data verification is enabled reset the affected zones before
* writing any data to avoid that a zone reset has to be issued while
@@ -968,7 +992,7 @@ static bool zbd_open_zone(struct thread_data *td, const struct fio_file *f,
uint32_t zone_idx)
{
const uint32_t min_bs = td->o.min_bs[DDIR_WRITE];
- struct fio_zone_info *z = &f->zbd_info->zone_info[zone_idx];
+ struct fio_zone_info *z = get_zone(f, zone_idx);
bool res = true;
if (z->cond == ZBD_ZONE_COND_OFFLINE)
@@ -1019,7 +1043,8 @@ static uint32_t pick_random_zone_idx(const struct fio_file *f,
/*
* Modify the offset of an I/O unit that does not refer to an open zone such
* that it refers to an open zone. Close an open zone and open a new zone if
- * necessary. This algorithm can only work correctly if all write pointers are
+ * necessary. The open zone is searched across sequential zones.
+ * This algorithm can only work correctly if all write pointers are
* a multiple of the fio block size. The caller must neither hold z->mutex
* nor f->zbd_info->mutex. Returns with z->mutex held upon success.
*/
@@ -1061,16 +1086,19 @@ static struct fio_zone_info *zbd_convert_to_open_zone(struct thread_data *td,
for (;;) {
uint32_t tmp_idx;
- z = &f->zbd_info->zone_info[zone_idx];
-
- zone_lock(td, f, z);
+ z = get_zone(f, zone_idx);
+ if (z->has_wp)
+ zone_lock(td, f, z);
pthread_mutex_lock(&f->zbd_info->mutex);
- if (td->o.max_open_zones == 0 && td->o.job_max_open_zones == 0)
- goto examine_zone;
- if (f->zbd_info->num_open_zones == 0) {
- dprint(FD_ZBD, "%s(%s): no zones are open\n",
- __func__, f->file_name);
- goto open_other_zone;
+ if (z->has_wp) {
+ if (z->cond != ZBD_ZONE_COND_OFFLINE &&
+ td->o.max_open_zones == 0 && td->o.job_max_open_zones == 0)
+ goto examine_zone;
+ if (f->zbd_info->num_open_zones == 0) {
+ dprint(FD_ZBD, "%s(%s): no zones are open\n",
+ __func__, f->file_name);
+ goto open_other_zone;
+ }
}
/*
@@ -1079,7 +1107,8 @@ static struct fio_zone_info *zbd_convert_to_open_zone(struct thread_data *td,
* Ignore zones which don't belong to thread's offset/size area.
*/
open_zone_idx = pick_random_zone_idx(f, io_u);
- assert(open_zone_idx < f->zbd_info->num_open_zones);
+ assert(!open_zone_idx ||
+ open_zone_idx < f->zbd_info->num_open_zones);
tmp_idx = open_zone_idx;
for (i = 0; i < f->zbd_info->num_open_zones; i++) {
uint32_t tmpz;
@@ -1098,7 +1127,8 @@ static struct fio_zone_info *zbd_convert_to_open_zone(struct thread_data *td,
dprint(FD_ZBD, "%s(%s): no candidate zone\n",
__func__, f->file_name);
pthread_mutex_unlock(&f->zbd_info->mutex);
- pthread_mutex_unlock(&z->mutex);
+ if (z->has_wp)
+ zone_unlock(z);
return NULL;
found_candidate_zone:
@@ -1107,7 +1137,8 @@ found_candidate_zone:
break;
zone_idx = new_zone_idx;
pthread_mutex_unlock(&f->zbd_info->mutex);
- pthread_mutex_unlock(&z->mutex);
+ if (z->has_wp)
+ zone_unlock(z);
}
/* Both z->mutex and f->zbd_info->mutex are held. */
@@ -1144,14 +1175,17 @@ open_other_zone:
/* Zone 'z' is full, so try to open a new zone. */
for (i = f->io_size / f->zbd_info->zone_size; i > 0; i--) {
zone_idx++;
- pthread_mutex_unlock(&z->mutex);
+ if (z->has_wp)
+ zone_unlock(z);
z++;
if (!is_valid_offset(f, z->start)) {
/* Wrap-around. */
zone_idx = f->min_zone;
- z = &f->zbd_info->zone_info[zone_idx];
+ z = get_zone(f, zone_idx);
}
assert(is_valid_offset(f, z->start));
+ if (!z->has_wp)
+ continue;
zone_lock(td, f, z);
if (z->open)
continue;
@@ -1168,9 +1202,9 @@ open_other_zone:
if (zone_idx < f->min_zone || zone_idx >= f->max_zone)
continue;
pthread_mutex_unlock(&f->zbd_info->mutex);
- pthread_mutex_unlock(&z->mutex);
+ zone_unlock(z);
- z = &f->zbd_info->zone_info[zone_idx];
+ z = get_zone(f, zone_idx);
zone_lock(td, f, z);
if (z->wp + min_bs <= zbd_zone_capacity_end(z))
@@ -1178,7 +1212,7 @@ open_other_zone:
pthread_mutex_lock(&f->zbd_info->mutex);
}
pthread_mutex_unlock(&f->zbd_info->mutex);
- pthread_mutex_unlock(&z->mutex);
+ zone_unlock(z);
dprint(FD_ZBD, "%s(%s): did not open another zone\n", __func__,
f->file_name);
return NULL;
@@ -1187,6 +1221,8 @@ out:
dprint(FD_ZBD, "%s(%s): returning zone %d\n", __func__, f->file_name,
zone_idx);
io_u->offset = z->start;
+ assert(z->has_wp);
+ assert(z->cond != ZBD_ZONE_COND_OFFLINE);
return z;
}
@@ -1198,26 +1234,39 @@ static struct fio_zone_info *zbd_replay_write_order(struct thread_data *td,
const struct fio_file *f = io_u->file;
const uint32_t min_bs = td->o.min_bs[DDIR_WRITE];
- if (!zbd_open_zone(td, f, z - f->zbd_info->zone_info)) {
- pthread_mutex_unlock(&z->mutex);
+ if (!zbd_open_zone(td, f, zbd_zone_nr(f, z))) {
+ zone_unlock(z);
z = zbd_convert_to_open_zone(td, io_u);
assert(z);
}
- if (z->verify_block * min_bs >= z->capacity)
+ if (z->verify_block * min_bs >= z->capacity) {
log_err("%s: %d * %d >= %llu\n", f->file_name, z->verify_block,
min_bs, (unsigned long long)z->capacity);
- io_u->offset = z->start + z->verify_block++ * min_bs;
+ /*
+ * If the assertion below fails during a test run, adding
+ * "--experimental_verify=1" to the command line may help.
+ */
+ assert(false);
+ }
+ io_u->offset = z->start + z->verify_block * min_bs;
+ if (io_u->offset + io_u->buflen >= zbd_zone_capacity_end(z)) {
+ log_err("%s: %llu + %llu >= %llu\n", f->file_name, io_u->offset,
+ io_u->buflen, (unsigned long long) zbd_zone_capacity_end(z));
+ assert(false);
+ }
+ z->verify_block += io_u->buflen / min_bs;
+
return z;
}
/*
- * Find another zone for which @io_u fits below the write pointer. Start
- * searching in zones @zb + 1 .. @zl and continue searching in zones
- * @zf .. @zb - 1.
+ * Find another zone for which @io_u fits in the readable data in the zone.
+ * Search in zones @zb + 1 .. @zl. For random workload, also search in zones
+ * @zb - 1 .. @zf.
*
- * Either returns NULL or returns a zone pointer and holds the mutex for that
- * zone.
+ * Either returns NULL or returns a zone pointer. When the zone has write
+ * pointer, hold the mutex for the zone.
