[Qemu-devel] [PATCH 0/7 V7] nvdimm: guarantee persistence of QEMU writes to persistent memory

[Qemu-devel] [PATCH 0/7 V7] nvdimm: guarantee persistence of QEMU writes to persistent memory

[junyan.he]

From: Junyan He <junyan.he@intel.com>

QEMU writes to vNVDIMM backends in the vNVDIMM label emulation and
live migration. If the backend is on the persistent memory, QEMU needs
to take proper operations to ensure its writes persistent on the
persistent memory. Otherwise, a host power failure may result in the
loss the guest data on the persistent memory.

This v3 patch series is based on Marcel's patch "mem: add share
parameter to memory-backend-ram" [1] because of the changes in patch 1.

[1] https://lists.gnu.org/archive/html/qemu-devel/2018-02/msg03858.html

Previous versions can be found at:
v6: https://lists.gnu.org/archive/html/qemu-devel/2018-06/msg00061.html
v5: https://lists.gnu.org/archive/html/qemu-devel/2018-05/msg02258.html
V4: https://lists.gnu.org/archive/html/qemu-devel/2018-02/msg06993.html
v3: https://lists.gnu.org/archive/html/qemu-devel/2018-02/msg04365.html
v2: https://lists.gnu.org/archive/html/qemu-devel/2018-02/msg01579.html
v1: https://lists.gnu.org/archive/html/qemu-devel/2017-12/msg05040.html

Changes in v7:
The v6 patch set has already reviewed by Stefan Hajnoczi <stefanha@redhat.com>
No logic change in this v7 version, just:
* Spelling check and some document words refined.
* Rebase to "ram is migratable" patch set.

Changes in v6:
* (Patch 1) Expose all ram block flags rather than redefine the flags.
* (Patch 4) Use pkg-config rather the hard check when configure. 
* (Patch 7) Sync and flush all the pmem data when migration completes,
rather than sync pages one by one in previous version.

Changes in v5:
* (Patch 9) Add post copy check and output some messages for nvdimm.

Changes in v4:
* (Patch 2) Fix compilation errors found by patchew.

Changes in v3:
* (Patch 5) Add a is_pmem flag to ram_handle_compressed() and handle
PMEM writes in it, so we don't need the _common function.
* (Patch 6) Expose qemu_get_buffer_common so we can remove the
unnecessary qemu_get_buffer_to_pmem wrapper.
* (Patch 8) Add a is_pmem flag to xbzrle_decode_buffer() and handle
PMEM writes in it, so we can remove the unnecessary
xbzrle_decode_buffer_{common, to_pmem}.
* Move libpmem stubs to stubs/pmem.c and fix the compilation failures
of test-{xbzrle,vmstate}.c.

Changes in v2:
* (Patch 1) Use a flags parameter in file ram allocation functions.
* (Patch 2) Add a new option 'pmem' to hostmem-file.
* (Patch 3) Use libpmem to operate on the persistent memory, rather
than re-implementing those operations in QEMU.
* (Patch 5-8) Consider the write persistence in the migration path.

Junyan:
[1/7] memory, exec: Expose all memory block related flags.
[6/7] migration/ram: Add check and info message to nvdimm post copy.
[7/7] migration/ram: ensure write persistence on loading all date to PMEM.

Haozhong:
[5/7] mem/nvdimm: ensure write persistence to PMEM in label emulation

Haozhong & Junyan:
[2/7] memory, exec: switch file ram allocation functions to 'flags' parameters
[3/7] hostmem-file: add the 'pmem' option
[4/7] configure: add libpmem support


Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>

---
backends/hostmem-file.c | 28 +++++++++++++++++++++++++++-
configure               | 29 +++++++++++++++++++++++++++++
docs/nvdimm.txt         | 18 ++++++++++++++++++
exec.c                  | 39 ++++++++++++++-------------------------
hw/mem/nvdimm.c         |  9 ++++++++-
include/exec/memory.h   | 31 +++++++++++++++++++++++++++++--
include/exec/ram_addr.h | 28 ++++++++++++++++++++++++++--
include/qemu/pmem.h     | 24 ++++++++++++++++++++++++
memory.c                |  8 +++++---
migration/ram.c         | 17 +++++++++++++++++
numa.c                  |  2 +-
qemu-options.hx         |  7 +++++++
stubs/Makefile.objs     |  1 +
stubs/pmem.c            | 23 +++++++++++++++++++++++
14 files changed, 229 insertions(+), 35 deletions(-)
--
2.7.4

[Qemu-devel] [PATCH 1/7 V7] memory, exec: Expose all memory block related flags.

[junyan.he]

From: Junyan He <junyan.he@intel.com>

We need to use these flags in other files rather than just in exec.c,
For example, RAM_SHARED should be used when create a ram block from file.
We expose them the exec/memory.h

Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 exec.c                | 20 --------------------
 include/exec/memory.h | 20 ++++++++++++++++++++
 2 files changed, 20 insertions(+), 20 deletions(-)

diff --git a/exec.c b/exec.c
index f6645ed..9246722 100644
--- a/exec.c
+++ b/exec.c
@@ -87,26 +87,6 @@ AddressSpace address_space_memory;
 
 MemoryRegion io_mem_rom, io_mem_notdirty;
 static MemoryRegion io_mem_unassigned;
-
-/* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */
-#define RAM_PREALLOC   (1 << 0)
-
-/* RAM is mmap-ed with MAP_SHARED */
-#define RAM_SHARED     (1 << 1)
-
-/* Only a portion of RAM (used_length) is actually used, and migrated.
- * This used_length size can change across reboots.
- */
-#define RAM_RESIZEABLE (1 << 2)
-
-/* UFFDIO_ZEROPAGE is available on this RAMBlock to atomically
- * zero the page and wake waiting processes.
- * (Set during postcopy)
- */
-#define RAM_UF_ZEROPAGE (1 << 3)
-
-/* RAM can be migrated */
-#define RAM_MIGRATABLE (1 << 4)
 #endif
 
 #ifdef TARGET_PAGE_BITS_VARY
diff --git a/include/exec/memory.h b/include/exec/memory.h
index eb2ba06..1bb9172 100644
--- a/include/exec/memory.h
+++ b/include/exec/memory.h
@@ -102,6 +102,26 @@ struct IOMMUNotifier {
 };
 typedef struct IOMMUNotifier IOMMUNotifier;
 
+/* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */
+#define RAM_PREALLOC   (1 << 0)
+
+/* RAM is mmap-ed with MAP_SHARED */
+#define RAM_SHARED     (1 << 1)
+
+/* Only a portion of RAM (used_length) is actually used, and migrated.
+ * This used_length size can change across reboots.
+ */
+#define RAM_RESIZEABLE (1 << 2)
+
+/* UFFDIO_ZEROPAGE is available on this RAMBlock to atomically
+ * zero the page and wake waiting processes.
+ * (Set during postcopy)
+ */
+#define RAM_UF_ZEROPAGE (1 << 3)
+
+/* RAM can be migrated */
+#define RAM_MIGRATABLE (1 << 4)
+
 static inline void iommu_notifier_init(IOMMUNotifier *n, IOMMUNotify fn,
                                        IOMMUNotifierFlag flags,
                                        hwaddr start, hwaddr end)
-- 
2.7.4

[Qemu-devel] [PATCH 2/7 V7] memory, exec: switch file ram allocation functions to 'flags' parameters

[junyan.he]

From: Junyan He <junyan.he@intel.com>

As more flag parameters besides the existing 'share' are going to be
added to following functions
  memory_region_init_ram_from_file
  qemu_ram_alloc_from_fd
  qemu_ram_alloc_from_file
let's switch them to use the 'flags' parameters so as to ease future
flag additions.

The existing 'share' flag is converted to the RAM_SHARED bit in ram_flags,
and other flag bits are ignored by above functions right now.

