Fixed some potential buffer-related concurrency bugs

1 files changed, 63 insertions, 43 deletions

diff --git a/include/framework/structure/MutableBuffer.h b/include/framework/structure/MutableBuffer.h
index a065154..3a06f0d 100644
--- a/include/framework/structure/MutableBuffer.h
+++ b/include/framework/structure/MutableBuffer.h
@@ -6,12 +6,22 @@
  *
  * Distributed under the Modified BSD License.
  *
+ * NOTE: Concerning the tombstone count. One possible approach
+ * would be to track the number of tombstones below and above the
+ * low water mark--this would be straightforward to do. Then, if we
+ * *require* that the head only advance up to the LWM, we can get a
+ * correct view on the number of tombstones in the active buffer at
+ * any point in time, and the BufferView will have a pretty good
+ * approximation as well (potentially with a few extra if new inserts
+ * happen between when the tail pointer and tombstone count are fetched)
+ *
  */
 #pragma once
 
 #include <cstdlib>
 #include <atomic>
 #include <cassert>
+#include <immintrin.h>
 
 #include "psu-util/alignment.h"
 #include "util/bf_config.h"
@@ -28,20 +38,22 @@ class MutableBuffer {
     friend class BufferView<R>;
 public:
     MutableBuffer(size_t low_watermark, size_t high_watermark, size_t capacity=0) 
-    : m_lwm(low_watermark), m_hwm(high_watermark), m_cap(capacity), m_head(0), m_tail(0),
-    m_old_head(0), m_head_refcnt(0), m_old_head_refcnt(0) {
-        /* 
-         * default capacity is twice the high water mark, to account for the worst-case
-         * memory requirements.
-         */
-        if (m_cap == 0) {
-            m_cap = m_hwm * 2;
-        }
-
-        m_data = (Wrapped<R> *) psudb::sf_aligned_alloc(CACHELINE_SIZE, m_cap * sizeof(Wrapped<R>));
-        
-        // FIXME: need to figure out how to detail with tombstones at some point...
-        m_tombstone_filter = new psudb::BloomFilter<R>(BF_FPR, m_hwm, BF_HASH_FUNCS);
+        : m_lwm(low_watermark)
+        , m_hwm(high_watermark)
+        , m_cap((capacity == 0) ? 2 * high_watermark : capacity)
+        , m_tail(0)
+        , m_head(0)
+        , m_head_refcnt(0)
+        , m_old_head(0)
+        , m_old_head_refcnt(0)
+        , m_data((Wrapped<R> *) psudb::sf_aligned_alloc(CACHELINE_SIZE, m_cap * sizeof(Wrapped<R>)))
+        , m_tombstone_filter(new psudb::BloomFilter<R>(BF_FPR, m_hwm, BF_HASH_FUNCS))
+        , m_tscnt(0)
+        , m_old_tscnt(0)
+        , m_active_head_advance(false) 
+    {
+        assert(m_cap > m_hwm);
+        assert(m_hwm > m_lwm);
     }
 
     ~MutableBuffer() {
@@ -62,7 +74,7 @@ public:
         m_data[pos].header |= (pos << 2);
 
         if (tombstone) {
-            m_tombstonecnt.fetch_add(1);
+            m_tscnt.fetch_add(1);
             if (m_tombstone_filter) m_tombstone_filter->insert(rec);
         }
 
@@ -70,7 +82,7 @@ public:
     }
 
     bool truncate() {
-        m_tombstonecnt.store(0);
+        m_tscnt.store(0);
         m_tail.store(0);
         if (m_tombstone_filter) m_tombstone_filter->clear();
 