*/
static struct fio_zone_info *
zbd_find_zone(struct thread_data *td, struct io_u *io_u,
@@ -1226,8 +1275,7 @@ zbd_find_zone(struct thread_data *td, struct io_u *io_u,
const uint32_t min_bs = td->o.min_bs[io_u->ddir];
struct fio_file *f = io_u->file;
struct fio_zone_info *z1, *z2;
- const struct fio_zone_info *const zf =
- &f->zbd_info->zone_info[f->min_zone];
+ const struct fio_zone_info *const zf = get_zone(f, f->min_zone);
/*
* Skip to the next non-empty zone in case of sequential I/O and to
@@ -1235,19 +1283,23 @@ zbd_find_zone(struct thread_data *td, struct io_u *io_u,
*/
for (z1 = zb + 1, z2 = zb - 1; z1 < zl || z2 >= zf; z1++, z2--) {
if (z1 < zl && z1->cond != ZBD_ZONE_COND_OFFLINE) {
- zone_lock(td, f, z1);
+ if (z1->has_wp)
+ zone_lock(td, f, z1);
if (z1->start + min_bs <= z1->wp)
return z1;
- pthread_mutex_unlock(&z1->mutex);
+ if (z1->has_wp)
+ zone_unlock(z1);
} else if (!td_random(td)) {
break;
}
if (td_random(td) && z2 >= zf &&
z2->cond != ZBD_ZONE_COND_OFFLINE) {
- zone_lock(td, f, z2);
+ if (z2->has_wp)
+ zone_lock(td, f, z2);
if (z2->start + min_bs <= z2->wp)
return z2;
- pthread_mutex_unlock(&z2->mutex);
+ if (z2->has_wp)
+ zone_unlock(z2);
}
}
dprint(FD_ZBD, "%s: adjusting random read offset failed\n",
@@ -1272,7 +1324,7 @@ static void zbd_end_zone_io(struct thread_data *td, const struct io_u *io_u,
if (io_u->ddir == DDIR_WRITE &&
io_u->offset + io_u->buflen >= zbd_zone_capacity_end(z)) {
pthread_mutex_lock(&f->zbd_info->mutex);
- zbd_close_zone(td, f, z - f->zbd_info->zone_info);
+ zbd_close_zone(td, f, zbd_zone_nr(f, z));
pthread_mutex_unlock(&f->zbd_info->mutex);
}
}
@@ -1300,10 +1352,9 @@ static void zbd_queue_io(struct thread_data *td, struct io_u *io_u, int q,
zone_idx = zbd_zone_idx(f, io_u->offset);
assert(zone_idx < zbd_info->nr_zones);
- z = &zbd_info->zone_info[zone_idx];
+ z = get_zone(f, zone_idx);
- if (!zbd_zone_swr(z))
- return;
+ assert(z->has_wp);
if (!success)
goto unlock;
@@ -1321,8 +1372,10 @@ static void zbd_queue_io(struct thread_data *td, struct io_u *io_u, int q,
* z->wp > zone_end means that one or more I/O errors
* have occurred.
*/
- if (z->wp <= zone_end)
+ if (z->wp <= zone_end) {
zbd_info->sectors_with_data += zone_end - z->wp;
+ zbd_info->wp_sectors_with_data += zone_end - z->wp;
+ }
pthread_mutex_unlock(&zbd_info->mutex);
z->wp = zone_end;
break;
@@ -1339,7 +1392,7 @@ static void zbd_queue_io(struct thread_data *td, struct io_u *io_u, int q,
unlock:
if (!success || q != FIO_Q_QUEUED) {
/* BUSY or COMPLETED: unlock the zone */
- pthread_mutex_unlock(&z->mutex);
+ zone_unlock(z);
io_u->zbd_put_io = NULL;
}
}
@@ -1354,17 +1407,15 @@ static void zbd_put_io(struct thread_data *td, const struct io_u *io_u)
struct zoned_block_device_info *zbd_info = f->zbd_info;
struct fio_zone_info *z;
uint32_t zone_idx;
- int ret;
if (!zbd_info)
return;
zone_idx = zbd_zone_idx(f, io_u->offset);
assert(zone_idx < zbd_info->nr_zones);
- z = &zbd_info->zone_info[zone_idx];
+ z = get_zone(f, zone_idx);
- if (!zbd_zone_swr(z))
- return;
+ assert(z->has_wp);
dprint(FD_ZBD,
"%s: terminate I/O (%lld, %llu) for zone %u\n",
@@ -1372,9 +1423,8 @@ static void zbd_put_io(struct thread_data *td, const struct io_u *io_u)
zbd_end_zone_io(td, io_u, z);
- ret = pthread_mutex_unlock(&z->mutex);
- assert(ret == 0);
- zbd_check_swd(f);
+ zone_unlock(z);
+ zbd_check_swd(td, f);
}
/*
@@ -1417,7 +1467,7 @@ void setup_zbd_zone_mode(struct thread_data *td, struct io_u *io_u)
assert(td->o.zone_size);
zone_idx = zbd_zone_idx(f, f->last_pos[ddir]);
- z = &f->zbd_info->zone_info[zone_idx];
+ z = get_zone(f, zone_idx);
/*
* When the zone capacity is smaller than the zone size and the I/O is
@@ -1431,8 +1481,7 @@ void setup_zbd_zone_mode(struct thread_data *td, struct io_u *io_u)
"%s: Jump from zone capacity limit to zone end:"
" (%llu -> %llu) for zone %u (%llu)\n",
f->file_name, (unsigned long long) f->last_pos[ddir],
- (unsigned long long) zbd_zone_end(z),
- zbd_zone_nr(f->zbd_info, z),
+ (unsigned long long) zbd_zone_end(z), zone_idx,
(unsigned long long) z->capacity);
td->io_skip_bytes += zbd_zone_end(z) - f->last_pos[ddir];
f->last_pos[ddir] = zbd_zone_end(z);
@@ -1526,12 +1575,34 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
assert(is_valid_offset(f, io_u->offset));
assert(io_u->buflen);
zone_idx_b = zbd_zone_idx(f, io_u->offset);
- zb = &f->zbd_info->zone_info[zone_idx_b];
+ zb = get_zone(f, zone_idx_b);
orig_zb = zb;
- /* Accept the I/O offset for conventional zones. */
- if (!zbd_zone_swr(zb))
+ if (!zb->has_wp) {
+ /* Accept non-write I/Os for conventional zones. */
+ if (io_u->ddir != DDIR_WRITE)
+ return io_u_accept;
+ /*
+ * Make sure that writes to conventional zones
+ * don't cross over to any sequential zones.
+ */
+ if (!(zb + 1)->has_wp ||
+ io_u->offset + io_u->buflen <= (zb + 1)->start)
+ return io_u_accept;
+
+ if (io_u->offset + min_bs > (zb + 1)->start) {
+ dprint(FD_IO,
+ "%s: off=%llu + min_bs=%u > next zone %llu\n",
+ f->file_name, io_u->offset,
+ min_bs, (unsigned long long) (zb + 1)->start);
+ io_u->offset = zb->start + (zb + 1)->start - io_u->offset;
+ new_len = min(io_u->buflen, (zb + 1)->start - io_u->offset);
+ } else {
+ new_len = (zb + 1)->start - io_u->offset;
+ }
+ io_u->buflen = new_len / min_bs * min_bs;
return io_u_accept;
+ }
/*
* Accept the I/O offset for reads if reading beyond the write pointer
@@ -1541,7 +1612,7 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
io_u->ddir == DDIR_READ && td->o.read_beyond_wp)
return io_u_accept;
- zbd_check_swd(f);
+ zbd_check_swd(td, f);
zone_lock(td, f, zb);
@@ -1549,7 +1620,6 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
case DDIR_READ:
if (td->runstate == TD_VERIFYING && td_write(td)) {
zb = zbd_replay_write_order(td, io_u, zb);
- pthread_mutex_unlock(&zb->mutex);
goto accept;
}
/*
@@ -1561,8 +1631,8 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
zb->wp - zb->start : 0;
if (range < min_bs ||
((!td_random(td)) && (io_u->offset + min_bs > zb->wp))) {
- pthread_mutex_unlock(&zb->mutex);
- zl = &f->zbd_info->zone_info[f->max_zone];
+ zone_unlock(zb);
+ zl = get_zone(f, f->max_zone);
zb = zbd_find_zone(td, io_u, zb, zl);
if (!zb) {
dprint(FD_ZBD,
@@ -1591,6 +1661,12 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
io_u->offset = zb->start +
((io_u->offset - orig_zb->start) %
(range - io_u->buflen)) / min_bs * min_bs;
+ /*
+ * When zbd_find_zone() returns a conventional zone,
+ * we can simply accept the new i/o offset here.
+ */
+ if (!zb->has_wp)
+ return io_u_accept;
/*
* Make sure the I/O does not cross over the zone wp position.
