From d223e3abb87035eef0fc5a97aa2f8710a7e48510 Mon Sep 17 00:00:00 2001
From: mutouyun <mark.lonr@tom.com>
Date: Mon, 25 Mar 2019 16:25:14 +0800
Subject: [PATCH] fix bugs; modify shm interfaces

---
 include/def.h              |  3 +-
 include/shm.h              |  7 ++--
 src/circ/elem_array.h      | 10 +++---
 src/ipc.cpp                | 68 +++++++++++++++++++++-----------------
 src/platform/shm_linux.cpp | 61 +++++++++++++++++-----------------
 src/platform/shm_win.cpp   | 46 ++++++++------------------
 src/policy.h               |  7 ++--
 src/prod_cons.h            | 59 +++++++++++++++++----------------
 src/queue.h                | 21 ++++++------
 src/shm.cpp                | 15 +++++----
 test/test_circ.cpp         |  4 +--
 test/test_shm.cpp          |  2 ++
 test/test_waiter.cpp       |  2 ++
 13 files changed, 151 insertions(+), 154 deletions(-)

diff --git a/include/def.h b/include/def.h
index fd3a608..6764d31 100644
--- a/include/def.h
+++ b/include/def.h
@@ -28,7 +28,8 @@ using uint_t = typename uint<N>::type;
 
 enum : std::size_t {
     invalid_value = (std::numeric_limits<std::size_t>::max)(),
-    data_length   = 64
+    data_length   = 64,
+    name_length   = 64
 };
 
 enum class relat { // multiplicity of the relationship
diff --git a/include/shm.h b/include/shm.h
index 02f0c0c..163d3bd 100644
--- a/include/shm.h
+++ b/include/shm.h
@@ -7,8 +7,11 @@
 namespace ipc {
 namespace shm {
 
-IPC_EXPORT void* acquire(char const * name, std::size_t size);
-IPC_EXPORT void  release(void* mem, std::size_t size);
+using id_t = void*;
+
+IPC_EXPORT id_t   acquire(char const * name, std::size_t size);
+IPC_EXPORT void * to_mem (id_t id);
+IPC_EXPORT void   release(id_t id, void * mem, std::size_t size);
 
 class IPC_EXPORT handle {
 public:
diff --git a/src/circ/elem_array.h b/src/circ/elem_array.h
index 33c8fab..b5ef71f 100644
--- a/src/circ/elem_array.h
+++ b/src/circ/elem_array.h
@@ -13,12 +13,12 @@ namespace ipc {
 namespace circ {
 namespace detail {
 
-template <typename Policy, std::size_t DataSize>
+template <typename Policy, std::size_t DataSize, std::size_t AlignSize>
 class elem_array {
 public:
     using policy_t = Policy;
     using cursor_t = decltype(std::declval<policy_t>().cursor());
-    using elem_t   = typename policy_t::template elem_t<DataSize>;
+    using elem_t   = typename policy_t::template elem_t<DataSize, AlignSize>;
 
     enum : std::size_t {
         data_size  = DataSize,
@@ -29,7 +29,7 @@ public:
 
 private:
     policy_t head_;
-    elem_t   block_[elem_max];
+    elem_t block_[elem_max];
 
 public:
     cursor_t cursor() const noexcept {
@@ -50,11 +50,11 @@ public:
 
 } // namespace detail
 
-template <typename Policy, std::size_t DataSize>
+template <typename Policy, std::size_t DataSize, std::size_t AlignSize>
 class elem_array : public ipc::circ::conn_head {
 public:
     using base_t   = ipc::circ::conn_head;
-    using array_t  = detail::elem_array<Policy, DataSize>;
+    using array_t  = detail::elem_array<Policy, DataSize, AlignSize>;
     using policy_t = typename array_t::policy_t;
     using cursor_t = typename array_t::cursor_t;
     using elem_t   = typename array_t::elem_t;
diff --git a/src/ipc.cpp b/src/ipc.cpp
index 8d85411..8c9e9bf 100644
--- a/src/ipc.cpp
+++ b/src/ipc.cpp
@@ -5,6 +5,7 @@
 #include <algorithm>
 #include <utility>
 #include <atomic>
+#include <type_traits>
 
