VHDX log replay and write support
cover VHDX log replay and write support
This patch series contains the initial VHDX log parsing, replay, and write support.
This will allow an existing log in a VHDX image to be replayed (e.g., a VHDX image from a Hyper-V host that crashed). In addition, metadata writes are enabled through the log. This allows write support to be enabled for VHDX, as the BAT can be updated safely via the log journal.
The patches are also available from github, for testing:
https://github.com/codyprime/qemu-kvm-jtc/tree/jtc-vhdx-latest
Jeff Cody (9):
block: vhdx - minor comments and typo correction.
block: vhdx - add header update capability.
block: vhdx code movement - VHDXMetadataEntries and BDRVVHDXState to
header.
block: vhdx - log support struct and defines
block: vhdx - break endian translation functions out
block: vhdx - update log guid in header, and first write tracker
block: vhdx - log parsing, replay, and flush support
block: vhdx - add log write support
block: vhdx write support
block/Makefile.objs | 2 +- block/vhdx-endian.c | 141 ++++++++ block/vhdx-log.c | 1007 +++++++++++++++++++++++++++++++++++++++++++++++++++ block/vhdx.c | 394 ++++++++++++++------ block/vhdx.h | 110 +++++- configure | 13 + 6 files changed, 1556 insertions(+), 111 deletions(-) create mode 100644 block/vhdx-endian.c create mode 100644 block/vhdx-log.c
Signature · 1 line
1.8.1.4
PATCH 1/9 block: vhdx - minor comments and typo correction.
Just a couple of minor comments to help note where allocated buffers are freed, and a typo fix.
--- block/vhdx.c | 6 ++++-- block/vhdx.h | 6 +++--- 2 files changed, 7 insertions(+), 5 deletions(-)
block/vhdx.c +4 -2
@@ -6,9 +6,9 @@* Authors: * Jeff Cody <[email protected]> * -* This is based on the "VHDX Format Specification v0.95", published 4/12/2012 +* This is based on the "VHDX Format Specification v1.00", published 8/25/2012 * by Microsoft: -* https://www.microsoft.com/en-us/download/details.aspx?id=29681 +* https://www.microsoft.com/en-us/download/details.aspx?id=34750 * * This work is licensed under the terms of the GNU LGPL, version 2 or later. * See the COPYING.LIB file in the top-level directory.
@@ -262,6 +262,7 @@ static int vhdx_parse_header(BlockDriverState *bs, BDRVVHDXState *s)uint64_t h2_seq =0 ;uint8_t *buffer; +/* header1 & header2 are freed in vhdx_close() */ header1 =qemu_blockalign (bs,sizeof (VHDXHeader)); header2 =qemu_blockalign (bs,sizeof (VHDXHeader));
@@ -787,6 +788,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags) goto fail; } +/* s->bat is freed in vhdx_close() */ s->bat =qemu_blockalign (bs, s->bat_rt.length); ret =bdrv_pread (bs->file, s->bat_offset, s->bat, s->bat_rt.length);
block/vhdx.h +3 -3
@@ -6,9 +6,9 @@* Authors: * Jeff Cody <[email protected]> * -* This is based on the "VHDX Format Specification v0.95", published 4/12/2012 +* This is based on the "VHDX Format Specification v1.00", published 8/25/2012 * by Microsoft: -* https://www.microsoft.com/en-us/download/details.aspx?id=29681 +* https://www.microsoft.com/en-us/download/details.aspx?id=34750 * * This work is licensed under the terms of the GNU LGPL, version 2 or later. * See the COPYING.LIB file in the top-level directory.
@@ -116,7 +116,7 @@ typedef struct QEMU_PACKED VHDXHeader {valid. */ uint16_t log_version;/* version of the log format. Mustn't be zero, unless log_guid is also zero */ -uint16_t version;/* version of th evhdx file. Currently, +uint16_t version;/* version of the vhdx file. Currently, only supported version is "1" */ uint32_t log_length;/* length of the log. Must be multiple of 1MB */
Signature · 1 line
1.8.1.4
PATCH 2/9 block: vhdx - add header update capability.
In reply to Jeff CodyThis adds the ability to update the headers in a VHDX image, including generating a new MS-compatible GUID.
As VHDX depends on uuid.h, VHDX is now a configurable build option. If VHDX support is enabled, that will also enable uuid as well. The default is to have VHDX enabled.
To enable/disable VHDX: --enable-vhdx, --disable-vhdx
--- block/Makefile.objs | 2 +- block/vhdx.c | 157 +++++++++++++++++++++++++++++++++++++++++++++++++++- block/vhdx.h | 12 +++- configure | 13 +++++ 4 files changed, 180 insertions(+), 4 deletions(-)
block/Makefile.objs +1 -1
@@ -2,7 +2,7 @@ block-obj-y += raw.o cow.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o block-obj-y += qed-check.o -block-obj-y += vhdx.o +block-obj-$(CONFIG_VHDX) += vhdx.o block-obj-y += parallels.o blkdebug.o blkverify.o block-obj-y += snapshot.o qapi.o block-obj-$(CONFIG_WIN32) += raw-win32.o win32-aio.o
block/vhdx.c +155 -2
198 lines not on this page · raw message
block/vhdx.h +11 -1
@@ -309,17 +309,27 @@ typedef struct QEMU_PACKED VHDXParentLocatorEntry {/* ----- END VHDX SPECIFICATION STRUCTURES ---- */ +void vhdx_guid_generate (MSGUID *guid); + +uint32_t vhdx_update_checksum (uint8_t *buf,size_t size,int crc_offset);uint32_t vhdx_checksum_calc (uint32_t crc,uint8_t *buf,size_t size,int crc_offset);bool vhdx_checksum_is_valid (uint8_t *buf,size_t size,int crc_offset); -static void leguid_to_cpus (MSGUID *guid) +static inline void leguid_to_cpus (MSGUID *guid) {le32_to_cpus (&guid->data1);le16_to_cpus (&guid->data2);le16_to_cpus (&guid->data3); } +static inline void cpu_to_leguids (MSGUID *guid) +{ +cpu_to_le32s (&guid->data1); +cpu_to_le16s (&guid->data2); +cpu_to_le16s (&guid->data3); +} +#endif
configure +13 -0
@@ -244,6 +244,7 @@ gtk="" gtkabi="2.0" tpm="no" libssh2="" +vhdx="yes" # parse CC options first for opt do
@@ -950,6 +951,11 @@ for opt do ;; --enable-libssh2) libssh2="yes" ;; + --enable-vhdx) vhdx="yes" ; + uuid="yes" + ;; + --disable-vhdx) vhdx="no" + ;; *) echo "ERROR: unknown option $opt"; show_help="yes" ;; esac
@@ -1166,6 +1172,8 @@ echo " --gcov=GCOV use specified gcov [$gcov_tool]" echo " --enable-tpm enable TPM support" echo " --disable-libssh2 disable ssh block device support" echo " --enable-libssh2 enable ssh block device support" +echo " --disable-vhdx disables support for the Microsoft VHDX image format" +echo " --enable-vhdx enable support for the Microsoft VHDX image format" echo "" echo "NOTE: The object files are built at the place where configure is launched" exit 1
@@ -3622,6 +3630,7 @@ echo "TPM support $tpm" echo "libssh2 support $libssh2" echo "TPM passthrough $tpm_passthrough" echo "QOM debugging $qom_cast_debug" +echo "vhdx $vhdx" if test "$sdl_too_old" = "yes"; then echo "-> Your SDL version is too old - please upgrade to have SDL support"
@@ -4012,6 +4021,10 @@ if test "$virtio_blk_data_plane" = "yes" ; then echo 'CONFIG_VIRTIO_BLK_DATA_PLANE=$(CONFIG_VIRTIO)' >> $config_host_mak fi +if test "$vhdx" = "yes" ; then + echo "CONFIG_VHDX=y" >> $config_host_mak +fi + # USB host support case "$usb" in linux)
Signature · 1 line
1.8.1.4
PATCH 3/9 block: vhdx code movement - VHDXMetadataEntries and BDRVVHDXState to header.
In reply to Jeff CodyIn preparation for VHDX log support, move these structures to the header.
--- block/vhdx.c | 51 --------------------------------------------------- block/vhdx.h | 47 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 47 insertions(+), 51 deletions(-)
block/vhdx.c +0 -51
@@ -104,16 +104,6 @@ static const MSGUID parent_vhdx_guid = { .data1 = 0xb04aefb7, META_PAGE_83_PRESENT | META_LOGICAL_SECTOR_SIZE_PRESENT | \ META_PHYS_SECTOR_SIZE_PRESENT) -typedef struct VHDXMetadataEntries { -VHDXMetadataTableEntry file_parameters_entry; -VHDXMetadataTableEntry virtual_disk_size_entry; -VHDXMetadataTableEntry page83_data_entry; -VHDXMetadataTableEntry logical_sector_size_entry; -VHDXMetadataTableEntry phys_sector_size_entry; -VHDXMetadataTableEntry parent_locator_entry; -uint16_t present; -}VHDXMetadataEntries ; - typedef struct VHDXSectorInfo { uint32_t bat_idx; /* BAT entry index */
@@ -124,47 +114,6 @@ typedef struct VHDXSectorInfo {uint64_t block_offset;/* block offset, in bytes */ } VHDXSectorInfo; - - -typedef struct BDRVVHDXState { -CoMutex lock; - -int curr_header; -VHDXHeader *headers[2 ]; - -VHDXRegionTableHeader rt; -VHDXRegionTableEntry bat_rt;/* region table for the BAT */ -VHDXRegionTableEntry metadata_rt;/* region table for the metadata */ - -VHDXMetadataTableHeader metadata_hdr; -VHDXMetadataEntries metadata_entries; - -VHDXFileParameters params; -uint32_t block_size; -uint32_t block_size_bits; -uint32_t sectors_per_block; -uint32_t sectors_per_block_bits; - -uint64_t virtual_disk_size; -uint32_t logical_sector_size; -uint32_t physical_sector_size; - -uint64_t chunk_ratio; -uint32_t chunk_ratio_bits; -uint32_t logical_sector_size_bits; - -uint32_t bat_entries; -VHDXBatEntry *bat; -uint64_t bat_offset; - -MSGUID session_guid; - - -VHDXParentLocatorHeader parent_header; -VHDXParentLocatorEntry *parent_entries; - -}BDRVVHDXState ; -/* Calculates new checksum. * * Zero is substituted during crc calculation for the original crc field
block/vhdx.h +47 -0
53 lines not on this page · raw message
Signature · 1 line
1.8.1.4
PATCH 4/9 block: vhdx - log support struct and defines
In reply to Jeff CodyThis adds some magic number defines, and internal structure definitions for VHDX log replay support.
--- block/vhdx.h | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-)
block/vhdx.h +20 -1
@@ -151,7 +151,10 @@ typedef struct QEMU_PACKED VHDXRegionTableEntry {/* ---- LOG ENTRY STRUCTURES ---- */ +#define VHDX_LOG_MIN_SIZE (1024*1024) +#define VHDX_LOG_SECTOR_SIZE 4096#define VHDX_LOG_HDR_SIZE 64 +#define VHDX_LOG_SIGNATURE 0x65676f6c typedef struct QEMU_PACKED VHDXLogEntryHeader {uint32_t signature;/* "loge" in ASCII */ uint32_t checksum;/* CRC-32C hash of the 64KB table */
@@ -174,7 +177,8 @@ typedef struct QEMU_PACKED VHDXLogEntryHeader { } VHDXLogEntryHeader;#define VHDX_LOG_DESC_SIZE 32 - +#define VHDX_LOG_DESC_SIGNATURE 0x63736564 +#define VHDX_LOG_ZERO_SIGNATURE 0x6f72657a typedef struct QEMU_PACKED VHDXLogDescriptor {uint32_t signature;/* "zero" or "desc" in ASCII */ union {
@@ -194,6 +198,7 @@ typedef struct QEMU_PACKED VHDXLogDescriptor {vhdx_log_entry_header */ } VHDXLogDescriptor; +#define VHDX_LOG_DATA_SIGNATURE 0x61746164 typedef struct QEMU_PACKED VHDXLogDataSector {uint32_t data_signature;/* "data" in ASCII */ uint32_t sequence_high;/* 4 MSB of 8 byte sequence_number */
@@ -318,6 +323,18 @@ typedef struct VHDXMetadataEntries {uint16_t present; } VHDXMetadataEntries; +typedef struct VHDXLogEntries { +uint64_t offset; +uint64_t length; +uint32_t head; +uint32_t tail; +}VHDXLogEntries ; + +typedef struct VHDXLogEntryInfo { +uint64_t sector_start; +uint32_t desc_count; +}VHDXLogEntryInfo ; + typedef struct BDRVVHDXState { CoMutex lock;
@@ -351,6 +368,8 @@ typedef struct BDRVVHDXState {MSGUID session_guid; +VHDXLogEntries log; +VHDXParentLocatorHeader parent_header;VHDXParentLocatorEntry *parent_entries;
Signature · 1 line
1.8.1.4
PATCH 5/9 block: vhdx - break endian translation functions out
In reply to Jeff CodyThis moves the endian translation functions out from the vhdx.c source, into a separate source file. In addition to the previously defined endian functions, new endian translation functions for log support are added as well.