@@ -78,7 +90,7 @@ public:
     }
 
     size_t get_record_count() {
-        return (m_tail - m_head) % m_cap;
+        return m_tail - m_head;
     }
     
     size_t get_capacity() {
@@ -94,30 +106,15 @@ public:
     }
 
     size_t get_tombstone_count() {
-        return m_tombstonecnt.load();
+        return m_tscnt.load();
     }
 
     bool delete_record(const R& rec) {
-        for (size_t i=0; i<get_record_count(); i++) {
-            if (m_data[to_idx(i)].rec == rec) {
-                m_data[to_idx(i)].set_delete();
-                return true;
-            }
-        }
-
-        return false;
-    }
+        return get_buffer_view().delete_record(rec);
+   }
 
     bool check_tombstone(const R& rec) {
-        if (m_tombstone_filter && !m_tombstone_filter->lookup(rec)) return false;
-
-        for (size_t i=0; i<get_record_count(); i++) {
-            if (m_data[to_idx(i)].rec == rec && m_data[to_idx(i)].is_tombstone()) {
-                return true;
-            }
-        }
-
-        return false;
+        return get_buffer_view().check_tombstone(rec);
     }
 
     size_t get_memory_usage() {
@@ -130,7 +127,8 @@ public:
 
     BufferView<R> get_buffer_view() {
         m_head_refcnt.fetch_add(1);
-        return BufferView(m_data, m_head, m_tail.load(), m_tombstone_filter, (void*) this, release_head_reference);
+        auto f = std::bind(release_head_reference, (void *) this, m_head.load());
+        return BufferView<R>(m_data, m_cap, m_head.load(), m_tail.load(), m_tscnt.load(), m_tombstone_filter, f);
     }
 
     /*
@@ -147,6 +145,8 @@ public:
             return false;
         }
 
+        m_active_head_advance.store(true);
+
         /*
          * the order here is very important. We first store zero to the 
          * old_refcnt (should be zero anyway). Then we move the current 
@@ -157,12 +157,14 @@ public:
          * dropped. Only after this do we change to the new head
          */
         m_old_head_refcnt.store(0);
+
         m_old_head.store(m_head.load());
         m_old_head_refcnt.fetch_add(m_head_refcnt);
 
         m_head_refcnt.store(0);
         m_head.store(new_head);
 
+        m_active_head_advance.store(false);
         return true;
     }
 
@@ -212,29 +214,44 @@ private:
             if (m_tail.load() >= m_hwm) {
                 return -1;
             }
+
+            _mm_pause();
         }
 
         return old_value;
     }
 
-    size_t to_idx(size_t i) {
-        return (m_head + i) % m_cap;
+    size_t to_idx(size_t i, size_t head) {
+        return (head + i) % m_cap;
     }
 
     static void release_head_reference(void *buff, size_t head) {
         MutableBuffer<R> *buffer = (MutableBuffer<R> *) buff;
 
-        if (head == buffer->m_head.load()) {
-            buffer->m_head_refcnt.fetch_sub(1);
-        } else if (head == buffer->m_old_head.load()) {
+        /* 
+         * check old head first. During a head transition, the head being 
+         * retired will first be assigned to *both* head and old_head. As
+         * a result, any refcnt updates during this time should be applied
+         * to old_head, even if the current head and the head being released
+         * also match.
+         */
+        if (head == buffer->m_old_head.load()) {
             buffer->m_old_head_refcnt.fetch_sub(1);
             /* 
              * if the old head refcnt drops to 0, free
              * the records by setting old_head = head
+             * before this, spin while the two heads are equal to
+             * avoid 
              */
+            while (buffer->m_active_head_advance.load()) {
+                _mm_pause();
+            }
+
             if (buffer->m_old_head_refcnt.load() == 0) {
                 buffer->m_old_head.store(buffer->m_head);   
             }
+        } else if (head == buffer->m_head.load()) {
+            buffer->m_head_refcnt.fetch_sub(1);
         }
     }
 
@@ -252,7 +269,10 @@ private:
     
     Wrapped<R>* m_data;
     psudb::BloomFilter<R>* m_tombstone_filter;
-    alignas(64) std::atomic<size_t> m_tombstonecnt;
+    alignas(64) std::atomic<size_t> m_tscnt;
+    size_t m_old_tscnt;
+
+    alignas(64) std::atomic<bool> m_active_head_advance;
 };
 
 }