*/
@@ -1606,18 +1682,27 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
assert(io_u->offset + io_u->buflen <= zb->wp);
goto accept;
case DDIR_WRITE:
- if (io_u->buflen > f->zbd_info->zone_size)
+ if (io_u->buflen > f->zbd_info->zone_size) {
+ td_verror(td, EINVAL, "I/O buflen exceeds zone size");
+ dprint(FD_IO,
+ "%s: I/O buflen %llu exceeds zone size %llu\n",
+ f->file_name, io_u->buflen,
+ (unsigned long long) f->zbd_info->zone_size);
goto eof;
+ }
if (!zbd_open_zone(td, f, zone_idx_b)) {
- pthread_mutex_unlock(&zb->mutex);
+ zone_unlock(zb);
zb = zbd_convert_to_open_zone(td, io_u);
- if (!zb)
+ if (!zb) {
+ dprint(FD_IO, "%s: can't convert to open zone",
+ f->file_name);
goto eof;
- zone_idx_b = zb - f->zbd_info->zone_info;
+ }
+ zone_idx_b = zbd_zone_nr(f, zb);
}
/* Check whether the zone reset threshold has been exceeded */
if (td->o.zrf.u.f) {
- if (f->zbd_info->sectors_with_data >=
+ if (f->zbd_info->wp_sectors_with_data >=
f->io_size * td->o.zrt.u.f &&
zbd_dec_and_reset_write_cnt(td, f)) {
zb->reset_zone = 1;
@@ -1639,6 +1724,7 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
goto eof;
if (zb->capacity < min_bs) {
+ td_verror(td, EINVAL, "ZCAP is less min_bs");
log_err("zone capacity %llu smaller than minimum block size %d\n",
(unsigned long long)zb->capacity,
min_bs);
@@ -1649,8 +1735,9 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
assert(!zbd_zone_full(f, zb, min_bs));
io_u->offset = zb->wp;
if (!is_valid_offset(f, io_u->offset)) {
- dprint(FD_ZBD, "Dropped request with offset %llu\n",
- io_u->offset);
+ td_verror(td, EINVAL, "invalid WP value");
+ dprint(FD_ZBD, "%s: dropped request with offset %llu\n",
+ f->file_name, io_u->offset);
goto eof;
}
/*
@@ -1669,9 +1756,9 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
orig_len, io_u->buflen);
goto accept;
}
- log_err("Zone remainder %lld smaller than minimum block size %d\n",
- (zbd_zone_capacity_end(zb) - io_u->offset),
- min_bs);
+ td_verror(td, EIO, "zone remainder too small");
+ log_err("zone remainder %lld smaller than min block size %d\n",
+ (zbd_zone_capacity_end(zb) - io_u->offset), min_bs);
goto eof;
case DDIR_TRIM:
/* fall-through */
@@ -1687,17 +1774,23 @@ enum io_u_action zbd_adjust_block(struct thread_data *td, struct io_u *io_u)
assert(false);
accept:
- assert(zb);
+ assert(zb->has_wp);
assert(zb->cond != ZBD_ZONE_COND_OFFLINE);
assert(!io_u->zbd_queue_io);
assert(!io_u->zbd_put_io);
io_u->zbd_queue_io = zbd_queue_io;
io_u->zbd_put_io = zbd_put_io;
+ /*
+ * Since we return with the zone lock still held,
+ * add an annotation to let Coverity know that it
+ * is intentional.
+ */
+ /* coverity[missing_unlock] */
return io_u_accept;
eof:
- if (zb)
- pthread_mutex_unlock(&zb->mutex);
+ if (zb && zb->has_wp)
+ zone_unlock(zb);
return io_u_eof;
}
diff --git a/zbd.h b/zbd.h
index bff55f99..cc3ab624 100644
--- a/zbd.h
+++ b/zbd.h
@@ -28,6 +28,7 @@ enum io_u_action {
* @mutex: protects the modifiable members in this structure
* @type: zone type (BLK_ZONE_TYPE_*)
* @cond: zone state (BLK_ZONE_COND_*)
+ * @has_wp: whether or not this zone can have a valid write pointer
* @open: whether or not this zone is currently open. Only relevant if
* max_open_zones > 0.
* @reset_zone: whether or not this zone should be reset before writing to it
@@ -40,6 +41,7 @@ struct fio_zone_info {
uint32_t verify_block;
enum zbd_zone_type type:2;
enum zbd_zone_cond cond:4;
+ unsigned int has_wp:1;
unsigned int open:1;
unsigned int reset_zone:1;
};
@@ -53,6 +55,8 @@ struct fio_zone_info {
* num_open_zones).
* @zone_size: size of a single zone in bytes.
* @sectors_with_data: total size of data in all zones in units of 512 bytes
+ * @wp_sectors_with_data: total size of data in zones with write pointers in
+ * units of 512 bytes
* @zone_size_log2: log2 of the zone size in bytes if it is a power of 2 or 0
* if the zone size is not a power of 2.
* @nr_zones: number of zones
@@ -73,6 +77,7 @@ struct zoned_block_device_info {
pthread_mutex_t mutex;
uint64_t zone_size;
uint64_t sectors_with_data;
+ uint64_t wp_sectors_with_data;
uint32_t zone_size_log2;
uint32_t nr_zones;
uint32_t refcount;
next reply other threads:[~2021-01-30 13:00 UTC|newest]
Thread overview: 1435+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-30 13:00 Jens Axboe [this message]
-- strict thread matches above, loose matches on Subject: below --
2025-12-30 13:00 Recent changes (master) Jens Axboe
2025-12-19 13:00 Jens Axboe
2025-12-17 13:00 Jens Axboe
2025-12-14 13:00 Jens Axboe
2025-12-09 13:00 Jens Axboe
2025-11-25 13:00 Jens Axboe
2025-11-19 13:00 Jens Axboe
2025-11-18 13:00 Jens Axboe
2025-11-15 13:00 Jens Axboe
2025-11-06 13:00 Jens Axboe
2025-11-01 12:00 Jens Axboe
2025-10-31 12:00 Jens Axboe
2025-10-30 12:00 Jens Axboe
2025-10-29 12:00 Jens Axboe
2025-10-16 12:00 Jens Axboe
2025-10-11 12:00 Jens Axboe
2025-10-10 12:00 Jens Axboe
2025-10-09 12:00 Jens Axboe
2025-10-06 12:00 Jens Axboe
2025-10-05 12:00 Jens Axboe
2025-10-02 12:00 Jens Axboe
2025-09-26 12:00 Jens Axboe
2025-09-24 12:00 Jens Axboe
2025-09-19 12:00 Jens Axboe
2025-09-18 12:00 Jens Axboe
2025-09-17 12:00 Jens Axboe
2025-09-09 12:00 Jens Axboe
2025-09-06 12:00 Jens Axboe
2025-09-05 12:00 Jens Axboe
2025-09-04 12:00 Jens Axboe
2025-08-27 12:00 Jens Axboe
2025-08-26 12:00 Jens Axboe
2025-08-23 12:00 Jens Axboe
2025-08-22 12:00 Jens Axboe
2025-08-21 12:00 Jens Axboe
2025-08-20 12:00 Jens Axboe
2025-08-19 12:00 Jens Axboe
2025-08-12 12:00 Jens Axboe
2025-08-10 12:00 Jens Axboe
2025-08-08 12:00 Jens Axboe
2025-08-06 12:00 Jens Axboe
2025-08-03 12:00 Jens Axboe
2025-08-01 12:00 Jens Axboe
2025-07-24 12:00 Jens Axboe
2025-07-23 12:00 Jens Axboe
2025-07-19 12:00 Jens Axboe
2025-07-17 12:00 Jens Axboe
2025-07-10 12:00 Jens Axboe
2025-07-09 12:00 Jens Axboe
2025-07-01 12:00 Jens Axboe
2025-06-24 12:00 Jens Axboe
2025-06-05 12:00 Jens Axboe
2025-06-03 12:00 Jens Axboe
2025-06-01 12:00 Jens Axboe
2025-05-24 12:00 Jens Axboe
2025-05-21 12:00 Jens Axboe
2025-05-17 12:00 Jens Axboe
2025-05-14 12:00 Jens Axboe
2025-05-10 12:00 Jens Axboe
2025-05-09 12:00 Jens Axboe
2025-05-08 12:00 Jens Axboe
2025-05-07 12:00 Jens Axboe
2025-04-16 12:00 Jens Axboe
2025-04-15 12:00 Jens Axboe
2025-04-08 12:00 Jens Axboe
2025-04-05 12:00 Jens Axboe
2025-03-19 12:00 Jens Axboe
2025-03-08 13:00 Jens Axboe
2025-03-07 13:00 Jens Axboe
2025-03-06 13:00 Jens Axboe
2025-02-21 13:00 Jens Axboe
2025-02-19 13:00 Jens Axboe
2025-02-18 13:00 Jens Axboe
2025-02-15 13:00 Jens Axboe
2025-02-14 13:00 Jens Axboe
2025-01-24 13:00 Jens Axboe
2025-01-23 13:00 Jens Axboe
2025-01-22 13:00 Jens Axboe
2024-12-17 13:00 Jens Axboe
2024-12-10 13:00 Jens Axboe
2024-12-05 13:00 Jens Axboe
2024-11-23 13:00 Jens Axboe
2024-11-06 13:00 Jens Axboe
2024-11-05 13:00 Jens Axboe
2024-10-29 12:00 Jens Axboe
2024-10-17 12:00 Jens Axboe
2024-10-04 12:00 Jens Axboe
2024-10-03 12:00 Jens Axboe
2024-10-01 12:00 Jens Axboe
2024-09-28 12:00 Jens Axboe
2024-09-27 12:00 Jens Axboe
2024-09-17 12:00 Jens Axboe
2024-09-07 12:00 Jens Axboe
2024-09-06 12:00 Jens Axboe
2024-09-05 12:00 Jens Axboe
2024-09-04 12:00 Jens Axboe
2024-08-30 12:00 Jens Axboe
2024-08-29 12:00 Jens Axboe
2024-08-22 12:00 Jens Axboe
2024-08-17 12:00 Jens Axboe
2024-08-07 12:00 Jens Axboe
2024-08-06 12:00 Jens Axboe
2024-07-27 12:00 Jens Axboe
2024-07-18 12:00 Jens Axboe
2024-07-16 12:00 Jens Axboe
2024-07-13 12:00 Jens Axboe
2024-07-12 12:00 Jens Axboe
2024-06-29 12:00 Jens Axboe
2024-06-15 12:00 Jens Axboe
2024-06-13 12:00 Jens Axboe
2024-06-12 12:00 Jens Axboe
2024-06-08 12:00 Jens Axboe
2024-06-07 12:00 Jens Axboe
2024-06-05 12:00 Jens Axboe
2024-06-04 12:00 Jens Axboe
2024-06-04 12:11 ` Niklas Cassel
2024-06-04 12:53 ` Vincent Fu
2024-06-01 12:00 Jens Axboe
2024-05-29 12:00 Jens Axboe
2024-05-25 12:00 Jens Axboe
2024-05-22 12:00 Jens Axboe
2024-05-01 12:00 Jens Axboe
2024-04-26 12:00 Jens Axboe
2024-04-25 12:00 Jens Axboe
2024-04-20 12:00 Jens Axboe
2024-04-19 12:00 Jens Axboe
2024-04-18 12:00 Jens Axboe
2024-04-17 12:00 Jens Axboe
2024-04-16 12:00 Jens Axboe
2024-04-03 12:00 Jens Axboe
2024-03-27 12:00 Jens Axboe
2024-03-26 12:00 Jens Axboe
2024-03-23 12:00 Jens Axboe
2024-03-22 12:00 Jens Axboe
2024-03-21 12:00 Jens Axboe
2024-03-19 12:00 Jens Axboe
2024-03-08 13:00 Jens Axboe
2024-03-06 13:00 Jens Axboe
2024-03-05 13:00 Jens Axboe
2024-02-28 13:00 Jens Axboe
2024-02-23 13:00 Jens Axboe
2024-02-17 13:00 Jens Axboe
2024-02-16 13:00 Jens Axboe
2024-02-15 13:00 Jens Axboe
2024-02-14 13:00 Jens Axboe
2024-02-13 13:00 Jens Axboe
2024-02-09 13:00 Jens Axboe
2024-02-08 13:00 Jens Axboe
2024-01-28 13:00 Jens Axboe
2024-01-26 13:00 Jens Axboe
2024-01-25 13:00 Jens Axboe
2024-01-24 13:00 Jens Axboe
2024-01-23 13:00 Jens Axboe
2024-01-19 13:00 Jens Axboe
2024-01-18 13:00 Jens Axboe
2024-01-17 13:00 Jens Axboe
2023-12-30 13:00 Jens Axboe
2023-12-20 13:00 Jens Axboe
2023-12-16 13:00 Jens Axboe
2023-12-15 13:00 Jens Axboe
2023-12-13 13:00 Jens Axboe
2023-12-12 13:00 Jens Axboe
2023-11-20 13:00 Jens Axboe
2023-11-08 13:00 Jens Axboe
2023-11-07 13:00 Jens Axboe
2023-11-04 12:00 Jens Axboe
2023-11-03 12:00 Jens Axboe
2023-11-01 12:00 Jens Axboe
2023-10-26 12:00 Jens Axboe
2023-10-24 12:00 Jens Axboe
2023-10-23 12:00 Jens Axboe
2023-10-20 12:00 Jens Axboe
2023-10-17 12:00 Jens Axboe
2023-10-14 12:00 Jens Axboe
2023-10-07 12:00 Jens Axboe
2023-10-03 12:00 Jens Axboe
2023-09-30 12:00 Jens Axboe
2023-09-29 12:00 Jens Axboe
2023-09-27 12:00 Jens Axboe
2023-09-20 12:00 Jens Axboe
2023-09-16 12:00 Jens Axboe
2023-09-12 12:00 Jens Axboe
2023-09-03 12:00 Jens Axboe
2023-08-24 12:00 Jens Axboe
2023-08-17 12:00 Jens Axboe
2023-08-15 12:00 Jens Axboe
2023-08-04 12:00 Jens Axboe
2023-08-03 12:00 Jens Axboe
2023-08-01 12:00 Jens Axboe
2023-07-29 12:00 Jens Axboe
2023-07-28 12:00 Jens Axboe
2023-07-22 12:00 Jens Axboe
2023-07-21 12:00 Jens Axboe
2023-07-16 12:00 Jens Axboe
2023-07-15 12:00 Jens Axboe
2023-07-14 12:00 Jens Axboe
2023-07-06 12:00 Jens Axboe
2023-07-04 12:00 Jens Axboe
2023-06-22 12:00 Jens Axboe
2023-06-17 12:00 Jens Axboe
2023-06-10 12:00 Jens Axboe
2023-06-09 12:00 Jens Axboe
2023-06-02 12:00 Jens Axboe
2023-05-31 12:00 Jens Axboe
2023-05-25 12:00 Jens Axboe
2023-05-24 12:00 Jens Axboe
2023-05-20 12:00 Jens Axboe
2023-05-19 12:00 Jens Axboe
2023-05-18 12:00 Jens Axboe
2023-05-17 12:00 Jens Axboe
2023-05-16 12:00 Jens Axboe
2023-05-12 12:00 Jens Axboe
2023-05-11 12:00 Jens Axboe
2023-04-28 12:00 Jens Axboe
2023-04-27 12:00 Jens Axboe
2023-04-21 12:00 Jens Axboe
2023-04-14 12:00 Jens Axboe
2023-04-11 12:00 Jens Axboe
2023-04-08 12:00 Jens Axboe
2023-04-05 12:00 Jens Axboe
2023-04-01 12:00 Jens Axboe
2023-03-28 12:00 Jens Axboe
2023-03-22 12:00 Jens Axboe
2023-03-21 12:00 Jens Axboe
2023-03-16 12:00 Jens Axboe
2023-03-15 12:00 Jens Axboe
2023-03-08 13:00 Jens Axboe
2023-03-04 13:00 Jens Axboe
2023-03-03 13:00 Jens Axboe
2023-03-01 13:00 Jens Axboe
2023-02-28 13:00 Jens Axboe
2023-02-24 13:00 Jens Axboe