--- block/Makefile.objs | 2 +- block/vhdx-endian.c | 141 ++++++++++++++++++++++++++++++++++++++++++++++++++++ block/vhdx.c | 43 ---------------- block/vhdx.h | 13 +++++ 4 files changed, 155 insertions(+), 44 deletions(-) create mode 100644 block/vhdx-endian.c
block/Makefile.objs +1 -1
@@ -2,7 +2,7 @@ block-obj-y += raw.o cow.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o block-obj-y += qed-check.o -block-obj-$(CONFIG_VHDX) += vhdx.o +block-obj-$(CONFIG_VHDX) += vhdx.o vhdx-endian.o block-obj-y += parallels.o blkdebug.o blkverify.o block-obj-y += snapshot.o qapi.o block-obj-$(CONFIG_WIN32) += raw-win32.o win32-aio.o
block/vhdx-endian.c new file +141 -0
141 lines not on this page · raw message
block/vhdx.c +0 -43
49 lines not on this page · raw message
block/vhdx.h +13 -0
@@ -398,4 +398,17 @@ static inline void cpu_to_leguids(MSGUID *guid) cpu_to_le16s(&guid->data3); } +void vhdx_header_le_import (VHDXHeader *h); +void vhdx_header_le_export (VHDXHeader *orig_h,VHDXHeader *new_h); +void vhdx_log_desc_le_import (VHDXLogDescriptor *d); +void vhdx_log_desc_le_export (VHDXLogDescriptor *d); +void vhdx_log_data_le_export (VHDXLogDataSector *d); +void vhdx_log_entry_hdr_le_import (VHDXLogEntryHeader *hdr); +void vhdx_log_entry_hdr_le_export (VHDXLogEntryHeader *hdr); + + + + + +#endif
Signature · 1 line
1.8.1.4
PATCH 6/9 block: vhdx - update log guid in header, and first write tracker
In reply to Jeff CodyAllow tracking of first file write in the VHDX image, as well as the ability to update the GUID in the header. This is in preparation for log support.
--- block/vhdx.c | 28 +++++++++++++++++++++++----- block/vhdx.h | 7 +++++-- 2 files changed, 28 insertions(+), 7 deletions(-)
block/vhdx.c +23 -5
@@ -233,7 +233,8 @@ static int vhdx_probe(const uint8_t *buf, int buf_size, const char *filename)* * - non-current header is updated with largest sequence number */ -static int vhdx_update_header (BlockDriverState *bs,BDRVVHDXState *s,bool rw) +static int vhdx_update_header (BlockDriverState *bs,BDRVVHDXState *s,bool rw, +MSGUID *log_guid) {int ret =0 ;int hdr_idx =0 ;
@@ -266,6 +267,11 @@ static int vhdx_update_header(BlockDriverState *bs, BDRVVHDXState *s, bool rw)vhdx_guid_generate (&inactive_header->data_write_guid); } +/* update the log guid if present */ +if (log_guid) { +memcpy (&inactive_header->log_guid, log_guid,sizeof (MSGUID )); + } +/* the header checksum is not over just the packed size of VHDXHeader, * but rather over the entire 'reserved' range for the header, which is * 4KB (VHDX_HEADER_SIZE). */
@@ -293,15 +299,16 @@ fail:* The VHDX spec calls for header updates to be performed twice, so that both * the current and non-current header have valid info */ -static int vhdx_update_headers (BlockDriverState *bs,BDRVVHDXState *s,bool rw) +int vhdx_update_headers (BlockDriverState *bs,BDRVVHDXState *s,bool rw, +MSGUID *log_guid) {int ret; - ret =vhdx_update_header (bs, s, rw); + ret =vhdx_update_header (bs, s, rw, log_guid);if (ret <0 ) {return ret; } - ret =vhdx_update_header (bs, s, rw); + ret =vhdx_update_header (bs, s, rw, log_guid);return ret; }
@@ -781,6 +788,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags) s->bat =NULL ; + s->first_visible_write = true;qemu_co_mutex_init (&s->lock);
@@ -861,7 +869,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags) }if (flags &BDRV_O_RDWR ) { -vhdx_update_headers (bs, s, false); +vhdx_update_headers (bs, s, false,NULL ); }/* TODO: differencing files, write */
@@ -998,6 +1006,16 @@ exit: +/* Per the spec, on the first write of guest-visible data to the file the +* data write guid must be updated in the header */ +void vhdx_user_visible_write (BlockDriverState *bs,BDRVVHDXState *s) +{ +if (s->first_visible_write) { + s->first_visible_write = false; +vhdx_update_headers (bs, s, true,NULL ); + } +} + static coroutine_fn int vhdx_co_writev(BlockDriverState *bs, int64_t sector_num,int nb_sectors,QEMUIOVector *qiov) {
block/vhdx.h +5 -2
@@ -366,6 +366,7 @@ typedef struct BDRVVHDXState {VHDXBatEntry *bat;uint64_t bat_offset; +bool first_visible_write;MSGUID session_guid;VHDXLogEntries log;
@@ -377,6 +378,9 @@ typedef struct BDRVVHDXState {void vhdx_guid_generate (MSGUID *guid); +int vhdx_update_headers (BlockDriverState *bs,BDRVVHDXState *s,bool rw, +MSGUID *log_guid); +uint32_t vhdx_update_checksum (uint8_t *buf,size_t size,int crc_offset);uint32_t vhdx_checksum_calc (uint32_t crc,uint8_t *buf,size_t size,int crc_offset);
@@ -406,8 +410,7 @@ void vhdx_log_data_le_export(VHDXLogDataSector *d);void vhdx_log_entry_hdr_le_import (VHDXLogEntryHeader *hdr);void vhdx_log_entry_hdr_le_export (VHDXLogEntryHeader *hdr); - - +void vhdx_user_visible_write (BlockDriverState *bs,BDRVVHDXState *s);
Signature · 1 line
1.8.1.4
PATCH 7/9 block: vhdx - log parsing, replay, and flush support
In reply to Jeff CodyThis adds support for VHDX v0 logs, as specified in Microsoft's VHDX Specification Format v1.00:
https://www.microsoft.com/en-us/download/details.aspx?id=34750
The following support is added:
* Log parsing, and validation - validate that an existing log is correct.
* Log search - search through an existing log, to find any valid sequence of entries.
* Log replay and flush - replay an existing log, and flush/clear the log when complete.
The VHDX log is a circular buffer, with elements (sectors) of 4KB.
A log entry is a variably-length number of sectors, that is comprised of a header and 'descriptors', that describe each sector.
A log may contain multiple entries, know as a log sequence. In a log sequence, each log entry immediately follows the previous entry, with an incrementing sequence number. There can only ever be one active and valid sequence in the log.
Each log entry must match the file log GUID in order to be valid (along with other criteria). Once we have flushed all valid log entries, we marked the file log GUID to be zero, which indicates a buffer with no valid entries.
--- block/Makefile.objs | 2 +- block/vhdx-log.c | 734 ++++++++++++++++++++++++++++++++++++++++++++++++++++ block/vhdx.c | 44 +--- block/vhdx.h | 7 +- 4 files changed, 743 insertions(+), 44 deletions(-) create mode 100644 block/vhdx-log.c
block/Makefile.objs +1 -1
@@ -2,7 +2,7 @@ block-obj-y += raw.o cow.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o block-obj-y += qed-check.o -block-obj-$(CONFIG_VHDX) += vhdx.o vhdx-endian.o +block-obj-$(CONFIG_VHDX) += vhdx.o vhdx-endian.o vhdx-log.o block-obj-y += parallels.o blkdebug.o blkverify.o block-obj-y += snapshot.o qapi.o block-obj-$(CONFIG_WIN32) += raw-win32.o win32-aio.o
block/vhdx-log.c new file +734 -0
734 lines not on this page · raw message
block/vhdx.c +2 -42
62 lines not on this page · raw message
block/vhdx.h +6 -1
@@ -326,7 +326,11 @@ typedef struct VHDXMetadataEntries {typedef struct VHDXLogEntries {uint64_t offset;uint64_t length; -uint32_t head; +uint32_t write; +uint32_t read; +VHDXLogEntryHeader *hdr; +void *desc_buffer; +uint64_t sequence;uint32_t tail; }VHDXLogEntries ;
@@ -387,6 +391,7 @@ uint32_t vhdx_checksum_calc(uint32_t crc, uint8_t *buf, size_t size,bool vhdx_checksum_is_valid (uint8_t *buf,size_t size,int crc_offset); +int vhdx_parse_log (BlockDriverState *bs,BDRVVHDXState *s);static inline void leguid_to_cpus (MSGUID *guid) {
Signature · 1 line
1.8.1.4
PATCH 8/9 block: vhdx - add log write support
In reply to Jeff CodyThis adds support for writing to the VHDX log.
For spec details, see VHDX Specification Format v1.00:
https://www.microsoft.com/en-us/download/details.aspx?id=34750
There are a few limitations to this log support:
1.) There is no caching yet
2.) The log is flushed after each entry
The primary write interface, vhdx_log_write_and_flush(), performs a log write followed by an immediate flush of the log.
As each log entry sector is a minimum of 4KB, partial sector writes are filled in with data from the disk write destination.
If the current file log GUID is 0, a new GUID is generated and updated in the header.
