Re: [PATCH v3] libceph: print fsid and client gid with osd id

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 2022/04/26 13:59, Daichi Mukai wrote:
Print fsid and client gid in libceph log messages to distinct from which
each message come.

Signed-off-by: Satoru Takeuchi <satoru.takeuchi@xxxxxxxxx>
Signed-off-by: Daichi Mukai <daichi-mukai@xxxxxxxxxxxx>
---
  include/linux/ceph/osdmap.h |  2 +-
  net/ceph/osd_client.c       |  3 ++-
  net/ceph/osdmap.c           | 43 ++++++++++++++++++++++++++-----------
  3 files changed, 34 insertions(+), 14 deletions(-)

diff --git a/include/linux/ceph/osdmap.h b/include/linux/ceph/osdmap.h
index 5553019c3f07..a9216c64350c 100644
--- a/include/linux/ceph/osdmap.h
+++ b/include/linux/ceph/osdmap.h
@@ -253,7 +253,7 @@ static inline int ceph_decode_pgid(void **p, void *end, struct ceph_pg *pgid)
  struct ceph_osdmap *ceph_osdmap_alloc(void);
  struct ceph_osdmap *ceph_osdmap_decode(void **p, void *end, bool msgr2);
  struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, bool msgr2,
-                         struct ceph_osdmap *map);
+                         struct ceph_osdmap *map, u64 gid);
  extern void ceph_osdmap_destroy(struct ceph_osdmap *map);

  struct ceph_osds {
diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c
index 1c5815530e0d..e7c01fcc1d16 100644
--- a/net/ceph/osd_client.c
+++ b/net/ceph/osd_client.c
@@ -3920,7 +3920,8 @@ static int handle_one_map(struct ceph_osd_client *osdc,
      if (incremental)
          newmap = osdmap_apply_incremental(&p, end,
                            ceph_msgr2(osdc->client),
-                          osdc->osdmap);
+                          osdc->osdmap,
+                          ceph_client_gid(osdc->client));
      else
          newmap = ceph_osdmap_decode(&p, end, ceph_msgr2(osdc->client));
      if (IS_ERR(newmap))
diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c
index 2823bb3cff55..cd65677baff3 100644
--- a/net/ceph/osdmap.c
+++ b/net/ceph/osdmap.c
@@ -1549,8 +1549,23 @@ static int decode_primary_affinity(void **p, void *end,
      return -EINVAL;
  }

+static __printf(3, 4)
+void print_osd_info(struct ceph_fsid *fsid, u64 gid, const char *fmt, ...)
+{
+    struct va_format vaf;
+    va_list args;
+
+    va_start(args, fmt);
+    vaf.fmt = fmt;
+    vaf.va = &args;
+
+    printk(KERN_INFO "%s (%pU %lld): %pV", KBUILD_MODNAME, fsid, gid, &vaf);
+
+    va_end(args);
+}
+
  static int decode_new_primary_affinity(void **p, void *end,
-                       struct ceph_osdmap *map)
+                       struct ceph_osdmap *map, u64 gid)
  {
      u32 n;

@@ -1566,7 +1581,8 @@ static int decode_new_primary_affinity(void **p, void *end,
          if (ret)
              return ret;

-        pr_info("osd%d primary-affinity 0x%x\n", osd, aff);
+        print_osd_info(&map->fsid, gid, "osd%d primary-affinity 0x%x\n",
+                   osd, aff);
      }

      return 0;
@@ -1825,7 +1841,8 @@ struct ceph_osdmap *ceph_osdmap_decode(void **p, void *end, bool msgr2)
   *     new_state: { osd=6, xorstate=EXISTS } # clear osd_state
   */
  static int decode_new_up_state_weight(void **p, void *end, u8 struct_v,
-                      bool msgr2, struct ceph_osdmap *map)
+                      bool msgr2, struct ceph_osdmap *map,
+                      u64 gid)
  {
      void *new_up_client;
      void *new_state;
@@ -1864,9 +1881,10 @@ static int decode_new_up_state_weight(void **p, void *end, u8 struct_v,
          osd = ceph_decode_32(p);
          w = ceph_decode_32(p);
          BUG_ON(osd >= map->max_osd);
-        pr_info("osd%d weight 0x%x %s\n", osd, w,
-             w == CEPH_OSD_IN ? "(in)" :
-             (w == CEPH_OSD_OUT ? "(out)" : ""));
+        print_osd_info(&map->fsid, gid, "osd%d weight 0x%x %s\n",
+                   osd, w,
+                   w == CEPH_OSD_IN ? "(in)" :
+                   (w == CEPH_OSD_OUT ? "(out)" : ""));
          map->osd_weight[osd] = w;

          /*
@@ -1898,10 +1916,11 @@ static int decode_new_up_state_weight(void **p, void *end, u8 struct_v,
          BUG_ON(osd >= map->max_osd);
          if ((map->osd_state[osd] & CEPH_OSD_UP) &&
              (xorstate & CEPH_OSD_UP))
-            pr_info("osd%d down\n", osd);
+            print_osd_info(&map->fsid, gid, "osd%d down\n", osd);
          if ((map->osd_state[osd] & CEPH_OSD_EXISTS) &&
              (xorstate & CEPH_OSD_EXISTS)) {
-            pr_info("osd%d does not exist\n", osd);
+            print_osd_info(&map->fsid, gid, "osd%d does not exist\n",
+                       osd);
              ret = set_primary_affinity(map, osd,
                             CEPH_OSD_DEFAULT_PRIMARY_AFFINITY);
              if (ret)
@@ -1931,7 +1950,7 @@ static int decode_new_up_state_weight(void **p, void *end, u8 struct_v,

          dout("%s osd%d addr %s\n", __func__, osd, ceph_pr_addr(&addr));

-        pr_info("osd%d up\n", osd);
+        print_osd_info(&map->fsid, gid, "osd%d up\n", osd);
          map->osd_state[osd] |= CEPH_OSD_EXISTS | CEPH_OSD_UP;
          map->osd_addr[osd] = addr;
      }
@@ -1947,7 +1966,7 @@ static int decode_new_up_state_weight(void **p, void *end, u8 struct_v,
   * decode and apply an incremental map update.
   */
  struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, bool msgr2,
-                         struct ceph_osdmap *map)
+                         struct ceph_osdmap *map, u64 gid)
  {
      struct ceph_fsid fsid;
      u32 epoch = 0;
@@ -2033,7 +2052,7 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, bool msgr2,
      }

      /* new_up_client, new_state, new_weight */
-    err = decode_new_up_state_weight(p, end, struct_v, msgr2, map);
+    err = decode_new_up_state_weight(p, end, struct_v, msgr2, map, gid);
      if (err)
          goto bad;

@@ -2051,7 +2070,7 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, bool msgr2,

      /* new_primary_affinity */
      if (struct_v >= 2) {
-        err = decode_new_primary_affinity(p, end, map);
+        err = decode_new_primary_affinity(p, end, map, gid);
          if (err)
              goto bad;
      }

Would you please review this patch?

Thanks,
Daichi



[Index of Archives]     [CEPH Users]     [Ceph Large]     [Ceph Dev]     [Information on CEPH]     [Linux BTRFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux