Tag 2.3.0-rc2

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>

.github/workflows/codeql.yml

@@ -11,7 +11,7 @@ concurrency:
 jobs:
   analyze:
     name: Analyze
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     permissions:
       actions: read
       contents: read

.github/workflows/scripts/qemu-1-setup.sh

@@ -18,19 +18,21 @@ ssh-keygen -t ed25519 -f ~/.ssh/id_ed25519 -q -N ""
 
 # we expect RAM shortage
 cat << EOF | sudo tee /etc/ksmtuned.conf > /dev/null
+# /etc/ksmtuned.conf - Configuration file for ksmtuned
 # https://docs.redhat.com/en/documentation/red_hat_enterprise_linux/7/html/virtualization_tuning_and_optimization_guide/chap-ksm
 KSM_MONITOR_INTERVAL=60
 
 # Millisecond sleep between ksm scans for 16Gb server.
 # Smaller servers sleep more, bigger sleep less.
-KSM_SLEEP_MSEC=10
+KSM_SLEEP_MSEC=30
-KSM_NPAGES_BOOST=300
-KSM_NPAGES_DECAY=-50
-KSM_NPAGES_MIN=64
-KSM_NPAGES_MAX=2048
 
-KSM_THRES_COEF=25
+KSM_NPAGES_BOOST=0
-KSM_THRES_CONST=2048
+KSM_NPAGES_DECAY=0
+KSM_NPAGES_MIN=1000
+KSM_NPAGES_MAX=25000
+
+KSM_THRES_COEF=80
+KSM_THRES_CONST=8192
 
 LOGFILE=/var/log/ksmtuned.log
 DEBUG=1

.github/workflows/scripts/qemu-2-start.sh

@@ -14,7 +14,7 @@ OSv=$OS
 
 # compressed with .zst extension
 REPO="https://github.com/mcmilk/openzfs-freebsd-images"
-FREEBSD="$REPO/releases/download/v2024-09-16"
+FREEBSD="$REPO/releases/download/v2024-10-05"
 URLzs=""
 
 # Ubuntu mirrors
@@ -62,33 +62,45 @@ case "$OS" in
     OSv="fedora39"
     URL="https://download.fedoraproject.org/pub/fedora/linux/releases/40/Cloud/x86_64/images/Fedora-Cloud-Base-Generic.x86_64-40-1.14.qcow2"
     ;;
-  freebsd13r)
+  freebsd13-3r)
+    OSNAME="FreeBSD 13.3-RELEASE"
+    OSv="freebsd13.0"
+    URLzs="$FREEBSD/amd64-freebsd-13.3-RELEASE.qcow2.zst"
+    BASH="/usr/local/bin/bash"
+    NIC="rtl8139"
+    ;;
+  freebsd13-4r)
     OSNAME="FreeBSD 13.4-RELEASE"
     OSv="freebsd13.0"
     URLzs="$FREEBSD/amd64-freebsd-13.4-RELEASE.qcow2.zst"
     BASH="/usr/local/bin/bash"
     NIC="rtl8139"
     ;;
-  freebsd13)
+  freebsd14-0r)
-    OSNAME="FreeBSD 13.4-STABLE"
+    OSNAME="FreeBSD 14.0-RELEASE"
-    OSv="freebsd13.0"
+    OSv="freebsd14.0"
-    URLzs="$FREEBSD/amd64-freebsd-13.4-STABLE.qcow2.zst"
+    URLzs="$FREEBSD/amd64-freebsd-14.0-RELEASE.qcow2.zst"
     BASH="/usr/local/bin/bash"
-    NIC="rtl8139"
     ;;
-  freebsd14r)
+  freebsd14-1r)
     OSNAME="FreeBSD 14.1-RELEASE"
     OSv="freebsd14.0"
     URLzs="$FREEBSD/amd64-freebsd-14.1-RELEASE.qcow2.zst"
     BASH="/usr/local/bin/bash"
     ;;
-  freebsd14)
+  freebsd13-4s)
+    OSNAME="FreeBSD 13.4-STABLE"
+    OSv="freebsd13.0"
+    URLzs="$FREEBSD/amd64-freebsd-13.4-STABLE.qcow2.zst"
+    BASH="/usr/local/bin/bash"
+    ;;
+  freebsd14-1s)
     OSNAME="FreeBSD 14.1-STABLE"
     OSv="freebsd14.0"
     URLzs="$FREEBSD/amd64-freebsd-14.1-STABLE.qcow2.zst"
     BASH="/usr/local/bin/bash"
     ;;
-  freebsd15)
+  freebsd15-0c)
     OSNAME="FreeBSD 15.0-CURRENT"
     OSv="freebsd14.0"
     URLzs="$FREEBSD/amd64-freebsd-15.0-CURRENT.qcow2.zst"

.github/workflows/scripts/qemu-5-setup.sh

@@ -14,17 +14,21 @@ PID=$(pidof /usr/bin/qemu-system-x86_64)
 tail --pid=$PID -f /dev/null
 sudo virsh undefine openzfs
 
-# definitions of per operating system
+# default values per test vm:
+VMs=2
+CPU=2
+
+# cpu pinning
+CPUSET=("0,1" "2,3")
+
 case "$OS" in
   freebsd*)
-    VMs=2
+    # FreeBSD can't be optimized via ksmtuned
-    CPU=3
     RAM=6
     ;;
   *)
-    VMs=2
+    # Linux can be optimized via ksmtuned
-    CPU=3
+    RAM=8
-    RAM=7
     ;;
 esac
 
@@ -73,6 +77,7 @@ EOF
     --cpu host-passthrough \
     --virt-type=kvm --hvm \
     --vcpus=$CPU,sockets=1 \
+    --cpuset=${CPUSET[$((i-1))]} \
     --memory $((1024*RAM)) \
     --memballoon model=virtio \
     --graphics none \

.github/workflows/scripts/qemu-9-summary-page.sh

@@ -11,12 +11,10 @@ function output() {
 }
 
 function outfile() {
-  test -s "$1" || return
   cat "$1" >> "out-$logfile.md"
 }
 
 function outfile_plain() {
-  test -s "$1" || return
   output "<pre>"
   cat "$1" >> "out-$logfile.md"
   output "</pre>"
@@ -45,6 +43,8 @@ if [ ! -f out-1.md ]; then
     tar xf "$tarfile"
     test -s env.txt || continue
     source env.txt
+    # when uname.txt is there, the other files are also ok
+    test -s uname.txt || continue
     output "\n## Functional Tests: $OSNAME\n"
     outfile_plain uname.txt
     outfile_plain summary.txt

.github/workflows/zfs-qemu.yml

@@ -22,8 +22,8 @@ jobs:
       - name: Generate OS config and CI type
         id: os
         run: |
-          FULL_OS='["almalinux8", "almalinux9", "centos-stream9", "debian11", "debian12", "fedora39", "fedora40", "freebsd13", "freebsd13r", "freebsd14", "freebsd14r", "ubuntu20", "ubuntu22", "ubuntu24"]'
+          FULL_OS='["almalinux8", "almalinux9", "centos-stream9", "debian11", "debian12", "fedora39", "fedora40", "freebsd13-4r", "freebsd14-0r", "freebsd14-1s", "ubuntu20", "ubuntu22", "ubuntu24"]'
-          QUICK_OS='["almalinux8", "almalinux9", "debian12", "fedora40", "freebsd13", "freebsd14", "ubuntu24"]'
+          QUICK_OS='["almalinux8", "almalinux9", "debian12", "fedora40", "freebsd13-3r", "freebsd14-1r", "ubuntu24"]'
           # determine CI type when running on PR
           ci_type="full"
           if ${{ github.event_name == 'pull_request' }}; then
@@ -46,10 +46,12 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        # all:
+        # rhl:     almalinux8, almalinux9, centos-stream9, fedora39, fedora40
-        # os:  [almalinux8, almalinux9, archlinux, centos-stream9, fedora39, fedora40, debian11, debian12, freebsd13, freebsd13r, freebsd14, freebsd14r, freebsd15, ubuntu20, ubuntu22, ubuntu24]
+        # debian:  debian11, debian12, ubuntu20, ubuntu22, ubuntu24
-        # openzfs:
+        # misc:    archlinux, tumbleweed
-        # os: [almalinux8, almalinux9, centos-stream9, debian11, debian12, fedora39, fedora40, freebsd13, freebsd13r, freebsd14, freebsd14r, ubuntu20, ubuntu22, ubuntu24]
+        # FreeBSD Release: freebsd13-3r, freebsd13-4r, freebsd14-0r, freebsd14-1r
+        # FreeBSD Stable:  freebsd13-4s, freebsd14-1s
+        # FreeBSD Current: freebsd15-0c
         os: ${{ fromJson(needs.test-config.outputs.test_os) }}
     runs-on: ubuntu-24.04
     steps:

META

@@ -2,7 +2,7 @@ Meta:          1
 Name:          zfs
 Branch:        1.0
 Version:       2.3.0
-Release:       rc1
+Release:       rc2
 Release-Tags:  relext
 License:       CDDL
 Author:        OpenZFS

cmd/zdb/zdb.c

@@ -1131,7 +1131,7 @@ dump_zap(objset_t *os, uint64_t object, void *data, size_t size)
 		    !!(zap_getflags(zc.zc_zap) & ZAP_FLAG_UINT64_KEY);
 
 		if (key64)
-			(void) printf("\t\t0x%010lx = ",
+			(void) printf("\t\t0x%010" PRIu64 "x = ",
 			    *(uint64_t *)attrp->za_name);
 		else
 			(void) printf("\t\t%s = ", attrp->za_name);

cmd/zfs/zfs_main.c

@@ -2162,6 +2162,7 @@ zfs_do_get(int argc, char **argv)
 	cb.cb_type = ZFS_TYPE_DATASET;
 
 	struct option long_options[] = {
+		{"json", no_argument, NULL, 'j'},
 		{"json-int", no_argument, NULL, ZFS_OPTION_JSON_NUMS_AS_INT},
 		{0, 0, 0, 0}
 	};
@@ -3852,6 +3853,7 @@ zfs_do_list(int argc, char **argv)
 	nvlist_t *data = NULL;
 
 	struct option long_options[] = {
+		{"json", no_argument, NULL, 'j'},
 		{"json-int", no_argument, NULL, ZFS_OPTION_JSON_NUMS_AS_INT},
 		{0, 0, 0, 0}
 	};
@@ -7436,9 +7438,15 @@ share_mount(int op, int argc, char **argv)
 	uint_t nthr;
 	jsobj = data = item = NULL;
 
+	struct option long_options[] = {
+		{"json", no_argument, NULL, 'j'},
+		{0, 0, 0, 0}
+	};
+
 	/* check options */
-	while ((c = getopt(argc, argv, op == OP_MOUNT ? ":ajRlvo:Of" : "al"))
+	while ((c = getopt_long(argc, argv,
-	    != -1) {
+	    op == OP_MOUNT ? ":ajRlvo:Of" : "al",
+	    op == OP_MOUNT ? long_options : NULL, NULL)) != -1) {
 		switch (c) {
 		case 'a':
 			do_all = 1;
@@ -8374,8 +8382,14 @@ zfs_do_channel_program(int argc, char **argv)
 	boolean_t sync_flag = B_TRUE, json_output = B_FALSE;
 	zpool_handle_t *zhp;
 
+	struct option long_options[] = {
+		{"json", no_argument, NULL, 'j'},
+		{0, 0, 0, 0}
+	};
+
 	/* check options */
-	while ((c = getopt(argc, argv, "nt:m:j")) != -1) {
+	while ((c = getopt_long(argc, argv, "nt:m:j", long_options,
+	    NULL)) != -1) {
 		switch (c) {
 		case 't':
 		case 'm': {
@@ -9083,7 +9097,13 @@ zfs_do_version(int argc, char **argv)
 	int c;
 	nvlist_t *jsobj = NULL, *zfs_ver = NULL;
 	boolean_t json = B_FALSE;
-	while ((c = getopt(argc, argv, "j")) != -1) {
+
+	struct option long_options[] = {
+		{"json", no_argument, NULL, 'j'},
+		{0, 0, 0, 0}
+	};
+
+	while ((c = getopt_long(argc, argv, "j", long_options, NULL)) != -1) {
 		switch (c) {
 		case 'j':
 			json = B_TRUE;
@@ -9187,7 +9207,7 @@ main(int argc, char **argv)
 	 * Special case '-V|--version'
 	 */
 	if ((strcmp(cmdname, "-V") == 0) || (strcmp(cmdname, "--version") == 0))
-		return (zfs_do_version(argc, argv));
+		return (zfs_version_print() != 0);
 
 	/*
 	 * Special case 'help'

cmd/zpool/zpool_main.c

@@ -7340,6 +7340,7 @@ zpool_do_list(int argc, char **argv)
 	current_prop_type = ZFS_TYPE_POOL;
 
 	struct option long_options[] = {
+		{"json", no_argument, NULL, 'j'},
 		{"json-int", no_argument, NULL, ZPOOL_OPTION_JSON_NUMS_AS_INT},
 		{"json-pool-key-guid", no_argument, NULL,
 		    ZPOOL_OPTION_POOL_KEY_GUID},
@@ -9224,6 +9225,12 @@ vdev_stats_nvlist(zpool_handle_t *zhp, status_cbdata_t *cb, nvlist_t *nv,
 		}
 	}
 
