Fix osd_peering_pg_test

mon/stats.js

@@ -37,6 +37,7 @@ function derive_osd_stats(st, prev, prev_diff)
         const n = c.count - BigInt(pr && pr.count||0);
         diff.recovery_stats[op] = { ...c, bps: n > 0 ? b*1000n/timediff : 0n, iops: n > 0 ? n*1000n/timediff : 0n };
     }
+    diff.inode_stats = {};
     for (const pool_id in st.inode_stats||{})
     {
         diff.inode_stats[pool_id] = {};

src/blockstore/blockstore_heap.cpp

@@ -334,19 +334,12 @@ corrupted_block:
                     block_num, block_offset, wr->size, sizeof(heap_entry_t));
                 goto corrupted_block;
             }
-            if (wr->is_garbage())
-            {
-                // Garbage collection is only performed when writing new entries into the block
-                // because it needs a fake LSN and modified blocks require consecutive modified LSNs
-                // That's why garbage entries may persist on disk
-                if (log_level > 5)
-                {
-                    fprintf(stderr, "Notice: skipping garbage entry %jx:%jx v%ju l%ju in metadata block %u at %u\n",
-                        wr->inode, wr->stripe, wr->version, wr->lsn, block_num, block_offset);
-                }
-                block_offset += wr->size;
-                continue;
-            }
+            // Garbage collection is only performed when writing new entries into the block
+            // because it needs a fake LSN and modified blocks require consecutive modified LSNs
+            // At the same time, further modifications _after_ putting new entries into the block,
+            // but _before_ writing it, may mark some entries in it as garbage. That's why garbage
+            // entries may still be present on disk.
+            wr->entry_type &= ~BS_HEAP_GARBAGE;
             if ((wr->entry_type & BS_HEAP_TYPE) < BS_HEAP_BIG_WRITE ||
                 (wr->entry_type & BS_HEAP_TYPE) > BS_HEAP_ROLLBACK ||
                 (wr->entry_type & ~(BS_HEAP_TYPE|BS_HEAP_STABLE)) ||

src/blockstore/blockstore_init.cpp

@@ -311,7 +311,7 @@ resume_8:
         uint32_t block_num = recheck_mod[i];
         uint64_t block_offset = bs->dsk.meta_offset + (uint64_t)(block_num+1) * bs->dsk.meta_block_size;
         data = ((ring_data_t*)sqe->user_data);
-        uint8_t *buf = (uint8_t*)malloc_or_die(bs->dsk.meta_block_size);
+        uint8_t *buf = (uint8_t*)memalign_or_die(MEM_ALIGNMENT, bs->dsk.meta_block_size);
         bs->heap->get_meta_block(block_num, buf);
         data->iov = { buf, bs->dsk.meta_block_size };
         data->callback = [this, buf, block_offset](ring_data_t *data)

src/client/cluster_client.cpp

@@ -590,7 +590,7 @@ void cluster_client_t::on_change_pool_config_hook()
         {
             if (log_level > 2 && pg_counts[pool_item.first])
             {
-                printf("Pool %u (%s) PG count changed from %lu to %lu\n", pool_item.first, pool_item.second.name.c_str(),
+                fprintf(stderr, "Pool %u (%s) PG count changed from %lu to %lu\n", pool_item.first, pool_item.second.name.c_str(),
                     pg_counts[pool_item.first], pool_item.second.real_pg_count);
             }
             // At this point, all pool operations should have been suspended

src/client/msgr_rdma.cpp

@@ -187,6 +187,8 @@ std::vector<msgr_rdma_context_t*> msgr_rdma_context_t::create_all(const std::vec
     ibv_device **raw_dev_list = NULL;
     ibv_device **dev_list = NULL;
     ibv_device *single_list[2] = {};
+    int up_ports = 0;
+    int single_port_num = 0;
 
     raw_dev_list = dev_list = ibv_get_device_list(NULL);
     if (!dev_list || !*dev_list)
@@ -221,6 +223,7 @@ std::vector<msgr_rdma_context_t*> msgr_rdma_context_t::create_all(const std::vec
         dev_list = single_list;
     }
 
