support MAP_SYNC for memory-backend-file
v9 cover support MAP_SYNC for memory-backend-file
Linux 4.15 introduces a new mmap flag MAP_SYNC, which can be used to guarantee the write persistence to mmap'ed files supporting DAX (e.g., files on ext4/xfs file system mounted with '-o dax').
A description of MAP_SYNC and MAP_SHARED_VALIDATE can be found at
https://patchwork.kernel.org/patch/10028151/
In order to make sure that the file metadata is in sync after a fault while we are writing a shared DAX supporting backend files, this patch-set enables QEMU to use MAP_SYNC flag for memory-backend-dax-file.
As the DAX vs DMA truncated issue was solved, we refined the code and send out this feature for the v5 version.
We will pass MAP_SYNC to mmap(2); if MAP_SYNC is supported and 'share=on' & 'pmem=on'.
Or QEMU will not pass this flag to mmap(2)
Changes in V9:
* 1/6: Reviewed-by: Eduardo Habkost <[email protected]>
* 2/6: New Added: Micheal: use sparse feature define RAM_FLAG. since I don't have much knowledge about the sparse feature, @Micheal Could you add some documentation/commit message on this patch? Thank you very much.
* 3/6: from 2/5: Eduardo: updated the commit message.
* 4/6: from 3/5: Micheal: don't ignore MAP_SYNC failures silently.
* 5/6: from 4/5: Eduardo: updated the commit message.
* 6/6: from 5/5: Micheal: Drop the sync option, document the MAP_SYNC.
Changes in v8:
* Micheal: 3/5, remove the duplicated define in the os_dep.h
* Micheal: 2/5, make type define safety.
* Micheal: 2/5, fixed the incorrect define MAP_SHARE on qemu_anon_ram_alloc.
* 4/6 removed, we remove the on/off/auto define of sync, as by now, MAP_SYNC only worked with pmem=on.
* @Micheal, I still reuse the RAM_SYNC flag, it is much straightforward to parse all the flags in one parameter.
Changes in v7:
* Micheal: [3,4,6]/6 limited the "sync" flag only on a nvdimm backend.(pmem=on)
Changes in v6:
* Pankaj: 3/7 are squashed with 2/7
* Pankaj: 7/7 update comments to "consistent filesystem metadata".
* Pankaj, Igor: 1/7 Added Reviewed-by in patch-1/7
* Stefan, 4/7 move the include header from "/linux/mman.h" to "osdep.h"
* Stefan, 5/7 Add missing "munmap"
* Stefan, 2/7 refine the shared/flag.
Changes in v5:
* Add patch 1 to fix a memory leak issue.
* Refine the patch 4-6
* Remove the patch 3 as we already change the parameter from "shared" to "flags"
Changes in v4:
* Add patch 1-3 to switch some functions to a single 'flags' parameters. (Michael S. Tsirkin)
* v3 patch 1-3 become v4 patch 4-6.
* Patch 4: move definitions of MAP_SYNC and MAP_SHARED_VALIDATE to a new header file under include/standard-headers/linux/. (Michael S. Tsirkin)
* Patch 6: refine the description of the 'sync' option. (Michael S. Tsirkin)
Changes in v3:
* Patch 1: add MAP_SHARED_VALIDATE in both sync=on and sync=auto cases, and add back the retry mechanism. MAP_SYNC will be ignored by Linux kernel 4.15 if MAP_SHARED_VALIDATE is missed.
* Patch 1: define MAP_SYNC and MAP_SHARED_VALIDATE as 0 on non-Linux platforms in order to make qemu_ram_mmap() compile on those platforms.
* Patch 2&3: include more information in error messages of memory-backend in hope to help user to identify the error. (Dr. David Alan Gilbert)
* Patch 3: fix typo in the commit message. (Dr. David Alan Gilbert)
Changes in v2:
* Add 'sync' option to control the use of MAP_SYNC. (Eduardo Habkost)
* Remove the unnecessary set of MAP_SHARED_VALIDATE in some cases and the retry mechanism in qemu_ram_mmap(). (Michael S. Tsirkin)
* Move OS dependent definitions of MAP_SYNC and MAP_SHARED_VALIDATE to osdep.h. (Michael S. Tsirkin)
Zhang Yi (6):
numa: Fixed the memory leak of numa error message
memory: use sparse feature define RAM_FLAG. util/mmap-alloc: switch 'shared' to 'flags' parameter util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
hostmem: add more information in error messages
docs: Added MAP_SYNC documentation
backends/hostmem-file.c | 6 ++++-- backends/hostmem.c | 8 +++++--- docs/nvdimm.txt | 21 ++++++++++++++++++++- exec.c | 7 ++++--- include/exec/memory.h | 12 ++++++------ include/qemu/mmap-alloc.h | 20 +++++++++++++++++++- include/qemu/osdep.h | 25 +++++++++++++++++++++++++ numa.c | 1 + qemu-options.hx | 4 ++++ util/mmap-alloc.c | 15 +++++++++++---- util/oslib-posix.c | 9 ++++++++- 11 files changed, 107 insertions(+), 21 deletions(-)
Signature · 1 line
2.7.4
PATCH v9 1/6 numa: Fixed the memory leak of numa error message
object_get_canonical_path_component() returns a string which must be freed using g_free().
--- numa.c | 1 + 1 file changed, 1 insertion(+)
numa.c +1 -0
@@ -533,6 +533,7 @@ void memory_region_allocate_system_memory(MemoryRegion *mr, Object *owner,error_report ("memory backend %s is used multiple times. Each " "-numa option must use a different memdev value." , path); +g_free (path);exit (1 ); }
Signature · 1 line
2.7.4
PATCH v9 2/6 memory: use sparse feature define RAM_FLAG.
In reply to Zhang Yi--- include/exec/memory.h | 12 ++++++------ include/qemu/osdep.h | 9 +++++++++ 2 files changed, 15 insertions(+), 6 deletions(-)
include/exec/memory.h +6 -6
@@ -104,27 +104,27 @@ struct IOMMUNotifier {typedef struct IOMMUNotifier IOMMUNotifier ;/* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */ -#define RAM_PREALLOC (1 << 0) +#define RAM_PREALLOC ((QEMU_FORCE QemuMmapFlags) (1 << 0))/* RAM is mmap-ed with MAP_SHARED */ -#define RAM_SHARED (1 << 1) +#define RAM_SHARED ((QEMU_FORCE QemuMmapFlags) (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) +#define RAM_RESIZEABLE ((QEMU_FORCE QemuMmapFlags) (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) +#define RAM_UF_ZEROPAGE ((QEMU_FORCE QemuMmapFlags) (1 << 3))/* RAM can be migrated */ -#define RAM_MIGRATABLE (1 << 4) +#define RAM_MIGRATABLE ((QEMU_FORCE QemuMmapFlags) (1 << 4))/* RAM is a persistent kind memory */ -#define RAM_PMEM (1 << 5) +#define RAM_PMEM ((QEMU_FORCE QemuMmapFlags) (1 << 5)) static inline void iommu_notifier_init(IOMMUNotifier *n, IOMMUNotify fn, IOMMUNotifierFlag flags,
include/qemu/osdep.h +9 -0
@@ -185,6 +185,15 @@ extern int daemon(int, int);#define ESHUTDOWN 4099#endif +#ifdef __CHECKER__ +#define QEMU_BITWISE __attribute__((bitwise)) +#define QEMU_FORCE __attribute__((force)) +#else +#define QEMU_BITWISE +#define QEMU_FORCE +#endif + +typedef unsigned QEMU_BITWISE QemuMmapFlags ;/* time_t may be either 32 or 64 bits depending on the host OS, and * can be either signed or unsigned, so we can't just hardcode a * specific maximum value. This is not a C preprocessor constant,
Signature · 1 line
2.7.4
PATCH v9 3/6 util/mmap-alloc: switch 'shared' to 'flags' parameter
In reply to Zhang YiAs more flag parameters besides the existing 'shared' are going to be added to qemu_ram_mmap() and qemu_ram_alloc_from_{file,fd}(), let's switch 'shared' to a 'flags' parameter in advance, so as to ease the further additions.
