Merge remote-tracking branch 'remotes/dgilbert/tags/pull-migration-20200703a' into staging

virtiofsd+migration pull 2020-07-03

A couple of small migration fixes, and some capability
rework for virtiofsd.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

# gpg: Signature made Fri 03 Jul 2020 16:26:35 BST
# 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-20200703a:
  migration: Count new_dirty instead of real_dirty
  migration: postcopy take proper error return
  virtiofsd: Allow addition or removal of capabilities
  virtiofsd: Check capability calls
  virtiofsd: Terminate capability list

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

6 files changed, 80 insertions, 13 deletions

diff --git a/docs/tools/virtiofsd.rst b/docs/tools/virtiofsd.rst
index 378594c422..824e713491 100644
--- a/docs/tools/virtiofsd.rst
+++ b/docs/tools/virtiofsd.rst
@@ -54,6 +54,11 @@ Options
   * flock|no_flock -
     Enable/disable flock.  The default is ``no_flock``.
 
+  * modcaps=CAPLIST
+    Modify the list of capabilities allowed; CAPLIST is a colon separated
+    list of capabilities, each preceded by either + or -, e.g.
+    ''+sys_admin:-chown''.
+
   * log_level=LEVEL -
     Print only log messages matching LEVEL or more severe.  LEVEL is one of
     ``err``, ``warn``, ``info``, or ``debug``.  The default is ``info``.
diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h
index 7b5c24e928..3ef729a23c 100644
--- a/include/exec/ram_addr.h
+++ b/include/exec/ram_addr.h
@@ -442,8 +442,7 @@ static inline void cpu_physical_memory_clear_dirty_range(ram_addr_t start,
 static inline
 uint64_t cpu_physical_memory_sync_dirty_bitmap(RAMBlock *rb,
                                                ram_addr_t start,
-                                               ram_addr_t length,
-                                               uint64_t *real_dirty_pages)
+                                               ram_addr_t length)
 {
     ram_addr_t addr;
     unsigned long word = BIT_WORD((start + rb->offset) >> TARGET_PAGE_BITS);
@@ -469,7 +468,6 @@ uint64_t cpu_physical_memory_sync_dirty_bitmap(RAMBlock *rb,
             if (src[idx][offset]) {
                 unsigned long bits = atomic_xchg(&src[idx][offset], 0);
                 unsigned long new_dirty;
-                *real_dirty_pages += ctpopl(bits);
                 new_dirty = ~dest[k];
                 dest[k] |= bits;
                 new_dirty &= bits;
@@ -502,7 +500,6 @@ uint64_t cpu_physical_memory_sync_dirty_bitmap(RAMBlock *rb,
                         start + addr + offset,
                         TARGET_PAGE_SIZE,
                         DIRTY_MEMORY_MIGRATION)) {
-                *real_dirty_pages += 1;
                 long k = (start + addr) >> TARGET_PAGE_BITS;
                 if (!test_and_set_bit(k, dest)) {
                     num_dirty++;
diff --git a/migration/postcopy-ram.c b/migration/postcopy-ram.c
index a36402722b..bef2a3afed 100644
--- a/migration/postcopy-ram.c
+++ b/migration/postcopy-ram.c
@@ -389,7 +389,7 @@ bool postcopy_ram_supported_by_host(MigrationIncomingState *mis)
      */
     if (munlockall()) {
         error_report("%s: munlockall: %s", __func__,  strerror(errno));
-        return -1;
+        goto out;
     }
 
     /*
diff --git a/migration/ram.c b/migration/ram.c
index 069b6e30bc..5554a7d2d8 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -859,9 +859,11 @@ static inline bool migration_bitmap_clear_dirty(RAMState *rs,
 /* Called with RCU critical section */
 static void ramblock_sync_dirty_bitmap(RAMState *rs, RAMBlock *rb)
 {
-    rs->migration_dirty_pages +=
-        cpu_physical_memory_sync_dirty_bitmap(rb, 0, rb->used_length,
-                                              &rs->num_dirty_pages_period);
+    uint64_t new_dirty_pages =
+        cpu_physical_memory_sync_dirty_bitmap(rb, 0, rb->used_length);
+
+    rs->migration_dirty_pages += new_dirty_pages;
+    rs->num_dirty_pages_period += new_dirty_pages;
 }
 