+	if (cb->cb_print_dio_verify) {
+		nice_num_str_nvlist(vds, "dio_verify_errors",
+		    vs->vs_dio_verify_errors, cb->cb_literal,
+		    cb->cb_json_as_int, ZFS_NICENUM_1024);
+	}
+
 	if (nvlist_lookup_uint64(nv, ZPOOL_CONFIG_NOT_PRESENT,
 	    &notpresent) == 0) {
 		nice_num_str_nvlist(vds, ZPOOL_CONFIG_NOT_PRESENT,
@@ -10975,6 +10982,7 @@ zpool_do_status(int argc, char **argv)
 
 	struct option long_options[] = {
 		{"power", no_argument, NULL, ZPOOL_OPTION_POWER},
+		{"json", no_argument, NULL, 'j'},
 		{"json-int", no_argument, NULL, ZPOOL_OPTION_JSON_NUMS_AS_INT},
 		{"json-flat-vdevs", no_argument, NULL,
 		    ZPOOL_OPTION_JSON_FLAT_VDEVS},
@@ -12583,6 +12591,7 @@ zpool_do_get(int argc, char **argv)
 	current_prop_type = cb.cb_type;
 
 	struct option long_options[] = {
+		{"json", no_argument, NULL, 'j'},
 		{"json-int", no_argument, NULL, ZPOOL_OPTION_JSON_NUMS_AS_INT},
 		{"json-pool-key-guid", no_argument, NULL,
 		    ZPOOL_OPTION_POOL_KEY_GUID},
@@ -13497,7 +13506,12 @@ zpool_do_version(int argc, char **argv)
 	int c;
 	nvlist_t *jsobj = NULL, *zfs_ver = NULL;
 	boolean_t json = B_FALSE;
-	while ((c = getopt(argc, argv, "j")) != -1) {
+
+	struct option long_options[] = {
+		{"json", no_argument, NULL, 'j'},
+	};
+
+	while ((c = getopt_long(argc, argv, "j", long_options, NULL)) != -1) {
 		switch (c) {
 		case 'j':
 			json = B_TRUE;
@@ -13613,7 +13627,7 @@ main(int argc, char **argv)
 	 * Special case '-V|--version'
 	 */
 	if ((strcmp(cmdname, "-V") == 0) || (strcmp(cmdname, "--version") == 0))
-		return (zpool_do_version(argc, argv));
+		return (zfs_version_print() != 0);
 
 	/*
 	 * Special case 'help'

cmd/ztest.c

@@ -6717,6 +6717,17 @@ out:
  *
  * Only after a full scrub has been completed is it safe to start injecting
  * data corruption.  See the comment in zfs_fault_inject().
+ *
+ * EBUSY may be returned for the following six cases.  It's the callers
+ * responsibility to handle them accordingly.
+ *
+ * Current state                Requested
+ * 1. Normal Scrub Running      Normal Scrub or Error Scrub
+ * 2. Normal Scrub Paused       Error Scrub
+ * 3. Normal Scrub Paused       Pause Normal Scrub
+ * 4. Error Scrub Running       Normal Scrub or Error Scrub
+ * 5. Error Scrub Paused        Pause Error Scrub
+ * 6. Resilvering               Anything else
  */
 static int
 ztest_scrub_impl(spa_t *spa)
@@ -8082,8 +8093,14 @@ ztest_raidz_expand_check(spa_t *spa)
 		(void) printf("verifying an interrupted raidz "
 		    "expansion using a pool scrub ...\n");
 	}
+
 	/* Will fail here if there is non-recoverable corruption detected */
-	VERIFY0(ztest_scrub_impl(spa));
+	int error = ztest_scrub_impl(spa);
+	if (error == EBUSY)
+		error = 0;
+
+	VERIFY0(error);
+
 	if (ztest_opts.zo_verbose >= 1) {
 		(void) printf("raidz expansion scrub check complete\n");
 	}

config/deb.am

@@ -58,9 +58,9 @@ deb-utils: deb-local rpm-utils-initramfs
 	pkg1=$${name}-$${version}.$${arch}.rpm; \
 	pkg2=libnvpair3-$${version}.$${arch}.rpm; \
 	pkg3=libuutil3-$${version}.$${arch}.rpm; \
-	pkg4=libzfs5-$${version}.$${arch}.rpm; \
+	pkg4=libzfs6-$${version}.$${arch}.rpm; \
-	pkg5=libzpool5-$${version}.$${arch}.rpm; \
+	pkg5=libzpool6-$${version}.$${arch}.rpm; \
-	pkg6=libzfs5-devel-$${version}.$${arch}.rpm; \
+	pkg6=libzfs6-devel-$${version}.$${arch}.rpm; \
 	pkg7=$${name}-test-$${version}.$${arch}.rpm; \
 	pkg8=$${name}-dracut-$${version}.noarch.rpm; \
 	pkg9=$${name}-initramfs-$${version}.$${arch}.rpm; \
@@ -72,7 +72,7 @@ deb-utils: deb-local rpm-utils-initramfs
 	path_prepend=`mktemp -d /tmp/intercept.XXXXXX`; \
 	echo "#!$(SHELL)" > $${path_prepend}/dh_shlibdeps; \
 	echo "`which dh_shlibdeps` -- \
-	 -xlibuutil3linux -xlibnvpair3linux -xlibzfs5linux -xlibzpool5linux" \
+	 -xlibuutil3linux -xlibnvpair3linux -xlibzfs6linux -xlibzpool6linux" \
 	 >> $${path_prepend}/dh_shlibdeps; \
 ## These -x arguments are passed to dpkg-shlibdeps, which exclude the
 ## Debianized packages from the auto-generated dependencies of the new debs,
@@ -93,13 +93,17 @@ debian:
 	cp -r contrib/debian debian; chmod +x debian/rules;
 
 native-deb-utils: native-deb-local debian
+	while [ -f debian/deb-build.lock ]; do sleep 1; done; \
+	echo "native-deb-utils" > debian/deb-build.lock; \
 	cp contrib/debian/control debian/control; \
-	$(DPKGBUILD) -b -rfakeroot -us -uc;
+	$(DPKGBUILD) -b -rfakeroot -us -uc; \
+	$(RM) -f debian/deb-build.lock
 
 native-deb-kmod: native-deb-local debian
+	while [ -f debian/deb-build.lock ]; do sleep 1; done; \
+	echo "native-deb-kmod" > debian/deb-build.lock; \
 	sh scripts/make_gitrev.sh; \
-	fakeroot debian/rules override_dh_binary-modules;
+	fakeroot debian/rules override_dh_binary-modules; \
+	$(RM) -f debian/deb-build.lock
 
 native-deb: native-deb-utils native-deb-kmod
-
-.NOTPARALLEL: native-deb native-deb-utils native-deb-kmod

config/user.m4

@@ -33,7 +33,7 @@ AC_DEFUN([ZFS_AC_CONFIG_USER], [
 	ZFS_AC_CONFIG_USER_MAKEDEV_IN_MKDEV
 	ZFS_AC_CONFIG_USER_ZFSEXEC
 
-	AC_CHECK_FUNCS([execvpe issetugid mlockall strlcat strlcpy gettid])
+	AC_CHECK_FUNCS([execvpe issetugid mlockall strerror_l strlcat strlcpy gettid])
 
 	AC_SUBST(RM)
 ])

contrib/debian/Makefile.am

@@ -12,14 +12,14 @@ dist_noinst_DATA += %D%/openzfs-libpam-zfs.postinst
 dist_noinst_DATA += %D%/openzfs-libpam-zfs.prerm
 dist_noinst_DATA += %D%/openzfs-libuutil3.docs
 dist_noinst_DATA += %D%/openzfs-libuutil3.install.in
-dist_noinst_DATA += %D%/openzfs-libzfs4.docs
+dist_noinst_DATA += %D%/openzfs-libzfs6.docs
-dist_noinst_DATA += %D%/openzfs-libzfs4.install.in
+dist_noinst_DATA += %D%/openzfs-libzfs6.install.in
 dist_noinst_DATA += %D%/openzfs-libzfsbootenv1.docs
 dist_noinst_DATA += %D%/openzfs-libzfsbootenv1.install.in
 dist_noinst_DATA += %D%/openzfs-libzfs-dev.docs
 dist_noinst_DATA += %D%/openzfs-libzfs-dev.install.in
-dist_noinst_DATA += %D%/openzfs-libzpool5.docs
+dist_noinst_DATA += %D%/openzfs-libzpool6.docs
-dist_noinst_DATA += %D%/openzfs-libzpool5.install.in
+dist_noinst_DATA += %D%/openzfs-libzpool6.install.in
 dist_noinst_DATA += %D%/openzfs-python3-pyzfs.install
 dist_noinst_DATA += %D%/openzfs-zfs-dkms.config
 dist_noinst_DATA += %D%/openzfs-zfs-dkms.dkms

contrib/debian/clean

@@ -6,6 +6,6 @@ contrib/pyzfs/libzfs_core/bindings/__pycache__/
 contrib/pyzfs/pyzfs.egg-info/
 debian/openzfs-libnvpair3.install
 debian/openzfs-libuutil3.install
-debian/openzfs-libzfs4.install
+debian/openzfs-libzfs6.install
 debian/openzfs-libzfs-dev.install
-debian/openzfs-libzpool5.install
+debian/openzfs-libzpool6.install

contrib/debian/control

@@ -78,9 +78,9 @@ Architecture: linux-any
 Depends: libssl-dev | libssl1.0-dev,
          openzfs-libnvpair3 (= ${binary:Version}),
          openzfs-libuutil3 (= ${binary:Version}),
-         openzfs-libzfs4 (= ${binary:Version}),
+         openzfs-libzfs6 (= ${binary:Version}),
          openzfs-libzfsbootenv1 (= ${binary:Version}),
-         openzfs-libzpool5 (= ${binary:Version}),
+         openzfs-libzpool6 (= ${binary:Version}),
          ${misc:Depends}
 Replaces: libzfslinux-dev
 Conflicts: libzfslinux-dev
@@ -90,18 +90,18 @@ Description: OpenZFS filesystem development files for Linux
  libraries of OpenZFS filesystem.
  .
  This package includes the development files of libnvpair3, libuutil3,
- libzpool5 and libzfs4.
+ libzpool6 and libzfs6.
 
-Package: openzfs-libzfs4
+Package: openzfs-libzfs6
 Section: contrib/libs
 Architecture: linux-any
 Depends: ${misc:Depends}, ${shlibs:Depends}
 # The libcurl4 is loaded through dlopen("libcurl.so.4").
 # https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=988521
 Recommends: libcurl4
-Breaks: libzfs2, libzfs4
+Breaks: libzfs2, libzfs4, libzfs4linux, libzfs6linux
-Replaces: libzfs2, libzfs4, libzfs4linux
+Replaces: libzfs2, libzfs4, libzfs4linux, libzfs6linux
-Conflicts: libzfs4linux
+Conflicts: libzfs6linux
 Description: OpenZFS filesystem library for Linux - general support
  OpenZFS is a storage platform that encompasses the functionality of
  traditional filesystems and volume managers. It supports data checksums,
@@ -123,13 +123,13 @@ Description: OpenZFS filesystem library for Linux - label info support
  .
  The zfsbootenv library provides support for modifying ZFS label information.
 