--- exec.c | 7 ++++--- include/qemu/mmap-alloc.h | 19 ++++++++++++++++++- util/mmap-alloc.c | 8 +++++--- util/oslib-posix.c | 9 ++++++++- 4 files changed, 35 insertions(+), 8 deletions(-)
exec.c +4 -3
@@ -1810,6 +1810,7 @@ static void *file_ram_alloc(RAMBlock *block,ram_addr_t memory, int fd, bool truncate, + uint32_t flags, Error **errp) { void *area;
@@ -1859,8 +1860,7 @@ static void *file_ram_alloc(RAMBlock *block,perror ("ftruncate" ); } - area =qemu_ram_mmap (fd, memory, block->mr->align, - block->flags &RAM_SHARED ); + area =qemu_ram_mmap (fd, memory, block->mr->align, flags);if (area ==MAP_FAILED ) {error_setg_errno (errp, errno,"unable to map backing store for guest RAM" );
@@ -2279,7 +2279,8 @@ RAMBlock *qemu_ram_alloc_from_fd(ram_addr_t size, MemoryRegion *mr, new_block->used_length = size; new_block->max_length = size; new_block->flags = ram_flags; - new_block->host =file_ram_alloc (new_block, size, fd, !file_size, errp); + new_block->host =file_ram_alloc (new_block, size, fd, !file_size, + ram_flags, errp);if (!new_block->host) {g_free (new_block);return NULL ;
include/qemu/mmap-alloc.h +18 -1
@@ -7,7 +7,24 @@ size_t qemu_fd_getpagesize(int fd);size_t qemu_mempath_getpagesize (const char *mem_path); -void *qemu_ram_mmap (int fd,size_t size,size_t align,bool shared); +/** +* qemu_ram_mmap: mmap the specified file or device. +* +* Parameters: +* @fd: the file or the device to mmap +* @size: the number of bytes to be mmaped +* @align: if not zero, specify the alignment of the starting mapping address; +* otherwise, the alignment in use will be determined by QEMU. +* @flags: specifies additional properties of the mapping, which can be one or +* bit-or of following values +* - RAM_SHARED: mmap with MAP_SHARED flag +* Other bits are ignored. +* +* Return: +* On success, return a pointer to the mapped area. +* On failure, return MAP_FAILED. +*/ +void *qemu_ram_mmap (int fd,size_t size,size_t align,uint32_t flags);void qemu_ram_munmap (void *ptr,size_t size);
util/mmap-alloc.c +5 -3
@@ -13,6 +13,7 @@#include "qemu/osdep.h" #include "qemu/mmap-alloc.h" #include "qemu/host-utils.h" +#include "exec/memory.h" #define HUGETLBFS_MAGIC 0x958458f6
@@ -75,7 +76,7 @@ size_t qemu_mempath_getpagesize(const char *mem_path)return getpagesize (); } -void *qemu_ram_mmap (int fd,size_t size,size_t align,bool shared) +void *qemu_ram_mmap (int fd,size_t size,size_t align,uint32_t flags) {/* * Note: this always allocates at least one extra page of virtual address
@@ -92,11 +93,12 @@ void *qemu_ram_mmap(int fd, size_t size, size_t align, bool shared)* anonymous memory is OK. */ int anonfd = fd ==-1 ||qemu_fd_getpagesize (fd) ==getpagesize () ?-1 : fd; -int flags = anonfd ==-1 ?MAP_ANONYMOUS :MAP_NORESERVE ; -void *ptr =mmap (0 , total,PROT_NONE , flags |MAP_PRIVATE , anonfd,0 ); +int mmap_flags = anonfd ==-1 ?MAP_ANONYMOUS :MAP_NORESERVE ; +void *ptr =mmap (0 , total,PROT_NONE , mmap_flags |MAP_PRIVATE , anonfd,0 );#else void *ptr =mmap (0 , total,PROT_NONE ,MAP_ANONYMOUS |MAP_PRIVATE ,-1 ,0 );#endif +bool shared = flags &RAM_SHARED ;size_t offset;void *ptr1;
util/oslib-posix.c +8 -1
@@ -54,6 +54,7 @@#endif #include "qemu/mmap-alloc.h" +#include "exec/memory.h" #ifdef CONFIG_DEBUG_STACK_USAGE #include "qemu/error-report.h"
@@ -203,7 +204,13 @@ void *qemu_memalign(size_t alignment, size_t size)void *qemu_anon_ram_alloc (size_t size,uint64_t *alignment,bool shared) {size_t align =QEMU_VMALLOC_ALIGN ; -void *ptr =qemu_ram_mmap (-1 , size, align, shared); +uint32_t flags =0 ; +void *ptr; + +if (shared) { + flags =RAM_SHARED ; + } + ptr =qemu_ram_mmap (-1 , size, align, flags);if (ptr ==MAP_FAILED ) {return NULL ;
Signature · 1 line
2.7.4
PATCH v9 6/6 docs: Added MAP_SYNC documentation
In reply to Zhang Yi--- docs/nvdimm.txt | 21 ++++++++++++++++++++- qemu-options.hx | 4 ++++ 2 files changed, 24 insertions(+), 1 deletion(-)
docs/nvdimm.txt +20 -1
@@ -142,11 +142,30 @@ backend of vNVDIMM: Guest Data Persistence ---------------------- +vNVDIMM is designed and implemented to guarantee the guest data +persistence on the backends even on the host crash and power +failures. However, there are still some requirements and limitations +as explained below. + Though QEMU supports multiple types of vNVDIMM backends on Linux, -currently the only one that can guarantee the guest write persistence +if MAP_SYNC is not supported by the host kernel and the backends, +the only backend that can guarantee the guest write persistence is the device DAX on the real NVDIMM device (e.g., /dev/dax0.0), to which all guest access do not involve any host-side kernel cache. +mmap(2) flag MAP_SYNC is added since Linux kernel 4.15. On such +systems, QEMU can mmap(2) the backend with MAP_SYNC, which can ensure +filesystem metadata consistent even after a system crash or power +failure. Besides the host kernel support, enabling MAP_SYNC in QEMU +also requires: + + - the backend is a file supporting DAX, e.g., a file on an ext4 or + xfs file system mounted with '-o dax', + + - 'share' option of memory-backend-file is 'on'. + + - 'pmem' option of memory-backend-file is 'on' + When using other types of backends, it's suggested to set 'unarmed' option of '-device nvdimm' to 'on', which sets the unarmed flag of the guest NVDIMM region mapping structure. This unarmed flag indicates
qemu-options.hx +4 -0
@@ -4002,6 +4002,10 @@ using the SNIA NVM programming model (e.g. Intel NVDIMM). If @option{pmem} is set to 'on', 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). +Also, we will map the backend-file with MAP_SYNC flag, which can ensure +the file metadata is in sync to @option{mem-path} even on the host crash +and power failures. MAP_SYNC requires supports from both the host kernel +(since Linux kernel 4.15) and @option{mem-path} (only files supporting DAX). @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}
Signature · 1 line
2.7.4
PATCH v9 4/6 util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
In reply to Zhang YiWhen a file supporting DAX is used as vNVDIMM backend, mmap it with MAP_SYNC flag in addition which can ensure file system metadata synced in each guest writes to the backend file, without other QEMU actions (e.g., periodic fsync() by QEMU).
--- include/qemu/mmap-alloc.h | 1 + include/qemu/osdep.h | 16 ++++++++++++++++ util/mmap-alloc.c | 7 ++++++- 3 files changed, 23 insertions(+), 1 deletion(-)
include/qemu/mmap-alloc.h +1 -0
@@ -18,6 +18,7 @@ size_t qemu_mempath_getpagesize(const char *mem_path);* @flags: specifies additional properties of the mapping, which can be one or * bit-or of following values * - RAM_SHARED: mmap with MAP_SHARED flag +* - RAM_PMEM: mmap with MAP_SYNC flag * Other bits are ignored. * * Return:
include/qemu/osdep.h +16 -0
@@ -419,6 +419,22 @@ void qemu_anon_ram_free(void *ptr, size_t size);# define QEMU_VMALLOC_ALIGN getpagesize()#endif +/* +* MAP_SHARED_VALIDATE and MAP_SYNC are introduced in Linux kernel +* 4.15, so they may not be defined when compiling on older kernels. +*/ +#ifdef CONFIG_LINUX + +#include <asm-generic/mman.h> + +#ifndef MAP_SYNC +#define MAP_SYNC 0x0 +#endif + +#else /* !CONFIG_LINUX */ +#define MAP_SYNC 0x0 +#endif /* CONFIG_LINUX */ + #ifdef CONFIG_POSIX struct qemu_signalfd_siginfo { uint32_t ssi_signo; /* Signal number */
util/mmap-alloc.c +6 -1
@@ -99,6 +99,8 @@ void *qemu_ram_mmap(int fd, size_t size, size_t align, uint32_t flags)void *ptr =mmap (0 , total,PROT_NONE ,MAP_ANONYMOUS |MAP_PRIVATE ,-1 ,0 );#endif bool shared = flags &RAM_SHARED ; +bool is_pmem = flags &RAM_PMEM ; +int mmap_xflags =0 ;size_t offset;void *ptr1;
@@ -109,12 +111,15 @@ void *qemu_ram_mmap(int fd, size_t size, size_t align, uint32_t flags)assert (is_power_of_2 (align));/* Always align to host page size */ assert (align >=getpagesize ()); +if (shared && is_pmem) { + mmap_xflags |=MAP_SYNC ; + } offset =QEMU_ALIGN_UP ((uintptr_t )ptr, align) - (uintptr_t )ptr; ptr1 =mmap (ptr + offset, size,PROT_READ |PROT_WRITE ,MAP_FIXED | (fd ==-1 ?MAP_ANONYMOUS :0 ) | - (shared ?MAP_SHARED :MAP_PRIVATE ), + (shared ?MAP_SHARED :MAP_PRIVATE ) | mmap_xflags, fd,0 );if (ptr1 ==MAP_FAILED ) {munmap (ptr, total);
Signature · 1 line
2.7.4
PATCH v9 5/6 hostmem: add more information in error messages
In reply to Zhang YiWhen there are multiple memory backends in use, including the object type name and the property name in the error message can help users to locate the error.