--- block/vhdx-log.c | 273 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ block/vhdx.h | 3 + 2 files changed, 276 insertions(+)
block/vhdx-log.c +273 -0
@@ -170,6 +170,53 @@ exit:return ret; } +/* Writes num_sectors to the log (all log sectors are 4096 bytes), +* from buffer 'buffer'. Upon return, *sectors_written will contain +* the number of sectors successfully written. +* +* It is assumed that 'buffer' is at least 4096*num_sectors large. +* +* 0 is returned on success, -errno otherwise */ +static int vhdx_log_write_sectors (BlockDriverState *bs,VHDXLogEntries *log, +uint32_t *sectors_written,void *buffer, +uint32_t num_sectors) +{ +int ret =0 ; +uint64_t offset; +uint32_t write; +void *buffer_tmp; +BDRVVHDXState *s = bs->opaque; + +vhdx_user_visible_write (bs, s); + + write = log->write; + + buffer_tmp = buffer; +while (num_sectors) { + + offset = log->offset + write; + write =vhdx_log_inc_idx (write, log->length); +if (write == log->read) { +/* full */ +break ; + } + ret =bdrv_pwrite_sync (bs->file, offset, buffer_tmp, +VHDX_LOG_SECTOR_SIZE ); +if (ret <0 ) { + goto exit; + } + buffer_tmp +=VHDX_LOG_SECTOR_SIZE ; + + log->write = write; + *sectors_written = *sectors_written +1 ; + num_sectors--; + } + +exit: +return ret; +} + +/* Validates a log entry header */ static bool vhdx_log_hdr_is_valid (VHDXLogEntries *log,VHDXLogEntryHeader *hdr,BDRVVHDXState *s)
@@ -732,3 +779,229 @@ exit:return ret; } + + +static void vhdx_log_raw_to_le_sector (VHDXLogDescriptor *desc, +VHDXLogDataSector *sector,void *data, +uint64_t seq) +{ +memcpy (&desc->leading_bytes, data,8 ); + data +=8 ; +cpu_to_le64s (&desc->leading_bytes); +memcpy (sector->data, data,4084 ); + data +=4084 ; +memcpy (&desc->trailing_bytes, data,4 ); +cpu_to_le32s (&desc->trailing_bytes); + data +=4 ; + + sector->sequence_high = (uint32_t ) (seq >>32 ); + sector->sequence_low = (uint32_t ) (seq &0xffffffff ); + sector->data_signature =VHDX_LOG_DATA_SIGNATURE ; + +vhdx_log_desc_le_export (desc); +vhdx_log_data_le_export (sector); +} + + +static int vhdx_log_write (BlockDriverState *bs,BDRVVHDXState *s, +void *data,uint32_t length,uint64_t offset) +{ +int ret =0 ; +void *buffer =NULL ; +void *merged_sector =NULL ; +void *data_tmp, *sector_write; +unsigned int i; +int sector_offset; +uint32_t desc_sectors, sectors, total_length; +uint32_t sectors_written =0 ; +uint32_t aligned_length; +uint32_t leading_length =0 ; +uint32_t trailing_length =0 ; +uint32_t partial_sectors =0 ; +uint32_t bytes_written =0 ; +uint64_t file_offset; +VHDXHeader *header; +VHDXLogEntryHeader new_hdr; +VHDXLogDescriptor *new_desc =NULL ; +VHDXLogDataSector *data_sector =NULL ; +MSGUID new_guid = {0 }; + + header = s->headers[s->curr_header]; + +/* need to have offset read data, and be on 4096 byte boundary */ + +if (length > header->log_length) { +/* no log present. we could create a log here instead of failing */ + ret = -EINVAL ; + goto exit; + } + +if (vhdx_log_guid_is_zero (&header->log_guid)) { +vhdx_guid_generate (&new_guid); +vhdx_update_headers (bs, s, false, &new_guid); + }else { +/* currently, we require that the log be flushed after +* every write. */ + ret = -ENOTSUP ; + } + +/* 0 is an invalid sequence number, but may also represent the first +* log write (or a wrapped seq) */ +if (s->log.sequence ==0 ) { + s->log.sequence =1 ; + } + + sector_offset = offset %VHDX_LOG_SECTOR_SIZE ; + file_offset = (offset /VHDX_LOG_SECTOR_SIZE ) *VHDX_LOG_SECTOR_SIZE ; + + aligned_length = length; + +/* add in the unaligned head and tail bytes */ +if (sector_offset) { + leading_length = (VHDX_LOG_SECTOR_SIZE - sector_offset); + leading_length = leading_length > length ? length : leading_length; + aligned_length -= leading_length; + partial_sectors++; + } + + sectors = aligned_length /VHDX_LOG_SECTOR_SIZE ; + trailing_length = aligned_length - (sectors *VHDX_LOG_SECTOR_SIZE ); +if (trailing_length) { + partial_sectors++; + } + + sectors += partial_sectors; + +/* sectors is now how many sectors the data itself takes, not +* including the header and descriptor metadata */ + + new_hdr = (VHDXLogEntryHeader ) { + .signature =VHDX_LOG_SIGNATURE , + .tail = s->log.tail, + .sequence_number = s->log.sequence, + .descriptor_count = sectors, + .reserved =0 , + .flushed_file_offset =bdrv_getlength (bs->file), + .last_file_offset =bdrv_getlength (bs->file), + }; + +memcpy (&new_hdr.log_guid, &header->log_guid,sizeof (MSGUID )); + + desc_sectors =vhdx_compute_desc_sectors (new_hdr.descriptor_count); + + total_length = (desc_sectors + sectors) *VHDX_LOG_SECTOR_SIZE ; + new_hdr.entry_length = total_length; + +vhdx_log_entry_hdr_le_export (&new_hdr); + + buffer =qemu_blockalign (bs, total_length); +memcpy (buffer, &new_hdr,sizeof (new_hdr)); + + new_desc = (VHDXLogDescriptor *) (buffer +sizeof (new_hdr)); + data_sector = buffer + (desc_sectors *VHDX_LOG_SECTOR_SIZE ); + data_tmp = data; + +/* All log sectors are 4KB, so for any partial sectors we must +* merge the data with preexisting data from the final file +* destination */ + merged_sector =qemu_blockalign (bs,VHDX_LOG_SECTOR_SIZE ); + +for (i =0 ; i < sectors; i++) { + new_desc->signature =VHDX_LOG_DESC_SIGNATURE ; + new_desc->sequence_number = s->log.sequence; + new_desc->file_offset = file_offset; + +if (i ==0 && leading_length) { +/* partial sector at the front of the buffer */ + ret =bdrv_pread (bs->file, file_offset, merged_sector, +VHDX_LOG_SECTOR_SIZE ); +if (ret <0 ) { + goto exit; + } +memcpy (merged_sector + sector_offset, data_tmp, leading_length); + bytes_written = leading_length; + sector_write = merged_sector; + }else if (i == sectors -1 && trailing_length) { +/* partial sector at the end of the buffer */ + ret =bdrv_pread (bs->file, + file_offset, + merged_sector + trailing_length, +VHDX_LOG_SECTOR_SIZE - trailing_length); +if (ret <0 ) { + goto exit; + } +memcpy (merged_sector, data_tmp, trailing_length); + bytes_written = trailing_length; + sector_write = merged_sector; + }else { + bytes_written =VHDX_LOG_SECTOR_SIZE ; + sector_write = data_tmp; + } + +/* populate the raw sector data into the proper structures, +* as well as update the descriptor, and convert to proper +* endianness */ +vhdx_log_raw_to_le_sector (new_desc, data_sector, sector_write, + s->log.sequence); + + data_tmp += bytes_written; + data_sector++; + new_desc++; + file_offset +=VHDX_LOG_SECTOR_SIZE ; + } + +/* checksum covers entire entry, from the log header through the +* last data sector */ +vhdx_update_checksum (buffer, total_length,4 ); +cpu_to_le32s ((uint32_t *)(buffer +4 )); + +/* now write to the log */ +vhdx_log_write_sectors (bs, &s->log, §ors_written, buffer, + desc_sectors + sectors); +if (ret <0 ) { + goto exit; + } + +if (sectors_written != desc_sectors + sectors) { +/* instead of failing, we could flush the log here */ + ret = -EINVAL ; + goto exit; + } + + s->log.sequence++; +/* write new tail */ + s->log.tail = s->log.write; + +exit: +qemu_vfree (buffer); +qemu_vfree (merged_sector); +return ret; +} + +/* Perform a log write, and then immediately flush the entire log */ +int vhdx_log_write_and_flush (BlockDriverState *bs,BDRVVHDXState *s, +void *data,uint32_t length,uint64_t offset) +{ +int ret =0 ; +VHDXLogSequence logs = { .valid = true, + .count =1 , + .hdr = {0 } }; + + + ret =vhdx_log_write (bs, s, data, length, offset); +if (ret <0 ) { + goto exit; + } + logs.log = s->log; + + ret =vhdx_log_flush (bs, s, &logs); +if (ret <0 ) { + goto exit; + } + + s->log = logs.log; + +exit: +return ret; +} +
block/vhdx.h +3 -0
@@ -393,6 +393,9 @@ bool vhdx_checksum_is_valid(uint8_t *buf, size_t size, int crc_offset);int vhdx_parse_log (BlockDriverState *bs,BDRVVHDXState *s); +int vhdx_log_write_and_flush (BlockDriverState *bs,BDRVVHDXState *s, +void *data,uint32_t length,uint64_t offset); +static inline void leguid_to_cpus (MSGUID *guid) {le32_to_cpus (&guid->data1);
Signature · 1 line
1.8.1.4
PATCH 9/9 block: vhdx write support
In reply to Jeff CodyThis adds support for writing to VHDX image files, using coroutines. Writes into the BAT table goes through the VHDX log. Currently, BAT table writes occur when expanding a dynamic VHDX file, and allocating a new BAT entry.
--- block/vhdx.c | 149 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 147 insertions(+), 2 deletions(-)
block/vhdx.c +147 -2
@@ -831,7 +831,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags)vhdx_update_headers (bs, s, false,NULL ); } -/* TODO: differencing files, write */ +/* TODO: differencing files */ return 0 ; fail:
@@ -963,7 +963,45 @@ exit:return ret; } +/* +* Allocate a new payload block at the end of the file. +* +* Allocation will happen at 1MB alignment inside the file +* +* Returns the file offset start of the new payload block +*/ +static int vhdx_allocate_block (BlockDriverState *bs,BDRVVHDXState *s, +uint64_t *new_offset) +{ + *new_offset =bdrv_getlength (bs->file); +/* per the spec, the address for a block is in units of 1MB */ +if (*new_offset % (1024 *1024 )) { + *new_offset = ((*new_offset >>20 ) +1 ) <<20 ;/* round up to 1MB */ + } + +return bdrv_truncate (bs->file, *new_offset + s->block_size); +} + +/* +* Update the BAT tablet entry with the new file offset, and the new entry +* state */ +static void vhdx_update_bat_table_entry (BlockDriverState *bs,BDRVVHDXState *s, +VHDXSectorInfo *sinfo, +uint64_t *bat_entry, +uint64_t *bat_offset,int state) +{ +/* The BAT entry is a uint64, with 44 bits for the file offset in units of +* 1MB, and 3 bits for the block state. */ + s->bat[sinfo->bat_idx] = ((sinfo->file_offset>>20 ) << +VHDX_BAT_FILE_OFF_BITS ); + + s->bat[sinfo->bat_idx] |= state &VHDX_BAT_STATE_BIT_MASK ; + + *bat_entry =cpu_to_le64 (s->bat[sinfo->bat_idx]); + *bat_offset = s->bat_offset + sinfo->bat_idx *sizeof (VHDXBatEntry); + +}/* Per the spec, on the first write of guest-visible data to the file the * data write guid must be updated in the header */
@@ -978,7 +1016,114 @@ void vhdx_user_visible_write(BlockDriverState *bs, BDRVVHDXState *s) static coroutine_fn int vhdx_co_writev(BlockDriverState *bs, int64_t sector_num,int nb_sectors,QEMUIOVector *qiov) { -return -ENOTSUP ; +int ret = -ENOTSUP ; +BDRVVHDXState *s = bs->opaque; +VHDXSectorInfo sinfo; +uint64_t bytes_done =0 ; +uint64_t bat_entry =0 ; +uint64_t bat_entry_offset =0 ; +bool bat_update; +QEMUIOVector hd_qiov; + +qemu_iovec_init (&hd_qiov, qiov->niov); + +qemu_co_mutex_lock (&s->lock); + +vhdx_user_visible_write (bs, s); + +while (nb_sectors >0 ) { +if (s->params.data_bits &VHDX_PARAMS_HAS_PARENT ) { +/* not supported yet */ + ret = -ENOTSUP ; + goto exit; + }else { + bat_update = false; +vhdx_block_translate (s, sector_num, nb_sectors, &sinfo); + +qemu_iovec_reset (&hd_qiov); +qemu_iovec_concat (&hd_qiov, qiov, bytes_done, sinfo.bytes_avail); +/* check the payload block state */ +switch (s->bat[sinfo.bat_idx] &VHDX_BAT_STATE_BIT_MASK ) { +case PAYLOAD_BLOCK_ZERO : +/* in this case, we need to preserve zero writes for +* data that is not part of this write, so we must pad +* the rest of the buffer to zeroes */ + +/* if we are on a posix system with ftruncate() that extends +* a file, then it is zero-filled for us. On Win32, the raw +* layer uses SetFilePointer and SetFileEnd, which does not +* zero fill AFAIK */ + +/* TODO: queue another write of zero buffers if the host OS does +* not zero-fill on file extension */ + +/* fall through */ +case PAYLOAD_BLOCK_NOT_PRESENT :/* fall through */ +case PAYLOAD_BLOCK_UNMAPPED :/* fall through */ +case PAYLOAD_BLOCK_UNDEFINED :/* fall through */ + ret =vhdx_allocate_block (bs, s, &sinfo.file_offset); +if (ret <0 ) { + goto exit; + } +/* once we support differencing files, this may also be +* partially present */ +/* update block state to the newly specified state */ +vhdx_update_bat_table_entry (bs, s, &sinfo, &bat_entry, + &bat_entry_offset, +PAYLOAD_BLOCK_FULL_PRESENT ); + bat_update = true; +/* since we just allocated a block, file_offset is the +* beginning of the payload block. It needs to be the +* write address, which includes the offset into the block */ + sinfo.file_offset += sinfo.block_offset; +/* fall through */ +case PAYLOAD_BLOCK_FULL_PRESENT : +/* if the file offset address is in the header zone, +* there is a problem */ +if (sinfo.file_offset < (1024 *1024 )) { + ret = -EFAULT ; + goto exit; + } +/* block exists, so we can just overwrite it */ +qemu_co_mutex_unlock (&s->lock); + ret =bdrv_co_writev (bs->file, + sinfo.file_offset>>BDRV_SECTOR_BITS , + sinfo.sectors_avail, &hd_qiov); +qemu_co_mutex_lock (&s->lock); +if (ret <0 ) { + goto exit; + } +break ; +case PAYLOAD_BLOCK_PARTIALLY_PRESENT : +/* we don't yet support difference files, fall through +* to error */ +default : + ret = -EIO ; + goto exit; +break ; + } + +if (bat_update) { +/* this will update the BAT entry into the log journal, and +* then flush the log journal out to disk */ + ret =vhdx_log_write_and_flush (bs, s, &bat_entry, +sizeof (VHDXBatEntry), + bat_entry_offset); +if (ret <0 ) { + goto exit; + } + } + + nb_sectors -= sinfo.sectors_avail; + sector_num += sinfo.sectors_avail; + bytes_done += sinfo.bytes_avail; + + } + } + +exit: +qemu_co_mutex_unlock (&s->lock); +return ret; }
Signature · 1 line
1.8.1.4
Re: [Qemu-devel] [PATCH 2/9] block: vhdx - add header update capability.
In reply to Jeff CodyQuoted 184 lines · Jeff Cody
This adds the ability to update the headers in a VHDX image, including generating a new MS-compatible GUID.
As VHDX depends on
uuid.h, VHDX is now a configurable build option. If VHDX support is enabled, that will also enable uuid as well. The default is to have VHDX enabled.To enable/disable VHDX: --enable-vhdx, --disable-vhdx
Signed-off-by Jeff Cody <[email protected]>--- block/Makefile.objs | 2 +- block/vhdx.c | 157 +++++++++++++++++++++++++++++++++++++++++++++++++++- block/vhdx.h | 12 +++- configure | 13 +++++ 4 files changed, 180 insertions(+), 4 deletions(-)block/Makefile.objs +1
-1@@ -2,7 +2,7 @@ block-obj-y += raw.o cow.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o block-obj-y += qed-check.o -block-obj-y += vhdx.o +block-obj-$(CONFIG_VHDX) += vhdx.o block-obj-y += parallels.o blkdebug.o blkverify.o block-obj-y += snapshot.o qapi.o block-obj-$(CONFIG_WIN32) += raw-win32.o win32-aio.oblock/vhdx.c +120
-0@@ -21,6 +21,7 @@#include "qemu/crc32c.h" #include "block/vhdx.h" +#include <uuid/uuid.h> /* Several metadata and region table data entries are identified by * guids in a MS-specific GUID format. */ @@ -156,11 +157,40 @@ typedef struct BDRVVHDXState {VHDXBatEntry *bat;uint64_t bat_offset; +MSGUID session_guid; + +VHDXParentLocatorHeader parent_header;VHDXParentLocatorEntry *parent_entries; } BDRVVHDXState; +/* Calculates new checksum. +* +* Zero is substituted during crc calculation for the original crc field +* crc_offset: byte offset in buf of the buffer crc +* buf: buffer pointer +* size: size of buffer (must be > crc_offset+4) +* +* Note: The resulting checksum is in the CPU endianness, not necessarily +* in the file format endianness (LE). Any header export to disk should +* make sure that vhdx_header_le_export() is used to convert to the +* correct endianness +*/ +uint32_t vhdx_update_checksum (uint8_t *buf,size_t size,int crc_offset) +{ +uint32_t crc; + +assert (buf !=NULL ); +assert (size > (crc_offset +4 )); + +memset (buf + crc_offset,0 ,sizeof (crc)); + crc =crc32c (0xffffffff , buf, size); +memcpy (buf + crc_offset, &crc,sizeof (crc)); + +return crc; +} +uint32_t vhdx_checksum_calc (uint32_t crc,uint8_t *buf,size_t size,int crc_offset) {@@ -212,6 +242,24 @@ bool vhdx_checksum_is_valid(uint8_t *buf, size_t size, int crc_offset)/* +* This generates a UUID that is compliant with the MS GUIDs used +* in the VHDX spec (and elsewhere). +* +* We can do this with uuid_generate if uuid.h is present, +* however not all systems have uuid and the generation is +* pretty straightforward for the DCE + random usage case +* +*/ +void vhdx_guid_generate (MSGUID *guid) +{ +uuid_t uuid; +assert (guid !=NULL ); + +uuid_generate (uuid); +memcpy (guid, uuid,16 ); +} + +/* * Per the MS VHDX Specification, for every VHDX file: * - The header section is fixed size - 1 MB * - The header section is always the first "object" @@ -249,6 +297,107 @@ static void vhdx_header_le_import(VHDXHeader *h)le64_to_cpus (&h->log_offset); } +/* All VHDX structures on disk are little endian */ +static void vhdx_header_le_export (VHDXHeader *orig_h,VHDXHeader *new_h) +{ +assert (orig_h !=NULL ); +assert (new_h !=NULL ); + + new_h->signature =cpu_to_le32 (orig_h->signature); + new_h->checksum =cpu_to_le32 (orig_h->checksum); + new_h->sequence_number =cpu_to_le64 (orig_h->sequence_number); + +memcpy (&new_h->file_write_guid, &orig_h->file_write_guid,sizeof (MSGUID )); +memcpy (&new_h->data_write_guid, &orig_h->data_write_guid,sizeof (MSGUID )); +memcpy (&new_h->log_guid, &orig_h->log_guid,sizeof (MSGUID )); + +cpu_to_leguids (&new_h->file_write_guid); +cpu_to_leguids (&new_h->data_write_guid); +cpu_to_leguids (&new_h->log_guid); + + new_h->log_version =cpu_to_le16 (orig_h->log_version); + new_h->version =cpu_to_le16 (orig_h->version); + new_h->log_length =cpu_to_le32 (orig_h->log_length); + new_h->log_offset =cpu_to_le64 (orig_h->log_offset); +} + +/* Update the VHDX headers +* +* This follows the VHDX spec procedures for header updates. +* +* - non-current header is updated with largest sequence number +*/ +static int vhdx_update_header (BlockDriverState *bs,BDRVVHDXState *s,bool rw) +{ +int ret =0 ; +int hdr_idx =0 ; +uint64_t header_offset =VHDX_HEADER1_OFFSET ; + +VHDXHeader *active_header; +VHDXHeader *inactive_header; +VHDXHeader header_le; +uint8_t *buffer; + +/* operate on the non-current header */ +if (s->curr_header ==0 ) { + hdr_idx =1 ; + header_offset =VHDX_HEADER2_OFFSET ; + } + + active_header = s->headers[s->curr_header]; + inactive_header = s->headers[hdr_idx]; + + inactive_header->sequence_number = active_header->sequence_number +1 ; + +/* a new file guid must be generate before any file write, including +* headers */ +memcpy (&inactive_header->file_write_guid, &s->session_guid, +sizeof (MSGUID )); + +/* a new data guid only needs to be generate before any guest-visible +* writes, so update it if the image is opened r/w. */ +if (rw) { +vhdx_guid_generate (&inactive_header->data_write_guid); + } + +/* the header checksum is not over just the packed size of VHDXHeader, +* but rather over the entire 'reserved' range for the header, which is +* 4KB (VHDX_HEADER_SIZE). */ + + buffer =qemu_blockalign (bs,VHDX_HEADER_SIZE ); +/* we can't assume the extra reserved bytes are 0 */ + ret =bdrv_pread (bs->file, header_offset, buffer,VHDX_HEADER_SIZE ); +if (ret <0 ) { + goto fail;
block/vhdx.c static void vhdx_header_le_import(VHDXHeader *h) · line 381
Quoted 10 lines
block/vhdx.c +10
-0+ } +/* overwrite the actual VHDXHeader portion */ +memcpy (buffer, inactive_header,sizeof (VHDXHeader)); + inactive_header->checksum =vhdx_update_checksum (buffer, +VHDX_HEADER_SIZE ,4 ); +vhdx_header_le_export (inactive_header, &header_le); +bdrv_pwrite_sync (bs->file, header_offset, &header_le,sizeof (VHDXHeader)); + s->curr_header = hdr_idx; + +fail:
Labeling it "fail:" is a bit misleading as normal path ends here too.
Quoted 133 lines
block/vhdx.c +25
-2+qemu_vfree (buffer); +return ret; +} + +/* +* The VHDX spec calls for header updates to be performed twice, so that both +* the current and non-current header have valid info +*/ +static int vhdx_update_headers (BlockDriverState *bs,BDRVVHDXState *s,bool rw) +{ +int ret; + + ret =vhdx_update_header (bs, s, rw); +if (ret <0 ) { +return ret; + } + ret =vhdx_update_header (bs, s, rw); +return ret; +}/* opens the specified header block from the VHDX file header section */ static int vhdx_parse_header (BlockDriverState *bs,BDRVVHDXState *s)@@ -739,6 +888,11 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags) goto fail; } +/* This is used for any header updates, for the file_write_guid. +* The spec dictates that a new value should be used for the first +* header update */ +vhdx_guid_generate (&s->session_guid); + ret =vhdx_parse_header (bs, s);if (ret) { goto fail;@@ -801,8 +955,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags) }if (flags &BDRV_O_RDWR ) { - ret = -ENOTSUP ; - goto fail; +vhdx_update_headers (bs, s, false); }/* TODO: differencing files, write */ block/vhdx.h +11
-1@@ -309,17 +309,27 @@ typedef struct QEMU_PACKED VHDXParentLocatorEntry {/* ----- END VHDX SPECIFICATION STRUCTURES ---- */ +void vhdx_guid_generate (MSGUID *guid); + +uint32_t vhdx_update_checksum (uint8_t *buf,size_t size,int crc_offset);uint32_t vhdx_checksum_calc (uint32_t crc,uint8_t *buf,size_t size,int crc_offset);bool vhdx_checksum_is_valid (uint8_t *buf,size_t size,int crc_offset); -static void leguid_to_cpus (MSGUID *guid) +static inline void leguid_to_cpus (MSGUID *guid) {le32_to_cpus (&guid->data1);le16_to_cpus (&guid->data2);le16_to_cpus (&guid->data3); } +static inline void cpu_to_leguids (MSGUID *guid) +{ +cpu_to_le32s (&guid->data1); +cpu_to_le16s (&guid->data2); +cpu_to_le16s (&guid->data3); +} +#endif configure +13
-0@@ -244,6 +244,7 @@ gtk="" gtkabi="2.0" tpm="no" libssh2="" +vhdx="yes" # parse CC options first for opt do@@ -950,6 +951,11 @@ for opt do ;; --enable-libssh2) libssh2="yes" ;; + --enable-vhdx) vhdx="yes" ; + uuid="yes" + ;; + --disable-vhdx) vhdx="no" + ;; *) echo "ERROR: unknown option $opt"; show_help="yes" ;; esac@@ -1166,6 +1172,8 @@ echo " --gcov=GCOV use specified gcov [$gcov_tool]" echo " --enable-tpm enable TPM support" echo " --disable-libssh2 disable ssh block device support" echo " --enable-libssh2 enable ssh block device support" +echo " --disable-vhdx disables support for the Microsoft VHDX image format" +echo " --enable-vhdx enable support for the Microsoft VHDX image format" echo "" echo "NOTE: The object files are built at the place where configure is launched" exit 1@@ -3622,6 +3630,7 @@ echo "TPM support $tpm" echo "libssh2 support $libssh2" echo "TPM passthrough $tpm_passthrough" echo "QOM debugging $qom_cast_debug" +echo "vhdx $vhdx" if test "$sdl_too_old" = "yes"; then echo "-> Your SDL version is too old - please upgrade to have SDL support"@@ -4012,6 +4021,10 @@ if test "$virtio_blk_data_plane" = "yes" ; then echo 'CONFIG_VIRTIO_BLK_DATA_PLANE=$(CONFIG_VIRTIO)' >> $config_host_mak fi +if test "$vhdx" = "yes" ; then + echo "CONFIG_VHDX=y" >> $config_host_mak +fi + # USB host support case "$usb" in linux)Signature · 1 line
1.8.1.4
Signature · 1 line
Fam
Re: [Qemu-devel] [PATCH 2/9] block: vhdx - add header update capability.