 #include "def.h"
 #include "shm.h"
@@ -38,8 +39,16 @@ struct msg_t<0, AlignSize> {
 
 template <std::size_t DataSize, std::size_t AlignSize>
 struct msg_t {
-    msg_t<0, AlignSize> head_;
-    alignas(AlignSize) byte_t data_[DataSize];
+    msg_t<0, AlignSize> head_ {};
+    std::aligned_storage_t<DataSize, AlignSize> data_ {};
+
+    msg_t() = default;
+    msg_t(void* q, msg_id_t i, int r, void const * d, std::size_t s) {
+        head_.que_    = q;
+        head_.id_     = i;
+        head_.remain_ = r;
+        std::memcpy(&data_, d, s);
+    }
 };
 
 buff_t make_cache(void const * data, std::size_t size) {
@@ -62,20 +71,6 @@ struct cache_t {
     }
 };
 
-auto& recv_cache() {
-    /*
-        <Remarks> thread_local may have some bugs.
-        See: https://sourceforge.net/p/mingw-w64/bugs/727/
-             https://sourceforge.net/p/mingw-w64/bugs/527/
-             https://github.com/Alexpux/MINGW-packages/issues/2519
-             https://github.com/ChaiScript/ChaiScript/issues/402
-             https://developercommunity.visualstudio.com/content/problem/124121/thread-local-variables-fail-to-be-initialized-when.html
-             https://software.intel.com/en-us/forums/intel-c-compiler/topic/684827
-    */
-    static tls::pointer<mem::unordered_map<msg_id_t, cache_t>> rc;
-    return *rc.create();
-}
-
 template <typename Policy>
 struct detail_impl {
 
@@ -94,6 +89,20 @@ static auto& queues_cache() {
     return *qc.create();
 }
 
+static auto& recv_cache() {
+    /*
+        <Remarks> thread_local may have some bugs.
+        See: https://sourceforge.net/p/mingw-w64/bugs/727/
+             https://sourceforge.net/p/mingw-w64/bugs/527/
+             https://github.com/Alexpux/MINGW-packages/issues/2519
+             https://github.com/ChaiScript/ChaiScript/issues/402
+             https://developercommunity.visualstudio.com/content/problem/124121/thread-local-variables-fail-to-be-initialized-when.html
+             https://software.intel.com/en-us/forums/intel-c-compiler/topic/684827
+    */
+    static tls::pointer<mem::unordered_map<msg_id_t, cache_t>> rc;
+    return *rc.create();
+}
+
 /* API implementations */
 