--- backends/hostmem-file.c | 6 ++++-- backends/hostmem.c | 8 +++++--- 2 files changed, 9 insertions(+), 5 deletions(-)
backends/hostmem-file.c +4 -2
@@ -82,7 +82,8 @@ static void set_mem_path(Object *o, const char *str, Error **errp)HostMemoryBackendFile *fb =MEMORY_BACKEND_FILE (o);if (host_memory_backend_mr_inited (backend)) { -error_setg (errp,"cannot change property value" ); +error_setg (errp,"cannot change property 'mem-path' of %s" , +object_get_typename (o));return ; }g_free (fb->mem_path);
@@ -120,7 +121,8 @@ static void file_memory_backend_set_align(Object *o, Visitor *v,uint64_t val;if (host_memory_backend_mr_inited (backend)) { -error_setg (&local_err,"cannot change property value" ); +error_setg (&local_err,"cannot change property '%s' of %s" , + name,object_get_typename (o)); goto out; }
backends/hostmem.c +5 -3
@@ -47,7 +47,8 @@ host_memory_backend_set_size(Object *obj, Visitor *v, const char *name,uint64_t value;if (host_memory_backend_mr_inited (backend)) { -error_setg (&local_err,"cannot change property value" ); +error_setg (&local_err,"cannot change property %s of %s " , + name,object_get_typename (obj)); goto out; }
@@ -56,8 +57,9 @@ host_memory_backend_set_size(Object *obj, Visitor *v, const char *name, goto out; }if (!value) { -error_setg (&local_err,"Property '%s.%s' doesn't take value '%" - PRIu64"'" ,object_get_typename (obj), name, value); +error_setg (&local_err, +"property '%s' of %s doesn't take value '%" PRIu64"'" , + name,object_get_typename (obj), value); goto out; } backend->size = value;
Signature · 1 line
2.7.4
Re: [Qemu-devel] [PATCH V9 5/6] hostmem: add more information in error messages
Quoted 5 lines · Zhang Yi
When there are multiple memory backends in use, including the object type name and the property name in the error message can help users to locate the error.
Quoted 8 lines
Signed-off-by Haozhong Zhang <[email protected]>Signed-off-by Zhang Yi <[email protected]>Reviewed-by Eduardo Habkost <[email protected]>--- backends/hostmem-file.c | 6 ++++-- backends/hostmem.c | 8 +++++--- 2 files changed, 9 insertions(+), 5 deletions(-)
Re: [Qemu-devel] [PATCH V9 3/6] util/mmap-alloc: switch 'shared' to 'flags' parameter
In reply to Zhang YiQuoted 19 lines · Zhang Yi
As more flag parameters besides the existing 'shared' are going to be added to
qemu_ram_mmap()and qemu_ram_alloc_from_{file,fd}(), let's switch 'shared' to a 'flags' parameter in advance, so as to ease the further additions.Signed-off-by Haozhong Zhang <[email protected]>Signed-off-by Zhang Yi <[email protected]>--- exec.c | 7 ++++--- include/qemu/mmap-alloc.h | 19 ++++++++++++++++++- util/mmap-alloc.c | 8 +++++--- util/oslib-posix.c | 9 ++++++++- 4 files changed, 35 insertions(+), 8 deletions(-)exec.c +0
-0
exec.c static void *file_ram_alloc(RAMBlock *block, · line 1816
Quoted 7 lines
exec.c +1
-0@@ -1810,6 +1810,7 @@ static void *file_ram_alloc(RAMBlock *block,ram_addr_t memory, int fd, bool truncate, + uint32_t flags, Error **errp) { void *area;
But given that flags is an un-annotated uint32_t does sparse actually do any checks?
I'm lost ...
Quoted 115 lines
exec.c +3
-3@@ -1859,8 +1860,7 @@ static void *file_ram_alloc(RAMBlock *block,perror ("ftruncate" ); } - area =qemu_ram_mmap (fd, memory, block->mr->align, - block->flags &RAM_SHARED ); + area =qemu_ram_mmap (fd, memory, block->mr->align, flags);if (area ==MAP_FAILED ) {error_setg_errno (errp, errno,"unable to map backing store for guest RAM" );@@ -2279,7 +2279,8 @@ RAMBlock *qemu_ram_alloc_from_fd(ram_addr_t size, MemoryRegion *mr, new_block->used_length = size; new_block->max_length = size; new_block->flags = ram_flags; - new_block->host =file_ram_alloc (new_block, size, fd, !file_size, errp); + new_block->host =file_ram_alloc (new_block, size, fd, !file_size, + ram_flags, errp);if (!new_block->host) {g_free (new_block);return NULL ;include/qemu/mmap-alloc.h +18
-1@@ -7,7 +7,24 @@ size_t qemu_fd_getpagesize(int fd);size_t qemu_mempath_getpagesize (const char *mem_path); -void *qemu_ram_mmap (int fd,size_t size,size_t align,bool shared); +/** +* qemu_ram_mmap: mmap the specified file or device. +* +* Parameters: +* @fd: the file or the device to mmap +* @size: the number of bytes to be mmaped +* @align: if not zero, specify the alignment of the starting mapping address; +* otherwise, the alignment in use will be determined by QEMU. +* @flags: specifies additional properties of the mapping, which can be one or +* bit-or of following values +* - RAM_SHARED: mmap with MAP_SHARED flag +* Other bits are ignored. +* +* Return: +* On success, return a pointer to the mapped area. +* On failure, return MAP_FAILED. +*/ +void *qemu_ram_mmap (int fd,size_t size,size_t align,uint32_t flags);void qemu_ram_munmap (void *ptr,size_t size);util/mmap-alloc.c +5
-3@@ -13,6 +13,7 @@#include "qemu/osdep.h" #include "qemu/mmap-alloc.h" #include "qemu/host-utils.h" +#include "exec/memory.h" #define HUGETLBFS_MAGIC 0x958458f6@@ -75,7 +76,7 @@ size_t qemu_mempath_getpagesize(const char *mem_path)return getpagesize (); } -void *qemu_ram_mmap (int fd,size_t size,size_t align,bool shared) +void *qemu_ram_mmap (int fd,size_t size,size_t align,uint32_t flags) {/* * Note: this always allocates at least one extra page of virtual address @@ -92,11 +93,12 @@ void *qemu_ram_mmap(int fd, size_t size, size_t align, bool shared)* anonymous memory is OK. */ int anonfd = fd ==-1 ||qemu_fd_getpagesize (fd) ==getpagesize () ?-1 : fd; -int flags = anonfd ==-1 ?MAP_ANONYMOUS :MAP_NORESERVE ; -void *ptr =mmap (0 , total,PROT_NONE , flags |MAP_PRIVATE , anonfd,0 ); +int mmap_flags = anonfd ==-1 ?MAP_ANONYMOUS :MAP_NORESERVE ; +void *ptr =mmap (0 , total,PROT_NONE , mmap_flags |MAP_PRIVATE , anonfd,0 );#else void *ptr =mmap (0 , total,PROT_NONE ,MAP_ANONYMOUS |MAP_PRIVATE ,-1 ,0 );#endif +bool shared = flags &RAM_SHARED ;size_t offset;void *ptr1;util/oslib-posix.c +8
-1@@ -54,6 +54,7 @@#endif #include "qemu/mmap-alloc.h" +#include "exec/memory.h" #ifdef CONFIG_DEBUG_STACK_USAGE #include "qemu/error-report.h" @@ -203,7 +204,13 @@ void *qemu_memalign(size_t alignment, size_t size)void *qemu_anon_ram_alloc (size_t size,uint64_t *alignment,bool shared) {size_t align =QEMU_VMALLOC_ALIGN ; -void *ptr =qemu_ram_mmap (-1 , size, align, shared); +uint32_t flags =0 ; +void *ptr; + +if (shared) { + flags =RAM_SHARED ; + } + ptr =qemu_ram_mmap (-1 , size, align, flags);if (ptr ==MAP_FAILED ) {return NULL ;Signature · 1 line
2.7.4
Re: [Qemu-devel] [PATCH V9 2/6] memory: use sparse feature define RAM_FLAG.
In reply to Zhang YiQuoted 2 lines · Zhang Yi
Signed-off-by Zhang Yi <[email protected]>Signed-off-by Michael S. Tsirkin <[email protected]>
OK so if you apply this patch, do you get
any sparse warning at all?
If not how come?
And we need to ask patchew maintainers to
Overall I'd suggest that since you no longer need
a new RAM flag, split this effort out and finishMAP_SYNC work first.
Quoted 65 lines
--- include/exec/memory.h | 12 ++++++------ include/qemu/osdep.h | 9 +++++++++ 2 files changed, 15 insertions(+), 6 deletions(-)include/exec/memory.h +6
-6@@ -104,27 +104,27 @@ struct IOMMUNotifier {typedef struct IOMMUNotifier IOMMUNotifier ;/* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */ -#define RAM_PREALLOC (1 << 0) +#define RAM_PREALLOC ((QEMU_FORCE QemuMmapFlags) (1 << 0))/* RAM is mmap-ed with MAP_SHARED */ -#define RAM_SHARED (1 << 1) +#define RAM_SHARED ((QEMU_FORCE QemuMmapFlags) (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) +#define RAM_RESIZEABLE ((QEMU_FORCE QemuMmapFlags) (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) +#define RAM_UF_ZEROPAGE ((QEMU_FORCE QemuMmapFlags) (1 << 3))/* RAM can be migrated */ -#define RAM_MIGRATABLE (1 << 4) +#define RAM_MIGRATABLE ((QEMU_FORCE QemuMmapFlags) (1 << 4))/* RAM is a persistent kind memory */ -#define RAM_PMEM (1 << 5) +#define RAM_PMEM ((QEMU_FORCE QemuMmapFlags) (1 << 5)) static inline void iommu_notifier_init(IOMMUNotifier *n, IOMMUNotify fn, IOMMUNotifierFlag flags,include/qemu/osdep.h +9
-0@@ -185,6 +185,15 @@ extern int daemon(int, int);#define ESHUTDOWN 4099#endif +#ifdef __CHECKER__ +#define QEMU_BITWISE __attribute__((bitwise)) +#define QEMU_FORCE __attribute__((force)) +#else +#define QEMU_BITWISE +#define QEMU_FORCE +#endif + +typedef unsigned QEMU_BITWISE QemuMmapFlags ;/* time_t may be either 32 or 64 bits depending on the host OS, and * can be either signed or unsigned, so we can't just hardcode a * specific maximum value. This is not a C preprocessor constant, Signature · 1 line
2.7.4
Re: [Qemu-devel] [PATCH V9 1/6] numa: Fixed the memory leak of numa error message
In reply to Zhang YiQuoted 7 lines · Zhang Yi
object_get_canonical_path_component()returns a string which must be freed usingg_free().Signed-off-by Zhang Yi <[email protected]>Reviewed-by Pankaj gupta <[email protected]>Reviewed-by Igor Mammedov <[email protected]>Reviewed-by Eduardo Habkost <[email protected]>
Pls post this separately as it's just a bugfix that is not a dependency of map_sync.