Quoted 189 lines · Fam Zheng
Quoted 188 lines · Jeff Cody
This adds the ability to update the headers in a VHDX image, including generating a new MS-compatible GUID.
As VHDX depends on
uuid.h, VHDX is now a configurable build option. If VHDX support is enabled, that will also enable uuid as well. The default is to have VHDX enabled.To enable/disable VHDX: --enable-vhdx, --disable-vhdx
Signed-off-by Jeff Cody <[email protected]>--- block/Makefile.objs | 2 +- block/vhdx.c | 157 +++++++++++++++++++++++++++++++++++++++++++++++++++- block/vhdx.h | 12 +++- configure | 13 +++++ 4 files changed, 180 insertions(+), 4 deletions(-)block/Makefile.objs +1
-1@@ -2,7 +2,7 @@ block-obj-y += raw.o cow.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o block-obj-y += qed-check.o -block-obj-y += vhdx.o +block-obj-$(CONFIG_VHDX) += vhdx.o block-obj-y += parallels.o blkdebug.o blkverify.o block-obj-y += snapshot.o qapi.o block-obj-$(CONFIG_WIN32) += raw-win32.o win32-aio.oblock/vhdx.c +124
-0@@ -21,6 +21,7 @@#include "qemu/crc32c.h" #include "block/vhdx.h" +#include <uuid/uuid.h> /* Several metadata and region table data entries are identified by * guids in a MS-specific GUID format. */ @@ -156,11 +157,40 @@ typedef struct BDRVVHDXState {VHDXBatEntry *bat;uint64_t bat_offset; +MSGUID session_guid; + +VHDXParentLocatorHeader parent_header;VHDXParentLocatorEntry *parent_entries; } BDRVVHDXState; +/* Calculates new checksum. +* +* Zero is substituted during crc calculation for the original crc field +* crc_offset: byte offset in buf of the buffer crc +* buf: buffer pointer +* size: size of buffer (must be > crc_offset+4) +* +* Note: The resulting checksum is in the CPU endianness, not necessarily +* in the file format endianness (LE). Any header export to disk should +* make sure that vhdx_header_le_export() is used to convert to the +* correct endianness +*/ +uint32_t vhdx_update_checksum (uint8_t *buf,size_t size,int crc_offset) +{ +uint32_t crc; + +assert (buf !=NULL ); +assert (size > (crc_offset +4 )); + +memset (buf + crc_offset,0 ,sizeof (crc)); + crc =crc32c (0xffffffff , buf, size); +memcpy (buf + crc_offset, &crc,sizeof (crc)); + +return crc; +} +uint32_t vhdx_checksum_calc (uint32_t crc,uint8_t *buf,size_t size,int crc_offset) {@@ -212,6 +242,24 @@ bool vhdx_checksum_is_valid(uint8_t *buf, size_t size, int crc_offset)/* +* This generates a UUID that is compliant with the MS GUIDs used +* in the VHDX spec (and elsewhere). +* +* We can do this with uuid_generate if uuid.h is present, +* however not all systems have uuid and the generation is +* pretty straightforward for the DCE + random usage case +* +*/ +void vhdx_guid_generate (MSGUID *guid) +{ +uuid_t uuid; +assert (guid !=NULL ); + +uuid_generate (uuid); +memcpy (guid, uuid,16 ); +} + +/* * Per the MS VHDX Specification, for every VHDX file: * - The header section is fixed size - 1 MB * - The header section is always the first "object" @@ -249,6 +297,107 @@ static void vhdx_header_le_import(VHDXHeader *h)le64_to_cpus (&h->log_offset); } +/* All VHDX structures on disk are little endian */ +static void vhdx_header_le_export (VHDXHeader *orig_h,VHDXHeader *new_h) +{ +assert (orig_h !=NULL ); +assert (new_h !=NULL ); + + new_h->signature =cpu_to_le32 (orig_h->signature); + new_h->checksum =cpu_to_le32 (orig_h->checksum); + new_h->sequence_number =cpu_to_le64 (orig_h->sequence_number); + +memcpy (&new_h->file_write_guid, &orig_h->file_write_guid,sizeof (MSGUID )); +memcpy (&new_h->data_write_guid, &orig_h->data_write_guid,sizeof (MSGUID )); +memcpy (&new_h->log_guid, &orig_h->log_guid,sizeof (MSGUID )); + +cpu_to_leguids (&new_h->file_write_guid); +cpu_to_leguids (&new_h->data_write_guid); +cpu_to_leguids (&new_h->log_guid); + + new_h->log_version =cpu_to_le16 (orig_h->log_version); + new_h->version =cpu_to_le16 (orig_h->version); + new_h->log_length =cpu_to_le32 (orig_h->log_length); + new_h->log_offset =cpu_to_le64 (orig_h->log_offset); +} + +/* Update the VHDX headers +* +* This follows the VHDX spec procedures for header updates. +* +* - non-current header is updated with largest sequence number +*/ +static int vhdx_update_header (BlockDriverState *bs,BDRVVHDXState *s,bool rw) +{ +int ret =0 ; +int hdr_idx =0 ; +uint64_t header_offset =VHDX_HEADER1_OFFSET ; + +VHDXHeader *active_header; +VHDXHeader *inactive_header; +VHDXHeader header_le; +uint8_t *buffer; + +/* operate on the non-current header */ +if (s->curr_header ==0 ) { + hdr_idx =1 ; + header_offset =VHDX_HEADER2_OFFSET ; + } + + active_header = s->headers[s->curr_header]; + inactive_header = s->headers[hdr_idx]; + + inactive_header->sequence_number = active_header->sequence_number +1 ; + +/* a new file guid must be generate before any file write, including +* headers */ +memcpy (&inactive_header->file_write_guid, &s->session_guid, +sizeof (MSGUID )); + +/* a new data guid only needs to be generate before any guest-visible +* writes, so update it if the image is opened r/w. */ +if (rw) { +vhdx_guid_generate (&inactive_header->data_write_guid); + } + +/* the header checksum is not over just the packed size of VHDXHeader, +* but rather over the entire 'reserved' range for the header, which is +* 4KB (VHDX_HEADER_SIZE). */ + + buffer =qemu_blockalign (bs,VHDX_HEADER_SIZE ); +/* we can't assume the extra reserved bytes are 0 */ + ret =bdrv_pread (bs->file, header_offset, buffer,VHDX_HEADER_SIZE ); +if (ret <0 ) { + goto fail; + } +/* overwrite the actual VHDXHeader portion */ +memcpy (buffer, inactive_header,sizeof (VHDXHeader)); + inactive_header->checksum = vhdx_update_checksum(buffer,
Quoted 9 lines
block/vhdx.c static void vhdx_header_le_import(VHDXHeader *h) · line 381
Quoted 6 lines
block/vhdx.c +6
-0+ VHDX_HEADER_SIZE, 4); +vhdx_header_le_export (inactive_header, &header_le); +bdrv_pwrite_sync (bs->file, header_offset, &header_le,sizeof (VHDXHeader)); + s->curr_header = hdr_idx; + +fail:Labeling it "fail:" is a bit misleading as normal path ends here too.
Good point; I'll change that to just be 'exit'.
Quoted 136 lines
Quoted 133 lines
block/vhdx.c +25
-2+qemu_vfree (buffer); +return ret; +} + +/* +* The VHDX spec calls for header updates to be performed twice, so that both +* the current and non-current header have valid info +*/ +static int vhdx_update_headers (BlockDriverState *bs,BDRVVHDXState *s,bool rw) +{ +int ret; + + ret =vhdx_update_header (bs, s, rw); +if (ret <0 ) { +return ret; + } + ret =vhdx_update_header (bs, s, rw); +return ret; +}/* opens the specified header block from the VHDX file header section */ static int vhdx_parse_header (BlockDriverState *bs,BDRVVHDXState *s)@@ -739,6 +888,11 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags) goto fail; } +/* This is used for any header updates, for the file_write_guid. +* The spec dictates that a new value should be used for the first +* header update */ +vhdx_guid_generate (&s->session_guid); + ret =vhdx_parse_header (bs, s);if (ret) { goto fail;@@ -801,8 +955,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags) }if (flags &BDRV_O_RDWR ) { - ret = -ENOTSUP ; - goto fail; +vhdx_update_headers (bs, s, false); }/* TODO: differencing files, write */ block/vhdx.h +11
-1@@ -309,17 +309,27 @@ typedef struct QEMU_PACKED VHDXParentLocatorEntry {/* ----- END VHDX SPECIFICATION STRUCTURES ---- */ +void vhdx_guid_generate (MSGUID *guid); + +uint32_t vhdx_update_checksum (uint8_t *buf,size_t size,int crc_offset);uint32_t vhdx_checksum_calc (uint32_t crc,uint8_t *buf,size_t size,int crc_offset);bool vhdx_checksum_is_valid (uint8_t *buf,size_t size,int crc_offset); -static void leguid_to_cpus (MSGUID *guid) +static inline void leguid_to_cpus (MSGUID *guid) {le32_to_cpus (&guid->data1);le16_to_cpus (&guid->data2);le16_to_cpus (&guid->data3); } +static inline void cpu_to_leguids (MSGUID *guid) +{ +cpu_to_le32s (&guid->data1); +cpu_to_le16s (&guid->data2); +cpu_to_le16s (&guid->data3); +} +#endif configure +13
-0@@ -244,6 +244,7 @@ gtk="" gtkabi="2.0" tpm="no" libssh2="" +vhdx="yes" # parse CC options first for opt do@@ -950,6 +951,11 @@ for opt do ;; --enable-libssh2) libssh2="yes" ;; + --enable-vhdx) vhdx="yes" ; + uuid="yes" + ;; + --disable-vhdx) vhdx="no" + ;; *) echo "ERROR: unknown option $opt"; show_help="yes" ;; esac@@ -1166,6 +1172,8 @@ echo " --gcov=GCOV use specified gcov [$gcov_tool]" echo " --enable-tpm enable TPM support" echo " --disable-libssh2 disable ssh block device support" echo " --enable-libssh2 enable ssh block device support" +echo " --disable-vhdx disables support for the Microsoft VHDX image format" +echo " --enable-vhdx enable support for the Microsoft VHDX image format" echo "" echo "NOTE: The object files are built at the place where configure is launched" exit 1@@ -3622,6 +3630,7 @@ echo "TPM support $tpm" echo "libssh2 support $libssh2" echo "TPM passthrough $tpm_passthrough" echo "QOM debugging $qom_cast_debug" +echo "vhdx $vhdx" if test "$sdl_too_old" = "yes"; then echo "-> Your SDL version is too old - please upgrade to have SDL support"@@ -4012,6 +4021,10 @@ if test "$virtio_blk_data_plane" = "yes" ; then echo 'CONFIG_VIRTIO_BLK_DATA_PLANE=$(CONFIG_VIRTIO)' >> $config_host_mak fi +if test "$vhdx" = "yes" ; then + echo "CONFIG_VHDX=y" >> $config_host_mak +fi + # USB host support case "$usb" in linux)Signature · 1 line
1.8.1.4Signature · 1 line
Fam
Re: [Qemu-devel] [PATCH 4/9] block: vhdx - log support struct and defines
In reply to Jeff CodyQuoted 24 lines · Jeff Cody
This adds some magic number defines, and internal structure definitions for VHDX log replay support.
