diff options
author | Peter Maydell <peter.maydell@linaro.org> | 2020-11-02 20:29:50 +0000 |
---|---|---|
committer | Peter Maydell <peter.maydell@linaro.org> | 2020-11-02 20:29:50 +0000 |
commit | 8545ae485b1e8e43cc0137310c4c68dbece59990 (patch) | |
tree | e5c7a35c156719ca8a7e342e2f94746c00c3f171 /tools | |
parent | 8680d6e36468f1ca00e2fe749bef50585d632401 (diff) | |
parent | af1bb3fe7f146fafdaadb479975ca2b53b49df40 (diff) | |
download | qemu-8545ae485b1e8e43cc0137310c4c68dbece59990.zip qemu-8545ae485b1e8e43cc0137310c4c68dbece59990.tar.gz qemu-8545ae485b1e8e43cc0137310c4c68dbece59990.tar.bz2 |
Merge remote-tracking branch 'remotes/dgilbert/tags/pull-migration-20201102a' into staging
Migration and virtiofs fixes 2020-11-02
Fixes for postcopy migration test hang
A seccomp crash for virtiofsd on some !x86
Help message and minor CID fix
And another crack at Max's set.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
# gpg: Signature made Mon 02 Nov 2020 19:54:59 GMT
# gpg: using RSA key 45F5C71B4A0CB7FB977A9FA90516331EBC5BFDE7
# gpg: Good signature from "Dr. David Alan Gilbert (RH2) <dgilbert@redhat.com>" [full]
# Primary key fingerprint: 45F5 C71B 4A0C B7FB 977A 9FA9 0516 331E BC5B FDE7
* remotes/dgilbert/tags/pull-migration-20201102a:
tests/acceptance: Add virtiofs_submounts.py
tests/acceptance/boot_linux: Accept SSH pubkey
virtiofsd: Announce sub-mount points
virtiofsd: Add mount ID to the lo_inode key
meson.build: Check for statx()
virtiofsd: Add attr_flags to fuse_entry_param
virtiofsd: Check FUSE_SUBMOUNTS
virtiofsd: Fix the help message of posix lock
tools/virtiofsd: Check vu_init() return value (CID 1435958)
virtiofsd: Seccomp: Add 'send' for syslog
migration: Postpone the kick of the fault thread after recover
migration: Unify reset of last_rb on destination node when recover
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Diffstat (limited to 'tools')
-rw-r--r-- | tools/virtiofsd/fuse_common.h | 7 | ||||
-rw-r--r-- | tools/virtiofsd/fuse_lowlevel.c | 5 | ||||
-rw-r--r-- | tools/virtiofsd/fuse_lowlevel.h | 5 | ||||
-rw-r--r-- | tools/virtiofsd/fuse_virtio.c | 7 | ||||
-rw-r--r-- | tools/virtiofsd/helper.c | 3 | ||||
-rw-r--r-- | tools/virtiofsd/passthrough_ll.c | 117 | ||||
-rw-r--r-- | tools/virtiofsd/passthrough_seccomp.c | 2 |
7 files changed, 133 insertions, 13 deletions
diff --git a/tools/virtiofsd/fuse_common.h b/tools/virtiofsd/fuse_common.h index 686c42c..5aee519 100644 --- a/tools/virtiofsd/fuse_common.h +++ b/tools/virtiofsd/fuse_common.h @@ -353,6 +353,13 @@ struct fuse_file_info { #define FUSE_CAP_NO_OPENDIR_SUPPORT (1 << 24) /** + * Indicates that the kernel supports the FUSE_ATTR_SUBMOUNT flag. + * + * Setting (or unsetting) this flag in the `want` field has *no effect*. + */ +#define FUSE_CAP_SUBMOUNTS (1 << 27) + +/** * Ioctl flags * * FUSE_IOCTL_COMPAT: 32bit compat ioctl on 64bit machine diff --git a/tools/virtiofsd/fuse_lowlevel.c b/tools/virtiofsd/fuse_lowlevel.c index 4d1ba29..c70fb16 100644 --- a/tools/virtiofsd/fuse_lowlevel.c +++ b/tools/virtiofsd/fuse_lowlevel.c @@ -341,6 +341,8 @@ static void fill_entry(struct fuse_entry_out *arg, .attr_valid_nsec = calc_timeout_nsec(e->attr_timeout), }; convert_stat(&e->attr, &arg->attr); + + arg->attr.flags = e->attr_flags; } /* @@ -1988,6 +1990,9 @@ static void do_init(fuse_req_t req, fuse_ino_t nodeid, bufsize = max_bufsize; } } + if (arg->flags & FUSE_SUBMOUNTS) { + se->conn.capable |= FUSE_CAP_SUBMOUNTS; + } #ifdef HAVE_SPLICE #ifdef HAVE_VMSPLICE se->conn.capable |= FUSE_CAP_SPLICE_WRITE | FUSE_CAP_SPLICE_MOVE; diff --git a/tools/virtiofsd/fuse_lowlevel.h b/tools/virtiofsd/fuse_lowlevel.h index 562fd52..9c06240 100644 --- a/tools/virtiofsd/fuse_lowlevel.h +++ b/tools/virtiofsd/fuse_lowlevel.h @@ -102,6 +102,11 @@ struct fuse_entry_param { * large value. */ double entry_timeout; + + /** + * Flags for fuse_attr.flags that do not fit into attr. + */ + uint32_t attr_flags; }; /** diff --git a/tools/virtiofsd/fuse_virtio.c b/tools/virtiofsd/fuse_virtio.c index 3249369..83ba07c 100644 --- a/tools/virtiofsd/fuse_virtio.c +++ b/tools/virtiofsd/fuse_virtio.c @@ -1013,8 +1013,11 @@ int virtio_session_mount(struct fuse_session *se) se->vu_socketfd = data_sock; se->virtio_dev->se = se; pthread_rwlock_init(&se->virtio_dev->vu_dispatch_rwlock, NULL); - vu_init(&se->virtio_dev->dev, 2, se->vu_socketfd, fv_panic, NULL, - fv_set_watch, fv_remove_watch, &fv_iface); + if (!vu_init(&se->virtio_dev->dev, 2, se->vu_socketfd, fv_panic, NULL, + fv_set_watch, fv_remove_watch, &fv_iface)) { + fuse_log(FUSE_LOG_ERR, "%s: vu_init failed\n", __func__); + return -1; + } return 0; } diff --git a/tools/virtiofsd/helper.c b/tools/virtiofsd/helper.c index 2e181a4..75ac48d 100644 --- a/tools/virtiofsd/helper.c +++ b/tools/virtiofsd/helper.c @@ -161,7 +161,7 @@ void fuse_cmdline_help(void) " allowed (default: 10)\n" " -o posix_lock|no_posix_lock\n" " enable/disable remote posix lock\n" - " default: posix_lock\n" + " default: no_posix_lock\n" " -o readdirplus|no_readdirplus\n" " enable/disable readirplus\n" " default: readdirplus except with " @@ -190,6 +190,7 @@ void fuse_cmdline_help(void) " retain/discard O_DIRECT flags passed down\n" " to virtiofsd from guest applications.\n" " default: no_allow_direct_io\n" + " -o announce_submounts Announce sub-mount points to the guest\n" ); } diff --git a/tools/virtiofsd/passthrough_ll.c b/tools/virtiofsd/passthrough_ll.c index a0beb98..ec1008b 100644 --- a/tools/virtiofsd/passthrough_ll.c +++ b/tools/virtiofsd/passthrough_ll.c @@ -40,6 +40,7 @@ #include "fuse_virtio.h" #include "fuse_log.h" #include "fuse_lowlevel.h" +#include "standard-headers/linux/fuse.h" #include <assert.h> #include <cap-ng.h> #include <dirent.h> @@ -94,6 +95,7 @@ struct lo_map { struct lo_key { ino_t ino; dev_t dev; + uint64_t mnt_id; }; struct lo_inode { @@ -166,6 +168,8 @@ struct lo_data { int readdirplus_set; int readdirplus_clear; int allow_direct_io; + int announce_submounts; + bool use_statx; struct lo_inode root; GHashTable *inodes; /* protected by lo->mutex */ struct lo_map ino_map; /* protected by lo->mutex */ @@ -205,6 +209,7 @@ static const struct fuse_opt lo_opts[] = { { "no_readdirplus", offsetof(struct lo_data, readdirplus_clear), 1 }, { "allow_direct_io", offsetof(struct lo_data, allow_direct_io), 1 }, { "no_allow_direct_io", offsetof(struct lo_data, allow_direct_io), 0 }, + { "announce_submounts", offsetof(struct lo_data, announce_submounts), 1 }, FUSE_OPT_END }; static bool use_syslog = false; @@ -219,7 +224,8 @@ static struct { /* That we loaded cap-ng in the current thread from the saved */ static __thread bool cap_loaded = 0; -static struct lo_inode *lo_find(struct lo_data *lo, struct stat *st); +static struct lo_inode *lo_find(struct lo_data *lo, struct stat *st, + uint64_t mnt_id); static int is_dot_or_dotdot(const char *name) { @@ -598,6 +604,20 @@ static void lo_init(void *userdata, struct fuse_conn_info *conn) fuse_log(FUSE_LOG_DEBUG, "lo_init: disabling readdirplus\n"); conn->want &= ~FUSE_CAP_READDIRPLUS; } + + if (!