Quoted 18 lines
--- numa.c | 1 + 1 file changed, 1 insertion(+)numa.c +1
-0@@ -533,6 +533,7 @@ void memory_region_allocate_system_memory(MemoryRegion *mr, Object *owner,error_report ("memory backend %s is used multiple times. Each " "-numa option must use a different memdev value." , path); +g_free (path);exit (1 ); }Signature · 1 line
2.7.4
Re: [Qemu-devel] [PATCH V9 4/6] util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
In reply to Zhang YiQuoted 30 lines · Zhang Yi
When a file supporting DAX is used as vNVDIMM backend, mmap it with
MAP_SYNCflag in addition which can ensure file system metadata synced in each guest writes to the backend file, without other QEMU actions (e.g., periodicfsync()by QEMU).Signed-off-by Haozhong Zhang <[email protected]>Signed-off-by Zhang Yi <[email protected]>--- include/qemu/mmap-alloc.h | 1 + include/qemu/osdep.h | 16 ++++++++++++++++ util/mmap-alloc.c | 7 ++++++- 3 files changed, 23 insertions(+), 1 deletion(-)include/qemu/mmap-alloc.h +1
-0@@ -18,6 +18,7 @@ size_t qemu_mempath_getpagesize(const char *mem_path);* @flags: specifies additional properties of the mapping, which can be one or * bit-or of following values * - RAM_SHARED: mmap with MAP_SHARED flag +* - RAM_PMEM: mmap with MAP_SYNC flag * Other bits are ignored. * * Return: include/qemu/osdep.h +0
-0
include/qemu/osdep.h void qemu_anon_ram_free(void *ptr, size_t size); · line 428
Quoted 10 lines
include/qemu/osdep.h +7
-0@@ -419,6 +419,22 @@ void qemu_anon_ram_free(void *ptr, size_t size);# define QEMU_VMALLOC_ALIGN getpagesize()#endif +/* +* MAP_SHARED_VALIDATE and MAP_SYNC are introduced in Linux kernel +* 4.15, so they may not be defined when compiling on older kernels. +*/ +#ifdef CONFIG_LINUX + +#include <asm-generic/mman.h>
I suspect this is a wrong way to pull in this header.
You are normally supposed to use
#include <linux/mman.h>
but see below.
include/qemu/osdep.h void qemu_anon_ram_free(void *ptr, size_t size); · line 432
Quoted 4 lines
include/qemu/osdep.h +4
-0+ +#ifndef MAP_SYNC +#define MAP_SYNC 0x0 +#endif
Oh that's bad.
So if you run with a new kernel but
your installed headers are old, you get MAP_SYNC 0 and no persistence transparently with no warning.
Quoted 40 lines
include/qemu/osdep.h +5
-0+ +#else /* !CONFIG_LINUX */ +#define MAP_SYNC 0x0 +#endif /* CONFIG_LINUX */ + #ifdef CONFIG_POSIX struct qemu_signalfd_siginfo { uint32_t ssi_signo; /* Signal number */util/mmap-alloc.c +6
-1@@ -99,6 +99,8 @@ void *qemu_ram_mmap(int fd, size_t size, size_t align, uint32_t flags)void *ptr =mmap (0 , total,PROT_NONE ,MAP_ANONYMOUS |MAP_PRIVATE ,-1 ,0 );#endif bool shared = flags &RAM_SHARED ; +bool is_pmem = flags &RAM_PMEM ; +int mmap_xflags =0 ;size_t offset;void *ptr1;@@ -109,12 +111,15 @@ void *qemu_ram_mmap(int fd, size_t size, size_t align, uint32_t flags)assert (is_power_of_2 (align));/* Always align to host page size */ assert (align >=getpagesize ()); +if (shared && is_pmem) { + mmap_xflags |=MAP_SYNC ; + } offset =QEMU_ALIGN_UP ((uintptr_t )ptr, align) - (uintptr_t )ptr; ptr1 =mmap (ptr + offset, size,PROT_READ |PROT_WRITE ,MAP_FIXED | (fd ==-1 ?MAP_ANONYMOUS :0 ) | - (shared ?MAP_SHARED :MAP_PRIVATE ), + (shared ?MAP_SHARED :MAP_PRIVATE ) | mmap_xflags, fd,0 );if (ptr1 ==MAP_FAILED ) {munmap (ptr, total);Signature · 1 line
2.7.4
Re: [Qemu-devel] [PATCH V9 6/6] docs: Added MAP_SYNC documentation
In reply to Zhang YiQuoted 11 lines · Zhang Yi
Signed-off-by Zhang Yi <[email protected]>--- docs/nvdimm.txt | 21 ++++++++++++++++++++- qemu-options.hx | 4 ++++ 2 files changed, 24 insertions(+), 1 deletion(-)docs/nvdimm.txt +0
-0
docs/nvdimm.txt backend of vNVDIMM: · line 146
Quoted 5 lines
docs/nvdimm.txt +2
-0@@ -142,11 +142,30 @@ backend of vNVDIMM: Guest Data Persistence ---------------------- +vNVDIMM is designed and implemented to guarantee the guest data +persistence on the backends even on the host crash and power
in case of a host crash or a power failure
(be consistent)
docs/nvdimm.txt backend of vNVDIMM: · line 157
Quoted 12 lines
docs/nvdimm.txt +7
-1+failures. However, there are still some requirements and limitations +as explained below. + Though QEMU supports multiple types of vNVDIMM backends on Linux, -currently the only one that can guarantee the guest write persistence +if MAP_SYNC is not supported by the host kernel and the backends, +the only backend that can guarantee the guest write persistence is the device DAX on the real NVDIMM device (e.g., /dev/dax0.0), to which all guest access do not involve any host-side kernel cache. +mmap(2) flag MAP_SYNC is added since Linux kernel 4.15. On such +systems, QEMU can mmap(2) the backend with MAP_SYNC, which can ensure
which ensures
docs/nvdimm.txt backend of vNVDIMM: · line 158
Quoted 1 line
docs/nvdimm.txt +1
-0+filesystem metadata consistent
consistency
Quoted 1 line
even after a system crash or power
a power
docs/nvdimm.txt backend of vNVDIMM: · line 165
Quoted 7 lines
docs/nvdimm.txt +7
-0+failure. Besides the host kernel support, enabling MAP_SYNC in QEMU +also requires: + + - the backend is a file supporting DAX, e.g., a file on an ext4 or + xfs file system mounted with '-o dax', + + - 'share' option of memory-backend-file is 'on'.
why does share need to be on?
Quoted 11 lines
docs/nvdimm.txt +3
-0+ + - 'pmem' option of memory-backend-file is 'on' + When using other types of backends, it's suggested to set 'unarmed' option of '-device nvdimm' to 'on', which sets the unarmed flag of the guest NVDIMM region mapping structure. This unarmed flag indicatesqemu-options.hx +0
-0
qemu-options.hx using the SNIA NVM programming model (e.g. Intel NVDIMM). · line 4007
Quoted 6 lines
qemu-options.hx +3
-0@@ -4002,6 +4002,10 @@ using the SNIA NVM programming model (e.g. Intel NVDIMM). If @option{pmem} is set to 'on', 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). +Also, we will map the backend-file with MAP_SYNC flag, which can ensure +the file metadata is in sync to @option{mem-path} even on the host crash +and power failures. MAP_SYNC requires supports from both the host kernel
requires support
Quoted 6 lines
qemu-options.hx +1
-0+(since Linux kernel 4.15) and @option{mem-path} (only files supporting DAX). @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}Signature · 1 line
2.7.4
Re: [Qemu-devel] [PATCH V9 1/6] numa: Fixed the memory leak of numa error message
In reply to Michael S. TsirkinQuoted 9 lines · Michael S. Tsirkin
Quoted 7 lines · Zhang Yi
object_get_canonical_path_component()returns a string which must be freed usingg_free().Signed-off-by Zhang Yi <[email protected]>Reviewed-by Pankaj gupta <[email protected]>Reviewed-by Igor Mammedov <[email protected]>Reviewed-by Eduardo Habkost <[email protected]>
Quoted 4 lines
Reviewed-by Michael S. Tsirkin <[email protected]>Pls post this separately as it's just a bugfix that is not a dependency of
map_sync.
I have already queued it on machine-next, there's no need to resubmit.