 static ipc::handle_t connect(char const * name) {
@@ -141,21 +150,18 @@ static bool send(ipc::handle_t h, void const * data, std::size_t size) {
     // push message fragment
     int offset = 0;
     for (int i = 0; i < static_cast<int>(size / data_length); ++i, offset += data_length) {
-        msg_t<data_length> msg {
-            { que, msg_id, static_cast<int>(size) - offset - static_cast<int>(data_length) }, {}
-        };
-        std::memcpy(msg.data_, static_cast<byte_t const *>(data) + offset, data_length);
-        while (!que->push(msg)) std::this_thread::yield();
+        while (!que->push(que, msg_id, static_cast<int>(size) - offset - static_cast<int>(data_length),
+                          static_cast<byte_t const *>(data) + offset, data_length)) {
+            std::this_thread::yield();
+        }
     }
     // if remain > 0, this is the last message fragment
     int remain = static_cast<int>(size) - offset;
     if (remain > 0) {
-        msg_t<data_length> msg {
-            { que, msg_id, remain - static_cast<int>(data_length) }, {}
-        };
-        std::memcpy(msg.data_, static_cast<byte_t const *>(data) + offset,
-                               static_cast<std::size_t>(remain));
-        while (!que->push(msg)) std::this_thread::yield();
+        while (!que->push(que, msg_id, remain - static_cast<int>(data_length),
+                          static_cast<byte_t const *>(data) + offset, static_cast<std::size_t>(remain))) {
+            std::this_thread::yield();
+        }
     }
     return true;
 }
@@ -177,24 +183,24 @@ static buff_t recv(ipc::handle_t h) {
         auto cac_it = rc.find(msg.head_.id_);
         if (cac_it == rc.end()) {
             if (remain <= data_length) {
-                return make_cache(msg.data_, remain);
+                return make_cache(&(msg.data_), remain);
             }
             // cache the first message fragment
-            else rc.emplace(msg.head_.id_, cache_t { data_length, make_cache(msg.data_, remain) });
+            else rc.emplace(msg.head_.id_, cache_t { data_length, make_cache(&(msg.data_), remain) });
         }
         // has cached before this message
         else {
             auto& cac = cac_it->second;
             // this is the last message fragment
             if (msg.head_.remain_ <= 0) {
-                cac.append(msg.data_, remain);
+                cac.append(&(msg.data_), remain);
                 // finish this message, erase it from cache
                 auto buff = std::move(cac.buff_);
                 rc.erase(cac_it);
                 return buff;
             }
             // there are remain datas after this message
-            cac.append(msg.data_, data_length);
+            cac.append(&(msg.data_), data_length);
         }
     }
 }
diff --git a/src/platform/shm_linux.cpp b/src/platform/shm_linux.cpp
index 8f9f25d..a291da1 100644
--- a/src/platform/shm_linux.cpp
+++ b/src/platform/shm_linux.cpp
@@ -10,7 +10,7 @@
 #include <atomic>
 #include <string>
 #include <utility>
-#include <mutex>
+#include <cstring>
 
 #include "def.h"
 #include "log.h"
@@ -18,22 +18,21 @@
 
 namespace {
 
-using acc_t = std::atomic_size_t;
+struct info_t {
+    std::atomic_size_t acc_;
+    char name_[ipc::name_length];
+};
 
-constexpr acc_t* acc_of(void* mem) {
-    return static_cast<acc_t*>(mem);
+constexpr std::size_t calc_size(std::size_t size) {
+    return ((((size - 1) / alignof(info_t)) + 1) * alignof(info_t)) + sizeof(info_t);
 }
 
-constexpr void* mem_of(void* mem) {
-    return static_cast<acc_t*>(mem) - 1;
+inline auto& acc_of(void* mem, std::size_t size) {
+    return reinterpret_cast<info_t*>(static_cast<ipc::byte_t*>(mem) + size - sizeof(info_t))->acc_;
 }
 
-inline auto* m2h() {
-    static struct {
-        std::mutex lc_;
-        ipc::mem::unordered_map<void*, std::string> cache_;
-    } m2h_;
-    return &m2h_;
+inline auto& str_of(void* mem, std::size_t size) {
+    return reinterpret_cast<info_t*>(static_cast<ipc::byte_t*>(mem) + size - sizeof(info_t))->name_;
 }
 
 } // internal-linkage
@@ -41,11 +40,15 @@ inline auto* m2h() {
 namespace ipc {
 namespace shm {
 
-void* acquire(char const * name, std::size_t size) {
+id_t acquire(char const * name, std::size_t size) {
     if (name == nullptr || name[0] == '\0' || size == 0) {
         return nullptr;
     }
     std::string op_name = std::string{"__IPC_SHM__"} + name;
+    if (op_name.size() >= ipc::name_length) {
+        ipc::error("name is too long!: [%d]%s\n", static_cast<int>(op_name.size()), op_name.c_str());
+        return nullptr;
+    }
     int fd = ::shm_open(op_name.c_str(), O_CREAT | O_RDWR,
                                          S_IRUSR | S_IWUSR |
                                          S_IRGRP | S_IWGRP |
@@ -54,7 +57,7 @@ void* acquire(char const * name, std::size_t size) {
         ipc::error("fail shm_open[%d]: %s\n", errno, name);
         return nullptr;
     }
-    size += sizeof(acc_t);
+    size = calc_size(size);
     if (::ftruncate(fd, static_cast<off_t>(size)) != 0) {
         ipc::error("fail ftruncate[%d]: %s\n", errno, name);
         ::close(fd);
@@ -68,33 +71,31 @@ void* acquire(char const * name, std::size_t size) {
         ::shm_unlink(op_name.c_str());
         return nullptr;
     }
-    auto acc = acc_of(mem);
-    acc->fetch_add(1, std::memory_order_release);
-    {
-        IPC_UNUSED_ auto guard = ipc::detail::unique_lock(m2h()->lc_);
-        m2h()->cache_.emplace(++acc, std::move(op_name));
+    if (acc_of(mem, size).fetch_add(1, std::memory_order_release) == 0) {
+        std::memcpy(&str_of(mem, size), op_name.c_str(), op_name.size());
     }
-    return acc;
+    return static_cast<id_t>(mem);
 }
 