Signed-off-by: Junyan He <junyan.he@intel.com>
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 backends/hostmem-file.c |  3 ++-
 exec.c                  | 10 +++++-----
 include/exec/memory.h   |  7 +++++--
 include/exec/ram_addr.h | 25 +++++++++++++++++++++++--
 memory.c                |  8 +++++---
 numa.c                  |  2 +-
 6 files changed, 41 insertions(+), 14 deletions(-)

diff --git a/backends/hostmem-file.c b/backends/hostmem-file.c
index 134b08d..34c68bb 100644
--- a/backends/hostmem-file.c
+++ b/backends/hostmem-file.c
@@ -58,7 +58,8 @@ file_backend_memory_alloc(HostMemoryBackend *backend, Error **errp)
         path = object_get_canonical_path(OBJECT(backend));
         memory_region_init_ram_from_file(&backend->mr, OBJECT(backend),
                                  path,
-                                 backend->size, fb->align, backend->share,
+                                 backend->size, fb->align,
+                                 backend->share ? RAM_SHARED : 0,
                                  fb->mem_path, errp);
         g_free(path);
     }
diff --git a/exec.c b/exec.c
index 9246722..8e079df 100644
--- a/exec.c
+++ b/exec.c
@@ -2070,7 +2070,7 @@ static void ram_block_add(RAMBlock *new_block, Error **errp, bool shared)
 
 #ifdef __linux__
 RAMBlock *qemu_ram_alloc_from_fd(ram_addr_t size, MemoryRegion *mr,
-                                 bool share, int fd,
+                                 uint64_t ram_flags, int fd,
                                  Error **errp)
 {
     RAMBlock *new_block;
@@ -2112,14 +2112,14 @@ RAMBlock *qemu_ram_alloc_from_fd(ram_addr_t size, MemoryRegion *mr,
     new_block->mr = mr;
     new_block->used_length = size;
     new_block->max_length = size;
-    new_block->flags = share ? RAM_SHARED : 0;
+    new_block->flags = ram_flags;
     new_block->host = file_ram_alloc(new_block, size, fd, !file_size, errp);
     if (!new_block->host) {
         g_free(new_block);
         return NULL;
     }
 
-    ram_block_add(new_block, &local_err, share);
+    ram_block_add(new_block, &local_err, ram_flags & RAM_SHARED);
     if (local_err) {
         g_free(new_block);
         error_propagate(errp, local_err);
@@ -2131,7 +2131,7 @@ RAMBlock *qemu_ram_alloc_from_fd(ram_addr_t size, MemoryRegion *mr,
 
 
 RAMBlock *qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
-                                   bool share, const char *mem_path,
+                                   uint64_t ram_flags, const char *mem_path,
                                    Error **errp)
 {
     int fd;
@@ -2143,7 +2143,7 @@ RAMBlock *qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
         return NULL;
     }
 
-    block = qemu_ram_alloc_from_fd(size, mr, share, fd, errp);
+    block = qemu_ram_alloc_from_fd(size, mr, ram_flags, fd, errp);
     if (!block) {
         if (created) {
             unlink(mem_path);
diff --git a/include/exec/memory.h b/include/exec/memory.h
index 1bb9172..3769c06 100644
--- a/include/exec/memory.h
+++ b/include/exec/memory.h
@@ -599,6 +599,7 @@ void memory_region_init_resizeable_ram(MemoryRegion *mr,
                                                        void *host),
                                        Error **errp);
 #ifdef __linux__
+
 /**
  * memory_region_init_ram_from_file:  Initialize RAM memory region with a
  *                                    mmap-ed backend.
@@ -610,7 +611,9 @@ void memory_region_init_resizeable_ram(MemoryRegion *mr,
  * @size: size of the region.
  * @align: alignment of the region base address; if 0, the default alignment
  *         (getpagesize()) will be used.
- * @share: %true if memory must be mmaped with the MAP_SHARED flag
+ * @ram_flags: Memory region features:
+ *             - RAM_SHARED: memory must be mmaped with the MAP_SHARED flag
+ *             Other bits are ignored now.
  * @path: the path in which to allocate the RAM.
  * @errp: pointer to Error*, to store an error if it happens.
  *
@@ -622,7 +625,7 @@ void memory_region_init_ram_from_file(MemoryRegion *mr,
                                       const char *name,
                                       uint64_t size,
                                       uint64_t align,
-                                      bool share,
+                                      uint64_t ram_flags,
                                       const char *path,
                                       Error **errp);
 
diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h
index cf2446a..5545cf3 100644
--- a/include/exec/ram_addr.h
+++ b/include/exec/ram_addr.h
@@ -72,12 +72,33 @@ static inline unsigned long int ramblock_recv_bitmap_offset(void *host_addr,
 
 long qemu_getrampagesize(void);
 unsigned long last_ram_page(void);
+
+/**
+ * qemu_ram_alloc_from_file,
+ * qemu_ram_alloc_from_fd:  Allocate a ram block from the specified backing
+ *                          file or device
+ *
+ * Parameters:
+ *  @size: the size in bytes of the ram block
+ *  @mr: the memory region where the ram block is
+ *  @ram_flags: specify the properties of the ram block, which can be one
+ *              or bit-or of following values
+ *              - RAM_SHARED: mmap the backing file or device with MAP_SHARED
+ *              Other bits are ignored.
+ *  @mem_path or @fd: specify the backing file or device
+ *  @errp: pointer to Error*, to store an error if it happens
+ *
+ * Return:
+ *  On success, return a pointer to the ram block.
+ *  On failure, return NULL.
+ */
 RAMBlock *qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
-                                   bool share, const char *mem_path,
+                                   uint64_t ram_flags, const char *mem_path,
                                    Error **errp);
 RAMBlock *qemu_ram_alloc_from_fd(ram_addr_t size, MemoryRegion *mr,
-                                 bool share, int fd,
+                                 uint64_t ram_flags, int fd,
                                  Error **errp);
+
 RAMBlock *qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
                                   MemoryRegion *mr, Error **errp);
 RAMBlock *qemu_ram_alloc(ram_addr_t size, bool share, MemoryRegion *mr,
diff --git a/memory.c b/memory.c
index 3212acc..240e47a 100644
--- a/memory.c
+++ b/memory.c
@@ -1551,7 +1551,7 @@ void memory_region_init_ram_from_file(MemoryRegion *mr,
                                       const char *name,
                                       uint64_t size,
                                       uint64_t align,
-                                      bool share,
+                                      uint64_t ram_flags,
                                       const char *path,
                                       Error **errp)
 {
@@ -1560,7 +1560,7 @@ void memory_region_init_ram_from_file(MemoryRegion *mr,
     mr->terminates = true;
     mr->destructor = memory_region_destructor_ram;
     mr->align = align;
-    mr->ram_block = qemu_ram_alloc_from_file(size, mr, share, path, errp);
+    mr->ram_block = qemu_ram_alloc_from_file(size, mr, ram_flags, path, errp);
     mr->dirty_log_mask = tcg_enabled() ? (1 << DIRTY_MEMORY_CODE) : 0;
 }
 
@@ -1576,7 +1576,9 @@ void memory_region_init_ram_from_fd(MemoryRegion *mr,
     mr->ram = true;
     mr->terminates = true;
     mr->destructor = memory_region_destructor_ram;
-    mr->ram_block = qemu_ram_alloc_from_fd(size, mr, share, fd, errp);
+    mr->ram_block = qemu_ram_alloc_from_fd(size, mr,
+                                           share ? RAM_SHARED : 0,
+                                           fd, errp);
     mr->dirty_log_mask = tcg_enabled() ? (1 << DIRTY_MEMORY_CODE) : 0;
 }
 #endif
diff --git a/numa.c b/numa.c
index 33572bf..ba72267 100644
--- a/numa.c
+++ b/numa.c
@@ -479,7 +479,7 @@ static void allocate_system_memory_nonnuma(MemoryRegion *mr, Object *owner,
     if (mem_path) {
 #ifdef __linux__
         Error *err = NULL;
-        memory_region_init_ram_from_file(mr, owner, name, ram_size, 0, false,
+        memory_region_init_ram_from_file(mr, owner, name, ram_size, 0, 0,
                                          mem_path, &err);
         if (err) {
             error_report_err(err);
-- 
2.7.4

[Qemu-devel] [PATCH 3/7 V7] hostmem-file: add the 'pmem' option

[junyan.he]

From: Junyan He <junyan.he@intel.com>

When QEMU emulates vNVDIMM labels and migrates vNVDIMM devices, it
needs to know whether the backend storage is a real persistent memory,
in order to decide whether special operations should be performed to
ensure the data persistence.