Signature · 1 line
Eduardo
Re: [Qemu-devel] [PATCH V9 4/6] util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
In reply to Michael S. TsirkinQuoted 57 lines · Michael S. Tsirkin
Quoted 30 lines · Zhang Yi
When a file supporting DAX is used as vNVDIMM backend, mmap it with
MAP_SYNCflag in addition which can ensure file system metadata synced in each guest writes to the backend file, without other QEMU actions (e.g., periodicfsync()by QEMU).Signed-off-by Haozhong Zhang <[email protected]>Signed-off-by Zhang Yi <[email protected]>--- include/qemu/mmap-alloc.h | 1 + include/qemu/osdep.h | 16 ++++++++++++++++ util/mmap-alloc.c | 7 ++++++- 3 files changed, 23 insertions(+), 1 deletion(-)include/qemu/mmap-alloc.h +1
-0@@ -18,6 +18,7 @@ size_t qemu_mempath_getpagesize(const char *mem_path);* @flags: specifies additional properties of the mapping, which can be one or * bit-or of following values * - RAM_SHARED: mmap with MAP_SHARED flag +* - RAM_PMEM: mmap with MAP_SYNC flag * Other bits are ignored. * * Return: include/qemu/osdep.h +0
-0include/qemu/osdep.h void qemu_anon_ram_free(void *ptr, size_t size); · line 428
Quoted 10 lines
include/qemu/osdep.h +7
-0@@ -419,6 +419,22 @@ void qemu_anon_ram_free(void *ptr, size_t size);# define QEMU_VMALLOC_ALIGN getpagesize()#endif +/* +* MAP_SHARED_VALIDATE and MAP_SYNC are introduced in Linux kernel +* 4.15, so they may not be defined when compiling on older kernels. +*/ +#ifdef CONFIG_LINUX + +#include <asm-generic/mman.h> I suspect this is a wrong way to pull in this header.
You are normally supposed to use
#include <linux/mman.h>but see below.
include/qemu/osdep.h void qemu_anon_ram_free(void *ptr, size_t size); · line 432
Quoted 4 lines
include/qemu/osdep.h +4
-0+ +#ifndef MAP_SYNC +#define MAP_SYNC 0x0 +#endif Oh that's bad.
Quoted 3 lines
So if you run with a new kernel but
your installed headers are old, you getMAP_SYNC0 and no persistence transparently with no warning.
Yes. The semantics of the command-line to not change depending on build time circumstances.
Anyway, I see a more fundamental problem in each version of this
patch: the semantics of the command-line options are not clearly documented.
We have at least 3 different possible use cases we might need to support:
1) pmem=on, MAP_SYNC not desired
2) pmem=on, MAP_SYNC desired but optional
3) pmem=on, MAP_SYNC required, not optional
Which cases from the list above we need to support?
Quoted 1 line
From the cases above, what's the expected semantics of "pmem=on"
with no extra options?
If these questions are not answered (in the commit message and user documentation), we won't be able to review and discuss the code.
Quoted 42 lines
Quoted 41 lines
include/qemu/osdep.h +5
-0+ +#else /* !CONFIG_LINUX */ +#define MAP_SYNC 0x0 +#endif /* CONFIG_LINUX */ + #ifdef CONFIG_POSIX struct qemu_signalfd_siginfo { uint32_t ssi_signo; /* Signal number */util/mmap-alloc.c +6
-1@@ -99,6 +99,8 @@ void *qemu_ram_mmap(int fd, size_t size, size_t align, uint32_t flags)void *ptr =mmap (0 , total,PROT_NONE ,MAP_ANONYMOUS |MAP_PRIVATE ,-1 ,0 );#endif bool shared = flags &RAM_SHARED ; +bool is_pmem = flags &RAM_PMEM ; +int mmap_xflags =0 ;size_t offset;void *ptr1;@@ -109,12 +111,15 @@ void *qemu_ram_mmap(int fd, size_t size, size_t align, uint32_t flags)assert (is_power_of_2 (align));/* Always align to host page size */ assert (align >=getpagesize ()); +if (shared && is_pmem) { + mmap_xflags |=MAP_SYNC ; + } offset =QEMU_ALIGN_UP ((uintptr_t )ptr, align) - (uintptr_t )ptr; ptr1 =mmap (ptr + offset, size,PROT_READ |PROT_WRITE ,MAP_FIXED | (fd ==-1 ?MAP_ANONYMOUS :0 ) | - (shared ?MAP_SHARED :MAP_PRIVATE ), + (shared ?MAP_SHARED :MAP_PRIVATE ) | mmap_xflags, fd,0 );if (ptr1 ==MAP_FAILED ) {munmap (ptr, total);Signature · 1 line
2.7.4
Signature · 1 line
Eduardo
Re: [Qemu-devel] [PATCH V9 4/6] util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
Quoted 69 lines · Eduardo Habkost
Quoted 57 lines · Michael S. Tsirkin
Quoted 30 lines · Zhang Yi
When a file supporting DAX is used as vNVDIMM backend, mmap it with
MAP_SYNCflag in addition which can ensure file system metadata synced in each guest writes to the backend file, without other QEMU actions (e.g., periodicfsync()by QEMU).Signed-off-by Haozhong Zhang <[email protected]>Signed-off-by Zhang Yi <[email protected]>--- include/qemu/mmap-alloc.h | 1 + include/qemu/osdep.h | 16 ++++++++++++++++ util/mmap-alloc.c | 7 ++++++- 3 files changed, 23 insertions(+), 1 deletion(-)include/qemu/mmap-alloc.h +1
-0@@ -18,6 +18,7 @@ size_t qemu_mempath_getpagesize(const char *mem_path);* @flags: specifies additional properties of the mapping, which can be one or * bit-or of following values * - RAM_SHARED: mmap with MAP_SHARED flag +* - RAM_PMEM: mmap with MAP_SYNC flag * Other bits are ignored. * * Return: include/qemu/osdep.h +0
-0include/qemu/osdep.h void qemu_anon_ram_free(void *ptr, size_t size); · line 428
Quoted 10 lines
include/qemu/osdep.h +7
-0@@ -419,6 +419,22 @@ void qemu_anon_ram_free(void *ptr, size_t size);# define QEMU_VMALLOC_ALIGN getpagesize()#endif +/* +* MAP_SHARED_VALIDATE and MAP_SYNC are introduced in Linux kernel +* 4.15, so they may not be defined when compiling on older kernels. +*/ +#ifdef CONFIG_LINUX + +#include <asm-generic/mman.h> I suspect this is a wrong way to pull in this header.
You are normally supposed to use
#include <linux/mman.h>but see below.
include/qemu/osdep.h void qemu_anon_ram_free(void *ptr, size_t size); · line 432
Quoted 4 lines
include/qemu/osdep.h +4
-0+ +#ifndef MAP_SYNC +#define MAP_SYNC 0x0 +#endif Oh that's bad.
Quoted 3 lines
So if you run with a new kernel but
your installed headers are old, you getMAP_SYNC0 and no persistence transparently with no warning.Yes. The semantics of the command-line to not change depending on build time circumstances.
Anyway, I see a more fundamental problem in each version of this
patch: the semantics of the command-line options are not clearly documented.
Quoted 5 lines
We have at least 3 different possible use cases we might need to support:
1) pmem=on,
MAP_SYNCnot desired
2) pmem=on,MAP_SYNCdesired but optional
Form V9, As Michael suggest, We removed the sync option, MAP_SYNC will force on while we set pmem=on. So we only have 2 user cases, Will update to user documentation.
1) pmem=on, MAP_SYNC not desired
We will not pass the flag to mmap2
2) pmem=on, MAP_SYNC desired
We will pass the flag to mmap2
Quoted 58 lines
3) pmem=on,
MAP_SYNCrequired, not optionalWhich cases from the list above we need to support?
From the cases above, what's the expected semantics of "pmem=on" with no extra options?
If these questions are not answered (in the commit message and user documentation), we won't be able to review and discuss the code.