-void release(void* mem, std::size_t size) {
+void * to_mem(id_t id) {
+    return static_cast<void *>(id);
+}
+
+void release(id_t id, void * mem, std::size_t size) {
     if (mem == nullptr) {
         return;
     }
-    IPC_UNUSED_ auto guard = ipc::detail::unique_lock(m2h()->lc_);
-    auto& cc = m2h()->cache_;
-    auto it = cc.find(mem);
-    if (it == cc.end()) {
+    if (mem != to_mem(id)) {
         return;
     }
-    mem = mem_of(mem);
-    size += sizeof(acc_t);
-    if (acc_of(mem)->fetch_sub(1, std::memory_order_acquire) == 1) {
+    size = calc_size(size);
+    if (acc_of(mem, size).fetch_sub(1, std::memory_order_acquire) == 1) {
+        char name[ipc::name_length] = {};
+        std::memcpy(name, &str_of(mem, size), sizeof(name));
         ::munmap(mem, size);
-        ::shm_unlink(it->second.c_str());
+        ::shm_unlink(name);
     }
     else ::munmap(mem, size);
-    cc.erase(it);
 }
 
 } // namespace shm
diff --git a/src/platform/shm_win.cpp b/src/platform/shm_win.cpp
index b2d6815..0d24bf4 100644
--- a/src/platform/shm_win.cpp
+++ b/src/platform/shm_win.cpp
@@ -4,29 +4,16 @@
 
 #include <string>
 #include <utility>
-#include <mutex>
 
 #include "def.h"
 #include "log.h"
 #include "memory/resource.h"
 #include "platform/to_tchar.h"
 
-namespace {
-
-inline auto* m2h() {
-    static struct {
-        std::mutex lc_;
-        ipc::mem::unordered_map<void*, HANDLE> cache_;
-    } m2h_;
-    return &m2h_;
-}
-
-} // internal-linkage
-
 namespace ipc {
 namespace shm {
 
-void* acquire(char const * name, std::size_t size) {
+id_t acquire(char const * name, std::size_t size) {
     if (name == nullptr || name[0] == '\0' || size == 0) {
         return nullptr;
     }
@@ -38,32 +25,25 @@ void* acquire(char const * name, std::size_t size) {
         ipc::error("fail CreateFileMapping[%d]: %s\n", static_cast<int>(::GetLastError()), name);
         return nullptr;
     }
+    return static_cast<id_t>(h);
+}
+
+void * to_mem(id_t id) {
+    if (id == nullptr) return nullptr;
+    HANDLE h = static_cast<HANDLE>(id);
     LPVOID mem = ::MapViewOfFile(h, FILE_MAP_ALL_ACCESS, 0, 0, 0);
     if (mem == NULL) {
         ipc::error("fail MapViewOfFile[%d]: %s\n", static_cast<int>(::GetLastError()), name);
-        ::CloseHandle(h);
         return nullptr;
     }
-    {
-        IPC_UNUSED_ auto guard = ipc::detail::unique_lock(m2h()->lc_);
-        m2h()->cache_.emplace(mem, h);
-    }
-    return mem;
+    return static_cast<void *>(mem);
 }
 