This boolean option 'pmem' allows users to specify whether the backend
storage of memory-backend-file is a real persistent memory. If
'pmem=on', QEMU will set the flag RAM_PMEM in the RAM block of the
corresponding memory region.

Signed-off-by: Junyan He <junyan.he@intel.com>
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 backends/hostmem-file.c | 27 ++++++++++++++++++++++++++-
 docs/nvdimm.txt         | 18 ++++++++++++++++++
 exec.c                  |  9 +++++++++
 include/exec/memory.h   |  4 ++++
 include/exec/ram_addr.h |  3 +++
 qemu-options.hx         |  7 +++++++
 6 files changed, 67 insertions(+), 1 deletion(-)

diff --git a/backends/hostmem-file.c b/backends/hostmem-file.c
index 34c68bb..6a861f0 100644
--- a/backends/hostmem-file.c
+++ b/backends/hostmem-file.c
@@ -34,6 +34,7 @@ struct HostMemoryBackendFile {
     bool discard_data;
     char *mem_path;
     uint64_t align;
+    bool is_pmem;
 };
 
 static void
@@ -59,7 +60,8 @@ file_backend_memory_alloc(HostMemoryBackend *backend, Error **errp)
         memory_region_init_ram_from_file(&backend->mr, OBJECT(backend),
                                  path,
                                  backend->size, fb->align,
-                                 backend->share ? RAM_SHARED : 0,
+                                 (backend->share ? RAM_SHARED : 0) |
+                                 (fb->is_pmem ? RAM_PMEM : 0),
                                  fb->mem_path, errp);
         g_free(path);
     }
@@ -131,6 +133,26 @@ static void file_memory_backend_set_align(Object *o, Visitor *v,
     error_propagate(errp, local_err);
 }
 
+static bool file_memory_backend_get_pmem(Object *o, Error **errp)
+{
+    return MEMORY_BACKEND_FILE(o)->is_pmem;
+}
+
+static void file_memory_backend_set_pmem(Object *o, bool value, Error **errp)
+{
+    HostMemoryBackend *backend = MEMORY_BACKEND(o);
+    HostMemoryBackendFile *fb = MEMORY_BACKEND_FILE(o);
+
+    if (host_memory_backend_mr_inited(backend)) {
+        error_setg(errp, "cannot change property 'pmem' of %s '%s'",
+                   object_get_typename(o),
+                   object_get_canonical_path_component(o));
+        return;
+    }
+
+    fb->is_pmem = value;
+}
+
 static void file_backend_unparent(Object *obj)
 {
     HostMemoryBackend *backend = MEMORY_BACKEND(obj);
@@ -162,6 +184,9 @@ file_backend_class_init(ObjectClass *oc, void *data)
         file_memory_backend_get_align,
         file_memory_backend_set_align,
         NULL, NULL, &error_abort);
+    object_class_property_add_bool(oc, "pmem",
+        file_memory_backend_get_pmem, file_memory_backend_set_pmem,
+        &error_abort);
 }
 
 static void file_backend_instance_finalize(Object *o)
diff --git a/docs/nvdimm.txt b/docs/nvdimm.txt
index 8b48fb4..2f7d348 100644
--- a/docs/nvdimm.txt
+++ b/docs/nvdimm.txt
@@ -180,3 +180,21 @@ supports CPU Cache Flush and Memory Controller Flush on Power Loss, etc.
 
 For a complete list of the flags available and for more detailed descriptions,
 please consult the ACPI spec.
+
+guest software that this vNVDIMM device contains a region that cannot
+accept persistent writes. In result, for example, the guest Linux
+NVDIMM driver, marks such vNVDIMM device as read-only.
+
+If the vNVDIMM backend is on the host persistent memory that can be
+accessed in SNIA NVM Programming Model [1] (e.g., Intel NVDIMM), it's
+suggested to set the 'pmem' option of memory-backend-file to 'on'. When
+'pmem=on' and QEMU is built with libpmem [2] support (configured with
+--enable-libpmem), QEMU will take necessary operations to guarantee
+the persistence of its own writes to the vNVDIMM backend (e.g., in
+vNVDIMM label emulation and live migration).
+
+References
+----------
+
+[1] SNIA NVM Programming Model: https://www.snia.org/sites/default/files/technical_work/final/NVMProgrammingModel_v1.2.pdf
+[2] PMDK: http://pmem.io/pmdk/
diff --git a/exec.c b/exec.c
index 8e079df..c42483e 100644
--- a/exec.c
+++ b/exec.c
@@ -2077,6 +2077,9 @@ RAMBlock *qemu_ram_alloc_from_fd(ram_addr_t size, MemoryRegion *mr,
     Error *local_err = NULL;
     int64_t file_size;
 
+    /* Just support these ram flags by now. */
+    assert(ram_flags == 0 || (ram_flags & (RAM_SHARED | RAM_PMEM)));
+
     if (xen_enabled()) {
         error_setg(errp, "-mem-path not supported with Xen");
         return NULL;
@@ -4007,6 +4010,11 @@ err:
     return ret;
 }
 
+bool ramblock_is_pmem(RAMBlock *rb)
+{
+    return rb->flags & RAM_PMEM;
+}
+
 #endif
 
 void page_size_init(void)
@@ -4105,3 +4113,4 @@ void mtree_print_dispatch(fprintf_function mon, void *f,
 }
 
 #endif
+
diff --git a/include/exec/memory.h b/include/exec/memory.h
index 3769c06..7dd43b0 100644
--- a/include/exec/memory.h
+++ b/include/exec/memory.h
@@ -122,6 +122,9 @@ typedef struct IOMMUNotifier IOMMUNotifier;
 /* RAM can be migrated */
 #define RAM_MIGRATABLE (1 << 4)
 
+/* RAM is a persistent kind memory */
+#define RAM_PMEM (1 << 5)
+
 static inline void iommu_notifier_init(IOMMUNotifier *n, IOMMUNotify fn,
                                        IOMMUNotifierFlag flags,
                                        hwaddr start, hwaddr end)
@@ -613,6 +616,7 @@ void memory_region_init_resizeable_ram(MemoryRegion *mr,
  *         (getpagesize()) will be used.
  * @ram_flags: Memory region features:
  *             - RAM_SHARED: memory must be mmaped with the MAP_SHARED flag
+ *             - RAM_PMEM: the memory is persistent memory
  *             Other bits are ignored now.
  * @path: the path in which to allocate the RAM.
  * @errp: pointer to Error*, to store an error if it happens.
diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h
index 5545cf3..3c65643 100644
--- a/include/exec/ram_addr.h
+++ b/include/exec/ram_addr.h
@@ -70,6 +70,8 @@ static inline unsigned long int ramblock_recv_bitmap_offset(void *host_addr,
     return host_addr_offset >> TARGET_PAGE_BITS;
 }
 
+bool ramblock_is_pmem(RAMBlock *rb);
+
 long qemu_getrampagesize(void);
 unsigned long last_ram_page(void);
 
@@ -84,6 +86,7 @@ unsigned long last_ram_page(void);
  *  @ram_flags: specify the properties of the ram block, which can be one
  *              or bit-or of following values
  *              - RAM_SHARED: mmap the backing file or device with MAP_SHARED
+ *              - RAM_PMEM: the backend @mem_path or @fd is persistent memory
  *              Other bits are ignored.
  *  @mem_path or @fd: specify the backing file or device
  *  @errp: pointer to Error*, to store an error if it happens
diff --git a/qemu-options.hx b/qemu-options.hx
index c0d3951..006ca44 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -4045,6 +4045,13 @@ requires an alignment different than the default one used by QEMU, eg
 the device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In
 such cases, users can specify the required alignment via this option.
 
+The @option{pmem} option specifies whether the backing file specified
+by @option{mem-path} is on the persistent memory that can be accessed
+using the SNIA NVM programming model (e.g. Intel NVDIMM).
+If @option{pmem}, QEMU will take necessary operations to
+guarantee the persistence of its own writes to @option{mem-path}
+(e.g. in vNVDIMM label emulation and live migration).
+
 @item -object memory-backend-ram,id=@var{id},merge=@var{on|off},dump=@var{on|off},share=@var{on|off},prealloc=@var{on|off},size=@var{size},host-nodes=@var{host-nodes},policy=@var{default|preferred|bind|interleave}
 
 Creates a memory backend object, which can be used to back the guest RAM.
-- 
2.7.4

[Qemu-devel] [PATCH 4/7 V7] configure: add libpmem support

[junyan.he]

From: Junyan He <junyan.he@intel.com>

Add a pair of configure options --{enable,disable}-libpmem to control
whether QEMU is compiled with PMDK libpmem [1].