 /**
diff --git a/tools/virtiofsd/helper.c b/tools/virtiofsd/helper.c
index 00a1ef666a..3105b6c23a 100644
--- a/tools/virtiofsd/helper.c
+++ b/tools/virtiofsd/helper.c
@@ -174,6 +174,8 @@ void fuse_cmdline_help(void)
            "                               default: no_writeback\n"
            "    -o xattr|no_xattr          enable/disable xattr\n"
            "                               default: no_xattr\n"
+           "    -o modcaps=CAPLIST         Modify the list of capabilities\n"
+           "                               e.g. -o modcaps=+sys_admin:-chown\n"
            "    --rlimit-nofile=<num>      set maximum number of file descriptors\n"
            "                               (0 leaves rlimit unchanged)\n"
            "                               default: min(1000000, fs.file-max - 16384)\n"
diff --git a/tools/virtiofsd/passthrough_ll.c b/tools/virtiofsd/passthrough_ll.c
index 2ce7c96085..94e0de2d2b 100644
--- a/tools/virtiofsd/passthrough_ll.c
+++ b/tools/virtiofsd/passthrough_ll.c
@@ -145,6 +145,7 @@ struct lo_data {
     int posix_lock;
     int xattr;
     char *source;
+    char *modcaps;
     double timeout;
     int cache;
     int timeout_set;
@@ -170,6 +171,7 @@ static const struct fuse_opt lo_opts[] = {
     { "no_posix_lock", offsetof(struct lo_data, posix_lock), 0 },
     { "xattr", offsetof(struct lo_data, xattr), 1 },
     { "no_xattr", offsetof(struct lo_data, xattr), 0 },
+    { "modcaps=%s", offsetof(struct lo_data, modcaps), 0 },
     { "timeout=%lf", offsetof(struct lo_data, timeout), 0 },
     { "timeout=", offsetof(struct lo_data, timeout_set), 1 },
     { "cache=none", offsetof(struct lo_data, cache), CACHE_NONE },
@@ -2570,9 +2572,11 @@ static void setup_mounts(const char *source)
 
 /*
  * Only keep whitelisted capabilities that are needed for file system operation
+ * The (possibly NULL) modcaps_in string passed in is free'd before exit.
  */
-static void setup_capabilities(void)
+static void setup_capabilities(char *modcaps_in)
 {
+    char *modcaps = modcaps_in;
     pthread_mutex_lock(&cap.mutex);
     capng_restore_state(&cap.saved);
 
@@ -2589,7 +2593,7 @@ static void setup_capabilities(void)
      */
     capng_setpid(syscall(SYS_gettid));
     capng_clear(CAPNG_SELECT_BOTH);
-    capng_updatev(CAPNG_ADD, CAPNG_PERMITTED | CAPNG_EFFECTIVE,
+    if (capng_updatev(CAPNG_ADD, CAPNG_PERMITTED | CAPNG_EFFECTIVE,
             CAP_CHOWN,
             CAP_DAC_OVERRIDE,
             CAP_DAC_READ_SEARCH,
@@ -2598,10 +2602,67 @@ static void setup_capabilities(void)
             CAP_SETGID,
             CAP_SETUID,
             CAP_MKNOD,
-            CAP_SETFCAP);
-    capng_apply(CAPNG_SELECT_BOTH);
+            CAP_SETFCAP,
+            -1)) {
+        fuse_log(FUSE_LOG_ERR, "%s: capng_updatev failed\n", __func__);
+        exit(1);
+    }
+
+    /*
+     * The modcaps option is a colon separated list of caps,
+     * each preceded by either + or -.
+     */
+    while (modcaps) {
+        capng_act_t action;
+        int cap;
+
+        char *next = strchr(modcaps, ':');
+        if (next) {
+            *next = '\0';
+            next++;
+        }
+
+        switch (modcaps[0]) {
+        case '+':
+            action = CAPNG_ADD;
+            break;
+
+        case '-':
+            action = CAPNG_DROP;
+            break;
+
+        default:
+            fuse_log(FUSE_LOG_ERR,
+                     "%s: Expecting '+'/'-' in modcaps but found '%c'\n",
+                     __func__, modcaps[0]);
+            exit(1);
+        }
+        cap = capng_name_to_capability(modcaps + 1);
+        if (cap < 0) {
+            fuse_log(FUSE_LOG_ERR, "%s: Unknown capability '%s'\n", __func__,
+                     modcaps);
+            exit(1);
+        }
+        if (capng_update(action, CAPNG_PERMITTED | CAPNG_EFFECTIVE, cap)) {
+            fuse_log(FUSE_LOG_ERR, "%s: capng_update failed for '%s'\n",
+                     __func__, modcaps);
+            exit(1);
+        }
+
+        modcaps = next;
+    }
+    g_free(modcaps_in);
+
+    if (capng_apply(CAPNG_SELECT_BOTH)) {
+        fuse_log(FUSE_LOG_ERR, "%s: capng_apply failed\n", __func__);
+        exit(1);
+    }
 
     cap.saved = capng_save_state();
+    if (!cap.saved) {
+        fuse_log(FUSE_LOG_ERR, "%s: capng_save_state failed\n", __func__);
+        exit(1);
+    }
     pthread_mutex_unlock(&cap.mutex);
 }
 
@@ -2615,7 +2676,7 @@ static void setup_sandbox(struct lo_data *lo, struct fuse_session *se,
     setup_namespaces(lo, se);
     setup_mounts(lo->source);
     setup_seccomp(enable_syslog);
-    setup_capabilities();
+    setup_capabilities(g_strdup(lo->modcaps));
 }
 
 /* Set the maximum number of open file descriptors */