-Package: openzfs-libzpool5
+Package: openzfs-libzpool6
 Section: contrib/libs
 Architecture: linux-any
 Depends: ${misc:Depends}, ${shlibs:Depends}
-Breaks: libzpool2, libzpool5
+Breaks: libzpool2, libzpool5, libzpool5linux, libzpool6linux
-Replaces: libzpool2, libzpool5, libzpool5linux
+Replaces: libzpool2, libzpool5, libzpool5linux, libzpool6linux
-Conflicts: libzpool5linux
+Conflicts: libzpool6linux
 Description: OpenZFS pool library for Linux
  OpenZFS is a storage platform that encompasses the functionality of
  traditional filesystems and volume managers. It supports data checksums,
@@ -246,8 +246,8 @@ Architecture: linux-any
 Pre-Depends: ${misc:Pre-Depends}
 Depends: openzfs-libnvpair3 (= ${binary:Version}),
          openzfs-libuutil3 (= ${binary:Version}),
-         openzfs-libzfs4 (= ${binary:Version}),
+         openzfs-libzfs6 (= ${binary:Version}),
-         openzfs-libzpool5 (= ${binary:Version}),
+         openzfs-libzpool6 (= ${binary:Version}),
          python3,
          ${misc:Depends},
          ${shlibs:Depends}

contrib/debian/openzfs-zfsutils.install

@@ -98,6 +98,7 @@ usr/share/man/man8/zpool-attach.8
 usr/share/man/man8/zpool-checkpoint.8
 usr/share/man/man8/zpool-clear.8
 usr/share/man/man8/zpool-create.8
+usr/share/man/man8/zpool-ddtprune.8
 usr/share/man/man8/zpool-destroy.8
 usr/share/man/man8/zpool-detach.8
 usr/share/man/man8/zpool-ddtprune.8
@@ -113,6 +114,7 @@ usr/share/man/man8/zpool-list.8
 usr/share/man/man8/zpool-offline.8
 usr/share/man/man8/zpool-online.8
 usr/share/man/man8/zpool-prefetch.8
+usr/share/man/man8/zpool-prefetch.8
 usr/share/man/man8/zpool-reguid.8
 usr/share/man/man8/zpool-remove.8
 usr/share/man/man8/zpool-reopen.8

include/libzutil.h

@@ -276,7 +276,11 @@ _LIBZUTIL_H void update_vdev_config_dev_sysfs_path(nvlist_t *nv,
  * Thread-safe strerror() for use in ZFS libraries
  */
 static inline char *zfs_strerror(int errnum) {
+#ifdef HAVE_STRERROR_L
 	return (strerror_l(errnum, uselocale(0)));
+#else
+	return (strerror(errnum));
+#endif
 }
 
 #ifdef	__cplusplus

include/os/freebsd/zfs/sys/abd_os.h

@@ -26,8 +26,10 @@
 #ifndef _ABD_OS_H
 #define	_ABD_OS_H
 
+#ifdef _KERNEL
 #include <sys/vm.h>
 #include <vm/vm_page.h>
+#endif
 
 #ifdef __cplusplus
 extern "C" {
@@ -47,8 +49,10 @@ struct abd_linear {
 #endif
 };
 
+#ifdef _KERNEL
 __attribute__((malloc))
 struct abd *abd_alloc_from_pages(vm_page_t *, unsigned long, uint64_t);
+#endif
 
 #ifdef __cplusplus
 }

include/sys/fm/fs/zfs.h

@@ -42,7 +42,8 @@ extern "C" {
 #define	FM_EREPORT_ZFS_DATA			"data"
 #define	FM_EREPORT_ZFS_DELAY			"delay"
 #define	FM_EREPORT_ZFS_DEADMAN			"deadman"
-#define	FM_EREPORT_ZFS_DIO_VERIFY		"dio_verify"
+#define	FM_EREPORT_ZFS_DIO_VERIFY_WR		"dio_verify_wr"
+#define	FM_EREPORT_ZFS_DIO_VERIFY_RD		"dio_verify_rd"
 #define	FM_EREPORT_ZFS_POOL			"zpool"
 #define	FM_EREPORT_ZFS_DEVICE_UNKNOWN		"vdev.unknown"
 #define	FM_EREPORT_ZFS_DEVICE_OPEN_FAILED	"vdev.open_failed"

include/sys/vdev_raidz.h

@@ -57,7 +57,7 @@ void vdev_raidz_reconstruct(struct raidz_map *, const int *, int);
 void vdev_raidz_child_done(zio_t *);
 void vdev_raidz_io_done(zio_t *);
 void vdev_raidz_checksum_error(zio_t *, struct raidz_col *, abd_t *);
-struct raidz_row *vdev_raidz_row_alloc(int);
+struct raidz_row *vdev_raidz_row_alloc(int, zio_t *);
 void vdev_raidz_reflow_copy_scratch(spa_t *);
 void raidz_dtl_reassessed(vdev_t *);
 

include/sys/zio.h

@@ -208,25 +208,25 @@ typedef uint64_t zio_flag_t;
 #define	ZIO_FLAG_PROBE		(1ULL << 16)
 #define	ZIO_FLAG_TRYHARD	(1ULL << 17)
 #define	ZIO_FLAG_OPTIONAL	(1ULL << 18)
-
+#define	ZIO_FLAG_DIO_READ	(1ULL << 19)
 #define	ZIO_FLAG_VDEV_INHERIT	(ZIO_FLAG_DONT_QUEUE - 1)
 
 	/*
 	 * Flags not inherited by any children.
 	 */
-#define	ZIO_FLAG_DONT_QUEUE	(1ULL << 19)	/* must be first for INHERIT */
+#define	ZIO_FLAG_DONT_QUEUE	(1ULL << 20)	/* must be first for INHERIT */
-#define	ZIO_FLAG_DONT_PROPAGATE	(1ULL << 20)
+#define	ZIO_FLAG_DONT_PROPAGATE	(1ULL << 21)
-#define	ZIO_FLAG_IO_BYPASS	(1ULL << 21)
+#define	ZIO_FLAG_IO_BYPASS	(1ULL << 22)
-#define	ZIO_FLAG_IO_REWRITE	(1ULL << 22)
+#define	ZIO_FLAG_IO_REWRITE	(1ULL << 23)
-#define	ZIO_FLAG_RAW_COMPRESS	(1ULL << 23)
+#define	ZIO_FLAG_RAW_COMPRESS	(1ULL << 24)
-#define	ZIO_FLAG_RAW_ENCRYPT	(1ULL << 24)
+#define	ZIO_FLAG_RAW_ENCRYPT	(1ULL << 25)
-#define	ZIO_FLAG_GANG_CHILD	(1ULL << 25)
+#define	ZIO_FLAG_GANG_CHILD	(1ULL << 26)
-#define	ZIO_FLAG_DDT_CHILD	(1ULL << 26)
+#define	ZIO_FLAG_DDT_CHILD	(1ULL << 27)
-#define	ZIO_FLAG_GODFATHER	(1ULL << 27)
+#define	ZIO_FLAG_GODFATHER	(1ULL << 28)
-#define	ZIO_FLAG_NOPWRITE	(1ULL << 28)
+#define	ZIO_FLAG_NOPWRITE	(1ULL << 29)
-#define	ZIO_FLAG_REEXECUTED	(1ULL << 29)
+#define	ZIO_FLAG_REEXECUTED	(1ULL << 30)
-#define	ZIO_FLAG_DELEGATED	(1ULL << 30)
+#define	ZIO_FLAG_DELEGATED	(1ULL << 31)
-#define	ZIO_FLAG_DIO_CHKSUM_ERR	(1ULL << 31)
+#define	ZIO_FLAG_DIO_CHKSUM_ERR	(1ULL << 32)
 
 #define	ZIO_ALLOCATOR_NONE	(-1)
 #define	ZIO_HAS_ALLOCATOR(zio)	((zio)->io_allocator != ZIO_ALLOCATOR_NONE)
@@ -647,6 +647,7 @@ extern void zio_vdev_io_redone(zio_t *zio);
 extern void zio_change_priority(zio_t *pio, zio_priority_t priority);
 
 extern void zio_checksum_verified(zio_t *zio);
+extern void zio_dio_chksum_verify_error_report(zio_t *zio);
 extern int zio_worst_error(int e1, int e2);
 
 extern enum zio_checksum zio_checksum_select(enum zio_checksum child,

lib/libspl/include/sys/uio.h

@@ -92,20 +92,20 @@ zfs_dio_page_aligned(void *buf)
 static inline boolean_t
 zfs_dio_offset_aligned(uint64_t offset, uint64_t blksz)
 {
-	return (IS_P2ALIGNED(offset, blksz));
+	return ((IS_P2ALIGNED(offset, blksz)) ? B_TRUE : B_FALSE);
 }
 
 static inline boolean_t
 zfs_dio_size_aligned(uint64_t size, uint64_t blksz)
 {
-	return ((size % blksz) == 0);
+	return (((size % blksz) == 0) ? B_TRUE : B_FALSE);
 }
 
 static inline boolean_t
 zfs_dio_aligned(uint64_t offset, uint64_t size, uint64_t blksz)
 {
-	return (zfs_dio_offset_aligned(offset, blksz) &&
+	return ((zfs_dio_offset_aligned(offset, blksz) &&
-	    zfs_dio_size_aligned(size, blksz));
+	    zfs_dio_size_aligned(size, blksz)) ? B_TRUE : B_FALSE);
 }
 
 static inline void

lib/libzfs/Makefile.am

@@ -70,7 +70,7 @@ if BUILD_FREEBSD
 libzfs_la_LIBADD += -lutil -lgeom
 endif
 
-libzfs_la_LDFLAGS += -version-info 5:0:1
+libzfs_la_LDFLAGS += -version-info 6:0:0
 
 pkgconfig_DATA += %D%/libzfs.pc
 

lib/libzfs/libzfs.abi

@@ -1,4 +1,4 @@
-<abi-corpus version='2.0' architecture='elf-amd-x86_64' soname='libzfs.so.4'>
+<abi-corpus version='2.0' architecture='elf-amd-x86_64' soname='libzfs.so.6'>
   <elf-needed>
     <dependency name='libzfs_core.so.3'/>
     <dependency name='libnvpair.so.3'/>

lib/libzfs/libzfs_pool.c

@@ -2796,7 +2796,7 @@ zpool_scan(zpool_handle_t *zhp, pool_scan_func_t func, pool_scrub_cmd_t cmd)
 	}
 
 	/*
-	 * With EBUSY, five cases are possible:
+	 * With EBUSY, six cases are possible:
 	 *
 	 * Current state		Requested
 	 * 1. Normal Scrub Running	Normal Scrub or Error Scrub

lib/libzpool/Makefile.am

@@ -212,7 +212,7 @@ if BUILD_FREEBSD
 libzpool_la_LIBADD += -lgeom
 endif
 
-libzpool_la_LDFLAGS += -version-info 5:0:0
+libzpool_la_LDFLAGS += -version-info 6:0:0
 
 if TARGET_CPU_POWERPC
 module/zfs/libzpool_la-vdev_raidz_math_powerpc_altivec.$(OBJEXT) : CFLAGS += -maltivec

man/man4/zfs.4

@@ -436,7 +436,7 @@ write.
 It can also help to identify if reported checksum errors are tied to Direct I/O
 writes.
 Each verify error causes a
-.Sy dio_verify
+.Sy dio_verify_wr
 zevent.
 Direct Write I/O checkum verify errors can be seen with
 .Nm zpool Cm status Fl d .

man/man5/vdev_id.conf.5

@@ -92,6 +92,11 @@ before a generic mapping for the same slot.
 In this way a custom mapping may be applied to a particular channel
 and a default mapping applied to the others.
 .
+.It Sy zpad_slot Ar digits
+Pad slot numbers with zeros to make them
+.Ar digits
+long, which can help to make disk names a consistent length and easier to sort.
+.
 .It Sy multipath Sy yes Ns | Ns Sy no
 Specifies whether
 .Xr vdev_id 8
@@ -122,7 +127,7 @@ device is connected to.
 The default is
 .Sy 4 .
 .
-.It Sy slot Sy bay Ns | Ns Sy phy Ns | Ns Sy port Ns | Ns Sy id Ns | Ns Sy lun Ns | Ns Sy ses
+.It Sy slot Sy bay Ns | Ns Sy phy Ns | Ns Sy port Ns | Ns Sy id Ns | Ns Sy lun Ns | Ns Sy bay_lun Ns | Ns Sy ses
 Specifies from which element of a SAS identifier the slot number is
 taken.
 The default is
@@ -138,6 +143,9 @@ use the SAS port as the slot number.
 use the scsi id as the slot number.
 .It Sy lun
 use the scsi lun as the slot number.
+.It Sy bay_lun
+read the slot number from the bay identifier and append the lun number.
+Useful for multi-lun multi-actuator hard drives.
 .It Sy ses
 use the SCSI Enclosure Services (SES) enclosure device slot number,
 as reported by

man/man8/zfs-list.8

@@ -71,7 +71,7 @@ The following fields are displayed:
 Used for scripting mode.
 Do not print headers and separate fields by a single tab instead of arbitrary
 white space.
-.It Fl j Op Ar --json-int
+.It Fl j , -json Op Ar --json-int
 Print the output in JSON format.
 Specify
 .Sy --json-int

man/man8/zfs-mount.8

@@ -59,7 +59,7 @@
 .Xc
 Displays all ZFS file systems currently mounted.
 .Bl -tag -width "-j"
-.It Fl j
+.It Fl j , -json
 Displays all mounted file systems in JSON format.
 .El
 .It Xo

man/man8/zfs-program.8

@@ -50,7 +50,7 @@ and any attempts to access or modify other pools will cause an error.
 .
 .Sh OPTIONS
 .Bl -tag -width "-t"
-.It Fl j
+.It Fl j , -json
 Display channel program output in JSON format.
 When this flag is specified and standard output is empty -
 channel program encountered an error.