Quoted 42 lines
Quoted 41 lines
include/qemu/osdep.h +5
-0+ +#else /* !CONFIG_LINUX */ +#define MAP_SYNC 0x0 +#endif /* CONFIG_LINUX */ + #ifdef CONFIG_POSIX struct qemu_signalfd_siginfo { uint32_t ssi_signo; /* Signal number */util/mmap-alloc.c +6
-1@@ -99,6 +99,8 @@ void *qemu_ram_mmap(int fd, size_t size, size_t align, uint32_t flags)void *ptr =mmap (0 , total,PROT_NONE ,MAP_ANONYMOUS |MAP_PRIVATE ,-1 ,0 );#endif bool shared = flags &RAM_SHARED ; +bool is_pmem = flags &RAM_PMEM ; +int mmap_xflags =0 ;size_t offset;void *ptr1;@@ -109,12 +111,15 @@ void *qemu_ram_mmap(int fd, size_t size, size_t align, uint32_t flags)assert (is_power_of_2 (align));/* Always align to host page size */ assert (align >=getpagesize ()); +if (shared && is_pmem) { + mmap_xflags |=MAP_SYNC ; + } offset =QEMU_ALIGN_UP ((uintptr_t )ptr, align) - (uintptr_t )ptr; ptr1 =mmap (ptr + offset, size,PROT_READ |PROT_WRITE ,MAP_FIXED | (fd ==-1 ?MAP_ANONYMOUS :0 ) | - (shared ?MAP_SHARED :MAP_PRIVATE ), + (shared ?MAP_SHARED :MAP_PRIVATE ) | mmap_xflags, fd,0 );if (ptr1 ==MAP_FAILED ) {munmap (ptr, total);Signature · 1 line
2.7.4Signature · 1 line
Eduardo
Re: [Qemu-devel] [PATCH V9 6/6] docs: Added MAP_SYNC documentation
In reply to Michael S. TsirkinQuoted 45 lines · Michael S. Tsirkin
Quoted 11 lines · Zhang Yi
Signed-off-by Zhang Yi <[email protected]>--- docs/nvdimm.txt | 21 ++++++++++++++++++++- qemu-options.hx | 4 ++++ 2 files changed, 24 insertions(+), 1 deletion(-)docs/nvdimm.txt +0
-0docs/nvdimm.txt backend of vNVDIMM: · line 146
Quoted 5 lines
docs/nvdimm.txt +2
-0@@ -142,11 +142,30 @@ backend of vNVDIMM: Guest Data Persistence ---------------------- +vNVDIMM is designed and implemented to guarantee the guest data +persistence on the backends even on the host crash and powerin case of a host crash or a power failure
(be consistent)
docs/nvdimm.txt backend of vNVDIMM: · line 157
Quoted 12 lines
docs/nvdimm.txt +7
-1+failures. However, there are still some requirements and limitations +as explained below. + Though QEMU supports multiple types of vNVDIMM backends on Linux, -currently the only one that can guarantee the guest write persistence +if MAP_SYNC is not supported by the host kernel and the backends, +the only backend that can guarantee the guest write persistence is the device DAX on the real NVDIMM device (e.g., /dev/dax0.0), to which all guest access do not involve any host-side kernel cache. +mmap(2) flag MAP_SYNC is added since Linux kernel 4.15. On such +systems, QEMU can mmap(2) the backend with MAP_SYNC, which can ensurewhich ensures
docs/nvdimm.txt backend of vNVDIMM: · line 158
Quoted 1 line
docs/nvdimm.txt +1
-0+filesystem metadata consistentconsistency
Quoted 1 line
even after a system crash or power
a power
Quoted 9 lines
docs/nvdimm.txt backend of vNVDIMM: · line 165
Quoted 7 lines
docs/nvdimm.txt +7
-0+failure. Besides the host kernel support, enabling MAP_SYNC in QEMU +also requires: + + - the backend is a file supporting DAX, e.g., a file on an ext4 or + xfs file system mounted with '-o dax', + + - 'share' option of memory-backend-file is 'on'.why does share need to be on?
According to mmap(2) man page,MAP_SYNC is available only with the MAP_SHARED_VALIDATE map‐ping type. I will add this to documentation, Thanks.
Quoted 27 lines
Quoted 12 lines
docs/nvdimm.txt +3
-0+ + - 'pmem' option of memory-backend-file is 'on' + When using other types of backends, it's suggested to set 'unarmed' option of '-device nvdimm' to 'on', which sets the unarmed flag of the guest NVDIMM region mapping structure. This unarmed flag indicatesqemu-options.hx +0
-0qemu-options.hx using the SNIA NVM programming model (e.g. Intel NVDIMM). · line 4007
Quoted 6 lines
qemu-options.hx +3
-0@@ -4002,6 +4002,10 @@ using the SNIA NVM programming model (e.g. Intel NVDIMM). If @option{pmem} is set to 'on', 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). +Also, we will map the backend-file with MAP_SYNC flag, which can ensure +the file metadata is in sync to @option{mem-path} even on the host crash +and power failures. MAP_SYNC requires supports from both the host kernelrequires support
Quoted 6 lines
qemu-options.hx +1
-0+(since Linux kernel 4.15) and @option{mem-path} (only files supporting DAX). @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}Signature · 1 line
2.7.4
Re: [Qemu-devel] [PATCH V9 2/6] memory: use sparse feature define RAM_FLAG.
In reply to Michael S. TsirkinQuoted 6 lines · Michael S. Tsirkin
Quoted 2 lines · Zhang Yi
Signed-off-by Zhang Yi <[email protected]>Signed-off-by Michael S. Tsirkin <[email protected]>OK so if you apply this patch, do you get
any sparse warning at all?
Didn't get any sparse warning.
Quoted 8 lines
If not how come?
And we need to ask patchew maintainers to
Overall I'd suggest that since you no longer need
a new RAM flag, split this effort out and finishMAP_SYNCwork first.
Ok, Will
Quoted 67 lines
Quoted 67 lines
--- include/exec/memory.h | 12 ++++++------ include/qemu/osdep.h | 9 +++++++++ 2 files changed, 15 insertions(+), 6 deletions(-)include/exec/memory.h +6
-6@@ -104,27 +104,27 @@ struct IOMMUNotifier {typedef struct IOMMUNotifier IOMMUNotifier ;/* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */ -#define RAM_PREALLOC (1 << 0) +#define RAM_PREALLOC ((QEMU_FORCE QemuMmapFlags) (1 << 0))/* RAM is mmap-ed with MAP_SHARED */ -#define RAM_SHARED (1 << 1) +#define RAM_SHARED ((QEMU_FORCE QemuMmapFlags) (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) +#define RAM_RESIZEABLE ((QEMU_FORCE QemuMmapFlags) (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) +#define RAM_UF_ZEROPAGE ((QEMU_FORCE QemuMmapFlags) (1 << 3))/* RAM can be migrated */ -#define RAM_MIGRATABLE (1 << 4) +#define RAM_MIGRATABLE ((QEMU_FORCE QemuMmapFlags) (1 << 4))/* RAM is a persistent kind memory */ -#define RAM_PMEM (1 << 5) +#define RAM_PMEM ((QEMU_FORCE QemuMmapFlags) (1 << 5)) static inline void iommu_notifier_init(IOMMUNotifier *n, IOMMUNotify fn, IOMMUNotifierFlag flags,include/qemu/osdep.h +9
-0@@ -185,6 +185,15 @@ extern int daemon(int, int);#define ESHUTDOWN 4099#endif +#ifdef __CHECKER__ +#define QEMU_BITWISE __attribute__((bitwise)) +#define QEMU_FORCE __attribute__((force)) +#else +#define QEMU_BITWISE +#define QEMU_FORCE +#endif + +typedef unsigned QEMU_BITWISE QemuMmapFlags ;/* time_t may be either 32 or 64 bits depending on the host OS, and * can be either signed or unsigned, so we can't just hardcode a * specific maximum value. This is not a C preprocessor constant, Signature · 1 line
2.7.4
Re: [Qemu-devel] [PATCH V9 6/6] docs: Added MAP_SYNC documentation
In reply to Yi ZhangQuoted 53 lines
Quoted 44 lines
Quoted 11 lines
--- docs/nvdimm.txt | 21 ++++++++++++++++++++- qemu-options.hx | 4 ++++ 2 files changed, 24 insertions(+), 1 deletion(-)docs/nvdimm.txt +0
-0docs/nvdimm.txt backend of vNVDIMM: · line 146
Quoted 5 lines
docs/nvdimm.txt +2
-0@@ -142,11 +142,30 @@ backend of vNVDIMM: Guest Data Persistence ---------------------- +vNVDIMM is designed and implemented to guarantee the guest data +persistence on the backends even on the host crash and powerin case of a host crash or a power failure
(be consistent)
docs/nvdimm.txt backend of vNVDIMM: · line 157
Quoted 12 lines
docs/nvdimm.txt +7
-1+failures. However, there are still some requirements and limitations +as explained below. + Though QEMU supports multiple types of vNVDIMM backends on Linux, -currently the only one that can guarantee the guest write persistence +if MAP_SYNC is not supported by the host kernel and the backends, +the only backend that can guarantee the guest write persistence is the device DAX on the real NVDIMM device (e.g., /dev/dax0.0), to which all guest access do not involve any host-side kernel cache. +mmap(2) flag MAP_SYNC is added since Linux kernel 4.15. On such +systems, QEMU can mmap(2) the backend with MAP_SYNC, which can ensurewhich ensures
docs/nvdimm.txt backend of vNVDIMM: · line 158
Quoted 1 line
docs/nvdimm.txt +1
-0+filesystem metadata consistentconsistency
Quoted 1 line
even after a system crash or power
a power
Quoted 9 lines
docs/nvdimm.txt backend of vNVDIMM: · line 165
Quoted 7 lines
docs/nvdimm.txt +7
-0+failure. Besides the host kernel support, enabling MAP_SYNC in QEMU +also requires: + + - the backend is a file supporting DAX, e.g., a file on an ext4 or + xfs file system mounted with '-o dax', + + - 'share' option of memory-backend-file is 'on'.why does share need to be on?
Quoted 3 lines
According to mmap(2) man page,
MAP_SYNCis available only with theMAP_SHARED_VALIDATEmap‐ping type. I will add this to documentation, Thanks.
MAP_SHARE_VALIDATE should be used to validate MAP_SYNC with MAP_SHARED.
#define MAP_SHARED_VALIDATE 0x03
In contrast, MAP_SHARED is:
#define MAP_SHARED 0x01 /* Share changes */
I think we should use MAP_SHARED_VALIDATE with MAP_SYNC here?
Thanks,
Pankaj
Quoted 31 lines
Quoted 30 lines
Quoted 16 lines
docs/nvdimm.txt +3
-0+ + - 'pmem' option of memory-backend-file is 'on' + When using other types of backends, it's suggested to set 'unarmed' option of '-device nvdimm' to 'on', which sets the unarmed flag of the guest NVDIMM region mapping structure. This unarmed flag indicatesqemu-options.hx +0
-0@@ -4002,6 +4002,10 @@ using the SNIA NVM programming model (e.g. IntelNVDIMM).