+retry:
     for (int i = 0; dev_list[i]; ++i)
     {
         auto dev = dev_list[i];
@@ -258,6 +261,9 @@ std::vector<msgr_rdma_context_t*> msgr_rdma_context_t::create_all(const std::vec
                 fprintf(stderr, "RDMA device %s port %d GID %d does not exist\n", ibv_get_device_name(dev), port_num, sel_gid_index);
                 continue;
             }
+            up_ports++;
+            single_port_num = port_num;
+            single_list[0] = dev;
             uint32_t port_mtu = sel_mtu ? sel_mtu : ibv_mtu_to_bytes(portinfo.active_mtu);
 #ifdef IBV_ADVISE_MR_ADVICE_PREFETCH_NO_FAULT
             if (sel_gid_index < 0)
@@ -298,6 +304,14 @@ cleanup_dev:
         ibv_close_device(context);
     }
 
+    if (!ret.size() && up_ports == 1 && dev_list != single_list)
+    {
+        // Auto-select the only available device/port if there is only one
+        dev_list = single_list;
+        sel_port_num = single_port_num;
+        goto retry;
+    }
+
 cleanup:
     if (raw_dev_list)
         ibv_free_device_list(raw_dev_list);

src/osd/osd_peering_pg_test.cpp

@@ -1,8 +1,6 @@
 // Copyright (c) Vitaliy Filippov, 2019+
 // License: VNPL-1.1 (see README.md for details)
 
-#define _LARGEFILE64_SOURCE
-
 #include "malloc_or_die.h"
 #include "osd_peering_pg.h"
 #define STRIPE_SHIFT 12
@@ -22,13 +20,20 @@
  */
 int main(int argc, char *argv[])
 {
+    setvbuf(stdout, NULL, _IONBF, 0);
+    setvbuf(stderr, NULL, _IONBF, 0);
     pg_t pg = {
         .state = PG_PEERING,
+        .scheme = POOL_SCHEME_EC,
+        .pg_size = 3,
+        .pg_minsize = 2,
+        .pg_data_size = 2,
         .pg_num = 1,
         .target_set = { 1, 2, 3 },
         .cur_set = { 1, 2, 3 },
         .peering_state = new pg_peering_state_t(),
     };
+    printf("generating\n");
     for (uint64_t osd_num = 1; osd_num <= 3; osd_num++)
     {
         pg_list_result_t r = {
@@ -48,8 +53,15 @@ int main(int argc, char *argv[])
         }
         pg.peering_state->list_results[osd_num] = r;
     }
+    printf("calculating states\n");
+    timespec tv0;
+    clock_gettime(CLOCK_REALTIME, &tv0);
     pg.calc_object_states(0);
-    printf("deviation variants=%jd clean=%ju\n", pg.state_dict.size(), pg.clean_count);
+    timespec tv1;
+    clock_gettime(CLOCK_REALTIME, &tv1);
+    printf("%.2f deviation variants=%jd clean=%ju\n",
+        (tv1.tv_sec - tv0.tv_sec + tv1.tv_nsec/1000000000.0 - tv0.tv_nsec/1000000000.0),
+        pg.state_dict.size(), pg.clean_count);
     for (auto it: pg.state_dict)
     {
         printf("dev: state=%jx\n", it.second.state);

src/test/test_heap.cpp

@@ -1893,7 +1893,8 @@ void test_big_intent_csums()
         heap.finish_recheck();
 
         auto mod = heap.get_recheck_modified_blocks();
-        assert(mod.size() == 0);
+        assert(mod.size() == 1);
+        assert(mod[0] == 0);
 
         // read object 1 - big_intent should be there
         object_id oid = { .inode = INODE_WITH_POOL(1, 1), .stripe = 0 };