man/man8/zfs-set.8

@@ -130,7 +130,7 @@ The value
 can be used to display all properties that apply to the given dataset's type
 .Pq Sy filesystem , volume , snapshot , No or Sy bookmark .
 .Bl -tag -width "-s source"
-.It Fl j Op Ar --json-int
+.It Fl j , -json Op Ar --json-int
 Display the output in JSON format.
 Specify
 .Sy --json-int

man/man8/zpool-events.8

@@ -98,7 +98,10 @@ This can be an indicator of problems with the underlying storage device.
 The number of delay events is ratelimited by the
 .Sy zfs_slow_io_events_per_second
 module parameter.
-.It Sy dio_verify
+.It Sy dio_verify_rd
+Issued when there was a checksum verify error after a Direct I/O read has been
+issued.
+.It Sy dio_verify_wr
 Issued when there was a checksum verify error after a Direct I/O write has been
 issued.
 This event can only take place if the module parameter

man/man8/zpool-get.8

@@ -98,7 +98,7 @@ See the
 .Xr zpoolprops 7
 manual page for more information on the available pool properties.
 .Bl -tag -compact -offset Ds -width "-o field"
-.It Fl j Op Ar --json-int, --json-pool-key-guid
+.It Fl j , -json Op Ar --json-int, --json-pool-key-guid
 Display the list of properties in JSON format.
 Specify
 .Sy --json-int
@@ -157,7 +157,7 @@ See the
 .Xr vdevprops 7
 manual page for more information on the available pool properties.
 .Bl -tag -compact -offset Ds -width "-o field"
-.It Fl j Op Ar --json-int
+.It Fl j , -json Op Ar --json-int
 Display the list of properties in JSON format.
 Specify
 .Sy --json-int

man/man8/zpool-list.8

@@ -59,7 +59,7 @@ is specified, the command exits after
 .Ar count
 reports are printed.
 .Bl -tag -width Ds
-.It Fl j Op Ar --json-int, --json-pool-key-guid
+.It Fl j , -json Op Ar --json-int, --json-pool-key-guid
 Display the list of pools in JSON format.
 Specify
 .Sy --json-int

man/man8/zpool-status.8

@@ -70,7 +70,7 @@ See the
 option of
 .Nm zpool Cm iostat
 for complete details.
-.It Fl j Op Ar --json-int, --json-flat-vdevs, --json-pool-key-guid
+.It Fl j , -json Op Ar --json-int, --json-flat-vdevs, --json-pool-key-guid
 Display the status for ZFS pools in JSON format.
 Specify
 .Sy --json-int
@@ -82,14 +82,18 @@ Specify
 .Sy --json-pool-key-guid
 to set pool GUID as key for pool objects instead of pool names.
 .It Fl d
-Display the number of Direct I/O write checksum verify errors that have occured
+Display the number of Direct I/O read/write checksum verify errors that have
-on a top-level VDEV.
+occured on a top-level VDEV.
 See
 .Sx zfs_vdev_direct_write_verify
 in
 .Xr zfs 4
 for details about the conditions that can cause Direct I/O write checksum
 verify failures to occur.
+Direct I/O reads checksum verify errors can also occur if the contents of the
+buffer are being manipulated after the I/O has been issued and is in flight.
+In the case of Direct I/O read checksum verify errors, the I/O will be reissued
+through the ARC.
 .It Fl D
 Display a histogram of deduplication statistics, showing the allocated
 .Pq physically present on disk

module/os/freebsd/zfs/abd_os.c

@@ -620,9 +620,16 @@ abd_borrow_buf_copy(abd_t *abd, size_t n)
 
 /*
  * Return a borrowed raw buffer to an ABD. If the ABD is scattered, this will
- * no change the contents of the ABD and will ASSERT that you didn't modify
+ * not change the contents of the ABD. If you want any changes you made to
- * the buffer since it was borrowed. If you want any changes you made to buf to
+ * buf to be copied back to abd, use abd_return_buf_copy() instead. If the
- * be copied back to abd, use abd_return_buf_copy() instead.
+ * ABD is not constructed from user pages from Direct I/O then an ASSERT
+ * checks to make sure the contents of the buffer have not changed since it was
+ * borrowed. We can not ASSERT the contents of the buffer have not changed if
+ * it is composed of user pages. While Direct I/O write pages are placed under
+ * write protection and can not be changed, this is not the case for Direct I/O
+ * reads. The pages of a Direct I/O read could be manipulated at any time.
+ * Checksum verifications in the ZIO pipeline check for this issue and handle
+ * it by returning an error on checksum verification failure.
  */
 void
 abd_return_buf(abd_t *abd, void *buf, size_t n)
@@ -632,8 +639,34 @@ abd_return_buf(abd_t *abd, void *buf, size_t n)
 #ifdef ZFS_DEBUG
 	(void) zfs_refcount_remove_many(&abd->abd_children, n, buf);
 #endif
-	if (abd_is_linear(abd)) {
+	if (abd_is_from_pages(abd)) {
+		if (!abd_is_linear_page(abd))
+			zio_buf_free(buf, n);
+	} else if (abd_is_linear(abd)) {
 		ASSERT3P(buf, ==, abd_to_buf(abd));
+	} else if (abd_is_gang(abd)) {
+#ifdef ZFS_DEBUG
+		/*
+		 * We have to be careful with gang ABD's that we do not ASSERT
+		 * for any ABD's that contain user pages from Direct I/O. See
+		 * the comment above about Direct I/O read buffers possibly
+		 * being manipulated. In order to handle this, we jsut iterate
+		 * through the gang ABD and only verify ABD's that are not from
+		 * user pages.
+		 */
+		void *cmp_buf = buf;
+
+		for (abd_t *cabd = list_head(&ABD_GANG(abd).abd_gang_chain);
+		    cabd != NULL;
+		    cabd = list_next(&ABD_GANG(abd).abd_gang_chain, cabd)) {
+			if (!abd_is_from_pages(cabd)) {
+				ASSERT0(abd_cmp_buf(cabd, cmp_buf,
+				    cabd->abd_size));
+			}
+			cmp_buf = (char *)cmp_buf + cabd->abd_size;
+		}
+#endif
+		zio_buf_free(buf, n);
 	} else {
 		ASSERT0(abd_cmp_buf(abd, buf, n));
 		zio_buf_free(buf, n);

module/os/linux/zfs/abd_os.c

@@ -1008,7 +1008,9 @@ abd_borrow_buf_copy(abd_t *abd, size_t n)
  * borrowed. We can not ASSERT that the contents of the buffer have not changed
  * if it is composed of user pages because the pages can not be placed under
  * write protection and the user could have possibly changed the contents in
- * the pages at any time.
+ * the pages at any time. This is also an issue for Direct I/O reads. Checksum
+ * verifications in the ZIO pipeline check for this issue and handle it by
+ * returning an error on checksum verification failure.
  */
 void
 abd_return_buf(abd_t *abd, void *buf, size_t n)

module/zcommon/zfs_valstr.c

@@ -206,6 +206,7 @@ _VALSTR_BITFIELD_IMPL(zio_flag,
 	{ '.', "PR", "PROBE" },
 	{ '.', "TH", "TRYHARD" },
 	{ '.', "OP", "OPTIONAL" },
+	{ '.', "RD", "DIO_READ" },
 	{ '.', "DQ", "DONT_QUEUE" },
 	{ '.', "DP", "DONT_PROPAGATE" },
 	{ '.', "BY", "IO_BYPASS" },

module/zfs/dmu_direct.c

@@ -330,7 +330,7 @@ dmu_read_abd(dnode_t *dn, uint64_t offset, uint64_t size,
 		 */
 		zio_t *cio = zio_read(rio, spa, bp, mbuf, db->db.db_size,
 		    dmu_read_abd_done, NULL, ZIO_PRIORITY_SYNC_READ,
-		    ZIO_FLAG_CANFAIL, &zb);
+		    ZIO_FLAG_CANFAIL | ZIO_FLAG_DIO_READ, &zb);
 		mutex_exit(&db->db_mtx);
 
 		zfs_racct_read(spa, db->db.db_size, 1, flags);

module/zfs/dsl_dataset.c

@@ -2987,6 +2987,7 @@ dsl_dataset_rename_snapshot_sync_impl(dsl_pool_t *dp,
 	dsl_dataset_t *ds;
 	uint64_t val;
 	dmu_tx_t *tx = ddrsa->ddrsa_tx;
+	char *oldname, *newname;
 	int error;
 
 	error = dsl_dataset_snap_lookup(hds, ddrsa->ddrsa_oldsnapname, &val);
@@ -3011,8 +3012,14 @@ dsl_dataset_rename_snapshot_sync_impl(dsl_pool_t *dp,
 	VERIFY0(zap_add(dp->dp_meta_objset,
 	    dsl_dataset_phys(hds)->ds_snapnames_zapobj,
 	    ds->ds_snapname, 8, 1, &ds->ds_object, tx));
-	zvol_rename_minors(dp->dp_spa, ddrsa->ddrsa_oldsnapname,
+
-	    ddrsa->ddrsa_newsnapname, B_TRUE);
+	oldname = kmem_asprintf("%s@%s", ddrsa->ddrsa_fsname,
+	    ddrsa->ddrsa_oldsnapname);
+	newname = kmem_asprintf("%s@%s", ddrsa->ddrsa_fsname,
+	    ddrsa->ddrsa_newsnapname);
+	zvol_rename_minors(dp->dp_spa, oldname, newname, B_TRUE);
+	kmem_strfree(oldname);
+	kmem_strfree(newname);
 
 	dsl_dataset_rele(ds, FTAG);
 	return (0);

module/zfs/vdev_draid.c

@@ -1026,7 +1026,7 @@ vdev_draid_map_alloc_row(zio_t *zio, raidz_row_t **rrp, uint64_t io_offset,
 
 	ASSERT3U(vdc->vdc_nparity, >, 0);
 
-	raidz_row_t *rr = vdev_raidz_row_alloc(groupwidth);
+	raidz_row_t *rr = vdev_raidz_row_alloc(groupwidth, zio);
 	rr->rr_bigcols = bc;
 	rr->rr_firstdatacol = vdc->vdc_nparity;
 #ifdef ZFS_DEBUG

module/zfs/vdev_indirect.c

@@ -34,6 +34,7 @@
 #include <sys/zap.h>
 #include <sys/abd.h>
 #include <sys/zthr.h>
+#include <sys/fm/fs/zfs.h>
 
 /*
  * An indirect vdev corresponds to a vdev that has been removed.  Since
@@ -1832,6 +1833,19 @@ vdev_indirect_io_done(zio_t *zio)
 
 	zio_bad_cksum_t zbc;
 	int ret = zio_checksum_error(zio, &zbc);
+	/*
+	 * Any Direct I/O read that has a checksum error must be treated as
+	 * suspicious as the contents of the buffer could be getting
+	 * manipulated while the I/O is taking place. The checksum verify error
+	 * will be reported to the top-level VDEV.
+	 */
+	if (zio->io_flags & ZIO_FLAG_DIO_READ && ret == ECKSUM) {
+		zio->io_error = ret;
+		zio->io_flags |= ZIO_FLAG_DIO_CHKSUM_ERR;
+		zio_dio_chksum_verify_error_report(zio);
+		ret = 0;
+	}
+
 	if (ret == 0) {
 		zio_checksum_verified(zio);
 		return;

module/zfs/vdev_mirror.c

@@ -764,6 +764,27 @@ vdev_mirror_io_done(zio_t *zio)
 
 	ASSERT(zio->io_type == ZIO_TYPE_READ);
 
+	/*
+	 * Any Direct I/O read that has a checksum error must be treated as
+	 * suspicious as the contents of the buffer could be getting
+	 * manipulated while the I/O is taking place. The checksum verify error
+	 * will be reported to the top-level Mirror VDEV.
+	 *
+	 * There will be no attampt at reading any additional data copies. If
+	 * the buffer is still being manipulated while attempting to read from
+	 * another child, there exists a possibly that the checksum could be
+	 * verified as valid. However, the buffer contents could again get
+	 * manipulated after verifying the checksum. This would lead to bad data
+	 * being written out during self healing.
+	 */
+	if ((zio->io_flags & ZIO_FLAG_DIO_READ) &&
+	    (zio->io_flags & ZIO_FLAG_DIO_CHKSUM_ERR)) {
+		zio_dio_chksum_verify_error_report(zio);
+		zio->io_error = vdev_mirror_worst_error(mm);
+		ASSERT3U(zio->io_error, ==, ECKSUM);
+		return;
+	}
+
 	/*
 	 * If we don't have a good copy yet, keep trying other children.
 	 */

module/zfs/vdev_raidz.c

@@ -433,7 +433,7 @@ const zio_vsd_ops_t vdev_raidz_vsd_ops = {
 };
 
 raidz_row_t *
-vdev_raidz_row_alloc(int cols)
+vdev_raidz_row_alloc(int cols, zio_t *zio)
 {
 	raidz_row_t *rr =
 	    kmem_zalloc(offsetof(raidz_row_t, rr_col[cols]), KM_SLEEP);
@@ -445,7 +445,17 @@ vdev_raidz_row_alloc(int cols)
 		raidz_col_t *rc = &rr->rr_col[c];
 		rc->rc_shadow_devidx = INT_MAX;
 		rc->rc_shadow_offset = UINT64_MAX;
-		rc->rc_allow_repair = 1;
+		/*
+		 * We can not allow self healing to take place for Direct I/O
+		 * reads. There is nothing that stops the buffer contents from
+		 * being manipulated while the I/O is in flight. It is possible
+		 * that the checksum could be verified on the buffer and then
+		 * the contents of that buffer are manipulated afterwards. This
+		 * could lead to bad data being written out during self
+		 * healing.
+		 */
+		if (!(zio->io_flags & ZIO_FLAG_DIO_READ))
+			rc->rc_allow_repair = 1;
 	}
 	return (rr);
 }
@@ -619,7 +629,7 @@ vdev_raidz_map_alloc(zio_t *zio, uint64_t ashift, uint64_t dcols,
 	}
 