2023-02-22 13:00 Jens Axboe
2023-02-21 13:00 Jens Axboe
2023-02-18 13:00 Jens Axboe
2023-02-16 13:00 Jens Axboe
2023-02-15 13:00 Jens Axboe
2023-02-11 13:00 Jens Axboe
2023-02-10 13:00 Jens Axboe
2023-02-08 13:00 Jens Axboe
2023-02-07 13:00 Jens Axboe
2023-02-04 13:00 Jens Axboe
2023-02-01 13:00 Jens Axboe
2023-01-31 13:00 Jens Axboe
2023-01-26 13:00 Jens Axboe
2023-01-25 13:00 Jens Axboe
2023-01-24 13:00 Jens Axboe
2023-01-21 13:00 Jens Axboe
2023-01-19 13:00 Jens Axboe
2023-01-12 13:00 Jens Axboe
2022-12-23 13:00 Jens Axboe
2022-12-17 13:00 Jens Axboe
2022-12-16 13:00 Jens Axboe
2022-12-13 13:00 Jens Axboe
2022-12-03 13:00 Jens Axboe
2022-12-02 13:00 Jens Axboe
2022-12-01 13:00 Jens Axboe
2022-11-30 13:00 Jens Axboe
2022-11-29 13:00 Jens Axboe
2022-11-24 13:00 Jens Axboe
2022-11-19 13:00 Jens Axboe
2022-11-15 13:00 Jens Axboe
2022-11-08 13:00 Jens Axboe
2022-11-07 13:00 Jens Axboe
2022-11-05 12:00 Jens Axboe
2022-11-03 12:00 Jens Axboe
2022-11-02 12:00 Jens Axboe
2022-10-25 12:00 Jens Axboe
2022-10-22 12:00 Jens Axboe
2022-10-20 12:00 Jens Axboe
2022-10-19 12:00 Jens Axboe
2022-10-17 12:00 Jens Axboe
2022-10-16 12:00 Jens Axboe
2022-10-15 12:00 Jens Axboe
2022-10-08 12:00 Jens Axboe
2022-10-06 12:00 Jens Axboe
2022-10-05 12:00 Jens Axboe
2022-10-04 12:00 Jens Axboe
2022-09-29 12:00 Jens Axboe
2022-09-23 12:00 Jens Axboe
2022-09-20 12:00 Jens Axboe
2022-09-16 12:00 Jens Axboe
2022-09-14 12:00 Jens Axboe
2022-09-13 12:00 Jens Axboe
2022-09-07 12:00 Jens Axboe
2022-09-04 12:00 Jens Axboe
2022-09-03 12:00 Jens Axboe
2022-09-02 12:00 Jens Axboe
2022-09-01 12:00 Jens Axboe
2022-08-31 12:00 Jens Axboe
2022-08-30 12:00 Jens Axboe
2022-08-27 12:00 Jens Axboe
2022-08-26 12:00 Jens Axboe
2022-08-25 12:00 Jens Axboe
2022-08-24 12:00 Jens Axboe
2022-08-17 12:00 Jens Axboe
2022-08-16 12:00 Jens Axboe
2022-08-12 12:00 Jens Axboe
2022-08-11 12:00 Jens Axboe
2022-08-10 12:00 Jens Axboe
2022-08-08 12:00 Jens Axboe
2022-08-04 12:00 Jens Axboe
2022-08-03 12:00 Jens Axboe
2022-08-01 12:00 Jens Axboe
2022-07-29 12:00 Jens Axboe
2022-07-28 12:00 Jens Axboe
2022-07-23 12:00 Jens Axboe
2022-07-22 12:00 Jens Axboe
2022-07-20 12:00 Jens Axboe
2022-07-12 12:00 Jens Axboe
2022-07-08 12:00 Jens Axboe
2022-07-07 12:00 Jens Axboe
2022-07-06 12:00 Jens Axboe
2022-07-02 12:00 Jens Axboe
2022-06-24 12:00 Jens Axboe
2022-06-23 12:00 Jens Axboe
2022-06-20 12:00 Jens Axboe
2022-06-16 12:00 Jens Axboe
2022-06-14 12:00 Jens Axboe
2022-06-02 12:00 Jens Axboe
2022-06-01 12:00 Jens Axboe
2022-05-30 12:00 Jens Axboe
2022-05-26 12:00 Jens Axboe
2022-05-13 12:00 Jens Axboe
2022-05-02 12:00 Jens Axboe
2022-04-30 12:00 Jens Axboe
2022-04-18 12:00 Jens Axboe
2022-04-11 12:00 Jens Axboe
2022-04-09 12:00 Jens Axboe
2022-04-07 12:00 Jens Axboe
2022-04-06 12:00 Jens Axboe
2022-03-31 12:00 Jens Axboe
2022-03-30 12:00 Jens Axboe
2022-03-29 12:00 Jens Axboe
2022-03-25 12:00 Jens Axboe
2022-03-21 12:00 Jens Axboe
2022-03-16 12:00 Jens Axboe
2022-03-12 13:00 Jens Axboe
2022-03-11 13:00 Jens Axboe
2022-03-10 13:00 Jens Axboe
2022-03-09 13:00 Jens Axboe
2022-03-08 13:00 Jens Axboe
2022-02-27 13:00 Jens Axboe
2022-02-25 13:00 Jens Axboe
2022-02-22 13:00 Jens Axboe
2022-02-21 13:00 Jens Axboe
2022-02-19 13:00 Jens Axboe
2022-02-18 13:00 Jens Axboe
2022-02-16 13:00 Jens Axboe
2022-02-12 13:00 Jens Axboe
2022-02-09 13:00 Jens Axboe
2022-02-05 13:00 Jens Axboe
2022-02-04 13:00 Jens Axboe
2022-01-29 13:00 Jens Axboe
2022-01-27 13:00 Jens Axboe
2022-01-22 13:00 Jens Axboe
2022-01-21 13:00 Jens Axboe
2022-01-19 13:00 Jens Axboe
2022-01-18 13:00 Jens Axboe
2022-01-11 13:00 Jens Axboe
2022-01-10 13:00 Jens Axboe
2021-12-24 13:00 Jens Axboe
2021-12-19 13:00 Jens Axboe
2021-12-16 13:00 Jens Axboe
2021-12-15 13:00 Jens Axboe
2021-12-11 13:00 Jens Axboe
2021-12-10 13:00 Jens Axboe
2021-12-07 13:00 Jens Axboe
2021-12-03 13:00 Jens Axboe
2021-11-26 13:00 Jens Axboe
2021-11-25 13:00 Jens Axboe
2021-11-22 13:00 Jens Axboe
2021-11-21 13:00 Jens Axboe
2021-11-20 13:00 Jens Axboe
2021-11-18 13:00 Jens Axboe
2021-11-13 13:00 Jens Axboe
2021-11-11 13:00 Jens Axboe
2021-10-26 12:00 Jens Axboe
2021-10-23 12:00 Jens Axboe
2021-10-25 15:37 ` Rebecca Cran
2021-10-25 15:41 ` Jens Axboe
2021-10-25 15:42 ` Rebecca Cran
2021-10-25 15:43 ` Jens Axboe
2021-10-20 12:00 Jens Axboe
2021-10-19 12:00 Jens Axboe
2021-10-18 12:00 Jens Axboe
2021-10-16 12:00 Jens Axboe
2021-10-15 12:00 Jens Axboe
2021-10-14 12:00 Jens Axboe
2021-10-13 12:00 Jens Axboe
2021-10-12 12:00 Jens Axboe
2021-10-10 12:00 Jens Axboe
2021-10-08 12:00 Jens Axboe
2021-10-06 12:00 Jens Axboe
2021-10-05 12:00 Jens Axboe
2021-10-02 12:00 Jens Axboe
2021-10-01 12:00 Jens Axboe
2021-09-30 12:00 Jens Axboe
2021-09-29 12:00 Jens Axboe
2021-09-27 12:00 Jens Axboe
2021-09-26 12:00 Jens Axboe
2021-09-25 12:00 Jens Axboe
2021-09-24 12:00 Jens Axboe
2021-09-21 12:00 Jens Axboe
2021-09-17 12:00 Jens Axboe
2021-09-16 12:00 Jens Axboe
2021-09-14 12:00 Jens Axboe
2021-09-09 12:00 Jens Axboe
2021-09-06 12:00 Jens Axboe
2021-09-04 12:00 Jens Axboe
2021-09-04 12:00 ` Jens Axboe
2021-09-03 12:00 Jens Axboe
2021-08-29 12:00 Jens Axboe
2021-08-28 12:00 Jens Axboe
2021-08-27 12:00 Jens Axboe
2021-08-21 12:00 Jens Axboe
2021-08-19 12:00 Jens Axboe
2021-08-14 12:00 Jens Axboe
2021-08-12 12:00 Jens Axboe
2021-08-07 12:00 Jens Axboe
2021-08-05 12:00 Jens Axboe
2021-08-04 12:00 Jens Axboe
2021-08-03 12:00 Jens Axboe
2021-08-02 12:00 Jens Axboe
2021-07-29 12:00 Jens Axboe
2021-07-26 12:00 Jens Axboe
2021-07-16 12:00 Jens Axboe