QEMU may write to the host persistent memory (e.g. in vNVDIMM label
emulation and live migration), so it must take the proper operations
to ensure the persistence of its own writes. Depending on the CPU
models and available instructions, the optimal operation can vary [2].
PMDK libpmem have already implemented those operations on multiple CPU
models (x86 and ARM) and the logic to select the optimal ones, so QEMU
can just use libpmem rather than re-implement them.

[1] PMDK (formerly known as NMVL), https://github.com/pmem/pmdk/
[2] https://github.com/pmem/pmdk/blob/38bfa652721a37fd94c0130ce0e3f5d8baa3ed40/src/libpmem/pmem.c#L33

Signed-off-by: Junyan He <junyan.he@intel.com>
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 configure | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/configure b/configure
index 14b1113..c49b7b6 100755
--- a/configure
+++ b/configure
@@ -457,6 +457,7 @@ replication="yes"
 vxhs=""
 libxml2=""
 docker="no"
+libpmem=""
 
 supported_cpu="no"
 supported_os="no"
@@ -1382,6 +1383,10 @@ for opt do
   ;;
   --disable-git-update) git_update=no
   ;;
+  --enable-libpmem) libpmem=yes
+  ;;
+  --disable-libpmem) libpmem=no
+  ;;
   *)
       echo "ERROR: unknown option $opt"
       echo "Try '$0 --help' for more information"
@@ -1639,6 +1644,7 @@ disabled with --disable-FEATURE, default is enabled if available:
   crypto-afalg    Linux AF_ALG crypto backend driver
   vhost-user      vhost-user support
   capstone        capstone disassembler support
+  libpmem         libpmem support
 
 NOTE: The object files are built at the place where configure is launched
 EOF
@@ -5463,6 +5469,24 @@ if has "docker"; then
 fi
 
 ##########################################
+# check for libpmem
+
+if test "$libpmem" != "no"; then
+	if $pkg_config --exists "libpmem"; then
+		libpmem="yes"
+		libpmem_libs=$($pkg_config --libs libpmem)
+		libpmem_cflags=$($pkg_config --cflags libpmem)
+		libs_softmmu="$libs_softmmu $libpmem_libs"
+		QEMU_CFLAGS="$QEMU_CFLAGS $libpmem_cflags"
+	else
+		if test "$libpmem" = "yes" ; then
+			feature_not_found "libpmem" "Install nvml or pmdk"
+		fi
+		libpmem="no"
+	fi
+fi
+
+##########################################
 # End of CC checks
 # After here, no more $cc or $ld runs
 
@@ -5926,6 +5950,7 @@ echo "replication support $replication"
 echo "VxHS block device $vxhs"
 echo "capstone          $capstone"
 echo "docker            $docker"
+echo "libpmem support   $libpmem"
 
 if test "$sdl_too_old" = "yes"; then
 echo "-> Your SDL version is too old - please upgrade to have SDL support"
@@ -6673,6 +6698,10 @@ if test "$vxhs" = "yes" ; then
   echo "VXHS_LIBS=$vxhs_libs" >> $config_host_mak
 fi
 
+if test "$libpmem" = "yes" ; then
+  echo "CONFIG_LIBPMEM=y" >> $config_host_mak
+fi
+
 if test "$tcg_interpreter" = "yes"; then
   QEMU_INCLUDES="-iquote \$(SRC_PATH)/tcg/tci $QEMU_INCLUDES"
 elif test "$ARCH" = "sparc64" ; then
-- 
2.7.4

[Qemu-devel] [PATCH 5/7 V7] mem/nvdimm: ensure write persistence to PMEM in label emulation

[junyan.he]

From: Junyan He <junyan.he@intel.com>

Guest writes to vNVDIMM labels are intercepted and performed on the
backend by QEMU. When the backend is a real persistent memort, QEMU
needs to take proper operations to ensure its write persistence on the
persistent memory. Otherwise, a host power failure may result in the
loss of guest label configurations.

Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 hw/mem/nvdimm.c     |  9 ++++++++-
 include/qemu/pmem.h | 23 +++++++++++++++++++++++
 stubs/Makefile.objs |  1 +
 stubs/pmem.c        | 19 +++++++++++++++++++
 4 files changed, 51 insertions(+), 1 deletion(-)
 create mode 100644 include/qemu/pmem.h
 create mode 100644 stubs/pmem.c

diff --git a/hw/mem/nvdimm.c b/hw/mem/nvdimm.c
index 4087aca..03b478e 100644
--- a/hw/mem/nvdimm.c
+++ b/hw/mem/nvdimm.c
@@ -23,6 +23,7 @@
  */
 
 #include "qemu/osdep.h"
+#include "qemu/pmem.h"
 #include "qapi/error.h"
 #include "qapi/visitor.h"
 #include "hw/mem/nvdimm.h"
@@ -155,11 +156,17 @@ static void nvdimm_write_label_data(NVDIMMDevice *nvdimm, const void *buf,
 {
     MemoryRegion *mr;
     PCDIMMDevice *dimm = PC_DIMM(nvdimm);
+    bool is_pmem = object_property_get_bool(OBJECT(dimm->hostmem),
+                                            "pmem", NULL);
     uint64_t backend_offset;
 
     nvdimm_validate_rw_label_data(nvdimm, size, offset);
 
-    memcpy(nvdimm->label_data + offset, buf, size);
+    if (!is_pmem) {
+        memcpy(nvdimm->label_data + offset, buf, size);
+    } else {
+        pmem_memcpy_persist(nvdimm->label_data + offset, buf, size);
+    }
 
     mr = host_memory_backend_get_memory(dimm->hostmem, &error_abort);
     backend_offset = memory_region_size(mr) - nvdimm->label_size + offset;
diff --git a/include/qemu/pmem.h b/include/qemu/pmem.h
new file mode 100644
index 0000000..00d6680
--- /dev/null
+++ b/include/qemu/pmem.h
@@ -0,0 +1,23 @@
+/*
+ * QEMU header file for libpmem.
+ *
+ * Copyright (c) 2018 Intel Corporation.
+ *
+ * Author: Haozhong Zhang <address@hidden>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#ifndef QEMU_PMEM_H
+#define QEMU_PMEM_H
+
+#ifdef CONFIG_LIBPMEM
+#include <libpmem.h>
+#else  /* !CONFIG_LIBPMEM */
+
+void *pmem_memcpy_persist(void *pmemdest, const void *src, size_t len);
+
+#endif /* CONFIG_LIBPMEM */
+
+#endif /* !QEMU_PMEM_H */
diff --git a/stubs/Makefile.objs b/stubs/Makefile.objs
index 53d3f32..be9a042 100644
--- a/stubs/Makefile.objs
+++ b/stubs/Makefile.objs
@@ -43,3 +43,4 @@ stub-obj-y += xen-common.o
 stub-obj-y += xen-hvm.o
 stub-obj-y += pci-host-piix.o
 stub-obj-y += ram-block.o
+stub-obj-$(call lnot,$(CONFIG_LIBPMEM)) += pmem.o
\ No newline at end of file
diff --git a/stubs/pmem.c b/stubs/pmem.c
new file mode 100644
index 0000000..b4ec72d
--- /dev/null
+++ b/stubs/pmem.c
@@ -0,0 +1,19 @@
+/*
+ * Stubs for libpmem.
+ *
+ * Copyright (c) 2018 Intel Corporation.
+ *
+ * Author: Haozhong Zhang <address@hidden>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include <string.h>
+
+#include "qemu/pmem.h"
+
+void *pmem_memcpy_persist(void *pmemdest, const void *src, size_t len)
+{
+    return memcpy(pmemdest, src, len);
+}
-- 
2.7.4

[Qemu-devel] [PATCH 6/7 V7] migration/ram: Add check and info message to nvdimm post copy.