 	ASSERT3U(acols, <=, scols);
-	rr = vdev_raidz_row_alloc(scols);
+	rr = vdev_raidz_row_alloc(scols, zio);
 	rm->rm_row[0] = rr;
 	rr->rr_cols = acols;
 	rr->rr_bigcols = bc;
@@ -765,7 +775,7 @@ vdev_raidz_map_alloc_expanded(zio_t *zio,
 
 	for (uint64_t row = 0; row < rows; row++) {
 		boolean_t row_use_scratch = B_FALSE;
-		raidz_row_t *rr = vdev_raidz_row_alloc(cols);
+		raidz_row_t *rr = vdev_raidz_row_alloc(cols, zio);
 		rm->rm_row[row] = rr;
 
 		/* The starting RAIDZ (parent) vdev sector of the row. */
@@ -2633,6 +2643,20 @@ raidz_checksum_verify(zio_t *zio)
 	raidz_map_t *rm = zio->io_vsd;
 
 	int ret = zio_checksum_error(zio, &zbc);
+	/*
+	 * Any Direct I/O read that has a checksum error must be treated as
+	 * suspicious as the contents of the buffer could be getting
+	 * manipulated while the I/O is taking place. The checksum verify error
+	 * will be reported to the top-level RAIDZ VDEV.
+	 */
+	if (zio->io_flags & ZIO_FLAG_DIO_READ && ret == ECKSUM) {
+		zio->io_error = ret;
+		zio->io_flags |= ZIO_FLAG_DIO_CHKSUM_ERR;
+		zio_dio_chksum_verify_error_report(zio);
+		zio_checksum_verified(zio);
+		return (0);
+	}
+
 	if (ret != 0 && zbc.zbc_injected != 0)
 		rm->rm_ecksuminjected = 1;
 
@@ -2776,6 +2800,11 @@ vdev_raidz_io_done_verified(zio_t *zio, raidz_row_t *rr)
 			    (rc->rc_error == 0 || rc->rc_size == 0)) {
 				continue;
 			}
+			/*
+			 * We do not allow self healing for Direct I/O reads.
+			 * See comment in vdev_raid_row_alloc().
+			 */
+			ASSERT0(zio->io_flags & ZIO_FLAG_DIO_READ);
 
 			zfs_dbgmsg("zio=%px repairing c=%u devidx=%u "
 			    "offset=%llx",
@@ -2979,6 +3008,8 @@ raidz_reconstruct(zio_t *zio, int *ltgts, int ntgts, int nparity)
 
 	/* Check for success */
 	if (raidz_checksum_verify(zio) == 0) {
+		if (zio->io_flags & ZIO_FLAG_DIO_CHKSUM_ERR)
+			return (0);
 
 		/* Reconstruction succeeded - report errors */
 		for (int i = 0; i < rm->rm_nrows; i++) {
@@ -3379,7 +3410,6 @@ vdev_raidz_io_done_unrecoverable(zio_t *zio)
 			zio_bad_cksum_t zbc;
 			zbc.zbc_has_cksum = 0;
 			zbc.zbc_injected = rm->rm_ecksuminjected;
-
 			mutex_enter(&cvd->vdev_stat_lock);
 			cvd->vdev_stat.vs_checksum_errors++;
 			mutex_exit(&cvd->vdev_stat_lock);
@@ -3444,6 +3474,9 @@ vdev_raidz_io_done(zio_t *zio)
 		}
 
 		if (raidz_checksum_verify(zio) == 0) {
+			if (zio->io_flags & ZIO_FLAG_DIO_CHKSUM_ERR)
+				goto done;
+
 			for (int i = 0; i < rm->rm_nrows; i++) {
 				raidz_row_t *rr = rm->rm_row[i];
 				vdev_raidz_io_done_verified(zio, rr);
@@ -3538,6 +3571,7 @@ vdev_raidz_io_done(zio_t *zio)
 			}
 		}
 	}
+done:
 	if (rm->rm_lr != NULL) {
 		zfs_rangelock_exit(rm->rm_lr);
 		rm->rm_lr = NULL;

module/zfs/zfs_vnops.c

@@ -303,6 +303,7 @@ zfs_read(struct znode *zp, zfs_uio_t *uio, int ioflag, cred_t *cr)
 	(void) cr;
 	int error = 0;
 	boolean_t frsync = B_FALSE;
+	boolean_t dio_checksum_failure = B_FALSE;
 
 	zfsvfs_t *zfsvfs = ZTOZSB(zp);
 	if ((error = zfs_enter_verify_zp(zfsvfs, zp, FTAG)) != 0)
@@ -424,8 +425,26 @@ zfs_read(struct znode *zp, zfs_uio_t *uio, int ioflag, cred_t *cr)
 
 		if (error) {
 			/* convert checksum errors into IO errors */
-			if (error == ECKSUM)
+			if (error == ECKSUM) {
-				error = SET_ERROR(EIO);
+				/*
+				 * If a Direct I/O read returned a checksum
+				 * verify error, then it must be treated as
+				 * suspicious. The contents of the buffer could
+				 * have beeen manipulated while the I/O was in
+				 * flight. In this case, the remainder of I/O
+				 * request will just be reissued through the
+				 * ARC.
+				 */
+				if (uio->uio_extflg & UIO_DIRECT) {
+					dio_checksum_failure = B_TRUE;
+					uio->uio_extflg &= ~UIO_DIRECT;
+					n += dio_remaining_resid;
+					dio_remaining_resid = 0;
+					continue;
+				} else {
+					error = SET_ERROR(EIO);
+				}
+			}
 
 #if defined(__linux__)
 			/*
@@ -472,6 +491,9 @@ zfs_read(struct znode *zp, zfs_uio_t *uio, int ioflag, cred_t *cr)
 out:
 	zfs_rangelock_exit(lr);
 
+	if (dio_checksum_failure == B_TRUE)
+		uio->uio_extflg |= UIO_DIRECT;
+
 	/*
 	 * Cleanup for Direct I/O if requested.
 	 */

module/zfs/zio.c

@@ -804,11 +804,11 @@ zio_notify_parent(zio_t *pio, zio_t *zio, enum zio_wait_type wait,
 	pio->io_reexecute |= zio->io_reexecute;
 	ASSERT3U(*countp, >, 0);
 
-	if (zio->io_flags & ZIO_FLAG_DIO_CHKSUM_ERR) {
+	/*
-		ASSERT3U(*errorp, ==, EIO);
+	 * Propogate the Direct I/O checksum verify failure to the parent.
-		ASSERT3U(pio->io_child_type, ==, ZIO_CHILD_LOGICAL);
+	 */
+	if (zio->io_flags & ZIO_FLAG_DIO_CHKSUM_ERR)
 		pio->io_flags |= ZIO_FLAG_DIO_CHKSUM_ERR;
-	}
 
 	(*countp)--;
 
@@ -1573,6 +1573,14 @@ zio_vdev_child_io(zio_t *pio, blkptr_t *bp, vdev_t *vd, uint64_t offset,
 		 */
 		pipeline |= ZIO_STAGE_CHECKSUM_VERIFY;
 		pio->io_pipeline &= ~ZIO_STAGE_CHECKSUM_VERIFY;
+		/*
+		 * We never allow the mirror VDEV to attempt reading from any
+		 * additional data copies after the first Direct I/O checksum
+		 * verify failure. This is to avoid bad data being written out
+		 * through the mirror during self healing. See comment in
+		 * vdev_mirror_io_done() for more details.
+		 */
+		ASSERT0(pio->io_flags & ZIO_FLAG_DIO_CHKSUM_ERR);
 	} else if (type == ZIO_TYPE_WRITE &&
 	    pio->io_prop.zp_direct_write == B_TRUE) {
 		/*
@@ -4555,18 +4563,18 @@ zio_vdev_io_assess(zio_t *zio)
 	}
 
 	/*
-	 * If a Direct I/O write checksum verify error has occurred then this
+	 * If a Direct I/O operation has a checksum verify error then this I/O
-	 * I/O should not attempt to be issued again. Instead the EIO will
+	 * should not attempt to be issued again.
-	 * be returned.
 	 */
 	if (zio->io_flags & ZIO_FLAG_DIO_CHKSUM_ERR) {
-		ASSERT3U(zio->io_child_type, ==, ZIO_CHILD_LOGICAL);
+		if (zio->io_type == ZIO_TYPE_WRITE) {
-		ASSERT3U(zio->io_error, ==, EIO);
+			ASSERT3U(zio->io_child_type, ==, ZIO_CHILD_LOGICAL);
+			ASSERT3U(zio->io_error, ==, EIO);
+		}
 		zio->io_pipeline = ZIO_INTERLOCK_PIPELINE;
 		return (zio);
 	}
 
-
 	if (zio_injection_enabled && zio->io_error == 0)
 		zio->io_error = zio_handle_fault_injection(zio, EIO);
 
@@ -4864,16 +4872,40 @@ zio_checksum_verify(zio_t *zio)
 		ASSERT3U(zio->io_prop.zp_checksum, ==, ZIO_CHECKSUM_LABEL);
 	}
 
+	ASSERT0(zio->io_flags & ZIO_FLAG_DIO_CHKSUM_ERR);
+	IMPLY(zio->io_flags & ZIO_FLAG_DIO_READ,
+	    !(zio->io_flags & ZIO_FLAG_SPECULATIVE));
+
 	if ((error = zio_checksum_error(zio, &info)) != 0) {
 		zio->io_error = error;
 		if (error == ECKSUM &&
 		    !(zio->io_flags & ZIO_FLAG_SPECULATIVE)) {
-			mutex_enter(&zio->io_vd->vdev_stat_lock);
+			if (zio->io_flags & ZIO_FLAG_DIO_READ) {
-			zio->io_vd->vdev_stat.vs_checksum_errors++;
+				zio->io_flags |= ZIO_FLAG_DIO_CHKSUM_ERR;
-			mutex_exit(&zio->io_vd->vdev_stat_lock);
+				zio_t *pio = zio_unique_parent(zio);
-			(void) zfs_ereport_start_checksum(zio->io_spa,
+				/*
-			    zio->io_vd, &zio->io_bookmark, zio,
+				 * Any Direct I/O read that has a checksum
-			    zio->io_offset, zio->io_size, &info);
+				 * error must be treated as suspicous as the
+				 * contents of the buffer could be getting
+				 * manipulated while the I/O is taking place.
+				 *
+				 * The checksum verify error will only be
+				 * reported here for disk and file VDEV's and
+				 * will be reported on those that the failure
+				 * occurred on. Other types of VDEV's report the
+				 * verify failure in their own code paths.
+				 */
+				if (pio->io_child_type == ZIO_CHILD_LOGICAL) {
+					zio_dio_chksum_verify_error_report(zio);
+				}
+			} else {
+				mutex_enter(&zio->io_vd->vdev_stat_lock);
+				zio->io_vd->vdev_stat.vs_checksum_errors++;
+				mutex_exit(&zio->io_vd->vdev_stat_lock);
+				(void) zfs_ereport_start_checksum(zio->io_spa,
+				    zio->io_vd, &zio->io_bookmark, zio,
+				    zio->io_offset, zio->io_size, &info);
+			}
 		}
 	}
 
@@ -4899,22 +4931,8 @@ zio_dio_checksum_verify(zio_t *zio)
 	if ((error = zio_checksum_error(zio, NULL)) != 0) {
 		zio->io_error = error;
 		if (error == ECKSUM) {
-			mutex_enter(&zio->io_vd->vdev_stat_lock);
-			zio->io_vd->vdev_stat.vs_dio_verify_errors++;
-			mutex_exit(&zio->io_vd->vdev_stat_lock);
-			zio->io_error = SET_ERROR(EIO);
 			zio->io_flags |= ZIO_FLAG_DIO_CHKSUM_ERR;
-
+			zio_dio_chksum_verify_error_report(zio);
-			/*
-			 * The EIO error must be propagated up to the logical
-			 * parent ZIO in zio_notify_parent() so it can be
-			 * returned to dmu_write_abd().
-			 */
-			zio->io_flags &= ~ZIO_FLAG_DONT_PROPAGATE;
-
-			(void) zfs_ereport_post(FM_EREPORT_ZFS_DIO_VERIFY,
-			    zio->io_spa, zio->io_vd, &zio->io_bookmark,
-			    zio, 0);
 		}
 	}
 