Signed-off-by Jeff Cody <[email protected]>--- block/vhdx.h | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-)block/vhdx.h +3
-0@@ -151,7 +151,10 @@ typedef struct QEMU_PACKED VHDXRegionTableEntry {/* ---- LOG ENTRY STRUCTURES ---- */ +#define VHDX_LOG_MIN_SIZE (1024*1024) +#define VHDX_LOG_SECTOR_SIZE 4096#define VHDX_LOG_HDR_SIZE 64 +#define VHDX_LOG_SIGNATURE 0x65676f6c typedef struct QEMU_PACKED VHDXLogEntryHeader {uint32_t signature;/* "loge" in ASCII */ uint32_t checksum;/* CRC-32C hash of the 64KB table */
block/vhdx.h typedef struct QEMU_PACKED VHDXLogEntryHeader { · line 181
Quoted 6 lines
block/vhdx.h +2
-1@@ -174,7 +177,8 @@ typedef struct QEMU_PACKED VHDXLogEntryHeader { } VHDXLogEntryHeader;#define VHDX_LOG_DESC_SIZE 32 - +#define VHDX_LOG_DESC_SIGNATURE 0x63736564 +#define VHDX_LOG_ZERO_SIGNATURE 0x6f72657a
Are these macros really used? I see "desc" and "zero" used to compare signatures.
Thanks
Fam
Re: [Qemu-devel] [PATCH 7/9] block: vhdx - log parsing, replay, and flush support
In reply to Jeff CodyQuoted 124 lines · Jeff Cody
This adds support for VHDX v0 logs, as specified in Microsoft's VHDX Specification Format v1.00:
https://www.microsoft.com/en-us/download/details.aspx?id=34750The following support is added:
* Log parsing, and validation - validate that an existing log is correct.
* Log search - search through an existing log, to find any valid sequence of entries.
* Log replay and flush - replay an existing log, and flush/clear the log when complete.
The VHDX log is a circular buffer, with elements (sectors) of 4KB.
A log entry is a variably-length number of sectors, that is comprised of a header and 'descriptors', that describe each sector.
A log may contain multiple entries, know as a log sequence. In a log sequence, each log entry immediately follows the previous entry, with an incrementing sequence number. There can only ever be one active and valid sequence in the log.
Each log entry must match the file log GUID in order to be valid (along with other criteria). Once we have flushed all valid log entries, we marked the file log GUID to be zero, which indicates a buffer with no valid entries.
Signed-off-by Jeff Cody <[email protected]>--- block/Makefile.objs | 2 +- block/vhdx-log.c | 734 ++++++++++++++++++++++++++++++++++++++++++++++++++++ block/vhdx.c | 44 +--- block/vhdx.h | 7 +- 4 files changed, 743 insertions(+), 44 deletions(-) create mode 100644 block/vhdx-log.cblock/Makefile.objs +1
-1@@ -2,7 +2,7 @@ block-obj-y += raw.o cow.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o block-obj-y += qed-check.o -block-obj-$(CONFIG_VHDX) += vhdx.o vhdx-endian.o +block-obj-$(CONFIG_VHDX) += vhdx.o vhdx-endian.o vhdx-log.o block-obj-y += parallels.o blkdebug.o blkverify.o block-obj-y += snapshot.o qapi.o block-obj-$(CONFIG_WIN32) += raw-win32.o win32-aio.oblock/vhdx-log.c new file +66
-0@@ -0,0 +1,734 @@ +/* +* Block driver for Hyper-V VHDX Images +* +* Copyright (c) 2013 Red Hat, Inc., +* +* Authors: +* Jeff Cody <[email protected]> +* +* This is based on the "VHDX Format Specification v1.00", published 8/25/2012 +* by Microsoft: +* https://www.microsoft.com/en-us/download/details.aspx?id=34750 +* +* This file covers the functionality of the metadata log writing, parsing, and +* replay. +* +* This work is licensed under the terms of the GNU LGPL, version 2 or later. +* See the COPYING.LIB file in the top-level directory. +* +*/ +#include "qemu-common.h" +#include "block/block_int.h" +#include "qemu/module.h" +#include "block/vhdx.h" + + +typedef struct VHDXLogSequence { +bool valid; +uint32_t count; +VHDXLogEntries log; +VHDXLogEntryHeader hdr; +}VHDXLogSequence ; + +typedef struct VHDXLogDescEntries { +VHDXLogEntryHeader hdr; +VHDXLogDescriptor desc[]; +}VHDXLogDescEntries ; + + +/* Returns true if the GUID is zero */ +static bool vhdx_log_guid_is_zero (MSGUID *guid) +{ +int i; +int ret =0 ; + +/* If either the log guid, or log length is zero, +* then a replay log is not present */ +for (i =0 ; i <sizeof (MSGUID ); i++) { + ret |= ((uint8_t *) guid)[i]; + } + +return ret ==0 ; +} + +/* The log located on the disk is circular buffer containing +* sectors of 4096 bytes each. +* +* It is assumed for the read/write functions below that the +* circular buffer scheme uses a 'one sector open' to indicate +* the buffer is full. Given the validation methods used for each +* sector, this method should be compatible with other methods that +* do not waste a sector. +*/ + + +/* Allow peeking at the hdr entry at the beginning of the current +* read index, without advancing the read index */
block/vhdx-log.c line 76
Quoted 10 lines
block/vhdx-log.c new file +10
-0+static int vhdx_log_peek_hdr (BlockDriverState *bs,VHDXLogEntries *log, +VHDXLogEntryHeader *hdr) +{ +int ret =0 ; +uint64_t offset; +uint32_t read; + +assert (hdr !=NULL ); + +/* peek is only support on sector boundaries */
s/support/supported/
Quoted 131 lines
block/vhdx-log.c new file +131
-0+if (log->read %VHDX_LOG_SECTOR_SIZE ) { + ret = -EFAULT ; + goto exit; + } + + read = log->read; +/* we are guaranteed that a) log sectors are 4096 bytes, +* and b) the log length is a multiple of 1MB. So, there +* is always a round number of sectors in the buffer */ +if ((read +sizeof (VHDXLogEntryHeader)) > log->length) { + read =0 ; + } + +if (read == log->write) { + ret = -EINVAL ; + goto exit; + } + + offset = log->offset + read; + + ret =bdrv_pread (bs->file, offset, hdr,sizeof (VHDXLogEntryHeader)); +if (ret <0 ) { + goto exit; + } + +exit: +return ret; +} + +/* Index increment for log, based on sector boundaries */ +static int vhdx_log_inc_idx (uint32_t idx,uint64_t length) +{ + idx +=VHDX_LOG_SECTOR_SIZE ; +/* we are guaranteed that a) log sectors are 4096 bytes, +* and b) the log length is a multiple of 1MB. So, there +* is always a round number of sectors in the buffer */ +return idx >= length ?0 : idx; +} + + +/* Reset the log to empty */ +static void vhdx_log_reset (BlockDriverState *bs,BDRVVHDXState *s) +{ +MSGUID guid = {0 }; + s->log.read = s->log.write =0 ; +/* a log guid of 0 indicates an empty log to any parser of v0 +* VHDX logs */ +vhdx_update_headers (bs, s, false, &guid); +} + +/* Reads num_sectors from the log (all log sectors are 4096 bytes), +* into buffer 'buffer'. Upon return, *sectors_read will contain +* the number of sectors successfully read. +* +* It is assumed that 'buffer' is already allocated, and of sufficient +* size (i.e. >= 4096*num_sectors). +* +* If 'peek' is true, then the tail (read) pointer for the circular buffer is +* not modified. +* +* 0 is returned on success, -errno otherwise. */ +static int vhdx_log_read_sectors (BlockDriverState *bs,VHDXLogEntries *log, +uint32_t *sectors_read,void *buffer, +uint32_t num_sectors,bool peek) +{ +int ret =0 ; +uint64_t offset; +uint32_t read; + + read = log->read; + + *sectors_read =0 ; +while (num_sectors) { +if (read == log->write) { +/* empty */ +break ; + } + offset = log->offset + read; + + ret =bdrv_pread (bs->file, offset, buffer,VHDX_LOG_SECTOR_SIZE ); +if (ret <0 ) { + goto exit; + } + read =vhdx_log_inc_idx (read, log->length); + + *sectors_read = *sectors_read +1 ; + num_sectors--; + } + +exit: +if (!peek) { + log->read = read; + } +return ret; +} + +/* Validates a log entry header */ +static bool vhdx_log_hdr_is_valid (VHDXLogEntries *log,VHDXLogEntryHeader *hdr, +BDRVVHDXState *s) +{ +int valid = false; + +if (memcmp (&hdr->signature,"loge" ,4 )) { + goto exit; + } + +/* if the individual entry length is larger than the whole log +* buffer, that is obviously invalid */ +if (log->length < hdr->entry_length) { + goto exit; + } + +/* length of entire entry must be in units of 4KB (log sector size) */ +if (hdr->entry_length % (VHDX_LOG_SECTOR_SIZE )) { + goto exit; + } + +/* per spec, sequence # must be > 0 */ +if (hdr->sequence_number ==0 ) { + goto exit; + } + +/* log entries are only valid if they match the file-wide log guid +* found in the active header */ +if (!guid_eq (hdr->log_guid, s->headers[s->curr_header]->log_guid)) { + goto exit; + } + + valid = true; + +exit:
block/vhdx-log.c line 217
Quoted 10 lines
block/vhdx-log.c new file +10
-0+return valid; +} + +/* +* Given a log header, this will validate that the descriptors and the +* corresponding data sectors (if applicable) +* +* Validation consists of: +* 1. Making sure the sequence numbers matches the entry header +* 2. Verifying a valid signature ('zero' or desc' for descriptors)
s/ desc'/ 'desc'/
block/vhdx-log.c line 223
Quoted 6 lines
block/vhdx-log.c new file +6
-0+* 3. File offset field is a multiple of 4KB +* 4. If a data descriptor, the corresponding data sector +* has its signature ('data') and matching sequence number +* +* 'desc' is the data buffer containing the descriptor +* hdr is the log entry header
Please use gtkdoc format:
@desc: the data buffer ...