-void release(void* mem, std::size_t /*size*/) {
-    if (mem == nullptr) {
-        return;
-    }
-    IPC_UNUSED_ auto guard = ipc::detail::unique_lock(m2h()->lc_);
-    auto& cc = m2h()->cache_;
-    auto it = cc.find(mem);
-    if (it == cc.end()) {
-        return;
-    }
-    ::UnmapViewOfFile(mem);
-    ::CloseHandle(it->second);
-    cc.erase(it);
+void release(id_t id, void * mem, std::size_t /*size*/) {
+    if (id  == nullptr) return;
+    if (mem == nullptr) return;
+    ::UnmapViewOfFile(static_cast<LPVOID>(mem));
+    ::CloseHandle(static_cast<HANDLE>(id));
 }
 
 } // namespace shm
diff --git a/src/policy.h b/src/policy.h
index 79da3ae..3cc5006 100644
--- a/src/policy.h
+++ b/src/policy.h
@@ -10,14 +10,13 @@
 namespace ipc {
 namespace policy {
 
-template <template <typename, std::size_t...> class Elems,
-          typename Flag>
+template <template <typename, std::size_t...> class Elems, typename Flag>
 struct choose;
 
 template <typename Flag>
 struct choose<circ::elem_array, Flag> {
-    template <std::size_t DataSize>
-    using elems_t = circ::elem_array<ipc::prod_cons_impl<Flag>, DataSize>;
+    template <std::size_t DataSize, std::size_t AlignSize>
+    using elems_t = circ::elem_array<ipc::prod_cons_impl<Flag>, DataSize, AlignSize>;
 };
 
 } // namespace policy
diff --git a/src/prod_cons.h b/src/prod_cons.h
index 5e95713..ff751e1 100644
--- a/src/prod_cons.h
+++ b/src/prod_cons.h
@@ -3,6 +3,7 @@
 #include <atomic>
 #include <utility>
 #include <cstring>
+#include <type_traits>
 
 #include "def.h"
 
@@ -20,9 +21,9 @@ struct prod_cons_impl;
 template <>
 struct prod_cons_impl<wr<relat::single, relat::single, trans::unicast>> {
 
-    template <std::size_t DataSize>
+    template <std::size_t DataSize, std::size_t AlignSize>
     struct elem_t {
-        byte_t data_[DataSize] {};
+        std::aligned_storage_t<DataSize, AlignSize> data_ {};
     };
 
     alignas(circ::cache_line_size) std::atomic<circ::u2_t> rd_; // read index
@@ -32,8 +33,8 @@ struct prod_cons_impl<wr<relat::single, relat::single, trans::unicast>> {
         return 0;
     }
 
-    template <typename W, typename F, template <std::size_t> class E, std::size_t DataSize>
-    bool push(W* /*wrapper*/, F&& f, E<DataSize>* elems) {
+    template <typename W, typename F, typename E>
+    bool push(W* /*wrapper*/, F&& f, E* elems) {
         auto cur_wt = circ::index_of(wt_.load(std::memory_order_relaxed));
         if (cur_wt == circ::index_of(rd_.load(std::memory_order_acquire) - 1)) {
             return false; // full
@@ -43,8 +44,8 @@ struct prod_cons_impl<wr<relat::single, relat::single, trans::unicast>> {
         return true;
     }
 
-    template <typename W, typename F, template <std::size_t> class E, std::size_t DataSize>
-    bool pop(W* /*wrapper*/, circ::u2_t& /*cur*/, F&& f, E<DataSize>* elems) {
+    template <typename W, typename F, typename E>
+    bool pop(W* /*wrapper*/, circ::u2_t& /*cur*/, F&& f, E* elems) {
         auto cur_rd = circ::index_of(rd_.load(std::memory_order_relaxed));
         if (cur_rd == circ::index_of(wt_.load(std::memory_order_acquire))) {
             return false; // empty
@@ -59,9 +60,9 @@ template <>
 struct prod_cons_impl<wr<relat::single, relat::multi , trans::unicast>>
      : prod_cons_impl<wr<relat::single, relat::single, trans::unicast>> {
 