@@ -4932,6 +4950,39 @@ zio_checksum_verified(zio_t *zio)
 	zio->io_pipeline &= ~ZIO_STAGE_CHECKSUM_VERIFY;
 }
 
+/*
+ * Report Direct I/O checksum verify error and create ZED event.
+ */
+void
+zio_dio_chksum_verify_error_report(zio_t *zio)
+{
+	ASSERT(zio->io_flags & ZIO_FLAG_DIO_CHKSUM_ERR);
+
+	if (zio->io_child_type == ZIO_CHILD_LOGICAL)
+		return;
+
+	mutex_enter(&zio->io_vd->vdev_stat_lock);
+	zio->io_vd->vdev_stat.vs_dio_verify_errors++;
+	mutex_exit(&zio->io_vd->vdev_stat_lock);
+	if (zio->io_type == ZIO_TYPE_WRITE) {
+		/*
+		 * Convert checksum error for writes into EIO.
+		 */
+		zio->io_error = SET_ERROR(EIO);
+		/*
+		 * Report dio_verify_wr ZED event.
+		 */
+		(void) zfs_ereport_post(FM_EREPORT_ZFS_DIO_VERIFY_WR,
+		    zio->io_spa,  zio->io_vd, &zio->io_bookmark, zio, 0);
+	} else {
+		/*
+		 * Report dio_verify_rd ZED event.
+		 */
+		(void) zfs_ereport_post(FM_EREPORT_ZFS_DIO_VERIFY_RD,
+		    zio->io_spa, zio->io_vd, &zio->io_bookmark, zio, 0);
+	}
+}
+
 /*
  * ==========================================================================
  * Error rank.  Error are ranked in the order 0, ENXIO, ECKSUM, EIO, other.
@@ -5343,10 +5394,9 @@ zio_done(zio_t *zio)
 
 	if (zio->io_reexecute) {
 		/*
-		 * A Direct I/O write that has a checksum verify error should
+		 * A Direct I/O operation that has a checksum verify error
-		 * not attempt to reexecute. Instead, EAGAIN should just be
+		 * should not attempt to reexecute. Instead, the error should
-		 * propagated back up so the write can be attempt to be issued
+		 * just be propagated back.
-		 * through the ARC.
 		 */
 		ASSERT(!(zio->io_flags & ZIO_FLAG_DIO_CHKSUM_ERR));
 

rpm/generic/zfs.spec.in

@@ -99,10 +99,10 @@ License:        @ZFS_META_LICENSE@
 URL:            https://github.com/openzfs/zfs
 Source0:        %{name}-%{version}.tar.gz
 BuildRoot:      %{_tmppath}/%{name}-%{version}-%{release}-root-%(%{__id_u} -n)
-Requires:       libzpool5%{?_isa} = %{version}-%{release}
+Requires:       libzpool6%{?_isa} = %{version}-%{release}
 Requires:       libnvpair3%{?_isa} = %{version}-%{release}
 Requires:       libuutil3%{?_isa} = %{version}-%{release}
-Requires:       libzfs5%{?_isa} = %{version}-%{release}
+Requires:       libzfs6%{?_isa} = %{version}-%{release}
 Requires:       %{name}-kmod = %{version}
 Provides:       %{name}-kmod-common = %{version}-%{release}
 Obsoletes:      spl <= %{version}
@@ -150,21 +150,22 @@ Requires:  sysstat
 %description
 This package contains the core ZFS command line utilities.
 
-%package -n libzpool5
+%package -n libzpool6
 Summary:        Native ZFS pool library for Linux
 Group:          System Environment/Kernel
 Obsoletes:      libzpool2 <= %{version}
 Obsoletes:      libzpool4 <= %{version}
+Obsoletes:      libzpool5 <= %{version}
 
-%description -n libzpool5
+%description -n libzpool6
 This package contains the zpool library, which provides support
 for managing zpools
 
 %if %{defined ldconfig_scriptlets}
-%ldconfig_scriptlets -n libzpool5
+%ldconfig_scriptlets -n libzpool6
 %else
-%post -n libzpool5 -p /sbin/ldconfig
+%post -n libzpool6 -p /sbin/ldconfig
-%postun -n libzpool5 -p /sbin/ldconfig
+%postun -n libzpool6 -p /sbin/ldconfig
 %endif
 
 %package -n libnvpair3
@@ -211,37 +212,39 @@ This library provides a variety of compatibility functions for OpenZFS:
 # The library version is encoded in the package name.  When updating the
 # version information it is important to add an obsoletes line below for
 # the previous version of the package.
-%package -n libzfs5
+%package -n libzfs6
 Summary:        Native ZFS filesystem library for Linux
 Group:          System Environment/Kernel
 Obsoletes:      libzfs2 <= %{version}
 Obsoletes:      libzfs4 <= %{version}
+Obsoletes:      libzfs5 <= %{version}
 
-%description -n libzfs5
+%description -n libzfs6
 This package provides support for managing ZFS filesystems
 
 %if %{defined ldconfig_scriptlets}
-%ldconfig_scriptlets -n libzfs5
+%ldconfig_scriptlets -n libzfs6
 %else
-%post -n libzfs5 -p /sbin/ldconfig
+%post -n libzfs6 -p /sbin/ldconfig
-%postun -n libzfs5 -p /sbin/ldconfig
+%postun -n libzfs6 -p /sbin/ldconfig
 %endif
 
-%package -n libzfs5-devel
+%package -n libzfs6-devel
 Summary:        Development headers
 Group:          System Environment/Kernel
-Requires:       libzfs5%{?_isa} = %{version}-%{release}
+Requires:       libzfs6%{?_isa} = %{version}-%{release}
-Requires:       libzpool5%{?_isa} = %{version}-%{release}
+Requires:       libzpool6%{?_isa} = %{version}-%{release}
 Requires:       libnvpair3%{?_isa} = %{version}-%{release}
 Requires:       libuutil3%{?_isa} = %{version}-%{release}
-Provides:       libzpool5-devel = %{version}-%{release}
+Provides:       libzpool6-devel = %{version}-%{release}
 Provides:       libnvpair3-devel = %{version}-%{release}
 Provides:       libuutil3-devel = %{version}-%{release}
 Obsoletes:      zfs-devel <= %{version}
 Obsoletes:      libzfs2-devel <= %{version}
 Obsoletes:      libzfs4-devel <= %{version}
+Obsoletes:      libzfs5-devel <= %{version}
 
-%description -n libzfs5-devel
+%description -n libzfs6-devel
 This package contains the header files needed for building additional
 applications against the ZFS libraries.
 
@@ -290,7 +293,7 @@ Summary:        Python %{python_version} wrapper for libzfs_core
 Group:          Development/Languages/Python
 License:        Apache-2.0
 BuildArch:      noarch
-Requires:       libzfs5 = %{version}-%{release}
+Requires:       libzfs6 = %{version}-%{release}
 Requires:       libnvpair3 = %{version}-%{release}
 Requires:       libffi
 Requires:       python%{__python_pkg_version}
@@ -534,7 +537,7 @@ systemctl --system daemon-reload >/dev/null || true
 %config(noreplace) %{_bashcompletiondir}/zfs
 %config(noreplace) %{_bashcompletiondir}/zpool
 
-%files -n libzpool5
+%files -n libzpool6
 %{_libdir}/libzpool.so.*
 
 %files -n libnvpair3
@@ -543,10 +546,10 @@ systemctl --system daemon-reload >/dev/null || true
 %files -n libuutil3
 %{_libdir}/libuutil.so.*
 
-%files -n libzfs5
+%files -n libzfs6
 %{_libdir}/libzfs*.so.*
 
-%files -n libzfs5-devel
+%files -n libzfs6-devel
 %{_pkgconfigdir}/libzfs.pc
 %{_pkgconfigdir}/libzfsbootenv.pc
 %{_pkgconfigdir}/libzfs_core.pc

tests/runfiles/common.run

@@ -697,8 +697,8 @@ tags = ['functional', 'delegate']
 tests = ['dio_aligned_block', 'dio_async_always', 'dio_async_fio_ioengines',
     'dio_compression', 'dio_dedup', 'dio_encryption', 'dio_grow_block',
     'dio_max_recordsize', 'dio_mixed', 'dio_mmap', 'dio_overwrites',
-    'dio_property', 'dio_random', 'dio_recordsize', 'dio_unaligned_block',
+    'dio_property', 'dio_random', 'dio_read_verify', 'dio_recordsize',
-    'dio_unaligned_filesize']
+    'dio_unaligned_block', 'dio_unaligned_filesize']
 tags = ['functional', 'direct']
 
 [tests/functional/exec]

tests/zfs-tests/cmd/manipulate_user_buffer.c

@@ -20,7 +20,7 @@
  */
 
 /*
- * Copyright (c) 2022 by Triad National Security, LLC.
+ * Copyright (c) 2024 by Triad National Security, LLC.
  */
 
 #include <sys/types.h>
@@ -39,51 +39,59 @@
 #define	MIN(a, b)	((a) < (b)) ? (a) : (b)
 #endif
 
-static char *outputfile = NULL;
+static char *filename = NULL;
 static int blocksize = 131072; /* 128K */
-static int wr_err_expected = 0;
+static int err_expected = 0;
+static int read_op = 0;
+static int write_op = 0;
 static int numblocks = 100;
 static char *execname = NULL;
 static int print_usage = 0;
 static int randompattern = 0;
-static int ofd;
+static int fd;
 char *buf = NULL;
 
 typedef struct {
-	int entire_file_written;
+	int entire_file_completed;
 } pthread_args_t;
 
 static void
 usage(void)
 {
 	(void) fprintf(stderr,
-	    "usage %s -o outputfile [-b blocksize] [-e wr_error_expected]\n"
+	    "usage %s -f filename [-b blocksize] [-e wr_error_expected]\n"
-	    "         [-n numblocks] [-p randpattern] [-h help]\n"
+	    "         [-n numblocks] [-p randompattern] -r read_op \n"
+	    "         -w write_op [-h help]\n"
 	    "\n"
 	    "Testing whether checksum verify works correctly for O_DIRECT.\n"
 	    "when manipulating the contents of a userspace buffer.\n"
 	    "\n"
-	    "    outputfile:      File to write to.\n"
+	    "    filename:       File to read or write to.\n"
-	    "    blocksize:       Size of each block to write (must be at \n"
+	    "    blocksize:      Size of each block to write (must be at \n"
-	    "                     least >= 512).\n"
+	    "                    least >= 512).\n"
-	    "    wr_err_expected: Whether pwrite() is expected to return EIO\n"
+	    "    err_expected:   Whether write() is expected to return EIO\n"
-	    "                     while manipulating the contents of the\n"
+	    "                    while manipulating the contents of the\n"
-	    "                     buffer.\n"
+	    "                    buffer.\n"
-	    "    numblocks:       Total number of blocksized blocks to\n"
+	    "    numblocks:      Total number of blocksized blocks to\n"
-	    "                     write.\n"
+	    "                    write.\n"
-	    "    randpattern:     Fill data buffer with random data. Default\n"
+	    "    read_op:        Perform reads to the filename file while\n"
-	    "                     behavior is to fill the buffer with the \n"
+	    "                    while manipulating the buffer contents\n"
-	    "                     known data pattern (0xdeadbeef).\n"
+	    "    write_op:       Perform writes to the filename file while\n"
+	    "                    manipulating the buffer contents\n"
+	    "    randompattern:  Fill data buffer with random data for \n"
+	    "                    writes. Default behavior is to fill the \n"
+	    "                    buffer with known data pattern (0xdeadbeef)\n"
 	    "    help:           Print usage information and exit.\n"
 	    "\n"
 	    "    Required parameters:\n"
-	    "    outputfile\n"
+	    "    filename\n"
+	    "    read_op or write_op\n"
 	    "\n"
 	    "    Default Values:\n"
 	    "    blocksize       -> 131072\n"
 	    "    wr_err_expexted -> false\n"
 	    "    numblocks       -> 100\n"
-	    "    randpattern     -> false\n",
+	    "    randompattern   -> false\n",
 	    execname);
 	(void) exit(1);
 }
@@ -97,16 +105,21 @@ parse_options(int argc, char *argv[])
 	extern int optind, optopt;
 	execname = argv[0];
 
-	while ((c = getopt(argc, argv, "b:ehn:o:p")) != -1) {
+	while ((c = getopt(argc, argv, "b:ef:hn:rw")) != -1) {
 		switch (c) {
 			case 'b':
 				blocksize = atoi(optarg);
 				break;
 
 			case 'e':
-				wr_err_expected = 1;
+				err_expected = 1;
 				break;
 
+			case 'f':
+				filename = optarg;
+				break;
+
+
 			case 'h':
 				print_usage = 1;
 				break;
@@ -115,12 +128,12 @@ parse_options(int argc, char *argv[])
 				numblocks = atoi(optarg);
 				break;
 
-			case 'o':
+			case 'r':
-				outputfile = optarg;
+				read_op = 1;
 				break;
 