[junyan.he]

From: Junyan He <junyan.he@intel.com>

The nvdimm kind memory does not support post copy now.
We disable post copy if we have nvdimm memory and print some
log hint to user.

Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 migration/ram.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/migration/ram.c b/migration/ram.c
index a500015..d9093a7 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -3419,6 +3419,15 @@ static int ram_load(QEMUFile *f, void *opaque, int version_id)
 
 static bool ram_has_postcopy(void *opaque)
 {
+    RAMBlock *rb;
+    RAMBLOCK_FOREACH(rb) {
+        if (ramblock_is_pmem(rb)) {
+            info_report("Block: %s, host: %p is a nvdimm memory, postcopy"
+                         "is not supported now!", rb->idstr, rb->host);
+            return false;
+        }
+    }
+
     return migrate_postcopy_ram();
 }
 
-- 
2.7.4

[Qemu-devel] [PATCH 7/7 V7] migration/ram: ensure write persistence on loading all data to PMEM.

[junyan.he]

From: Junyan He <junyan.he@intel.com>

Because we need to make sure the pmem kind memory data is synced
after migration, we choose to call pmem_persist() when the migration
finish. This will make sure the data of pmem is safe and will not
lose if power is off.

Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 include/qemu/pmem.h | 1 +
 migration/ram.c     | 8 ++++++++
 stubs/pmem.c        | 4 ++++
 3 files changed, 13 insertions(+)

diff --git a/include/qemu/pmem.h b/include/qemu/pmem.h
index 00d6680..8f52b08 100644
--- a/include/qemu/pmem.h
+++ b/include/qemu/pmem.h
@@ -17,6 +17,7 @@
 #else  /* !CONFIG_LIBPMEM */
 
 void *pmem_memcpy_persist(void *pmemdest, const void *src, size_t len);
+void pmem_persist(const void *addr, size_t len);
 
 #endif /* CONFIG_LIBPMEM */
 
diff --git a/migration/ram.c b/migration/ram.c
index d9093a7..5603505 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -33,6 +33,7 @@
 #include "qemu/bitops.h"
 #include "qemu/bitmap.h"
 #include "qemu/main-loop.h"
+#include "qemu/pmem.h"
 #include "xbzrle.h"
 #include "ram.h"
 #include "migration.h"
@@ -3065,6 +3066,13 @@ static int ram_load_setup(QEMUFile *f, void *opaque)
 static int ram_load_cleanup(void *opaque)
 {
     RAMBlock *rb;
+
+    RAMBLOCK_FOREACH_MIGRATABLE(rb) {
+        if (ramblock_is_pmem(rb)) {
+            pmem_persist(rb->host, rb->used_length);
+        }
+    }
+
     xbzrle_load_cleanup();
     compress_threads_load_cleanup();
 
diff --git a/stubs/pmem.c b/stubs/pmem.c
index b4ec72d..f794262 100644
--- a/stubs/pmem.c
+++ b/stubs/pmem.c
@@ -17,3 +17,7 @@ void *pmem_memcpy_persist(void *pmemdest, const void *src, size_t len)
 {
     return memcpy(pmemdest, src, len);
 }
+
+void pmem_persist(const void *addr, size_t len)
+{
+}
-- 
2.7.4

Re: [Qemu-devel] [PATCH 3/7 V7] hostmem-file: add the 'pmem' option

[Igor Mammedov]

On Tue, 12 Jun 2018 14:28:16 +0800
junyan.he@gmx.com wrote:

> From: Junyan He <junyan.he@intel.com>
> 
> When QEMU emulates vNVDIMM labels and migrates vNVDIMM devices, it
> needs to know whether the backend storage is a real persistent memory,
> in order to decide whether special operations should be performed to
> ensure the data persistence.
> 
> This boolean option 'pmem' allows users to specify whether the backend
> storage of memory-backend-file is a real persistent memory. If
> 'pmem=on', QEMU will set the flag RAM_PMEM in the RAM block of the
> corresponding memory region.
> 
As were noted in v6 by Eduardo and me, we would prefer that pmem=on
would fail if qemu is built without libpmem support.