-    template <typename W, typename F, template <std::size_t> class E, std::size_t DataSize>
-    bool pop(W* /*wrapper*/, circ::u2_t& /*cur*/, F&& f, E<DataSize>* elems) {
-        byte_t buff[DataSize];
+    template <typename W, typename F, template <std::size_t, std::size_t> class E, std::size_t DS, std::size_t AS>
+    bool pop(W* /*wrapper*/, circ::u2_t& /*cur*/, F&& f, E<DS, AS>* elems) {
+        byte_t buff[DS];
         for (unsigned k = 0;;) {
             auto cur_rd = rd_.load(std::memory_order_relaxed);
             if (circ::index_of(cur_rd) ==
@@ -84,16 +85,16 @@ struct prod_cons_impl<wr<relat::multi , relat::multi, trans::unicast>>
 
     using flag_t = std::uint64_t;
 
-    template <std::size_t DataSize>
+    template <std::size_t DataSize, std::size_t AlignSize>
     struct elem_t {
-        byte_t data_[DataSize] {};
+        std::aligned_storage_t<DataSize, AlignSize> data_ {};
         std::atomic<flag_t> f_ct_ { 0 }; // commit flag
     };
 
     alignas(circ::cache_line_size) std::atomic<circ::u2_t> ct_; // commit index
 
-    template <typename W, typename F, template <std::size_t> class E, std::size_t DataSize>
-    bool push(W* /*wrapper*/, F&& f, E<DataSize>* elems) {
+    template <typename W, typename F, typename E>
+    bool push(W* /*wrapper*/, F&& f, E* elems) {
         circ::u2_t cur_ct, nxt_ct;
         for (unsigned k = 0;;) {
             cur_ct = ct_.load(std::memory_order_relaxed);
@@ -129,9 +130,9 @@ struct prod_cons_impl<wr<relat::multi , relat::multi, trans::unicast>>
         return true;
     }
 
-    template <typename W, typename F, template <std::size_t> class E, std::size_t DataSize>
-    bool pop(W* /*wrapper*/, circ::u2_t& /*cur*/, F&& f, E<DataSize>* elems) {
-        byte_t buff[DataSize];
+    template <typename W, typename F, template <std::size_t, std::size_t> class E, std::size_t DS, std::size_t AS>
+    bool pop(W* /*wrapper*/, circ::u2_t& /*cur*/, F&& f, E<DS, AS>* elems) {
+        byte_t buff[DS];
         for (unsigned k = 0;;) {
             auto cur_rd = rd_.load(std::memory_order_relaxed);
             auto cur_wt = wt_.load(std::memory_order_acquire);
@@ -165,9 +166,9 @@ struct prod_cons_impl<wr<relat::single, relat::multi, trans::broadcast>> {
 
     using rc_t = std::size_t;
 
-    template <std::size_t DataSize>
+    template <std::size_t DataSize, std::size_t AlignSize>
     struct elem_t {
-        byte_t data_[DataSize] {};
+        std::aligned_storage_t<DataSize, AlignSize> data_ {};
         std::atomic<rc_t> rc_ { 0 }; // read-counter
     };
 
@@ -177,8 +178,8 @@ struct prod_cons_impl<wr<relat::single, relat::multi, trans::broadcast>> {
         return wt_.load(std::memory_order_acquire);
     }
 
-    template <typename W, typename F, template <std::size_t> class E, std::size_t DataSize>
-    bool push(W* wrapper, F&& f, E<DataSize>* elems) {
+    template <typename W, typename F, typename E>
+    bool push(W* wrapper, F&& f, E* elems) {
         auto conn_cnt = wrapper->conn_count(std::memory_order_relaxed);
         if (conn_cnt == 0) return false;
         auto* el = elems + circ::index_of(wt_.load(std::memory_order_acquire));
@@ -193,8 +194,8 @@ struct prod_cons_impl<wr<relat::single, relat::multi, trans::broadcast>> {
         return true;
     }
 