2021-07-08 12:00 Jens Axboe
2021-07-02 12:00 Jens Axboe
2021-06-30 12:00 Jens Axboe
2021-06-21 12:00 Jens Axboe
2021-06-18 12:00 Jens Axboe
2021-06-15 12:00 Jens Axboe
2021-06-11 12:00 Jens Axboe
2021-06-09 12:00 Jens Axboe
2021-06-04 12:00 Jens Axboe
2021-05-28 12:00 Jens Axboe
2021-05-27 12:00 Jens Axboe
2021-05-26 12:00 Jens Axboe
2021-05-19 12:00 Jens Axboe
2021-05-15 12:00 Jens Axboe
2021-05-12 12:00 Jens Axboe
2021-05-11 12:00 Jens Axboe
2021-05-09 12:00 Jens Axboe
2021-05-07 12:00 Jens Axboe
2021-04-28 12:00 Jens Axboe
2021-04-26 12:00 Jens Axboe
2021-04-24 12:00 Jens Axboe
2021-04-23 12:00 Jens Axboe
2021-04-17 12:00 Jens Axboe
2021-04-16 12:00 Jens Axboe
2021-04-14 12:00 Jens Axboe
2021-04-13 12:00 Jens Axboe
2021-04-11 12:00 Jens Axboe
2021-03-31 12:00 Jens Axboe
2021-03-19 12:00 Jens Axboe
2021-03-18 12:00 Jens Axboe
2021-03-12 13:00 Jens Axboe
2021-03-11 13:00 Jens Axboe
2021-03-10 13:00 Jens Axboe
2021-03-09 13:00 Jens Axboe
2021-03-07 13:00 Jens Axboe
2021-02-22 13:00 Jens Axboe
2021-02-17 13:00 Jens Axboe
2021-02-15 13:00 Jens Axboe
2021-02-11 13:00 Jens Axboe
2021-01-28 13:00 Jens Axboe
2021-01-27 13:00 Jens Axboe
2021-01-26 13:00 Jens Axboe
2021-01-24 13:00 Jens Axboe
2021-01-17 13:00 Jens Axboe
2021-01-16 13:00 Jens Axboe
2021-01-13 13:00 Jens Axboe
2021-01-10 13:00 Jens Axboe
2021-01-08 13:00 Jens Axboe
2021-01-07 13:00 Jens Axboe
2021-01-06 13:00 Jens Axboe
2020-12-30 13:00 Jens Axboe
2020-12-25 13:00 Jens Axboe
2020-12-18 13:00 Jens Axboe
2020-12-16 13:00 Jens Axboe
2020-12-08 13:00 Jens Axboe
2020-12-06 13:00 Jens Axboe
2020-12-05 13:00 Jens Axboe
2020-12-04 13:00 Jens Axboe
2020-11-28 13:00 Jens Axboe
2020-11-26 13:00 Jens Axboe
2020-11-23 13:00 Jens Axboe
2020-11-14 13:00 Jens Axboe
2020-11-13 13:00 Jens Axboe
2020-11-10 13:00 Jens Axboe
2020-11-06 13:00 Jens Axboe
2020-11-12 20:51 ` Rebecca Cran
2020-11-05 13:00 Jens Axboe
2020-11-02 13:00 Jens Axboe
2020-10-31 12:00 Jens Axboe
2020-10-29 12:00 Jens Axboe
2020-10-15 12:00 Jens Axboe
2020-10-14 12:00 Jens Axboe
2020-10-11 12:00 Jens Axboe
2020-10-10 12:00 Jens Axboe
2020-09-15 12:00 Jens Axboe
2020-09-12 12:00 Jens Axboe
2020-09-10 12:00 Jens Axboe
2020-09-09 12:00 Jens Axboe
2020-09-08 12:00 Jens Axboe
2020-09-07 12:00 Jens Axboe
2020-09-06 12:00 Jens Axboe
2020-09-04 12:00 Jens Axboe
2020-09-02 12:00 Jens Axboe
2020-09-01 12:00 Jens Axboe
2020-08-30 12:00 Jens Axboe
2020-08-29 12:00 Jens Axboe
2020-08-28 12:00 Jens Axboe
2020-08-23 12:00 Jens Axboe
2020-08-22 12:00 Jens Axboe
2020-08-20 12:00 Jens Axboe
2020-08-19 12:00 Jens Axboe
2020-08-18 12:00 Jens Axboe
2020-08-17 12:00 Jens Axboe
2020-08-15 12:00 Jens Axboe
2020-08-14 12:00 Jens Axboe
2020-08-13 12:00 Jens Axboe
2020-08-12 12:00 Jens Axboe
2020-08-11 12:00 Jens Axboe
2020-08-08 12:00 Jens Axboe
2020-08-02 12:00 Jens Axboe
2020-07-28 12:00 Jens Axboe
2020-07-27 12:00 Jens Axboe
2020-07-26 12:00 Jens Axboe
2020-07-25 12:00 Jens Axboe
2020-07-22 12:00 Jens Axboe
2020-07-21 12:00 Jens Axboe
2020-07-19 12:00 Jens Axboe
2020-07-18 12:00 Jens Axboe
2020-07-15 12:00 Jens Axboe
2020-07-14 12:00 Jens Axboe
2020-07-09 12:00 Jens Axboe
2020-07-05 12:00 Jens Axboe
2020-07-04 12:00 Jens Axboe
2020-07-03 12:00 Jens Axboe
2020-06-29 12:00 Jens Axboe
2020-06-25 12:00 Jens Axboe
2020-06-24 12:00 Jens Axboe
2020-06-22 12:00 Jens Axboe
2020-06-13 12:00 Jens Axboe
2020-06-10 12:00 Jens Axboe
2020-06-08 12:00 Jens Axboe
2020-06-06 12:00 Jens Axboe
2020-06-04 12:00 Jens Axboe
2020-06-03 12:00 Jens Axboe
2020-05-30 12:00 Jens Axboe
2020-05-29 12:00 Jens Axboe
2020-05-26 12:00 Jens Axboe
2020-05-25 12:00 Jens Axboe
2020-05-24 12:00 Jens Axboe
2020-05-22 12:00 Jens Axboe
2020-05-21 12:00 Jens Axboe
2020-05-20 12:00 Jens Axboe
2020-05-19 12:00 Jens Axboe
2020-05-15 12:00 Jens Axboe
2020-05-14 12:00 Jens Axboe
2020-05-12 12:00 Jens Axboe
2020-04-30 12:00 Jens Axboe
2020-04-22 12:00 Jens Axboe
2020-04-21 12:00 Jens Axboe
2020-04-18 12:00 Jens Axboe
2020-04-17 12:00 Jens Axboe
2020-04-16 12:00 Jens Axboe
2020-04-14 12:00 Jens Axboe
2020-04-09 12:00 Jens Axboe
2020-04-08 12:00 Jens Axboe
2020-04-07 12:00 Jens Axboe
2020-04-03 12:00 Jens Axboe
2020-04-01 12:00 Jens Axboe
2020-03-27 12:00 Jens Axboe
2020-03-18 12:00 Jens Axboe
2020-03-17 12:00 Jens Axboe
2020-03-16 12:00 Jens Axboe
2020-03-13 12:00 Jens Axboe
2020-03-04 13:00 Jens Axboe
2020-03-03 13:00 Jens Axboe
2020-03-02 13:00 Jens Axboe
2020-02-27 13:00 Jens Axboe
2020-02-25 13:00 Jens Axboe
2020-02-07 13:00 Jens Axboe
2020-02-06 13:00 Jens Axboe
2020-02-05 13:00 Jens Axboe
2020-01-29 13:00 Jens Axboe
2020-01-24 13:00 Jens Axboe
2020-01-23 13:00 Jens Axboe
2020-01-19 13:00 Jens Axboe
2020-01-17 13:00 Jens Axboe
2020-01-15 13:00 Jens Axboe
2020-01-14 13:00 Jens Axboe
2020-01-10 13:00 Jens Axboe
2020-01-07 13:00 Jens Axboe
2020-01-06 13:00 Jens Axboe
2020-01-05 13:00 Jens Axboe
2020-01-04 13:00 Jens Axboe
2019-12-26 13:00 Jens Axboe
2019-12-24 13:00 Jens Axboe
2019-12-22 13:00 Jens Axboe
2019-12-19 13:00 Jens Axboe
2019-12-17 13:00 Jens Axboe
2019-12-12 13:00 Jens Axboe
2019-12-07 13:00 Jens Axboe
2019-11-28 13:00 Jens Axboe
2019-11-27 13:00 Jens Axboe
2019-11-26 13:00 Jens Axboe
2019-11-15 13:00 Jens Axboe
2019-11-07 15:25 Jens Axboe
2019-11-07 13:00 Jens Axboe
2019-11-06 13:00 Jens Axboe
2019-11-04 13:00 Jens Axboe
2019-11-03 13:00 Jens Axboe
2019-10-30 12:00 Jens Axboe
2019-10-25 12:00 Jens Axboe
2019-10-22 12:00 Jens Axboe
2019-10-16 12:00 Jens Axboe
2019-10-15 12:00 Jens Axboe