If @option{pmem} is set to 'on', 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).qemu-options.hx using the SNIA NVM programming model (e.g. Intel NVDIMM). · line 4007
Quoted 3 lines
qemu-options.hx +3
-0+Also, we will map the backend-file with MAP_SYNC flag, which can ensure +the file metadata is in sync to @option{mem-path} even on the host crash +and power failures. MAP_SYNC requires supports from both the host kernelrequires support
Quoted 8 lines
+(since Linux kernel 4.15) and @option{mem-path} (only files supporting DAX).
@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}
--
2.7.4
Re: [Qemu-devel] [PATCH V9 6/6] docs: Added MAP_SYNC documentation
Quoted 61 lines · Pankaj Gupta
Quoted 53 lines
Quoted 44 lines
Quoted 11 lines
--- docs/nvdimm.txt | 21 ++++++++++++++++++++- qemu-options.hx | 4 ++++ 2 files changed, 24 insertions(+), 1 deletion(-)docs/nvdimm.txt +0
-0docs/nvdimm.txt backend of vNVDIMM: · line 146
Quoted 5 lines
docs/nvdimm.txt +2
-0@@ -142,11 +142,30 @@ backend of vNVDIMM: Guest Data Persistence ---------------------- +vNVDIMM is designed and implemented to guarantee the guest data +persistence on the backends even on the host crash and powerin case of a host crash or a power failure
(be consistent)
docs/nvdimm.txt backend of vNVDIMM: · line 157
Quoted 12 lines
docs/nvdimm.txt +7
-1+failures. However, there are still some requirements and limitations +as explained below. + Though QEMU supports multiple types of vNVDIMM backends on Linux, -currently the only one that can guarantee the guest write persistence +if MAP_SYNC is not supported by the host kernel and the backends, +the only backend that can guarantee the guest write persistence is the device DAX on the real NVDIMM device (e.g., /dev/dax0.0), to which all guest access do not involve any host-side kernel cache. +mmap(2) flag MAP_SYNC is added since Linux kernel 4.15. On such +systems, QEMU can mmap(2) the backend with MAP_SYNC, which can ensurewhich ensures
docs/nvdimm.txt backend of vNVDIMM: · line 158
Quoted 1 line
docs/nvdimm.txt +1
-0+filesystem metadata consistentconsistency
Quoted 1 line
even after a system crash or power
a power
Quoted 9 lines
docs/nvdimm.txt backend of vNVDIMM: · line 165
Quoted 7 lines
docs/nvdimm.txt +7
-0+failure. Besides the host kernel support, enabling MAP_SYNC in QEMU +also requires: + + - the backend is a file supporting DAX, e.g., a file on an ext4 or + xfs file system mounted with '-o dax', + + - 'share' option of memory-backend-file is 'on'.why does share need to be on?
Quoted 3 lines
According to mmap(2) man page,
MAP_SYNCis available only with theMAP_SHARED_VALIDATEmap‐ping type. I will add this to documentation, Thanks.
MAP_SHARE_VALIDATEshould be used to validateMAP_SYNCwithMAP_SHARED.#define
MAP_SHARED_VALIDATE0x03
Quoted 7 lines
In contrast,
MAP_SHAREDis:
#defineMAP_SHARED0x01/* Share changes */I think we should use
MAP_SHARED_VALIDATEwithMAP_SYNChere?Thanks,
Pankaj
Yes, Pankaj, Thanks point that out, I will add the MAP_SHARED_VALIDATE.
Quoted 32 lines
Quoted 32 lines
Quoted 31 lines
Quoted 17 lines
docs/nvdimm.txt +3
-0+ + - 'pmem' option of memory-backend-file is 'on' + When using other types of backends, it's suggested to set 'unarmed' option of '-device nvdimm' to 'on', which sets the unarmed flag of the guest NVDIMM region mapping structure. This unarmed flag indicatesqemu-options.hx +0
-0@@ -4002,6 +4002,10 @@ using the SNIA NVM programming model (e.g. IntelNVDIMM).
If @option{pmem} is set to 'on', 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).qemu-options.hx using the SNIA NVM programming model (e.g. Intel NVDIMM). · line 4007
Quoted 3 lines
qemu-options.hx +3
-0+Also, we will map the backend-file with MAP_SYNC flag, which can ensure +the file metadata is in sync to @option{mem-path} even on the host crash +and power failures. MAP_SYNC requires supports from both the host kernelrequires support
Quoted 8 lines
+(since Linux kernel 4.15) and @option{mem-path} (only files supporting DAX).
@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}
--
2.7.4
Re: [Qemu-devel] [PATCH V9 4/6] util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
In reply to Yi ZhangQuoted 1 line · Yi Zhang
On 2019-01-18 at 16:11:47 -0200, Eduardo Habkost wrote:
[...]
Quoted 7 lines
Quoted 7 lines
Anyway, I see a more fundamental problem in each version of this
patch: the semantics of the command-line options are not clearly documented.We have at least 3 different possible use cases we might need to support:
Quoted 8 lines
Quoted 2 lines
1) pmem=on,
MAP_SYNCnot desired
2) pmem=on,MAP_SYNCdesired but optionalForm V9, As Michael suggest, We removed the sync option,
MAP_SYNCwill force on while we set pmem=on. So we only have 2 user cases, Will update to user documentation.
1) pmem=on,MAP_SYNCnot desired
We will not pass the flag to mmap2
If this use case is supported, how the command-line should look like to enable it?
Quoted 2 lines
2) pmem=on,
MAP_SYNCdesired
We will pass the flag to mmap2
Same question as above: how the command-line should look like for this use case?
Quoted 7 lines
Quoted 7 lines
3) pmem=on,
MAP_SYNCrequired, not optionalWhich cases from the list above we need to support?
From the cases above, what's the expected semantics of "pmem=on" with no extra options?
We still need to answer that question.
The current semantics of pmem=on (with no extra options) is (1). It looks like we can't change it to (2) without breaking existing configurations. If you make existing configurations stop working on hosts where they currently work, you need to explain why it's OK to do that.
Quoted 6 lines
Quoted 6 lines
If these questions are not answered (in the commit message and user documentation), we won't be able to review and discuss the code.
[...]
Signature · 1 line
Eduardo
Re: [Qemu-devel] [PATCH V9 2/6] memory: use sparse feature define RAM_FLAG.
In reply to Yi ZhangQuoted 8 lines · Yi Zhang
Quoted 6 lines · Michael S. Tsirkin
Quoted 2 lines · Zhang Yi
Signed-off-by Zhang Yi <[email protected]>Signed-off-by Michael S. Tsirkin <[email protected]>OK so if you apply this patch, do you get
any sparse warning at all?Didn't get any sparse warning.
The reason I find it strange is because you then go on to assign these flags to an untagged unint32_t.
Did you forget to configure with sparse enabled maybe?
Quoted 76 lines
Quoted 8 lines
If not how come?
And we need to ask patchew maintainers to
Overall I'd suggest that since you no longer need
a new RAM flag, split this effort out and finishMAP_SYNCwork first.Ok, Will
Quoted 67 lines
Quoted 67 lines
--- include/exec/memory.h | 12 ++++++------ include/qemu/osdep.h | 9 +++++++++ 2 files changed, 15 insertions(+), 6 deletions(-)include/exec/memory.h +6
-6@@ -104,27 +104,27 @@ struct IOMMUNotifier {typedef struct IOMMUNotifier IOMMUNotifier ;/* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */ -#define RAM_PREALLOC (1 << 0) +#define RAM_PREALLOC ((QEMU_FORCE QemuMmapFlags) (1 << 0))/* RAM is mmap-ed with MAP_SHARED */ -#define RAM_SHARED (1 << 1) +#define RAM_SHARED ((QEMU_FORCE QemuMmapFlags) (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) +#define RAM_RESIZEABLE ((QEMU_FORCE QemuMmapFlags) (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) +#define RAM_UF_ZEROPAGE ((QEMU_FORCE QemuMmapFlags) (1 << 3))/* RAM can be migrated */ -#define RAM_MIGRATABLE (1 << 4) +#define RAM_MIGRATABLE ((QEMU_FORCE QemuMmapFlags) (1 << 4))/* RAM is a persistent kind memory */ -#define RAM_PMEM (1 << 5) +#define RAM_PMEM ((QEMU_FORCE QemuMmapFlags) (1 << 5)) static inline void iommu_notifier_init(IOMMUNotifier *n, IOMMUNotify fn, IOMMUNotifierFlag flags,include/qemu/osdep.h +9
-0@@ -185,6 +185,15 @@ extern int daemon(int, int);#define ESHUTDOWN 4099#endif +#ifdef __CHECKER__ +#define QEMU_BITWISE __attribute__((bitwise)) +#define QEMU_FORCE __attribute__((force)) +#else +#define QEMU_BITWISE +#define QEMU_FORCE +#endif + +typedef unsigned QEMU_BITWISE QemuMmapFlags ;/* time_t may be either 32 or 64 bits depending on the host OS, and * can be either signed or unsigned, so we can't just hardcode a * specific maximum value. This is not a C preprocessor constant, Signature · 1 line
2.7.4
Re: [Qemu-devel] [PATCH V9 4/6] util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
In reply to Eduardo HabkostQuoted 23 lines · Eduardo Habkost
Quoted 1 line · Yi Zhang
On 2019-01-18 at 16:11:47 -0200, Eduardo Habkost wrote:
[...]