-    template <typename W, typename F, template <std::size_t> class E, std::size_t DataSize>
-    bool pop(W* /*wrapper*/, circ::u2_t& cur, F&& f, E<DataSize>* elems) {
+    template <typename W, typename F, typename E>
+    bool pop(W* /*wrapper*/, circ::u2_t& cur, F&& f, E* elems) {
         if (cur == cursor()) return false; // acquire
         auto* el = elems + circ::index_of(cur++);
         std::forward<F>(f)(&(el->data_));
@@ -223,9 +224,9 @@ struct prod_cons_impl<wr<relat::multi , relat::multi, trans::broadcast>> {
         rc_incr = 0x0000000100000000ull
     };
 
-    template <std::size_t DataSize>
+    template <std::size_t DataSize, std::size_t AlignSize>
     struct elem_t {
-        byte_t data_[DataSize] {};
+        std::aligned_storage_t<DataSize, AlignSize> data_ {};
         std::atomic<rc_t  > rc_   { 0 }; // read-counter
         std::atomic<flag_t> f_ct_ { 0 }; // commit flag
     };
@@ -236,9 +237,9 @@ struct prod_cons_impl<wr<relat::multi , relat::multi, trans::broadcast>> {
         return ct_.load(std::memory_order_acquire);
     }
 
-    template <typename W, typename F, template <std::size_t> class E, std::size_t DataSize>
-    bool push(W* wrapper, F&& f, E<DataSize>* elems) {
-        E<DataSize>* el;
+    template <typename W, typename F, typename E>
+    bool push(W* wrapper, F&& f, E* elems) {
+        E* el;
         circ::u2_t cur_ct, nxt_ct;
         for (unsigned k = 0;;) {
             auto cc = wrapper->conn_count(std::memory_order_relaxed);
@@ -269,8 +270,8 @@ struct prod_cons_impl<wr<relat::multi , relat::multi, trans::broadcast>> {
         return true;
     }
 
-    template <typename W, typename F, template <std::size_t> class E, std::size_t DataSize, std::size_t N>
-    bool pop(W* /*wrapper*/, circ::u2_t& cur, F&& f, E<DataSize>(& elems)[N]) {
+    template <typename W, typename F, typename E, std::size_t N>
+    bool pop(W* /*wrapper*/, circ::u2_t& cur, F&& f, E(& elems)[N]) {
         auto* el = elems + circ::index_of(cur);
         auto cur_fl = el->f_ct_.load(std::memory_order_acquire);
         if (cur_fl != ~static_cast<flag_t>(cur)) {
diff --git a/src/queue.h b/src/queue.h
index 82cb02c..03ce4d0 100644
--- a/src/queue.h
+++ b/src/queue.h
@@ -27,16 +27,18 @@ protected:
     ipc::detail::waiter_wrapper cc_waiter_;
 
     bool connected_ = false;
-    bool dismiss_   = true;
+    shm::handle elems_h_;
 
     template <typename Elems>
     Elems* open(char const * name) {
-        auto elems = static_cast<Elems*>(shm::acquire(name, sizeof(Elems)));
+        if (!elems_h_.acquire(name, sizeof(Elems))) {
+            return nullptr;
+        }
+        auto elems = static_cast<Elems*>(elems_h_.get());
         if (elems == nullptr) {
             return nullptr;
         }
         elems->init();
-        dismiss_ = false;
         return elems;
     }
 
@@ -63,12 +65,9 @@ protected:
     }
 
     template <typename Elems>
-    void close(Elems* elems) {
+    void close(Elems* /*elems*/) {
         close();
-        if (!dismiss_ && (elems != nullptr)) {
-            shm::release(elems, sizeof(Elems));
-        }
-        dismiss_ = true;
+        elems_h_.release();
     }
 
 public:
@@ -131,7 +130,7 @@ public:
     using policy_t = typename elems_t::policy_t;
 
 protected:
-    elems_t* elems_ = nullptr;
+    elems_t * elems_ = nullptr;
     decltype(std::declval<elems_t>().cursor()) cursor_ = 0;
 
 public:
@@ -243,8 +242,8 @@ public:
 } // namespace detail
 
 template <typename T, typename Policy>
-class queue : public detail::queue_base<typename Policy::template elems_t<sizeof(T)>> {
-    using base_t = detail::queue_base<typename Policy::template elems_t<sizeof(T)>>;
+class queue : public detail::queue_base<typename Policy::template elems_t<sizeof(T), alignof(T)>> {
+    using base_t = detail::queue_base<typename Policy::template elems_t<sizeof(T), alignof(T)>>;
 