2019-10-14 12:00 Jens Axboe
2019-10-09 12:00 Jens Axboe
2019-10-08 12:00 Jens Axboe
2019-10-07 12:00 Jens Axboe
2019-10-03 12:00 Jens Axboe
2019-10-02 12:00 Jens Axboe
2019-09-28 12:00 Jens Axboe
2019-09-26 12:00 Jens Axboe
2019-09-25 12:00 Jens Axboe
2019-09-24 12:00 Jens Axboe
2019-09-20 12:00 Jens Axboe
2019-09-14 12:00 Jens Axboe
2019-09-13 12:00 Jens Axboe
2019-09-06 12:00 Jens Axboe
2019-09-04 12:00 Jens Axboe
2019-08-30 12:00 Jens Axboe
2019-08-29 12:00 Jens Axboe
2019-08-16 12:00 Jens Axboe
2019-08-15 12:00 Jens Axboe
2019-08-15 14:27 ` Rebecca Cran
2019-08-15 14:28 ` Jens Axboe
2019-08-15 15:05 ` Rebecca Cran
2019-08-15 15:17 ` Jens Axboe
2019-08-15 15:35 ` Rebecca Cran
2019-08-09 12:00 Jens Axboe
2019-08-06 12:00 Jens Axboe
2019-08-04 12:00 Jens Axboe
2019-08-03 12:00 Jens Axboe
2019-08-01 12:00 Jens Axboe
2019-07-27 12:00 Jens Axboe
2019-07-13 12:00 Jens Axboe
2019-07-10 12:00 Jens Axboe
2019-07-02 12:00 Jens Axboe
2019-06-01 12:00 Jens Axboe
2019-05-24 12:00 Jens Axboe
2019-05-23 12:00 Jens Axboe
2019-05-21 12:00 Jens Axboe
2019-05-17 12:00 Jens Axboe
2019-05-10 12:00 Jens Axboe
2019-05-09 12:00 Jens Axboe
2019-05-09 12:47 ` Erwan Velu
2019-05-09 14:07 ` Jens Axboe
2019-05-09 15:47 ` Elliott, Robert (Servers)
2019-05-09 15:52 ` Sebastien Boisvert
2019-05-09 16:12 ` Elliott, Robert (Servers)
2019-05-09 15:57 ` Jens Axboe
2019-05-07 12:00 Jens Axboe
2019-04-26 12:00 Jens Axboe
2019-04-23 12:00 Jens Axboe
2019-04-20 12:00 Jens Axboe
2019-04-19 12:00 Jens Axboe
2019-04-18 12:00 Jens Axboe
2019-04-02 12:00 Jens Axboe
2019-03-26 12:00 Jens Axboe
2019-03-22 12:00 Jens Axboe
2019-03-12 12:00 Jens Axboe
2019-03-09 13:00 Jens Axboe
2019-03-08 13:00 Jens Axboe
2019-03-07 13:00 Jens Axboe
2019-03-01 13:00 Jens Axboe
2019-02-25 13:00 Jens Axboe
2019-02-24 13:00 Jens Axboe
2019-02-22 13:00 Jens Axboe
2019-02-12 13:00 Jens Axboe
2019-02-11 13:00 Jens Axboe
2019-02-09 13:00 Jens Axboe
2019-02-08 13:00 Jens Axboe
2019-02-05 13:00 Jens Axboe
2019-02-01 13:00 Jens Axboe
2019-01-30 13:00 Jens Axboe
2019-01-29 13:00 Jens Axboe
2019-01-25 13:00 Jens Axboe
2019-01-24 13:00 Jens Axboe
2019-01-17 13:00 Jens Axboe
2019-01-16 13:00 Jens Axboe
2019-01-15 13:00 Jens Axboe
2019-01-14 13:00 Jens Axboe
2019-01-13 13:00 Jens Axboe
2019-01-12 13:00 Jens Axboe
2019-01-11 13:00 Jens Axboe
2019-01-10 13:00 Jens Axboe
2019-01-09 13:00 Jens Axboe
2019-01-08 13:00 Jens Axboe
2019-01-06 13:00 Jens Axboe
2019-01-05 13:00 Jens Axboe
2018-12-31 13:00 Jens Axboe
2018-12-22 13:00 Jens Axboe
2018-12-20 13:00 Jens Axboe
2018-12-15 13:00 Jens Axboe
2018-12-14 13:00 Jens Axboe
2018-12-13 13:00 Jens Axboe
2018-12-11 13:00 Jens Axboe
2018-12-05 13:00 Jens Axboe
2018-12-02 13:00 Jens Axboe
2018-12-01 13:00 Jens Axboe
2018-11-30 13:00 Jens Axboe
2018-11-28 13:00 Jens Axboe
2018-11-27 13:00 Jens Axboe
2018-11-26 13:00 Jens Axboe
2018-11-25 13:00 Jens Axboe
2018-11-22 13:00 Jens Axboe
2018-11-21 13:00 Jens Axboe
2018-11-20 13:00 Jens Axboe
2018-11-16 13:00 Jens Axboe
2018-11-07 13:00 Jens Axboe
2018-11-03 12:00 Jens Axboe
2018-10-27 12:00 Jens Axboe
2018-10-24 12:00 Jens Axboe
2018-10-20 12:00 Jens Axboe
2018-10-19 12:00 Jens Axboe
2018-10-16 12:00 Jens Axboe
2018-10-09 12:00 Jens Axboe
2018-10-06 12:00 Jens Axboe
2018-10-05 12:00 Jens Axboe
2018-10-04 12:00 Jens Axboe
2018-10-02 12:00 Jens Axboe
2018-10-01 12:00 Jens Axboe
2018-09-30 12:00 Jens Axboe
2018-09-28 12:00 Jens Axboe
2018-09-27 12:00 Jens Axboe
2018-09-26 12:00 Jens Axboe
2018-09-23 12:00 Jens Axboe
2018-09-22 12:00 Jens Axboe
2018-09-21 12:00 Jens Axboe
2018-09-20 12:00 Jens Axboe
2018-09-18 12:00 Jens Axboe
2018-09-17 12:00 Jens Axboe
2018-09-13 12:00 Jens Axboe
2018-09-12 12:00 Jens Axboe
2018-09-11 12:00 Jens Axboe
2018-09-10 12:00 Jens Axboe
2018-09-09 12:00 Jens Axboe
2018-09-08 12:00 Jens Axboe
2018-09-07 12:00 Jens Axboe
2018-09-06 12:00 Jens Axboe
2018-09-04 12:00 Jens Axboe
2018-09-01 12:00 Jens Axboe
2018-08-31 12:00 Jens Axboe
2018-08-26 12:00 Jens Axboe
2018-08-25 12:00 Jens Axboe
2018-08-24 12:00 Jens Axboe
2018-08-23 12:00 Jens Axboe
2018-08-22 12:00 Jens Axboe
2018-08-21 12:00 Jens Axboe
2018-08-18 12:00 Jens Axboe
2018-08-17 12:00 Jens Axboe
2018-08-16 12:00 Jens Axboe
2018-08-15 12:00 Jens Axboe
2018-08-14 12:00 Jens Axboe
2018-08-13 12:00 Jens Axboe
2018-08-11 12:00 Jens Axboe
2018-08-10 12:00 Jens Axboe
2018-08-08 12:00 Jens Axboe
2018-08-06 12:00 Jens Axboe
2018-08-04 12:00 Jens Axboe
2018-08-03 12:00 Jens Axboe
2018-07-31 12:00 Jens Axboe
2018-07-27 12:00 Jens Axboe
2018-07-26 12:00 Jens Axboe
2018-07-25 12:00 Jens Axboe
2018-07-24 12:00 Jens Axboe
2018-07-13 12:00 Jens Axboe
2018-07-12 12:00 Jens Axboe
2018-07-11 12:00 Jens Axboe
2018-07-05 12:00 Jens Axboe
2018-06-30 12:00 Jens Axboe
2018-06-22 12:00 Jens Axboe
2018-06-19 12:00 Jens Axboe
2018-06-16 12:00 Jens Axboe
2018-06-13 12:00 Jens Axboe
2018-06-12 12:00 Jens Axboe
2018-06-09 12:00 Jens Axboe
2018-06-08 12:00 Jens Axboe
2018-06-06 12:00 Jens Axboe
2018-06-05 12:00 Jens Axboe
2018-06-02 12:00 Jens Axboe
2018-06-01 12:00 Jens Axboe
2018-05-26 12:00 Jens Axboe
2018-05-19 12:00 Jens Axboe
2018-05-17 12:00 Jens Axboe
2018-05-15 12:00 Jens Axboe
2018-04-27 12:00 Jens Axboe
2018-04-25 12:00 Jens Axboe
2018-04-21 12:00 Jens Axboe
2018-04-19 12:00 Jens Axboe
2018-04-18 12:00 Jens Axboe
2018-04-17 12:00 Jens Axboe
2018-04-15 12:00 Jens Axboe
2018-04-14 12:00 Jens Axboe
2018-04-11 12:00 Jens Axboe
2018-04-10 12:00 Jens Axboe
2018-04-09 12:00 Jens Axboe