@hdr: the log entry header
Quoted 158 lines
block/vhdx-log.c new file +158
-0+* +* Returns true if valid +*/ +static bool vhdx_log_desc_is_valid (VHDXLogDescriptor *desc, +VHDXLogEntryHeader *hdr) +{ +bool ret = false; + +if (desc->sequence_number != hdr->sequence_number) { + goto exit; + } +if (desc->file_offset %VHDX_LOG_SECTOR_SIZE ) { + goto exit; + } + +if (!memcmp (&desc->signature,"zero" ,4 )) { +if (!desc->zero_length %VHDX_LOG_SECTOR_SIZE ) { +/* valid */ + ret = true; + } + }else if (!memcmp (&desc->signature,"desc" ,4 )) { +/* valid */ + ret = true; + } + +exit: +return ret; +} + + +/* Prior to sector data for a log entry, there is the header +* and the descriptors referenced in the header: +* +* [] = 4KB sector +* +* [ hdr, desc ][ desc ][ ... ][ data ][ ... ] +* +* The first sector in a log entry has a 64 byte header, and +* up to 126 32-byte descriptors. If more descriptors than +* 126 are required, then subsequent sectors can have up to 128 +* descriptors. Each sector is 4KB. Data follows the descriptor +* sectors. +* +* This will return the number of sectors needed to encompass +* the passed number of descriptors in desc_cnt. +* +* This will never return 0, even if desc_cnt is 0. +*/ +static int vhdx_compute_desc_sectors (uint32_t desc_cnt) +{ +uint32_t desc_sectors; + + desc_cnt +=2 ;/* account for header in first sector */ + desc_sectors = desc_cnt /128 ; +if (desc_cnt %128 ) { + desc_sectors++; + } + +return desc_sectors; +} + + +/* Reads the log header, and subsequent descriptors (if any). This +* will allocate all the space for buffer, which must be NULL when +* passed into this function. Each descriptor will also be validated, +* and error returned if any are invalid. */ +static int vhdx_log_read_desc (BlockDriverState *bs,BDRVVHDXState *s, +VHDXLogEntries *log,VHDXLogDescEntries **buffer) +{ +int ret =0 ; +uint32_t desc_sectors; +uint32_t sectors_read; +VHDXLogEntryHeader hdr; +VHDXLogDescEntries *desc_entries =NULL ; +int i; + +assert (*buffer ==NULL ); + + ret =vhdx_log_peek_hdr (bs, log, &hdr); +if (ret <0 ) { + goto exit; + } +vhdx_log_entry_hdr_le_import (&hdr); +if (vhdx_log_hdr_is_valid (log, &hdr, s) == false) { + ret = -EINVAL ; + goto exit; + } + + desc_sectors =vhdx_compute_desc_sectors (hdr.descriptor_count); + desc_entries =qemu_blockalign (bs, desc_sectors *VHDX_LOG_SECTOR_SIZE ); + + ret =vhdx_log_read_sectors (bs, log, §ors_read, desc_entries, + desc_sectors, false); +if (ret <0 ) { + goto free_and_exit; + } +if (sectors_read != desc_sectors) { + ret = -EINVAL ; + goto free_and_exit; + } + +/* put in proper endianness, and validate each desc */ +for (i =0 ; i < hdr.descriptor_count; i++) { +vhdx_log_desc_le_import (&desc_entries->desc[i]); +if (vhdx_log_desc_is_valid (&desc_entries->desc[i], &hdr) == false) { + ret = -EINVAL ; + goto free_and_exit; + } + } + + *buffer = desc_entries; + goto exit; + +free_and_exit: +qemu_vfree (desc_entries); +exit: +return ret; +} + + +/* Flushes the descriptor described by desc to the VHDX image file. +* If the descriptor is a data descriptor, than 'data' must be non-NULL, +* and >= 4096 bytes (VHDX_LOG_SECTOR_SIZE), containing the data to be +* written. +* +* Verification is performed to make sure the sequence numbers of a data +* descriptor match the sequence number in the desc. +* +* For a zero descriptor, it may describe multiple sectors to fill with zeroes. +* In this case, it should be noted that zeroes are written to disk, and the +* image file is not extended as a sparse file. */ +static int vhdx_log_flush_desc (BlockDriverState *bs,VHDXLogDescriptor *desc, +VHDXLogDataSector *data) +{ +int ret =0 ; +uint64_t seq, file_offset; +uint32_t offset =0 ; +void *buffer =NULL ; +uint64_t count =1 ; +int i; + + buffer =qemu_blockalign (bs,VHDX_LOG_SECTOR_SIZE ); + +if (!memcmp (&desc->signature,"desc" ,4 )) { +/* data sector */ +if (data ==NULL ) { + ret = -EFAULT ; + goto exit; + } + +/* The sequence number of the data sector must match that +* in the descriptor */ + seq = data->sequence_high; + seq <<=32 ; + seq |= data->sequence_low &0xffffffff ; + +if (seq != desc->sequence_number) { + ret = -EINVAL ;
block/vhdx-log.c line 391
Quoted 10 lines
block/vhdx-log.c new file +10
-0+ goto exit; + } + +/* Each data sector is in total 4096 bytes, however the first +* 8 bytes, and last 4 bytes, are located in the descriptor */ +memcpy (buffer, &desc->leading_bytes,sizeof (desc->leading_bytes)); + offset +=sizeof (desc->leading_bytes); + +memcpy (buffer+offset, data->data,4084 ); + offset +=4084 ;
Could you use sizeof(data->data) instead of 4084?
Quoted 304 lines
block/vhdx-log.c new file +304
-0+ +memcpy (buffer+offset, &desc->trailing_bytes, +sizeof (desc->trailing_bytes)); + + }else if (!memcmp (&desc->signature,"zero" ,4 )) { +/* write 'count' sectors of sector */ +memset (buffer,0 ,VHDX_LOG_SECTOR_SIZE ); + count = desc->zero_length /VHDX_LOG_SECTOR_SIZE ; + } + + file_offset = desc->file_offset; + +/* count is only > 1 if we are writing zeroes */ +for (i =0 ; i < count; i++) { + ret =bdrv_pwrite_sync (bs->file, file_offset, buffer, +VHDX_LOG_SECTOR_SIZE ); +if (ret <0 ) { + goto exit; + } + file_offset +=VHDX_LOG_SECTOR_SIZE ; + } + +exit: +qemu_vfree (buffer); +return ret; +} + +/* Flush the entire log (as described by 'logs') to the VHDX image +* file, and then set the log to 'empty' status once complete. +* +* The log entries should be validate prior to flushing */ +static int vhdx_log_flush (BlockDriverState *bs,BDRVVHDXState *s, +VHDXLogSequence *logs) +{ +int ret =0 ; +int i; +uint32_t cnt, sectors_read; +uint64_t new_file_size; +void *data =NULL ; +VHDXLogDescEntries *desc_entries =NULL ; +VHDXLogEntryHeader hdr_tmp = {0 }; + + cnt = logs->count; + + data =qemu_blockalign (bs,VHDX_LOG_SECTOR_SIZE ); + +vhdx_user_visible_write (bs, s); + +/* each iteration represents one log sequence, which may span multiple +* sectors */ +while (cnt--) { + ret =vhdx_log_peek_hdr (bs, &logs->log, &hdr_tmp); +if (ret <0 ) { + goto exit; + } +/* if the log shows a FlushedFileOffset larger than our current file +* size, then that means the file has been truncated / corrupted, and +* we must refused to open it / use it */ +if (hdr_tmp.flushed_file_offset >bdrv_getlength (bs->file)) { + ret = -EINVAL ; + goto exit; + } + + ret =vhdx_log_read_desc (bs, s, &logs->log, &desc_entries); +if (ret <0 ) { + goto exit; + } + +for (i =0 ; i < desc_entries->hdr.descriptor_count; i++) { +if (!memcmp (&desc_entries->desc[i].signature,"desc" ,4 )) { +/* data sector, so read a sector to flush */ + ret =vhdx_log_read_sectors (bs, &logs->log, §ors_read, + data,1 , false); +if (ret <0 ) { + goto exit; + } +if (sectors_read !=1 ) { + ret = -EINVAL ; + goto exit; + } + } + + ret =vhdx_log_flush_desc (bs, &desc_entries->desc[i], data); +if (ret <0 ) { + goto exit; + } + } +if (bdrv_getlength (bs->file) < desc_entries->hdr.last_file_offset) { + new_file_size = desc_entries->hdr.last_file_offset; +if (new_file_size % (1024 *1024 )) { +/* round up to nearest 1MB boundary */ + new_file_size = ((new_file_size >>20 ) +1 ) <<20 ; +bdrv_truncate (bs->file, new_file_size); + } + } +qemu_vfree (desc_entries); + desc_entries =NULL ; + } + +/* once the log is fully flushed, indicate that we have an empty log +* now. This also sets the log guid to 0, to indicate an empty log */ +vhdx_log_reset (bs, s); + +exit: +qemu_vfree (data); +qemu_vfree (desc_entries); +return ret; +} + +static int vhdx_validate_log_entry (BlockDriverState *bs,BDRVVHDXState *s, +VHDXLogEntries *log,uint64_t seq, +bool *valid,VHDXLogEntryHeader *entry) +{ +int ret =0 ; +VHDXLogEntryHeader hdr; +void *buffer =NULL ; +uint32_t i, desc_sectors, total_sectors, crc; +uint32_t sectors_read =0 ; +VHDXLogDescEntries *desc_buffer =NULL ; + + *valid = false; + + ret =vhdx_log_peek_hdr (bs, log, &hdr); +if (ret <0 ) { + goto inc_and_exit; + } + +vhdx_log_entry_hdr_le_import (&hdr); + + +if (vhdx_log_hdr_is_valid (log, &hdr, s) == false) { + goto inc_and_exit; + } + +if (seq >0 ) { +if (hdr.sequence_number != seq +1 ) { + goto inc_and_exit; + } + } + + desc_sectors =vhdx_compute_desc_sectors (hdr.descriptor_count); + +/* Read desc sectors, and calculate log checksum */ + + total_sectors = hdr.entry_length /VHDX_LOG_SECTOR_SIZE ; + + +/* read_desc() will incrememnt the read idx */ + ret =vhdx_log_read_desc (bs, s, log, &desc_buffer); +if (ret <0 ) { + goto free_and_exit; + } + + crc =vhdx_checksum_calc (0xffffffff , (void *)desc_buffer, + desc_sectors *VHDX_LOG_SECTOR_SIZE ,4 ); + crc ^=0xffffffff ; + + buffer =qemu_blockalign (bs,VHDX_LOG_SECTOR_SIZE ); +if (total_sectors > desc_sectors) { +for (i =0 ; i < total_sectors - desc_sectors; i++) { + sectors_read =0 ; + ret =vhdx_log_read_sectors (bs, log, §ors_read, buffer, +1 , false); +if (ret <0 || sectors_read !=1 ) { + goto free_and_exit; + } + crc =vhdx_checksum_calc (crc, buffer,VHDX_LOG_SECTOR_SIZE ,-1 ); + crc ^=0xffffffff ; + } + } + crc ^=0xffffffff ; +if (crc != desc_buffer->hdr.checksum) { + goto free_and_exit; + } + + *valid = true; + *entry = hdr; + goto free_and_exit; + +inc_and_exit: + log->read =vhdx_log_inc_idx (log->read, log->length); + +free_and_exit: +qemu_vfree (buffer); +qemu_vfree (desc_buffer); +return ret; +} + +/* Search through the log circular buffer, and find the valid, active +* log sequence, if any exists +* */ +static int vhdx_log_search (BlockDriverState *bs,BDRVVHDXState *s, +VHDXLogSequence *logs) +{ +int ret =0 ; + +uint64_t curr_seq =0 ; +VHDXLogSequence candidate = {0 }; +VHDXLogSequence current = {0 }; + +uint32_t tail; +bool seq_valid = false; +VHDXLogEntryHeader hdr = {0 }; +VHDXLogEntries curr_log; + +memcpy (&curr_log, &s->log,sizeof (VHDXLogEntries)); + curr_log.write = curr_log.length;/* assume log is full */ + curr_log.read =0 ; + + +/* now we will go through the whole log sector by sector, until +* we find a valid, active log sequence, or reach the end of the +* log buffer */ +for (;;) { + tail = curr_log.read; + + curr_seq =0 ; +memset (¤t,0 ,sizeof (current)); + + ret =vhdx_validate_log_entry (bs, s, &curr_log, curr_seq, + &seq_valid, &hdr); +if (ret <0 ) { + goto exit; + } + +if (seq_valid) { + current.valid = true; + current.log = curr_log; + current.log.read = tail; + current.log.write = curr_log.read; + current.count =1 ; + current.hdr = hdr; + + +for (;;) { + ret =vhdx_validate_log_entry (bs, s, &curr_log, curr_seq, + &seq_valid, &hdr); +if (ret <0 ) { + goto exit; + } +if (seq_valid == false) { +break ; + } + current.log.write = curr_log.read; + current.count++; + + curr_seq = hdr.sequence_number; + } + } + +if (current.valid) { +if (candidate.valid == false || + current.hdr.sequence_number > candidate.hdr.sequence_number) { + candidate = current; + } + } + +if (curr_log.read < tail) { +break ; + } + } + + *logs = candidate; + +if (candidate.valid) { +/* this is the next sequence number, for writes */ + s->log.sequence = candidate.hdr.sequence_number +1 ; + } + + +exit: +return ret; +} + +/* Parse the replay log. Per the VHDX spec, if the log is present +* it must be replayed prior to opening the file, even read-only. +* +* If read-only, we must replay the log in RAM (or refuse to open +* a dirty VHDX file read-only */ +int vhdx_parse_log (BlockDriverState *bs,BDRVVHDXState *s) +{ +int ret =0 ; +VHDXHeader *hdr; +VHDXLogSequence logs = {0 }; + + hdr = s->headers[s->curr_header]; + +/* s->log.hdr is freed in vhdx_close() */ +if (s->log.hdr ==NULL ) { + s->log.hdr =qemu_blockalign (bs,sizeof (VHDXLogEntryHeader)); + } + + s->log.offset = hdr->log_offset; + s->log.length = hdr->log_length; + +if (s->log.offset <VHDX_LOG_MIN_SIZE || + s->log.offset %VHDX_LOG_MIN_SIZE ) { + ret = -EINVAL ; + goto exit; + } + +/* per spec, only log version of 0 is supported */ +if (hdr->log_version !=0 ) { + ret = -EINVAL ;
block/vhdx-log.c line 705
Quoted 10 lines
block/vhdx-log.c new file +10
-0+ goto exit; + } + +/* If either the log guid, or log length is zero, +* then a replay log is not present */ +if (vhdx_log_guid_is_zero (&hdr->log_guid)) { + goto exit; + } + +
Too many blank lines here.
