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

From: Igor Mammedov <imammedo@redhat.com>
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, Junyan He <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
Date: Fri, 15 Jun 2018 11:03:52 +0200	[thread overview]
Message-ID: <20180615110352.4e65fe69@redhat.com> (raw)
In-Reply-To: <1528784900-7310-4-git-send-email-junyan.he@gmx.com>

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.