 public:
     using base_t::base_t;
diff --git a/src/shm.cpp b/src/shm.cpp
index 69e1635..0b5356e 100644
--- a/src/shm.cpp
+++ b/src/shm.cpp
@@ -10,7 +10,8 @@ namespace shm {
 
 class handle::handle_ : public pimpl<handle_> {
 public:
-    void* m_ = nullptr;
+    shm::id_t id_ = nullptr;
+    void*     m_  = nullptr;
 
     std::string n_;
     std::size_t s_ = 0;
@@ -58,16 +59,18 @@ char const * handle::name() const {
 
 bool handle::acquire(char const * name, std::size_t size) {
     release();
-    impl(p_)->m_ = shm::acquire((impl(p_)->n_ = name).c_str(),
-                                 impl(p_)->s_ = size);
+    impl(p_)->id_ = shm::acquire((impl(p_)->n_ = name).c_str(),
+                                  impl(p_)->s_ = size);
+    impl(p_)->m_  = shm::to_mem  (impl(p_)->id_);
     return valid();
 }
 
 void handle::release() {
     if (!valid()) return;
-    shm::release(impl(p_)->m_, impl(p_)->s_);
-    impl(p_)->m_ = nullptr;
-    impl(p_)->s_ = 0;
+    shm::release(impl(p_)->id_, impl(p_)->m_, impl(p_)->s_);
+    impl(p_)->id_ = nullptr;
+    impl(p_)->m_  = nullptr;
+    impl(p_)->s_  = 0;
     impl(p_)->n_.clear();
 }
 
diff --git a/test/test_circ.cpp b/test/test_circ.cpp
index df1a493..e377c8e 100644
--- a/test/test_circ.cpp
+++ b/test/test_circ.cpp
@@ -24,8 +24,8 @@ template <ipc::relat Rp, ipc::relat Rc, ipc::trans Ts>
 using pc_t = ipc::prod_cons_impl<ipc::wr<Rp, Rc, Ts>>;
 
 template <std::size_t DataSize, typename Policy>
-struct ea_t : public ipc::circ::elem_array<Policy, DataSize> {
-    ea_t() { std::memset(this, 0, sizeof(ipc::circ::elem_array<Policy, DataSize>)); }
+struct ea_t : public ipc::circ::elem_array<Policy, DataSize, 1> {
+    ea_t() { std::memset(this, 0, sizeof(ipc::circ::elem_array<Policy, DataSize, 1>)); }
 };
 
 using cq_t = ea_t<
diff --git a/test/test_shm.cpp b/test/test_shm.cpp
index 0535f37..5fa39c9 100644
--- a/test/test_shm.cpp
+++ b/test/test_shm.cpp
@@ -83,6 +83,8 @@ void Unit::test_hello() {
     QVERIFY(shm_hd__.get() == nullptr);
     QVERIFY(shm_hd__.acquire("my-test", 1024));
 
+    mem = shm_hd__.get();
+    QVERIFY(mem != nullptr);
     std::uint8_t buf[1024] = {};
     QVERIFY(memcmp(mem, buf, sizeof(buf)) == 0);
 
diff --git a/test/test_waiter.cpp b/test/test_waiter.cpp
index 5077add..6516100 100644
--- a/test/test_waiter.cpp
+++ b/test/test_waiter.cpp
@@ -28,6 +28,7 @@ void Unit::test_broadcast() {
             ipc::detail::waiter_wrapper wp { &w };
             QVERIFY(wp.open("test-ipc-waiter"));
             QVERIFY(wp.wait_if([] { return true; }));
+            wp.close();
         });
     }
 
@@ -39,6 +40,7 @@ void Unit::test_broadcast() {
     QVERIFY(wp.broadcast());
 
     for (auto& t : ts) t.join();
+    wp.close();
 }
 
 } // internal-linkage