> Signed-off-by: Junyan He <junyan.he@intel.com>
> Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
> ---
>  backends/hostmem-file.c | 27 ++++++++++++++++++++++++++-
>  docs/nvdimm.txt         | 18 ++++++++++++++++++
>  exec.c                  |  9 +++++++++
>  include/exec/memory.h   |  4 ++++
>  include/exec/ram_addr.h |  3 +++
>  qemu-options.hx         |  7 +++++++
>  6 files changed, 67 insertions(+), 1 deletion(-)
> 
> diff --git a/backends/hostmem-file.c b/backends/hostmem-file.c
> index 34c68bb..6a861f0 100644
> --- a/backends/hostmem-file.c
> +++ b/backends/hostmem-file.c
> @@ -34,6 +34,7 @@ struct HostMemoryBackendFile {
>      bool discard_data;
>      char *mem_path;
>      uint64_t align;
> +    bool is_pmem;
>  };
>  
>  static void
> @@ -59,7 +60,8 @@ file_backend_memory_alloc(HostMemoryBackend *backend, Error **errp)
>          memory_region_init_ram_from_file(&backend->mr, OBJECT(backend),
>                                   path,
>                                   backend->size, fb->align,
> -                                 backend->share ? RAM_SHARED : 0,
> +                                 (backend->share ? RAM_SHARED : 0) |
> +                                 (fb->is_pmem ? RAM_PMEM : 0),
>                                   fb->mem_path, errp);
>          g_free(path);
>      }
> @@ -131,6 +133,26 @@ static void file_memory_backend_set_align(Object *o, Visitor *v,
>      error_propagate(errp, local_err);
>  }
>  
> +static bool file_memory_backend_get_pmem(Object *o, Error **errp)
> +{
> +    return MEMORY_BACKEND_FILE(o)->is_pmem;
> +}
> +
> +static void file_memory_backend_set_pmem(Object *o, bool value, Error **errp)
> +{
> +    HostMemoryBackend *backend = MEMORY_BACKEND(o);
> +    HostMemoryBackendFile *fb = MEMORY_BACKEND_FILE(o);
> +
> +    if (host_memory_backend_mr_inited(backend)) {
> +        error_setg(errp, "cannot change property 'pmem' of %s '%s'",
> +                   object_get_typename(o),
> +                   object_get_canonical_path_component(o));
> +        return;
> +    }
> +
> +    fb->is_pmem = value;
> +}
> +
>  static void file_backend_unparent(Object *obj)
>  {
>      HostMemoryBackend *backend = MEMORY_BACKEND(obj);
> @@ -162,6 +184,9 @@ file_backend_class_init(ObjectClass *oc, void *data)
>          file_memory_backend_get_align,
>          file_memory_backend_set_align,
>          NULL, NULL, &error_abort);
> +    object_class_property_add_bool(oc, "pmem",
> +        file_memory_backend_get_pmem, file_memory_backend_set_pmem,
> +        &error_abort);
>  }
>  
>  static void file_backend_instance_finalize(Object *o)
> diff --git a/docs/nvdimm.txt b/docs/nvdimm.txt
> index 8b48fb4..2f7d348 100644
> --- a/docs/nvdimm.txt
> +++ b/docs/nvdimm.txt
> @@ -180,3 +180,21 @@ supports CPU Cache Flush and Memory Controller Flush on Power Loss, etc.
>  
>  For a complete list of the flags available and for more detailed descriptions,
>  please consult the ACPI spec.
> +
> +guest software that this vNVDIMM device contains a region that cannot
> +accept persistent writes. In result, for example, the guest Linux
> +NVDIMM driver, marks such vNVDIMM device as read-only.
> +
> +If the vNVDIMM backend is on the host persistent memory that can be
> +accessed in SNIA NVM Programming Model [1] (e.g., Intel NVDIMM), it's
> +suggested to set the 'pmem' option of memory-backend-file to 'on'. When
> +'pmem=on' and QEMU is built with libpmem [2] support (configured with
> +--enable-libpmem), QEMU will take necessary operations to guarantee
> +the persistence of its own writes to the vNVDIMM backend (e.g., in
> +vNVDIMM label emulation and live migration).
> +
> +References
> +----------
> +
> +[1] SNIA NVM Programming Model: https://www.snia.org/sites/default/files/technical_work/final/NVMProgrammingModel_v1.2.pdf
> +[2] PMDK: http://pmem.io/pmdk/
> diff --git a/exec.c b/exec.c
> index 8e079df..c42483e 100644
> --- a/exec.c
> +++ b/exec.c
> @@ -2077,6 +2077,9 @@ RAMBlock *qemu_ram_alloc_from_fd(ram_addr_t size, MemoryRegion *mr,
>      Error *local_err = NULL;
>      int64_t file_size;
>  
> +    /* Just support these ram flags by now. */
> +    assert(ram_flags == 0 || (ram_flags & (RAM_SHARED | RAM_PMEM)));
> +
>      if (xen_enabled()) {
>          error_setg(errp, "-mem-path not supported with Xen");
>          return NULL;
> @@ -4007,6 +4010,11 @@ err:
>      return ret;
>  }
>  
> +bool ramblock_is_pmem(RAMBlock *rb)
> +{
> +    return rb->flags & RAM_PMEM;
> +}
> +
>  #endif
>  
>  void page_size_init(void)
> @@ -4105,3 +4113,4 @@ void mtree_print_dispatch(fprintf_function mon, void *f,
>  }
>  
>  #endif
> +
> diff --git a/include/exec/memory.h b/include/exec/memory.h
> index 3769c06..7dd43b0 100644
> --- a/include/exec/memory.h
> +++ b/include/exec/memory.h
> @@ -122,6 +122,9 @@ typedef struct IOMMUNotifier IOMMUNotifier;
>  /* RAM can be migrated */
>  #define RAM_MIGRATABLE (1 << 4)
>  
> +/* RAM is a persistent kind memory */
> +#define RAM_PMEM (1 << 5)
> +
>  static inline void iommu_notifier_init(IOMMUNotifier *n, IOMMUNotify fn,
>                                         IOMMUNotifierFlag flags,
>                                         hwaddr start, hwaddr end)
> @@ -613,6 +616,7 @@ void memory_region_init_resizeable_ram(MemoryRegion *mr,
>   *         (getpagesize()) will be used.
>   * @ram_flags: Memory region features:
>   *             - RAM_SHARED: memory must be mmaped with the MAP_SHARED flag
> + *             - RAM_PMEM: the memory is persistent memory
>   *             Other bits are ignored now.
>   * @path: the path in which to allocate the RAM.
>   * @errp: pointer to Error*, to store an error if it happens.
> diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h
> index 5545cf3..3c65643 100644
> --- a/include/exec/ram_addr.h
> +++ b/include/exec/ram_addr.h
> @@ -70,6 +70,8 @@ static inline unsigned long int ramblock_recv_bitmap_offset(void *host_addr,
>      return host_addr_offset >> TARGET_PAGE_BITS;
>  }
>  
> +bool ramblock_is_pmem(RAMBlock *rb);
> +
>  long qemu_getrampagesize(void);
>  unsigned long last_ram_page(void);
>  
> @@ -84,6 +86,7 @@ unsigned long last_ram_page(void);
>   *  @ram_flags: specify the properties of the ram block, which can be one
>   *              or bit-or of following values
>   *              - RAM_SHARED: mmap the backing file or device with MAP_SHARED
> + *              - RAM_PMEM: the backend @mem_path or @fd is persistent memory
>   *              Other bits are ignored.
>   *  @mem_path or @fd: specify the backing file or device
>   *  @errp: pointer to Error*, to store an error if it happens
> diff --git a/qemu-options.hx b/qemu-options.hx
> index c0d3951..006ca44 100644
> --- a/qemu-options.hx
> +++ b/qemu-options.hx
> @@ -4045,6 +4045,13 @@ requires an alignment different than the default one used by QEMU, eg
>  the device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In
>  such cases, users can specify the required alignment via this option.
>  
> +The @option{pmem} option specifies whether the backing file specified
> +by @option{mem-path} is on the persistent memory that can be accessed
> +using the SNIA NVM programming model (e.g. Intel NVDIMM).
> +If @option{pmem}, QEMU will take necessary operations to
> +guarantee the persistence of its own writes to @option{mem-path}
> +(e.g. in vNVDIMM label emulation and live migration).
> +
>  @item -object memory-backend-ram,id=@var{id},merge=@var{on|off},dump=@var{on|off},share=@var{on|off},prealloc=@var{on|off},size=@var{size},host-nodes=@var{host-nodes},policy=@var{default|preferred|bind|interleave}
>  
>  Creates a memory backend object, which can be used to back the guest RAM.

Re: [Qemu-devel] [PATCH 3/7 V7] hostmem-file: add the 'pmem' option

[He, Junyan]

Have already resent this, make qemu exit when meet this, please help to check,
thanks

-----Original Message-----
From: Igor Mammedov [mailto:imammedo@redhat.com] 
Sent: Friday, June 15, 2018 5:04 PM
To: junyan.he@gmx.com
Cc: qemu-devel@nongnu.org; Haozhong Zhang <haozhong.zhang@intel.com>; xiaoguangrong.eric@gmail.com; crosthwaite.peter@gmail.com; mst@redhat.com; dgilbert@redhat.com; ehabkost@redhat.com; quintela@redhat.com; He, Junyan <junyan.he@intel.com>; stefanha@redhat.com; pbonzini@redhat.com; rth@twiddle.net
Subject: Re: [Qemu-devel] [PATCH 3/7 V7] hostmem-file: add the 'pmem' option

On Tue, 12 Jun 2018 14:28:16 +0800
junyan.he@gmx.com wrote:

> From: Junyan He <junyan.he@intel.com>
> 
> When QEMU emulates vNVDIMM labels and migrates vNVDIMM devices, it 
> needs to know whether the backend storage is a real persistent memory, 
> in order to decide whether special operations should be performed to 
> ensure the data persistence.
> 
> This boolean option 'pmem' allows users to specify whether the backend 
> storage of memory-backend-file is a real persistent memory. If 
> 'pmem=on', QEMU will set the flag RAM_PMEM in the RAM block of the 
> corresponding memory region.
> 
As were noted in v6 by Eduardo and me, we would prefer that pmem=on would fail if qemu is built without libpmem support.