Quoted 7 lines
Quoted 7 lines
Anyway, I see a more fundamental problem in each version of this
patch: the semantics of the command-line options are not clearly documented.We have at least 3 different possible use cases we might need to support:
Quoted 8 lines
Quoted 2 lines
1) pmem=on,
MAP_SYNCnot desired
2) pmem=on,MAP_SYNCdesired but optionalForm V9, As Michael suggest, We removed the sync option,
MAP_SYNCwill force on while we set pmem=on. So we only have 2 user cases, Will update to user documentation.
1) pmem=on,MAP_SYNCnot desired
We will not pass the flag to mmap2If this use case is supported, how the command-line should look like to enable it?
Quoted 5 lines
Quoted 2 lines
2) pmem=on,
MAP_SYNCdesired
We will pass the flag to mmap2Same question as above: how the command-line should look like for this use case?
Sorry, I got some miss-understood of the MAP_SYNC desired. As we talk with Micheal:
we give up on a bit of flexibility, and just say
pmem=on forces MAP_SYNC. on a MAP_SYNC capable configrations(kernel+ backend dax)
Current user case is like below:
1. pmem=on is set, shared=on is set, MAP_SYNC supported in linux kernel:
a: backend is a dax supporting file.
- MAP_SYNC will active.
b: backedn is not a dax supporting file.
- mmap will result in an EOPNOTSUPP error.
2. The reset of cases:
- we will never pass the MAP_SYNC to mmap2
Quoted 30 lines
Quoted 8 lines
Quoted 8 lines
3) pmem=on,
MAP_SYNCrequired, not optionalWhich cases from the list above we need to support?
From the cases above, what's the expected semantics of "pmem=on" with no extra options?
We still need to answer that question.
The current semantics of pmem=on (with no extra options) is (1). It looks like we can't change it to (2) without breaking existing configurations. If you make existing configurations stop working on hosts where they currently work, you need to explain why it's OK to do that.
Quoted 6 lines
Quoted 6 lines
If these questions are not answered (in the commit message and user documentation), we won't be able to review and discuss the code.
[...]
Signature · 1 line
Eduardo
Re: [Qemu-devel] [PATCH V9 4/6] util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
Quoted 38 lines · Yi Zhang
Quoted 23 lines · Eduardo Habkost
Quoted 1 line · Yi Zhang
On 2019-01-18 at 16:11:47 -0200, Eduardo Habkost wrote:
[...]
Quoted 7 lines
Quoted 7 lines
Anyway, I see a more fundamental problem in each version of this
patch: the semantics of the command-line options are not clearly documented.We have at least 3 different possible use cases we might need to support:
Quoted 8 lines
Quoted 2 lines
1) pmem=on,
MAP_SYNCnot desired
2) pmem=on,MAP_SYNCdesired but optionalForm V9, As Michael suggest, We removed the sync option,
MAP_SYNCwill force on while we set pmem=on. So we only have 2 user cases, Will update to user documentation.
1) pmem=on,MAP_SYNCnot desired
We will not pass the flag to mmap2If this use case is supported, how the command-line should look like to enable it?
Quoted 5 lines
Quoted 2 lines
2) pmem=on,
MAP_SYNCdesired
We will pass the flag to mmap2Same question as above: how the command-line should look like for this use case?
Sorry, I got some miss-understood of the
MAP_SYNCdesired. As we talk with Micheal:we give up on a bit of flexibility, and just say
pmem=on forcesMAP_SYNC. on aMAP_SYNCcapable configrations(kernel+ backend dax)Current user case is like below:
Quoted 8 lines
1. pmem=on is set, shared=on is set,
MAP_SYNCsupported in linux kernel:
a: backend is a dax supporting file.
-MAP_SYNCwill active.
b: backedn is not a dax supporting file.
- mmap will result in an EOPNOTSUPP error.2. The reset of cases:
- we will never pass theMAP_SYNCto mmap2
I don't see code probing for MAP_SYNC support. Did I miss it? But if all you want is to have old linux ignore MAP_SYNC, I think you got your wish automatically - just do not set MAP_SHARED_VALIDATE.
Quoted 30 lines
Quoted 30 lines
Quoted 8 lines
Quoted 8 lines
3) pmem=on,
MAP_SYNCrequired, not optionalWhich cases from the list above we need to support?
From the cases above, what's the expected semantics of "pmem=on" with no extra options?
We still need to answer that question.
The current semantics of pmem=on (with no extra options) is (1). It looks like we can't change it to (2) without breaking existing configurations. If you make existing configurations stop working on hosts where they currently work, you need to explain why it's OK to do that.
Quoted 6 lines
Quoted 6 lines
If these questions are not answered (in the commit message and user documentation), we won't be able to review and discuss the code.
[...]
Signature · 1 line
Eduardo
Re: [Qemu-devel] [PATCH V9 2/6] memory: use sparse feature define RAM_FLAG.
In reply to Michael S. TsirkinQuoted 8 lines · Michael S. Tsirkin
Quoted 7 lines · Yi Zhang
Quoted 6 lines · Michael S. Tsirkin
Quoted 2 lines · Zhang Yi
Signed-off-by Zhang Yi <[email protected]>Signed-off-by Michael S. Tsirkin <[email protected]>OK so if you apply this patch, do you get
any sparse warning at all?
Quoted 7 lines
Quoted 1 line
Didn't get any sparse warning.
The reason I find it strange is because you then go on to assign these flags to an untagged
unint32_t.Did you forget to configure with sparse enabled maybe?
Ah.. Right, Sorry for that I don't have much knowledge about this feature and how to use that. I will split this effort out and focus MAP_SYNC work.
Thanks Micheal.
Quoted 79 lines
Quoted 78 lines
Quoted 10 lines
If not how come?
And we need to ask patchew maintainers to
Overall I'd suggest that since you no longer need
a new RAM flag, split this effort out and finishMAP_SYNCwork first.Ok, Will
Quoted 67 lines
Quoted 67 lines
--- include/exec/memory.h | 12 ++++++------ include/qemu/osdep.h | 9 +++++++++ 2 files changed, 15 insertions(+), 6 deletions(-)include/exec/memory.h +6
-6@@ -104,27 +104,27 @@ struct IOMMUNotifier {typedef struct IOMMUNotifier IOMMUNotifier ;/* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */ -#define RAM_PREALLOC (1 << 0) +#define RAM_PREALLOC ((QEMU_FORCE QemuMmapFlags) (1 << 0))/* RAM is mmap-ed with MAP_SHARED */ -#define RAM_SHARED (1 << 1) +#define RAM_SHARED ((QEMU_FORCE QemuMmapFlags) (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) +#define RAM_RESIZEABLE ((QEMU_FORCE QemuMmapFlags) (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) +#define RAM_UF_ZEROPAGE ((QEMU_FORCE QemuMmapFlags) (1 << 3))/* RAM can be migrated */ -#define RAM_MIGRATABLE (1 << 4) +#define RAM_MIGRATABLE ((QEMU_FORCE QemuMmapFlags) (1 << 4))/* RAM is a persistent kind memory */ -#define RAM_PMEM (1 << 5) +#define RAM_PMEM ((QEMU_FORCE QemuMmapFlags) (1 << 5)) static inline void iommu_notifier_init(IOMMUNotifier *n, IOMMUNotify fn, IOMMUNotifierFlag flags,include/qemu/osdep.h +9
-0@@ -185,6 +185,15 @@ extern int daemon(int, int);#define ESHUTDOWN 4099#endif +#ifdef __CHECKER__ +#define QEMU_BITWISE __attribute__((bitwise)) +#define QEMU_FORCE __attribute__((force)) +#else +#define QEMU_BITWISE +#define QEMU_FORCE +#endif + +typedef unsigned QEMU_BITWISE QemuMmapFlags ;/* time_t may be either 32 or 64 bits depending on the host OS, and * can be either signed or unsigned, so we can't just hardcode a * specific maximum value. This is not a C preprocessor constant, Signature · 1 line
2.7.4
Re: [Qemu-devel] [PATCH V9 4/6] util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
In reply to Dan WilliamsQuoted 11 lines · Dan Williams
On Mon, Jan 21, 2019 at 7:27 PM Michael S. Tsirkin <[email protected]> wrote:
[..]Quoted 7 lines
Quoted 2 lines
2. The reset of cases:
- we will never pass theMAP_SYNCto mmap2I don't see code probing for
MAP_SYNCsupport. Did I miss it? But if all you want is to have old linux ignoreMAP_SYNC, I think you got your wish automatically - just do not setMAP_SHARED_VALIDATE.That will also cause new Linux to ignore
MAP_SYNC.
Oh you are right. I missed this point.
And given that these patches do not seem to set MAP_SHARED_VALIDATE at all I conclude that even though thet set MAP_SYNC
it actually has no effect at all.
So I wonder how they were tested.
Would the contributors care to elaborate?
That would be good info to put in the commit log message.
Signature · 1 line
MST
Re: [Qemu-devel] [PATCH V9 4/6] util/mmap-alloc: support MAP_SYNC in qemu_ram_mmap()
In reply to Michael S. TsirkinOn Mon, Jan 21, 2019 at 7:27 PM Michael S. Tsirkin <[email protected]> wrote:
[..]
Quoted 7 lines
Quoted 2 lines
2. The reset of cases:
- we will never pass theMAP_SYNCto mmap2I don't see code probing for
MAP_SYNCsupport. Did I miss it? But if all you want is to have old linux ignoreMAP_SYNC, I think you got your wish automatically - just do not setMAP_SHARED_VALIDATE.
That will also cause new Linux to ignore MAP_SYNC.