-			case 'p':
+			case 'w':
-				randompattern = 1;
+				write_op = 1;
 				break;
 
 			case ':':
@@ -141,7 +154,8 @@ parse_options(int argc, char *argv[])
 	if (errflag || print_usage == 1)
 		(void) usage();
 
-	if (blocksize < 512 || outputfile == NULL || numblocks <= 0) {
+	if (blocksize < 512 || filename == NULL || numblocks <= 0 ||
+	    (read_op == 0 && write_op == 0)) {
 		(void) fprintf(stderr,
 		    "Required paramater(s) missing or invalid.\n");
 		(void) usage();
@@ -160,10 +174,10 @@ write_thread(void *arg)
 	ssize_t wrote = 0;
 	pthread_args_t *args = (pthread_args_t *)arg;
 
-	while (!args->entire_file_written) {
+	while (!args->entire_file_completed) {
-		wrote = pwrite(ofd, buf, blocksize, offset);
+		wrote = pwrite(fd, buf, blocksize, offset);
 		if (wrote != blocksize) {
-			if (wr_err_expected)
+			if (err_expected)
 				assert(errno == EIO);
 			else
 				exit(2);
@@ -173,7 +187,35 @@ write_thread(void *arg)
 		left -= blocksize;
 
 		if (left == 0)
-			args->entire_file_written = 1;
+			args->entire_file_completed = 1;
+	}
+
+	pthread_exit(NULL);
+}
+
+/*
+ * Read blocksize * numblocks to the file using O_DIRECT.
+ */
+static void *
+read_thread(void *arg)
+{
+	size_t offset = 0;
+	int total_data = blocksize * numblocks;
+	int left = total_data;
+	ssize_t read = 0;
+	pthread_args_t *args = (pthread_args_t *)arg;
+
+	while (!args->entire_file_completed) {
+		read = pread(fd, buf, blocksize, offset);
+		if (read != blocksize) {
+			exit(2);
+		}
+
+		offset = ((offset + blocksize) % total_data);
+		left -= blocksize;
+
+		if (left == 0)
+			args->entire_file_completed = 1;
 	}
 
 	pthread_exit(NULL);
@@ -189,7 +231,7 @@ manipulate_buf_thread(void *arg)
 	char rand_char;
 	pthread_args_t *args = (pthread_args_t *)arg;
 
-	while (!args->entire_file_written) {
+	while (!args->entire_file_completed) {
 		rand_offset = (rand() % blocksize);
 		rand_char = (rand() % (126 - 33) + 33);
 		buf[rand_offset] = rand_char;
@@ -202,9 +244,9 @@ int
 main(int argc, char *argv[])
 {
 	const char *datapattern = "0xdeadbeef";
-	int ofd_flags = O_WRONLY | O_CREAT | O_DIRECT;
+	int fd_flags = O_DIRECT;
 	mode_t mode = S_IRUSR | S_IWUSR;
-	pthread_t write_thr;
+	pthread_t io_thr;
 	pthread_t manipul_thr;
 	int left = blocksize;
 	int offset = 0;
@@ -213,9 +255,15 @@ main(int argc, char *argv[])
 
 	parse_options(argc, argv);
 
-	ofd = open(outputfile, ofd_flags, mode);
+	if (write_op) {
-	if (ofd == -1) {
+		fd_flags |= (O_WRONLY | O_CREAT);
-		(void) fprintf(stderr, "%s, %s\n", execname, outputfile);
+	} else {
+		fd_flags |= O_RDONLY;
+	}
+
+	fd = open(filename, fd_flags, mode);
+	if (fd == -1) {
+		(void) fprintf(stderr, "%s, %s\n", execname, filename);
 		perror("open");
 		exit(2);
 	}
@@ -228,24 +276,22 @@ main(int argc, char *argv[])
 		exit(2);
 	}
 
-	if (!randompattern) {
+	if (write_op) {
-		/* Putting known data pattern in buffer */
+		if (!randompattern) {
-		while (left) {
+			/* Putting known data pattern in buffer */
-			size_t amt = MIN(strlen(datapattern), left);
+			while (left) {
-			memcpy(&buf[offset], datapattern, amt);
+				size_t amt = MIN(strlen(datapattern), left);
-			offset += amt;
+				memcpy(&buf[offset], datapattern, amt);
-			left -= amt;
+				offset += amt;
+				left -= amt;
+			}
+		} else {
+			/* Putting random data in buffer */
+			for (int i = 0; i < blocksize; i++)
+				buf[i] = rand();
 		}
-	} else {
-		/* Putting random data in buffer */
-		for (int i = 0; i < blocksize; i++)
-			buf[i] = rand();
 	}
 
-	/*
-	 * Writing using O_DIRECT while manipulating the buffer contents until
-	 * the entire file is written.
-	 */
 	if ((rc = pthread_create(&manipul_thr, NULL, manipulate_buf_thread,
 	    &args))) {
 		fprintf(stderr, "error: pthreads_create, manipul_thr, "
@@ -253,18 +299,34 @@ main(int argc, char *argv[])
 		exit(2);
 	}
 
-	if ((rc = pthread_create(&write_thr, NULL, write_thread, &args))) {
+	if (write_op) {
-		fprintf(stderr, "error: pthreads_create, write_thr, "
+		/*
-		    "rc: %d\n", rc);
+		 * Writing using O_DIRECT while manipulating the buffer contents
-		exit(2);
+		 * until the entire file is written.
+		 */
+		if ((rc = pthread_create(&io_thr, NULL, write_thread, &args))) {
+			fprintf(stderr, "error: pthreads_create, io_thr, "
+			    "rc: %d\n", rc);
+			exit(2);
+		}
+	} else {
+		/*
+		 * Reading using O_DIRECT while manipulating the buffer contents
+		 * until the entire file is read.
+		 */
+		if ((rc = pthread_create(&io_thr, NULL, read_thread, &args))) {
+			fprintf(stderr, "error: pthreads_create, io_thr, "
+			    "rc: %d\n", rc);
+			exit(2);
+		}
 	}
 
-	pthread_join(write_thr, NULL);
+	pthread_join(io_thr, NULL);
 	pthread_join(manipul_thr, NULL);
 
-	assert(args.entire_file_written == 1);
+	assert(args.entire_file_completed == 1);
 
-	(void) close(ofd);
+	(void) close(fd);
 
 	free(buf);
 

tests/zfs-tests/tests/Makefile.am

@@ -1477,6 +1477,7 @@ nobase_dist_datadir_zfs_tests_tests_SCRIPTS += \
 	functional/direct/dio_overwrites.ksh \
 	functional/direct/dio_property.ksh \
 	functional/direct/dio_random.ksh \
+	functional/direct/dio_read_verify.ksh \
 	functional/direct/dio_recordsize.ksh \
 	functional/direct/dio_unaligned_block.ksh \
 	functional/direct/dio_unaligned_filesize.ksh \

tests/zfs-tests/tests/functional/cli_root/json/json_sanity.ksh

@@ -30,28 +30,39 @@
 # STRATEGY:
 # 1. Run different zfs/zpool -j commands and check for valid JSON
 
+#
+# -j and --json mean the same thing. Each command will be run twice, replacing
+# JSONFLAG with the flag under test.
 list=(
-    "zpool status -j -g --json-int --json-flat-vdevs --json-pool-key-guid"
+    "zpool status JSONFLAG -g --json-int --json-flat-vdevs --json-pool-key-guid"
-    "zpool status -p -j -g --json-int --json-flat-vdevs --json-pool-key-guid"
+    "zpool status -p JSONFLAG -g --json-int --json-flat-vdevs --json-pool-key-guid"
-    "zpool status -j -c upath"
+    "zpool status JSONFLAG -c upath"
-    "zpool status -j"
+    "zpool status JSONFLAG"
-    "zpool status -j testpool1"
+    "zpool status JSONFLAG testpool1"
-    "zpool list -j"
+    "zpool list JSONFLAG"
-    "zpool list -j -g"
+    "zpool list JSONFLAG -g"
-    "zpool list -j -o fragmentation"
+    "zpool list JSONFLAG -o fragmentation"
-    "zpool get -j size"
+    "zpool get JSONFLAG size"
-    "zpool get -j all"
+    "zpool get JSONFLAG all"
-    "zpool version -j"
+    "zpool version JSONFLAG"
-    "zfs list -j"
+    "zfs list JSONFLAG"
-    "zfs list -j testpool1"
+    "zfs list JSONFLAG testpool1"
-    "zfs get -j all"
+    "zfs get JSONFLAG all"
-    "zfs get -j available"
+    "zfs get JSONFLAG available"
-    "zfs mount -j"
+    "zfs mount JSONFLAG"
-    "zfs version -j"
+    "zfs version JSONFLAG"
 )
 
-for cmd in "${list[@]}" ; do
+function run_json_tests
-    log_must eval "$cmd | jq > /dev/null" 
+{
-done
+	typeset flag=$1
+	for cmd in "${list[@]}" ; do
+	    cmd=${cmd//JSONFLAG/$flag}
+	    log_must eval "$cmd | jq > /dev/null"
+	done
+}
+
+log_must run_json_tests -j
+log_must run_json_tests --json
 
 log_pass "zpool and zfs commands outputted valid JSON"

tests/zfs-tests/tests/functional/cli_root/zpool_import/zpool_import_parallel_pos.ksh

@@ -113,7 +113,7 @@ wait
 parallel_time=$SECONDS
 log_note "asyncronously imported 4 pools in $parallel_time seconds"
 
-log_must test $parallel_time -lt $(($sequential_time / 3))
+log_must test $parallel_time -lt $(($sequential_time / 2))
 
 #
 # export pools with import delay injectors
@@ -132,6 +132,6 @@ log_must zpool import -a -d $DEVICE_DIR -f
 parallel_time=$SECONDS
 log_note "asyncronously imported 4 pools in $parallel_time seconds"
 
-log_must test $parallel_time -lt $(($sequential_time / 3))
+log_must test $parallel_time -lt $(($sequential_time / 2))
 
 log_pass "Pool imports occur in parallel"

tests/zfs-tests/tests/functional/dedup/dedup_quota.ksh

@@ -221,7 +221,7 @@ function ddt_dedup_vdev_limit
 	# For here, we just set the entry count a little higher than what we
 	# expect to allow for some instability.
 	#
-	log_must test $(ddt_entries) -le 600000
+	log_must test $(ddt_entries) -le 650000
 
 	do_clean
 }

tests/zfs-tests/tests/functional/direct/dio.kshlib

@@ -84,8 +84,9 @@ function get_zpool_status_chksum_verify_failures # pool_name vdev_type
 function get_zed_dio_verify_events # pool
 {
 	typeset pool=$1
+	typeset op=$2
 
-	val=$(zpool events $pool | grep -c dio_verify)
+	val=$(zpool events $pool | grep -c "dio_verify_${op}")
 
 	echo "$val"
 }
@@ -96,11 +97,12 @@ function get_zed_dio_verify_events # pool
 # zpool events
 # After getting that counts will clear the out the ZPool errors and events
 #
-function check_dio_write_chksum_verify_failures # pool vdev_type expect_errors
+function check_dio_chksum_verify_failures # pool vdev_type op expect_errors
 {
 	typeset pool=$1
 	typeset vdev_type=$2
 	typeset expect_errors=$3
+	typeset op=$4
 	typeset note_str="expecting none"
 
 	if [[ $expect_errors -ne 0 ]]; then
@@ -108,10 +110,10 @@ function check_dio_write_chksum_verify_failures # pool vdev_type expect_errors
 	fi
 
 	log_note "Checking for Direct I/O write checksum verify errors \
-	    $note_str on ZPool: $pool"
+	    $note_str on ZPool: $pool with $vdev_type"
 
 	status_failures=$(get_zpool_status_chksum_verify_failures $pool $vdev_type)
-	zed_dio_verify_events=$(get_zed_dio_verify_events $pool)
+	zed_dio_verify_events=$(get_zed_dio_verify_events $pool $op)
 