> Signed-off-by: Junyan He <junyan.he@intel.com>
> Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
> ---
>  backends/hostmem-file.c | 27 ++++++++++++++++++++++++++-
>  docs/nvdimm.txt         | 18 ++++++++++++++++++
>  exec.c                  |  9 +++++++++
>  include/exec/memory.h   |  4 ++++
>  include/exec/ram_addr.h |  3 +++
>  qemu-options.hx         |  7 +++++++
>  6 files changed, 67 insertions(+), 1 deletion(-)
> 
> diff --git a/backends/hostmem-file.c b/backends/hostmem-file.c index 
> 34c68bb..6a861f0 100644
> --- a/backends/hostmem-file.c
> +++ b/backends/hostmem-file.c
> @@ -34,6 +34,7 @@ struct HostMemoryBackendFile {
>      bool discard_data;
>      char *mem_path;
>      uint64_t align;
> +    bool is_pmem;
>  };
>  
>  static void
> @@ -59,7 +60,8 @@ file_backend_memory_alloc(HostMemoryBackend *backend, Error **errp)
>          memory_region_init_ram_from_file(&backend->mr, OBJECT(backend),
>                                   path,
>                                   backend->size, fb->align,
> -                                 backend->share ? RAM_SHARED : 0,
> +                                 (backend->share ? RAM_SHARED : 0) |
> +                                 (fb->is_pmem ? RAM_PMEM : 0),
>                                   fb->mem_path, errp);
>          g_free(path);
>      }
> @@ -131,6 +133,26 @@ static void file_memory_backend_set_align(Object *o, Visitor *v,
>      error_propagate(errp, local_err);  }
>  
> +static bool file_memory_backend_get_pmem(Object *o, Error **errp) {
> +    return MEMORY_BACKEND_FILE(o)->is_pmem; }
> +
> +static void file_memory_backend_set_pmem(Object *o, bool value, Error 
> +**errp) {
> +    HostMemoryBackend *backend = MEMORY_BACKEND(o);
> +    HostMemoryBackendFile *fb = MEMORY_BACKEND_FILE(o);
> +
> +    if (host_memory_backend_mr_inited(backend)) {
> +        error_setg(errp, "cannot change property 'pmem' of %s '%s'",
> +                   object_get_typename(o),
> +                   object_get_canonical_path_component(o));
> +        return;
> +    }
> +
> +    fb->is_pmem = value;
> +}
> +
>  static void file_backend_unparent(Object *obj)  {
>      HostMemoryBackend *backend = MEMORY_BACKEND(obj); @@ -162,6 
> +184,9 @@ file_backend_class_init(ObjectClass *oc, void *data)
>          file_memory_backend_get_align,
>          file_memory_backend_set_align,
>          NULL, NULL, &error_abort);
> +    object_class_property_add_bool(oc, "pmem",
> +        file_memory_backend_get_pmem, file_memory_backend_set_pmem,
> +        &error_abort);
>  }
>  
>  static void file_backend_instance_finalize(Object *o) diff --git 
> a/docs/nvdimm.txt b/docs/nvdimm.txt index 8b48fb4..2f7d348 100644
> --- a/docs/nvdimm.txt
> +++ b/docs/nvdimm.txt
> @@ -180,3 +180,21 @@ supports CPU Cache Flush and Memory Controller Flush on Power Loss, etc.
>  
>  For a complete list of the flags available and for more detailed 
> descriptions,  please consult the ACPI spec.
> +
> +guest software that this vNVDIMM device contains a region that cannot 
> +accept persistent writes. In result, for example, the guest Linux 
> +NVDIMM driver, marks such vNVDIMM device as read-only.
> +
> +If the vNVDIMM backend is on the host persistent memory that can be 
> +accessed in SNIA NVM Programming Model [1] (e.g., Intel NVDIMM), it's 
> +suggested to set the 'pmem' option of memory-backend-file to 'on'. 
> +When 'pmem=on' and QEMU is built with libpmem [2] support (configured 
> +with --enable-libpmem), QEMU will take necessary operations to 
> +guarantee the persistence of its own writes to the vNVDIMM backend 
> +(e.g., in vNVDIMM label emulation and live migration).
> +
> +References
> +----------
> +
> +[1] SNIA NVM Programming Model: 
> +https://www.snia.org/sites/default/files/technical_work/final/NVMProg
> +rammingModel_v1.2.pdf
> +[2] PMDK: http://pmem.io/pmdk/
> diff --git a/exec.c b/exec.c
> index 8e079df..c42483e 100644
> --- a/exec.c
> +++ b/exec.c
> @@ -2077,6 +2077,9 @@ RAMBlock *qemu_ram_alloc_from_fd(ram_addr_t size, MemoryRegion *mr,
>      Error *local_err = NULL;
>      int64_t file_size;
>  
> +    /* Just support these ram flags by now. */
> +    assert(ram_flags == 0 || (ram_flags & (RAM_SHARED | RAM_PMEM)));
> +
>      if (xen_enabled()) {
>          error_setg(errp, "-mem-path not supported with Xen");
>          return NULL;
> @@ -4007,6 +4010,11 @@ err:
>      return ret;
>  }
>  
> +bool ramblock_is_pmem(RAMBlock *rb)
> +{
> +    return rb->flags & RAM_PMEM;
> +}
> +
>  #endif
>  
>  void page_size_init(void)
> @@ -4105,3 +4113,4 @@ void mtree_print_dispatch(fprintf_function mon, 
> void *f,  }
>  
>  #endif
> +
> diff --git a/include/exec/memory.h b/include/exec/memory.h index 
> 3769c06..7dd43b0 100644
> --- a/include/exec/memory.h
> +++ b/include/exec/memory.h
> @@ -122,6 +122,9 @@ typedef struct IOMMUNotifier IOMMUNotifier;
>  /* RAM can be migrated */
>  #define RAM_MIGRATABLE (1 << 4)
>  
> +/* RAM is a persistent kind memory */ #define RAM_PMEM (1 << 5)
> +
>  static inline void iommu_notifier_init(IOMMUNotifier *n, IOMMUNotify fn,
>                                         IOMMUNotifierFlag flags,
>                                         hwaddr start, hwaddr end) @@ 
> -613,6 +616,7 @@ void memory_region_init_resizeable_ram(MemoryRegion *mr,
>   *         (getpagesize()) will be used.
>   * @ram_flags: Memory region features:
>   *             - RAM_SHARED: memory must be mmaped with the MAP_SHARED flag
> + *             - RAM_PMEM: the memory is persistent memory
>   *             Other bits are ignored now.
>   * @path: the path in which to allocate the RAM.
>   * @errp: pointer to Error*, to store an error if it happens.
> diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h index 
> 5545cf3..3c65643 100644
> --- a/include/exec/ram_addr.h
> +++ b/include/exec/ram_addr.h
> @@ -70,6 +70,8 @@ static inline unsigned long int ramblock_recv_bitmap_offset(void *host_addr,
>      return host_addr_offset >> TARGET_PAGE_BITS;  }
>  
> +bool ramblock_is_pmem(RAMBlock *rb);
> +
>  long qemu_getrampagesize(void);
>  unsigned long last_ram_page(void);
>  
> @@ -84,6 +86,7 @@ unsigned long last_ram_page(void);
>   *  @ram_flags: specify the properties of the ram block, which can be one
>   *              or bit-or of following values
>   *              - RAM_SHARED: mmap the backing file or device with MAP_SHARED
> + *              - RAM_PMEM: the backend @mem_path or @fd is persistent memory
>   *              Other bits are ignored.
>   *  @mem_path or @fd: specify the backing file or device
>   *  @errp: pointer to Error*, to store an error if it happens diff 
> --git a/qemu-options.hx b/qemu-options.hx index c0d3951..006ca44 
> 100644
> --- a/qemu-options.hx
> +++ b/qemu-options.hx
> @@ -4045,6 +4045,13 @@ requires an alignment different than the 
> default one used by QEMU, eg  the device DAX /dev/dax0.0 requires 2M 
> alignment rather than 4K. In  such cases, users can specify the required alignment via this option.
>  
> +The @option{pmem} option specifies whether the backing file specified 
> +by @option{mem-path} is on the persistent memory that can be accessed 
> +using the SNIA NVM programming model (e.g. Intel NVDIMM).
> +If @option{pmem}, QEMU will take necessary operations to guarantee 
> +the persistence of its own writes to @option{mem-path} (e.g. in 
> +vNVDIMM label emulation and live migration).
> +
>  @item -object 
> memory-backend-ram,id=@var{id},merge=@var{on|off},dump=@var{on|off},sh
> are=@var{on|off},prealloc=@var{on|off},size=@var{size},host-nodes=@var
> {host-nodes},policy=@var{default|preferred|bind|interleave}
>  
>  Creates a memory backend object, which can be used to back the guest RAM.

[Qemu-devel] [PATCH 3/7 V7] hostmem-file: add the 'pmem' option

[junyan.he]

From: Junyan He <junyan.he@intel.com>

When QEMU emulates vNVDIMM labels and migrates vNVDIMM devices, it
needs to know whether the backend storage is a real persistent memory,
in order to decide whether special operations should be performed to
ensure the data persistence.

This boolean option 'pmem' allows users to specify whether the backend
storage of memory-backend-file is a real persistent memory. If
'pmem=on', QEMU will set the flag RAM_PMEM in the RAM block of the
corresponding memory region.

Signed-off-by: Junyan He <junyan.he@intel.com>
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>

*RESEND: If pmem is on while we lack of libpmem support, we just make qemu
exit and print some error message to user. This can prevent misusing of pmem
parameter while we can not really ensure the persistence.