(conn->capable & FUSE_CAP_SUBMOUNTS) && lo->announce_submounts) { + fuse_log(FUSE_LOG_WARNING, "lo_init: Cannot announce submounts, client " + "does not support it\n"); + lo->announce_submounts = false; + } + +#ifndef CONFIG_STATX + if (lo->announce_submounts) { + fuse_log(FUSE_LOG_WARNING, "lo_init: Cannot announce submounts, there " + "is no statx()\n"); + lo->announce_submounts = false; + } +#endif } static void lo_getattr(fuse_req_t req, fuse_ino_t ino, @@ -741,12 +761,14 @@ out_err: fuse_reply_err(req, saverr); } -static struct lo_inode *lo_find(struct lo_data *lo, struct stat *st) +static struct lo_inode *lo_find(struct lo_data *lo, struct stat *st, + uint64_t mnt_id) { struct lo_inode *p; struct lo_key key = { .ino = st->st_ino, .dev = st->st_dev, + .mnt_id = mnt_id, }; pthread_mutex_lock(&lo->mutex); @@ -774,6 +796,60 @@ static void posix_locks_value_destroy(gpointer data) free(plock); } +static int do_statx(struct lo_data *lo, int dirfd, const char *pathname, + struct stat *statbuf, int flags, uint64_t *mnt_id) +{ + int res; + +#if defined(CONFIG_STATX) && defined(STATX_MNT_ID) + if (lo->use_statx) { + struct statx statxbuf; + + res = statx(dirfd, pathname, flags, STATX_BASIC_STATS | STATX_MNT_ID, + &statxbuf); + if (!res) { + memset(statbuf, 0, sizeof(*statbuf)); + statbuf->st_dev = makedev(statxbuf.stx_dev_major, + statxbuf.stx_dev_minor); + statbuf->st_ino = statxbuf.stx_ino; + statbuf->st_mode = statxbuf.stx_mode; + statbuf->st_nlink = statxbuf.stx_nlink; + statbuf->st_uid = statxbuf.stx_uid; + statbuf->st_gid = statxbuf.stx_gid; + statbuf->st_rdev = makedev(statxbuf.stx_rdev_major, + statxbuf.stx_rdev_minor); + statbuf->st_size = statxbuf.stx_size; + statbuf->st_blksize = statxbuf.stx_blksize; + statbuf->st_blocks = statxbuf.stx_blocks; + statbuf->st_atim.tv_sec = statxbuf.stx_atime.tv_sec; + statbuf->st_atim.tv_nsec = statxbuf.stx_atime.tv_nsec; + statbuf->st_mtim.tv_sec = statxbuf.stx_mtime.tv_sec; + statbuf->st_mtim.tv_nsec = statxbuf.stx_mtime.tv_nsec; + statbuf->st_ctim.tv_sec = statxbuf.stx_ctime.tv_sec; + statbuf->st_ctim.tv_nsec = statxbuf.stx_ctime.tv_nsec; + + if (statxbuf.stx_mask & STATX_MNT_ID) { + *mnt_id = statxbuf.stx_mnt_id; + } else { + *mnt_id = 0; + } + return 0; + } else if (errno != ENOSYS) { + return -1; + } + lo->use_statx = false; + /* fallback */ + } +#endif + res = fstatat(dirfd, pathname, statbuf, flags); + if (res == -1) { + return -1; + } + *mnt_id = 0; + + return 0; +} + /* * Increments nlookup and caller must release refcount using * lo_inode_put(&parent). @@ -784,6 +860,7 @@ static int lo_do_lookup(fuse_req_t req, fuse_ino_t parent, const char *name, int newfd; int res; int saverr; + uint64_t mnt_id; struct lo_data *lo = lo_data(req); struct lo_inode *inode = NULL; struct lo_inode *dir = lo_inode(req, parent); @@ -811,12 +888,18 @@ static int lo_do_lookup(fuse_req_t req, fuse_ino_t parent, const char *name, goto out_err; } - res = fstatat(newfd, "", &e->attr, AT_EMPTY_PATH | AT_SYMLINK_NOFOLLOW); + res = do_statx(lo, newfd, "", &e->attr, AT_EMPTY_PATH | AT_SYMLINK_NOFOLLOW, + &mnt_id); if (res == -1) { goto out_err; } - inode = lo_find(lo, &e->attr); + if (S_ISDIR(e->attr.st_mode) && lo->announce_submounts && + (e->attr.st_dev != dir->key.dev || mnt_id != dir->key.mnt_id)) { + e->attr_flags |= FUSE_ATTR_SUBMOUNT; + } + + inode = lo_find(lo, &e->attr, mnt_id); if (inode) { close(newfd); } else { @@ -838,6 +921,7 @@ static int lo_do_lookup(fuse_req_t req, fuse_ino_t parent, const char *name, inode->fd = newfd; inode->key.ino = e->attr.st_ino; inode->key.dev = e->attr.st_dev; + inode->key.mnt_id = mnt_id; pthread_mutex_init(&inode->plock_mutex, NULL); inode->posix_locks = g_hash_table_new_full( g_direct_hash, g_direct_equal, NULL, posix_locks_value_destroy); @@ -1090,15 +1174,23 @@ static struct lo_inode *lookup_name(fuse_req_t req, fuse_ino_t parent, const char *name) { int res; + uint64_t mnt_id; struct stat attr; + struct lo_data *lo = lo_data(req); + struct lo_inode *dir = lo_inode(req, parent); - res = fstatat(lo_fd(req, parent), name, &attr, - AT_EMPTY_PATH | AT_SYMLINK_NOFOLLOW); + if (!dir) { + return NULL; + } + + res = do_statx(lo, dir->fd, name, &attr, + AT_EMPTY_PATH | AT_SYMLINK_NOFOLLOW, &mnt_id); + lo_inode_put(lo, &dir); if (res == -1) { return NULL; } - return lo_find(lo_data(req), &attr); + return lo_find(lo, &attr, mnt_id); } static void lo_rmdir(fuse_req_t req, fuse_ino_t parent, const char *name) @@ -3266,6 +3358,7 @@ static void setup_root(struct lo_data *lo, struct lo_inode *root) { int fd, res; struct stat stat; + uint64_t mnt_id; fd = open("/", O_PATH); if (fd == -1) { @@ -3273,7 +3366,8 @@ static void setup_root(struct lo_data *lo, struct lo_inode *root) exit(1); } - res = fstatat(fd, "", &stat, AT_EMPTY_PATH | AT_SYMLINK_NOFOLLOW); + res = do_statx(lo, fd, "", &stat, AT_EMPTY_PATH | AT_SYMLINK_NOFOLLOW, + &mnt_id); if (res == -1) { fuse_log(FUSE_LOG_ERR, "fstatat(%s): %m\n", lo->source); exit(1); @@ -3283,6 +3377,7 @@ static void setup_root(struct lo_data *lo, struct lo_inode *root) root->fd = fd; root->key.ino = stat.st_ino; root->key.dev = stat.st_dev; + root->key.mnt_id = mnt_id; root->nlookup = 2; g_atomic_int_set(&root->refcount, 2); } @@ -3291,7 +3386,7 @@ static guint lo_key_hash(gconstpointer key) { const struct lo_key *lkey = key; - return (guint)lkey->ino + (guint)lkey->dev; + return (guint)lkey->ino + (guint)lkey->dev + (guint)lkey->mnt_id; } static gboolean lo_key_equal(gconstpointer a, gconstpointer b) @@ -3299,7 +3394,7 @@ static gboolean lo_key_equal(gconstpointer a, gconstpointer b) const struct lo_key *la = a; const struct lo_key *lb = b; - return la->ino == lb->ino && la->dev == lb->dev; + return la->ino == lb->ino && la->dev == lb->dev && la->mnt_id == lb->mnt_id; } static void fuse_lo_data_cleanup(struct lo_data *lo) @@ -3445,6 +3540,8 @@ int main(int argc, char *argv[]) exit(1); } + lo.use_statx = true; + se = fuse_session_new(&args, &lo_oper, sizeof(lo_oper), &lo); if (se == NULL) { goto err_out1; diff --git a/tools/virtiofsd/passthrough_seccomp.c b/tools/virtiofsd/passthrough_seccomp.c index eb9af82..11623f5 100644 --- a/tools/virtiofsd/passthrough_seccomp.c +++ b/tools/virtiofsd/passthrough_seccomp.c @@ -76,6 +76,7 @@ static const int syscall_whitelist[] = { SCMP_SYS(mremap), SCMP_SYS(munmap), SCMP_SYS(newfstatat), + SCMP_SYS(statx), SCMP_SYS(open), SCMP_SYS(openat), SCMP_SYS(ppoll), @@ -118,6 +119,7 @@ static const int syscall_whitelist[] = { /* Syscalls used when --syslog is enabled */ static const int syscall_whitelist_syslog[] = { + SCMP_SYS(send), SCMP_SYS(sendto), }; |