 	if [[ $expect_errors -ne 0 ]]; then
 		if [[ $status_failures -eq 0 ||

tests/zfs-tests/tests/functional/direct/dio_read_verify.ksh

@@ -0,0 +1,107 @@
+#!/bin/ksh -p
+#
+# CDDL HEADER START
+#
+# The contents of this file are subject to the terms of the
+# Common Development and Distribution License (the "License").
+# You may not use this file except in compliance with the License.
+#
+# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+# or https://opensource.org/licenses/CDDL-1.0.
+# See the License for the specific language governing permissions
+# and limitations under the License.
+#
+# When distributing Covered Code, include this CDDL HEADER in each
+# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+# If applicable, add the following below this CDDL HEADER, with the
+# fields enclosed by brackets "[]" replaced with your own identifying
+# information: Portions Copyright [yyyy] [name of copyright owner]
+#
+# CDDL HEADER END
+#
+
+#
+# Copyright (c) 2024 by Triad National Security, LLC.
+#
+
+. $STF_SUITE/include/libtest.shlib
+. $STF_SUITE/tests/functional/direct/dio.cfg
+. $STF_SUITE/tests/functional/direct/dio.kshlib
+
+#
+# DESCRIPTION:
+# 	Verify checksum verify works for Direct I/O reads.
+#
+# STRATEGY:
+#	1. Create a zpool from each vdev type.
+#	2. Start a Direct I/O read workload while manipulating the user buffer
+#	   contents.
+#	3. Verify there are Direct I/O read verify failures using
+#	   zpool status -d and checking for zevents. We also make sure there
+#	   are reported no data errors.
+#
+
+verify_runnable "global"
+
+log_assert "Verify checksum verify works for Direct I/O reads."
+
+log_onexit dio_cleanup
+
+NUMBLOCKS=300
+BS=$((128 * 1024)) # 128k
+
+log_must  truncate -s $MINVDEVSIZE $DIO_VDEVS
+
+# We will verify that there are no checksum errors for every Direct I/O read
+# while manipulating the buffer contents while the I/O is still in flight and
+# also that Direct I/O checksum verify failures and dio_verify_rd zevents are
+# reported.
+
+
+for type in "" "mirror" "raidz" "draid"; do
+	typeset vdev_type=$type
+	if [[ "${vdev_type}" == "" ]]; then
+		vdev_type="stripe"
+	fi
+
+	log_note "Verifying every Direct I/O read verify with VDEV type \
+	    ${vdev_type}"
+
+	create_pool $TESTPOOL1 $type $DIO_VDEVS
+	log_must eval "zfs create -o recordsize=128k -o compression=off  \
+	    $TESTPOOL1/$TESTFS1"
+
+	mntpnt=$(get_prop mountpoint $TESTPOOL1/$TESTFS1)
+	prev_dio_rd=$(get_iostats_stat $TESTPOOL1 direct_read_count)
+	prev_arc_rd=$(get_iostats_stat $TESTPOOL1 arc_read_count)
+
+	# Create the file before trying to manipulate the contents
+	log_must stride_dd -o "$mntpnt/direct-write.iso" -i /dev/urandom \
+	    -b $BS -c $NUMBLOCKS -D
+	# Manipulate the buffer contents will reading the file with Direct I/O
+	log_must manipulate_user_buffer -f "$mntpnt/direct-write.iso" \
+	    -n $NUMBLOCKS -b $BS -r
+
+	# Getting new Direct I/O and ARC Write counts.
+	curr_dio_rd=$(get_iostats_stat $TESTPOOL1 direct_read_count)
+	curr_arc_rd=$(get_iostats_stat $TESTPOOL1 arc_read_count)
+	total_dio_rd=$((curr_dio_rd - prev_dio_rd))
+	total_arc_rd=$((curr_arc_rd - prev_arc_rd))
+
+	log_note "Making sure there are no checksum errors with the ZPool"
+	log_must check_pool_status $TESTPOOL "errors" "No known data errors"
+
+	log_note "Making sure we have Direct I/O and ARC reads logged"
+	if [[ $total_dio_rd -lt 1 ]]; then
+		log_fail "No Direct I/O reads $total_dio_rd"
+	fi
+	if [[ $total_arc_rd -lt 1 ]]; then
+		log_fail "No ARC reads $total_arc_rd"
+	fi
+
+	log_note "Making sure we have Direct I/O write checksum verifies with ZPool"
+	check_dio_chksum_verify_failures "$TESTPOOL1" "$vdev_type" 1 "rd"
+	destroy_pool $TESTPOOL1
+done
+
+log_pass "Verified checksum verify works for Direct I/O reads." 

tests/zfs-tests/tests/functional/direct/dio_write_stable_pages.ksh

@@ -46,7 +46,7 @@ verify_runnable "global"
 function cleanup
 {
 	log_must rm -f "$mntpnt/direct-write.iso"
-	check_dio_write_chksum_verify_failures $TESTPOOL "raidz" 0
+	check_dio_chksum_verify_failures $TESTPOOL "raidz" 0 "wr"
 }
 
 log_assert "Verify stable pages work for Direct I/O writes."
@@ -76,8 +76,8 @@ do
 
 		# Manipulate the user's buffer while running O_DIRECT write
 		# workload with the buffer.
-		log_must manipulate_user_buffer -o "$mntpnt/direct-write.iso" \
+		log_must manipulate_user_buffer -f "$mntpnt/direct-write.iso" \
-		    -n $NUMBLOCKS -b $BS 
+		    -n $NUMBLOCKS -b $BS -w 
 
 		# Reading back the contents of the file
 		log_must stride_dd -i $mntpnt/direct-write.iso -o /dev/null \

tests/zfs-tests/tests/functional/direct/dio_write_verify.ksh

@@ -91,8 +91,8 @@ log_must set_tunable32 VDEV_DIRECT_WR_VERIFY 0
 log_note "Verifying no panics for Direct I/O writes with compression"
 log_must zfs set compression=on $TESTPOOL/$TESTFS
 prev_dio_wr=$(get_iostats_stat $TESTPOOL direct_write_count)
-log_must manipulate_user_buffer -o "$mntpnt/direct-write.iso" -n $NUMBLOCKS \
+log_must manipulate_user_buffer -f "$mntpnt/direct-write.iso" -n $NUMBLOCKS \
-    -b $BS
+    -b $BS -w
 curr_dio_wr=$(get_iostats_stat $TESTPOOL direct_write_count)
 total_dio_wr=$((curr_dio_wr - prev_dio_wr))
 
@@ -116,8 +116,8 @@ for i in $(seq 1 $ITERATIONS); do
 	    $i of $ITERATIONS with zfs_vdev_direct_write_verify=0"
 
 	prev_dio_wr=$(get_iostats_stat $TESTPOOL direct_write_count)
-	log_must manipulate_user_buffer -o "$mntpnt/direct-write.iso" \
+	log_must manipulate_user_buffer -f "$mntpnt/direct-write.iso" \
-	    -n $NUMBLOCKS -b $BS
+	    -n $NUMBLOCKS -b $BS -w
 
 	# Reading file back to verify checksum errors
 	filesize=$(get_file_size "$mntpnt/direct-write.iso")
@@ -144,7 +144,7 @@ for i in $(seq 1 $ITERATIONS); do
 	fi
 	log_note "Making sure we have no Direct I/O write checksum verifies \
 	    with ZPool"
-	check_dio_write_chksum_verify_failures $TESTPOOL "raidz" 0
+	check_dio_chksum_verify_failures $TESTPOOL "raidz" 0 "wr"
 
 	log_must rm -f "$mntpnt/direct-write.iso"
 done
@@ -166,8 +166,8 @@ for i in $(seq 1 $ITERATIONS); do
 	    $ITERATIONS with zfs_vdev_direct_write_verify=1"
 
 	prev_dio_wr=$(get_iostats_stat $TESTPOOL direct_write_count)
-	log_must manipulate_user_buffer -o "$mntpnt/direct-write.iso" \
+	log_must manipulate_user_buffer -f "$mntpnt/direct-write.iso" \
-	    -n $NUMBLOCKS -b $BS -e
+	    -n $NUMBLOCKS -b $BS -e -w
 
 	# Reading file back to verify there no are checksum errors
 	filesize=$(get_file_size "$mntpnt/direct-write.iso")
@@ -175,7 +175,7 @@ for i in $(seq 1 $ITERATIONS); do
 	log_must stride_dd -i "$mntpnt/direct-write.iso" -o /dev/null -b $BS \
 	    -c $num_blocks
 
-	# Getting new Direct I/O and ARC Write counts.
+	# Getting new Direct I/O write counts.
 	curr_dio_wr=$(get_iostats_stat $TESTPOOL direct_write_count)
 	total_dio_wr=$((curr_dio_wr - prev_dio_wr))
 
@@ -188,7 +188,7 @@ for i in $(seq 1 $ITERATIONS); do
 	fi
 
 	log_note "Making sure we have Direct I/O write checksum verifies with ZPool"
-	check_dio_write_chksum_verify_failures "$TESTPOOL" "raidz" 1
+	check_dio_chksum_verify_failures "$TESTPOOL" "raidz" 1 "wr"
 done
 
 log_must rm -f "$mntpnt/direct-write.iso"

tests/zfs-tests/tests/functional/replacement/resilver_restart_001.ksh

@@ -48,6 +48,8 @@ function cleanup
 	log_must set_tunable32 RESILVER_MIN_TIME_MS $ORIG_RESILVER_MIN_TIME
 	log_must set_tunable32 SCAN_SUSPEND_PROGRESS \
 	    $ORIG_SCAN_SUSPEND_PROGRESS
+	log_must set_tunable32 RESILVER_DEFER_PERCENT \
+	    $ORIG_RESILVER_DEFER_PERCENT
 	log_must set_tunable32 ZEVENT_LEN_MAX $ORIG_ZFS_ZEVENT_LEN_MAX
 	log_must zinject -c all
 	destroy_pool $TESTPOOL1
@@ -90,6 +92,7 @@ log_assert "Check for unnecessary resilver restarts"
 
 ORIG_RESILVER_MIN_TIME=$(get_tunable RESILVER_MIN_TIME_MS)
 ORIG_SCAN_SUSPEND_PROGRESS=$(get_tunable SCAN_SUSPEND_PROGRESS)
+ORIG_RESILVER_DEFER_PERCENT=$(get_tunable RESILVER_DEFER_PERCENT)
 ORIG_ZFS_ZEVENT_LEN_MAX=$(get_tunable ZEVENT_LEN_MAX)
 
 set -A RESTARTS -- '1' '2' '2' '2'

tests/zfs-tests/tests/functional/zvol/zvol_misc/zvol_misc_snapdev.ksh

@@ -117,5 +117,18 @@ log_must zfs set snapdev=visible $TESTPOOL
 verify_inherited 'snapdev' 'hidden' $SUBZVOL $VOLFS
 blockdev_missing $SUBSNAPDEV
 blockdev_exists $SNAPDEV
+log_must zfs destroy $SNAP
+
+# 4. Verify "rename" is correctly reflected when "snapdev=visible"
+# 4.1 First create a snapshot and verify the device is present
+log_must zfs snapshot $SNAP
+log_must zfs set snapdev=visible $ZVOL
+blockdev_exists $SNAPDEV
+# 4.2 rename the snapshot and verify the devices are updated
+log_must zfs rename $SNAP $SNAP-new
+blockdev_missing $SNAPDEV
+blockdev_exists $SNAPDEV-new
+# 4.3 cleanup
+log_must zfs destroy $SNAP-new
 
 log_pass "ZFS volume property 'snapdev' works as expected"

udev/vdev_id

@@ -124,6 +124,7 @@ TOPOLOGY=
 BAY=
 ENCL_ID=""
 UNIQ_ENCL_ID=""
+ZPAD=1
 
 usage() {
 	cat << EOF
@@ -154,7 +155,7 @@ map_slot() {
 	if [ -z "$MAPPED_SLOT" ] ; then
 		MAPPED_SLOT=$LINUX_SLOT
 	fi
-	printf "%d" "${MAPPED_SLOT}"
+	printf "%0${ZPAD}d" "${MAPPED_SLOT}"
 }
 
 map_channel() {
@@ -430,6 +431,15 @@ sas_handler() {
 		d=$(eval echo '$'{$i})
 		SLOT=$(echo "$d" | sed -e 's/^.*://')
 		;;
+	"bay_lun")
+		# Like 'bay' but with the LUN number appened. Added for SAS 
+		# multi-actuator HDDs, where one physical drive has multiple
+		# LUNs, thus multiple logical drives share the same bay number
+		i=$((i + 2))
+		d=$(eval echo '$'{$i})
+		LUN="-lun$(echo "$d" | sed -e 's/^.*://')"
+		SLOT=$(cat "$end_device_dir/bay_identifier" 2>/dev/null)
+		;;
 	"ses")
 		# look for this SAS path in all SCSI Enclosure Services
 		# (SES) enclosures
@@ -460,7 +470,7 @@ sas_handler() {
 		if [ -z "$CHAN" ] ; then
 			return
 		fi
-		echo "${CHAN}"-"${JBOD}"-"${SLOT}${PART}"
+		echo "${CHAN}"-"${JBOD}"-"${SLOT}${LUN}${PART}"
 	else
 		CHAN=$(map_channel "$PCI_ID" "$PORT")
 		SLOT=$(map_slot "$SLOT" "$CHAN")
@@ -468,7 +478,7 @@ sas_handler() {
 		if [ -z "$CHAN" ] ; then
 			return
 		fi
-		echo "${CHAN}${SLOT}${PART}"
+		echo "${CHAN}${SLOT}${LUN}${PART}"
 	fi
 }
 
@@ -748,6 +758,8 @@ if [ -z "$BAY" ] ; then
 	BAY=$(awk '($1 == "slot") {print $2; exit}' "$CONFIG")
 fi
 
+ZPAD=$(awk '($1 == "zpad_slot") {print $2; exit}' "$CONFIG")
+
 TOPOLOGY=${TOPOLOGY:-sas_direct}
 
 # Should we create /dev/by-enclosure symlinks?