---
 backends/hostmem-file.c | 39 ++++++++++++++++++++++++++++++++++++++-
 docs/nvdimm.txt         | 18 ++++++++++++++++++
 exec.c                  |  9 +++++++++
 include/exec/memory.h   |  4 ++++
 include/exec/ram_addr.h |  3 +++
 qemu-options.hx         |  7 +++++++
 6 files changed, 79 insertions(+), 1 deletion(-)

diff --git a/backends/hostmem-file.c b/backends/hostmem-file.c
index 34c68bb..4607651 100644
--- a/backends/hostmem-file.c
+++ b/backends/hostmem-file.c
@@ -12,6 +12,7 @@
 #include "qemu/osdep.h"
 #include "qapi/error.h"
 #include "qemu-common.h"
+#include "qemu/error-report.h"
 #include "sysemu/hostmem.h"
 #include "sysemu/sysemu.h"
 #include "qom/object_interfaces.h"
@@ -34,6 +35,7 @@ struct HostMemoryBackendFile {
     bool discard_data;
     char *mem_path;
     uint64_t align;
+    bool is_pmem;
 };
 
 static void
@@ -59,7 +61,8 @@ file_backend_memory_alloc(HostMemoryBackend *backend, Error **errp)
         memory_region_init_ram_from_file(&backend->mr, OBJECT(backend),
                                  path,
                                  backend->size, fb->align,
-                                 backend->share ? RAM_SHARED : 0,
+                                 (backend->share ? RAM_SHARED : 0) |
+                                 (fb->is_pmem ? RAM_PMEM : 0),
                                  fb->mem_path, errp);
         g_free(path);
     }
@@ -131,6 +134,37 @@ static void file_memory_backend_set_align(Object *o, Visitor *v,
     error_propagate(errp, local_err);
 }
 
+static bool file_memory_backend_get_pmem(Object *o, Error **errp)
+{
+    return MEMORY_BACKEND_FILE(o)->is_pmem;
+}
+
+static void file_memory_backend_set_pmem(Object *o, bool value, Error **errp)
+{
+    HostMemoryBackend *backend = MEMORY_BACKEND(o);
+    HostMemoryBackendFile *fb = MEMORY_BACKEND_FILE(o);
+
+    if (host_memory_backend_mr_inited(backend)) {
+        error_setg(errp, "cannot change property 'pmem' of %s '%s'",
+                   object_get_typename(o),
+                   object_get_canonical_path_component(o));
+        return;
+    }
+
+#ifndef CONFIG_LIBPMEM
+    if (value) {
+        error_report("Lack of libpmem support while setting the 'pmem=on'"
+                     " of %s '%s'. We can not ensure the persistence of it"
+                     " without libpmem support, this may cause serious"
+                     " problems." , object_get_typename(o),
+                     object_get_canonical_path_component(o));
+        exit(1);
+    }
+#endif
+
+    fb->is_pmem = value;
+}
+
 static void file_backend_unparent(Object *obj)
 {
     HostMemoryBackend *backend = MEMORY_BACKEND(obj);
@@ -162,6 +196,9 @@ file_backend_class_init(ObjectClass *oc, void *data)
         file_memory_backend_get_align,
         file_memory_backend_set_align,
         NULL, NULL, &error_abort);
+    object_class_property_add_bool(oc, "pmem",
+        file_memory_backend_get_pmem, file_memory_backend_set_pmem,
+        &error_abort);
 }
 
 static void file_backend_instance_finalize(Object *o)
diff --git a/docs/nvdimm.txt b/docs/nvdimm.txt
index 8b48fb4..2f7d348 100644
--- a/docs/nvdimm.txt
+++ b/docs/nvdimm.txt
@@ -180,3 +180,21 @@ supports CPU Cache Flush and Memory Controller Flush on Power Loss, etc.
 
 For a complete list of the flags available and for more detailed descriptions,
 please consult the ACPI spec.
+
+guest software that this vNVDIMM device contains a region that cannot
+accept persistent writes. In result, for example, the guest Linux
+NVDIMM driver, marks such vNVDIMM device as read-only.
+
+If the vNVDIMM backend is on the host persistent memory that can be
+accessed in SNIA NVM Programming Model [1] (e.g., Intel NVDIMM), it's
+suggested to set the 'pmem' option of memory-backend-file to 'on'. When
+'pmem=on' and QEMU is built with libpmem [2] support (configured with
+--enable-libpmem), QEMU will take necessary operations to guarantee
+the persistence of its own writes to the vNVDIMM backend (e.g., in
+vNVDIMM label emulation and live migration).
+
+References
+----------
+
+[1] SNIA NVM Programming Model: https://www.snia.org/sites/default/files/technical_work/final/NVMProgrammingModel_v1.2.pdf
+[2] PMDK: http://pmem.io/pmdk/
diff --git a/exec.c b/exec.c
index 8e079df..c42483e 100644
--- a/exec.c
+++ b/exec.c
@@ -2077,6 +2077,9 @@ RAMBlock *qemu_ram_alloc_from_fd(ram_addr_t size, MemoryRegion *mr,
     Error *local_err = NULL;
     int64_t file_size;
 
+    /* Just support these ram flags by now. */
+    assert(ram_flags == 0 || (ram_flags & (RAM_SHARED | RAM_PMEM)));
+
     if (xen_enabled()) {
         error_setg(errp, "-mem-path not supported with Xen");
         return NULL;
@@ -4007,6 +4010,11 @@ err:
     return ret;
 }
 
+bool ramblock_is_pmem(RAMBlock *rb)
+{
+    return rb->flags & RAM_PMEM;
+}
+
 #endif
 
 void page_size_init(void)
@@ -4105,3 +4113,4 @@ void mtree_print_dispatch(fprintf_function mon, void *f,
 }
 
 #endif
+
diff --git a/include/exec/memory.h b/include/exec/memory.h
index 3769c06..7dd43b0 100644
--- a/include/exec/memory.h
+++ b/include/exec/memory.h
@@ -122,6 +122,9 @@ typedef struct IOMMUNotifier IOMMUNotifier;
 /* RAM can be migrated */
 #define RAM_MIGRATABLE (1 << 4)
 
+/* RAM is a persistent kind memory */
+#define RAM_PMEM (1 << 5)
+
 static inline void iommu_notifier_init(IOMMUNotifier *n, IOMMUNotify fn,
                                        IOMMUNotifierFlag flags,
                                        hwaddr start, hwaddr end)
@@ -613,6 +616,7 @@ void memory_region_init_resizeable_ram(MemoryRegion *mr,
  *         (getpagesize()) will be used.
  * @ram_flags: Memory region features:
  *             - RAM_SHARED: memory must be mmaped with the MAP_SHARED flag
+ *             - RAM_PMEM: the memory is persistent memory
  *             Other bits are ignored now.
  * @path: the path in which to allocate the RAM.
  * @errp: pointer to Error*, to store an error if it happens.
diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h
index 5545cf3..3c65643 100644
--- a/include/exec/ram_addr.h
+++ b/include/exec/ram_addr.h
@@ -70,6 +70,8 @@ static inline unsigned long int ramblock_recv_bitmap_offset(void *host_addr,
     return host_addr_offset >> TARGET_PAGE_BITS;
 }
 
+bool ramblock_is_pmem(RAMBlock *rb);
+
 long qemu_getrampagesize(void);
 unsigned long last_ram_page(void);
 
@@ -84,6 +86,7 @@ unsigned long last_ram_page(void);
  *  @ram_flags: specify the properties of the ram block, which can be one
  *              or bit-or of following values
  *              - RAM_SHARED: mmap the backing file or device with MAP_SHARED
+ *              - RAM_PMEM: the backend @mem_path or @fd is persistent memory
  *              Other bits are ignored.
  *  @mem_path or @fd: specify the backing file or device
  *  @errp: pointer to Error*, to store an error if it happens
diff --git a/qemu-options.hx b/qemu-options.hx
index c0d3951..006ca44 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -4045,6 +4045,13 @@ requires an alignment different than the default one used by QEMU, eg
 the device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In
 such cases, users can specify the required alignment via this option.
 
+The @option{pmem} option specifies whether the backing file specified
+by @option{mem-path} is on the persistent memory that can be accessed
+using the SNIA NVM programming model (e.g. Intel NVDIMM).
+If @option{pmem}, QEMU will take necessary operations to
+guarantee the persistence of its own writes to @option{mem-path}
+(e.g. in vNVDIMM label emulation and live migration).
+
 @item -object memory-backend-ram,id=@var{id},merge=@var{on|off},dump=@var{on|off},share=@var{on|off},prealloc=@var{on|off},size=@var{size},host-nodes=@var{host-nodes},policy=@var{default|preferred|bind|interleave}
 
 Creates a memory backend object, which can be used to back the guest RAM.
-- 
2.7.4