2018-04-07 12:00 Jens Axboe
2018-04-05 12:00 Jens Axboe
2018-04-04 12:00 Jens Axboe
2018-03-31 12:00 Jens Axboe
2018-03-30 12:00 Jens Axboe
2018-03-24 12:00 Jens Axboe
2018-03-23 12:00 Jens Axboe
2018-03-22 12:00 Jens Axboe
2018-03-21 12:00 Jens Axboe
2018-03-20 12:00 Jens Axboe
2018-03-14 12:00 Jens Axboe
2018-03-13 12:00 Jens Axboe
2018-03-10 13:00 Jens Axboe
2018-03-08 13:00 Jens Axboe
2018-03-07 13:00 Jens Axboe
2018-03-06 13:00 Jens Axboe
2018-03-03 13:00 Jens Axboe
2018-03-02 13:00 Jens Axboe
2018-03-01 13:00 Jens Axboe
2018-02-28 13:00 Jens Axboe
2018-02-27 13:00 Jens Axboe
2018-02-21 13:00 Jens Axboe
2018-02-15 13:00 Jens Axboe
2018-02-13 13:00 Jens Axboe
2018-02-11 13:00 Jens Axboe
2018-02-09 13:00 Jens Axboe
2018-02-08 13:00 Jens Axboe
2018-01-26 13:00 Jens Axboe
2018-01-25 13:00 Jens Axboe
2018-01-17 13:00 Jens Axboe
2018-01-13 13:00 Jens Axboe
2018-01-11 13:00 Jens Axboe
2018-01-07 13:00 Jens Axboe
2018-01-06 13:00 Jens Axboe
2018-01-03 13:00 Jens Axboe
2017-12-30 13:00 Jens Axboe
2017-12-29 13:00 Jens Axboe
2017-12-28 13:00 Jens Axboe
2017-12-22 13:00 Jens Axboe
2017-12-20 13:00 Jens Axboe
2017-12-16 13:00 Jens Axboe
2017-12-15 13:00 Jens Axboe
2017-12-14 13:00 Jens Axboe
2017-12-09 13:00 Jens Axboe
2017-12-08 13:00 Jens Axboe
2017-12-07 13:00 Jens Axboe
2017-12-04 13:00 Jens Axboe
2017-12-03 13:00 Jens Axboe
2017-12-02 13:00 Jens Axboe
2017-12-01 13:00 Jens Axboe
2017-11-30 13:00 Jens Axboe
2017-11-29 13:00 Jens Axboe
2017-11-24 13:00 Jens Axboe
2017-11-23 13:00 Jens Axboe
2017-11-18 13:00 Jens Axboe
2017-11-20 15:00 ` Elliott, Robert (Persistent Memory)
2017-11-17 13:00 Jens Axboe
2017-11-16 13:00 Jens Axboe
2017-11-07 13:00 Jens Axboe
2017-11-04 12:00 Jens Axboe
2017-11-03 12:00 Jens Axboe
2017-11-02 12:00 Jens Axboe
2017-11-01 12:00 Jens Axboe
2017-10-31 12:00 Jens Axboe
2017-10-27 12:00 Jens Axboe
2017-10-26 12:00 Jens Axboe
2017-10-21 12:00 Jens Axboe
2017-10-18 12:00 Jens Axboe
2017-10-13 12:00 Jens Axboe
2017-10-12 12:00 Jens Axboe
2017-10-11 12:00 Jens Axboe
2017-10-10 12:00 Jens Axboe
2017-10-07 12:00 Jens Axboe
2017-10-04 12:00 Jens Axboe
2017-09-29 12:00 Jens Axboe
2017-09-28 12:00 Jens Axboe
2017-09-27 12:00 Jens Axboe
2017-09-21 12:00 Jens Axboe
2017-09-19 12:00 Jens Axboe
2017-09-15 12:00 Jens Axboe
2017-09-14 12:00 Jens Axboe
2017-09-13 12:00 Jens Axboe
2017-09-12 12:00 Jens Axboe
2017-09-06 12:00 Jens Axboe
2017-09-03 12:00 Jens Axboe
2017-09-02 12:00 Jens Axboe
2017-09-01 12:00 Jens Axboe
2017-08-31 12:00 Jens Axboe
2017-08-30 12:00 Jens Axboe
2017-08-29 12:00 Jens Axboe
2017-08-28 12:00 Jens Axboe
2017-08-24 12:00 Jens Axboe
2017-08-23 12:00 Jens Axboe
2017-08-18 12:00 Jens Axboe
2017-08-17 12:00 Jens Axboe
2017-08-15 12:00 Jens Axboe
2017-08-10 12:00 Jens Axboe
2017-08-09 12:00 Jens Axboe
2017-08-08 12:00 Jens Axboe
2017-08-02 12:00 Jens Axboe
2017-08-01 12:00 Jens Axboe
2017-07-28 12:00 Jens Axboe
2017-07-26 12:00 Jens Axboe
2017-07-21 12:00 Jens Axboe
2017-07-17 12:00 Jens Axboe
2017-07-15 12:00 Jens Axboe
2017-07-14 12:00 Jens Axboe
2017-07-13 12:00 Jens Axboe
2017-07-11 12:00 Jens Axboe
2017-07-08 12:00 Jens Axboe
2017-07-07 12:00 Jens Axboe
2017-07-05 12:00 Jens Axboe
2017-07-04 12:00 Jens Axboe
2017-07-03 12:00 Jens Axboe
2017-06-29 12:00 Jens Axboe
2017-06-28 12:00 Jens Axboe
2017-06-27 12:00 Jens Axboe
2017-06-26 12:00 Jens Axboe
2017-06-24 12:00 Jens Axboe
2017-06-23 12:00 Jens Axboe
2017-06-20 12:00 Jens Axboe
2017-06-19 12:00 Jens Axboe
2017-06-16 12:00 Jens Axboe
2017-06-15 12:00 Jens Axboe
2017-06-13 12:00 Jens Axboe
2017-06-09 12:00 Jens Axboe
2017-06-08 12:00 Jens Axboe
2017-06-06 12:00 Jens Axboe
2017-06-03 12:00 Jens Axboe
2017-05-27 12:00 Jens Axboe
2017-05-25 12:00 Jens Axboe
2017-05-24 12:00 Jens Axboe
2017-05-23 12:00 Jens Axboe
2017-05-20 12:00 Jens Axboe
2017-05-19 12:00 Jens Axboe
2017-05-10 12:00 Jens Axboe
2017-05-05 12:00 Jens Axboe
2017-05-04 12:00 Jens Axboe
2017-05-02 12:00 Jens Axboe
2017-05-01 12:00 Jens Axboe
2017-04-27 12:00 Jens Axboe
2017-04-26 12:00 Jens Axboe
2017-04-20 12:00 Jens Axboe
2017-04-11 12:00 Jens Axboe
2017-04-09 12:00 Jens Axboe
2017-04-08 12:00 Jens Axboe
2017-04-05 12:00 Jens Axboe
2017-04-04 12:00 Jens Axboe
2017-04-03 12:00 Jens Axboe
2017-03-29 12:00 Jens Axboe
2017-03-22 12:00 Jens Axboe
2017-03-20 12:00 Jens Axboe
2017-03-18 12:00 Jens Axboe
2017-03-17 12:00 Jens Axboe
2017-03-15 12:00 Jens Axboe
2017-03-14 12:00 Jens Axboe
2017-03-13 12:00 Jens Axboe
2017-03-11 13:00 Jens Axboe
2017-03-09 13:00 Jens Axboe
2017-03-08 13:00 Jens Axboe
2017-02-25 13:00 Jens Axboe
2017-02-24 13:00 Jens Axboe
2017-02-23 13:00 Jens Axboe
2017-02-22 13:00 Jens Axboe
2017-02-21 13:00 Jens Axboe
2017-02-20 13:00 Jens Axboe
2017-02-18 13:00 Jens Axboe
2017-02-17 13:00 Jens Axboe
2017-02-16 13:00 Jens Axboe
2017-02-15 13:00 Jens Axboe
2017-02-14 13:00 Jens Axboe
2017-02-08 13:00 Jens Axboe
2017-02-05 13:00 Jens Axboe
2017-02-03 13:00 Jens Axboe
2017-01-31 13:00 Jens Axboe
2017-01-28 13:00 Jens Axboe
2017-01-27 13:00 Jens Axboe
2017-01-24 13:00 Jens Axboe
2017-01-21 13:00 Jens Axboe
2017-01-20 13:00 Jens Axboe
2017-01-19 13:00 Jens Axboe
2017-01-18 13:00 Jens Axboe
2017-01-13 13:00 Jens Axboe
2017-01-17 14:42 ` Elliott, Robert (Persistent Memory)
2017-01-17 15:51 ` Jens Axboe
2017-01-17 16:03 ` Jens Axboe
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=20210130130001.95E371BC0134@kernel.dk \
--to=axboe@kernel.dk \
--cc=fio@vger.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).