Quoted 12 lines
block/vhdx-log.c new file +12
-0+ +if (hdr->log_length ==0 ) { + goto exit; + } + +if (hdr->log_length %VHDX_LOG_MIN_SIZE ) { + ret = -EINVAL ; + goto exit; + } + + +/* The log is present, we need to find if and where there is an active
block/vhdx-log.c line 727
Quoted 10 lines
block/vhdx-log.c new file +10
-0+* sequence of valid entries present in the log. */ + + ret =vhdx_log_search (bs, s, &logs); +if (ret <0 ) { + goto exit; + } + +if (logs.valid) { +/* now flush the log */ + ret =vhdx_log_flush (bs, s, &logs);
Does it flush log regardless of read-only open?
Quoted 105 lines
block/vhdx-log.c new file +7
-0+ } + + +exit: +return ret; +} +block/vhdx.c +2
-42@@ -735,48 +735,6 @@ exit:return ret; } -/* Parse the replay log. Per the VHDX spec, if the log is present -* it must be replayed prior to opening the file, even read-only. -* -* If read-only, we must replay the log in RAM (or refuse to open -* a dirty VHDX file read-only */ -static int vhdx_parse_log (BlockDriverState *bs,BDRVVHDXState *s) -{ -int ret =0 ; -int i; -VHDXHeader *hdr; - - hdr = s->headers[s->curr_header]; - -/* either the log guid, or log length is zero, -* then a replay log is present */ -for (i =0 ; i <sizeof (hdr->log_guid.data4); i++) { - ret |= hdr->log_guid.data4[i]; - } -if (hdr->log_guid.data1 ==0 && - hdr->log_guid.data2 ==0 && - hdr->log_guid.data3 ==0 && - ret ==0 ) { - goto exit; - } - -/* per spec, only log version of 0 is supported */ -if (hdr->log_version !=0 ) { - ret = -EINVAL ; - goto exit; - } - -if (hdr->log_length ==0 ) { - goto exit; - } - -/* We currently do not support images with logs to replay */ - ret = -ENOTSUP ; - -exit: -return ret; -} -static int vhdx_open (BlockDriverState *bs,QDict *options,int flags) {@@ -789,6 +747,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags) s->bat =NULL ; s->first_visible_write = true; + s->log.write = s->log.read =0 ;qemu_co_mutex_init (&s->lock);@@ -1030,6 +989,7 @@ static void vhdx_close(BlockDriverState *bs)qemu_vfree (s->headers[1 ]);qemu_vfree (s->bat);qemu_vfree (s->parent_entries); +qemu_vfree (s->log.hdr); }static BlockDriver bdrv_vhdx = {block/vhdx.h +6
-1@@ -326,7 +326,11 @@ typedef struct VHDXMetadataEntries {typedef struct VHDXLogEntries {uint64_t offset;uint64_t length; -uint32_t head; +uint32_t write; +uint32_t read; +VHDXLogEntryHeader *hdr; +void *desc_buffer; +uint64_t sequence;uint32_t tail; }VHDXLogEntries ;@@ -387,6 +391,7 @@ uint32_t vhdx_checksum_calc(uint32_t crc, uint8_t *buf, size_t size,bool vhdx_checksum_is_valid (uint8_t *buf,size_t size,int crc_offset); +int vhdx_parse_log (BlockDriverState *bs,BDRVVHDXState *s);static inline void leguid_to_cpus (MSGUID *guid) {Signature · 1 line
1.8.1.4
Signature · 1 line
Fam
Re: [Qemu-devel] [PATCH 8/9] block: vhdx - add log write support
In reply to Jeff CodyQuoted 129 lines · Jeff Cody
This adds support for writing to the VHDX log.
For spec details, see VHDX Specification Format v1.00:
https://www.microsoft.com/en-us/download/details.aspx?id=34750There are a few limitations to this log support:
1.) There is no caching yet
2.) The log is flushed after each entryThe primary write interface,
vhdx_log_write_and_flush(), performs a log write followed by an immediate flush of the log.As each log entry sector is a minimum of 4KB, partial sector writes are filled in with data from the disk write destination.
If the current file log GUID is 0, a new GUID is generated and updated in the header.
Signed-off-by Jeff Cody <[email protected]>--- block/vhdx-log.c | 273 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ block/vhdx.h | 3 + 2 files changed, 276 insertions(+)block/vhdx-log.c +90
-0@@ -170,6 +170,53 @@ exit:return ret; } +/* Writes num_sectors to the log (all log sectors are 4096 bytes), +* from buffer 'buffer'. Upon return, *sectors_written will contain +* the number of sectors successfully written. +* +* It is assumed that 'buffer' is at least 4096*num_sectors large. +* +* 0 is returned on success, -errno otherwise */ +static int vhdx_log_write_sectors (BlockDriverState *bs,VHDXLogEntries *log, +uint32_t *sectors_written,void *buffer, +uint32_t num_sectors) +{ +int ret =0 ; +uint64_t offset; +uint32_t write; +void *buffer_tmp; +BDRVVHDXState *s = bs->opaque; + +vhdx_user_visible_write (bs, s); + + write = log->write; + + buffer_tmp = buffer; +while (num_sectors) { + + offset = log->offset + write; + write =vhdx_log_inc_idx (write, log->length); +if (write == log->read) { +/* full */ +break ; + } + ret =bdrv_pwrite_sync (bs->file, offset, buffer_tmp, +VHDX_LOG_SECTOR_SIZE ); +if (ret <0 ) { + goto exit; + } + buffer_tmp +=VHDX_LOG_SECTOR_SIZE ; + + log->write = write; + *sectors_written = *sectors_written +1 ; + num_sectors--; + } + +exit: +return ret; +} + +/* Validates a log entry header */ static bool vhdx_log_hdr_is_valid (VHDXLogEntries *log,VHDXLogEntryHeader *hdr,BDRVVHDXState *s)@@ -732,3 +779,229 @@ exit:return ret; } + + +static void vhdx_log_raw_to_le_sector (VHDXLogDescriptor *desc, +VHDXLogDataSector *sector,void *data, +uint64_t seq) +{ +memcpy (&desc->leading_bytes, data,8 ); + data +=8 ; +cpu_to_le64s (&desc->leading_bytes); +memcpy (sector->data, data,4084 ); + data +=4084 ; +memcpy (&desc->trailing_bytes, data,4 ); +cpu_to_le32s (&desc->trailing_bytes); + data +=4 ; + + sector->sequence_high = (uint32_t ) (seq >>32 ); + sector->sequence_low = (uint32_t ) (seq &0xffffffff ); + sector->data_signature =VHDX_LOG_DATA_SIGNATURE ; + +vhdx_log_desc_le_export (desc); +vhdx_log_data_le_export (sector); +} + + +static int vhdx_log_write (BlockDriverState *bs,BDRVVHDXState *s, +void *data,uint32_t length,uint64_t offset) +{ +int ret =0 ; +void *buffer =NULL ; +void *merged_sector =NULL ; +void *data_tmp, *sector_write; +unsigned int i; +int sector_offset; +uint32_t desc_sectors, sectors, total_length; +uint32_t sectors_written =0 ; +uint32_t aligned_length; +uint32_t leading_length =0 ; +uint32_t trailing_length =0 ; +uint32_t partial_sectors =0 ; +uint32_t bytes_written =0 ; +uint64_t file_offset; +VHDXHeader *header; +VHDXLogEntryHeader new_hdr;
block/vhdx-log.c exit: · line 834
Quoted 10 lines
block/vhdx-log.c +10
-0+VHDXLogDescriptor *new_desc =NULL ; +VHDXLogDataSector *data_sector =NULL ; +MSGUID new_guid = {0 }; + + header = s->headers[s->curr_header]; + +/* need to have offset read data, and be on 4096 byte boundary */ + +if (length > header->log_length) { +/* no log present. we could create a log here instead of failing */
Does newly created vhdx have allocated log sectors?
block/vhdx-log.c exit: · line 845
Quoted 11 lines
block/vhdx-log.c +11
-0+ ret = -EINVAL ; + goto exit; + } + +if (vhdx_log_guid_is_zero (&header->log_guid)) { +vhdx_guid_generate (&new_guid); +vhdx_update_headers (bs, s, false, &new_guid); + }else { +/* currently, we require that the log be flushed after +* every write. */ + ret = -ENOTSUP ;
Can we make an assertion here?
Quoted 180 lines
block/vhdx-log.c +162
-0+ } + +/* 0 is an invalid sequence number, but may also represent the first +* log write (or a wrapped seq) */ +if (s->log.sequence ==0 ) { + s->log.sequence =1 ; + } + + sector_offset = offset %VHDX_LOG_SECTOR_SIZE ; + file_offset = (offset /VHDX_LOG_SECTOR_SIZE ) *VHDX_LOG_SECTOR_SIZE ; + + aligned_length = length; + +/* add in the unaligned head and tail bytes */ +if (sector_offset) { + leading_length = (VHDX_LOG_SECTOR_SIZE - sector_offset); + leading_length = leading_length > length ? length : leading_length; + aligned_length -= leading_length; + partial_sectors++; + } + + sectors = aligned_length /VHDX_LOG_SECTOR_SIZE ; + trailing_length = aligned_length - (sectors *VHDX_LOG_SECTOR_SIZE ); +if (trailing_length) { + partial_sectors++; + } + + sectors += partial_sectors; + +/* sectors is now how many sectors the data itself takes, not +* including the header and descriptor metadata */ + + new_hdr = (VHDXLogEntryHeader ) { + .signature =VHDX_LOG_SIGNATURE , + .tail = s->log.tail, + .sequence_number = s->log.sequence, + .descriptor_count = sectors, + .reserved =0 , + .flushed_file_offset =bdrv_getlength (bs->file), + .last_file_offset =bdrv_getlength (bs->file), + }; + +memcpy (&new_hdr.log_guid, &header->log_guid,sizeof (MSGUID )); + + desc_sectors =vhdx_compute_desc_sectors (new_hdr.descriptor_count); + + total_length = (desc_sectors + sectors) *VHDX_LOG_SECTOR_SIZE ; + new_hdr.entry_length = total_length; + +vhdx_log_entry_hdr_le_export (&new_hdr); + + buffer =qemu_blockalign (bs, total_length); +memcpy (buffer, &new_hdr,sizeof (new_hdr)); + + new_desc = (VHDXLogDescriptor *) (buffer +sizeof (new_hdr)); + data_sector = buffer + (desc_sectors *VHDX_LOG_SECTOR_SIZE ); + data_tmp = data; + +/* All log sectors are 4KB, so for any partial sectors we must +* merge the data with preexisting data from the final file +* destination */ + merged_sector =qemu_blockalign (bs,VHDX_LOG_SECTOR_SIZE ); + +for (i =0 ; i < sectors; i++) { + new_desc->signature =VHDX_LOG_DESC_SIGNATURE ; + new_desc->sequence_number = s->log.sequence; + new_desc->file_offset = file_offset; + +if (i ==0 && leading_length) { +/* partial sector at the front of the buffer */ + ret =bdrv_pread (bs->file, file_offset, merged_sector, +VHDX_LOG_SECTOR_SIZE ); +if (ret <0 ) { + goto exit; + } +memcpy (merged_sector + sector_offset, data_tmp, leading_length); + bytes_written = leading_length; + sector_write = merged_sector; + }else if (i == sectors -1 && trailing_length) { +/* partial sector at the end of the buffer */ + ret =bdrv_pread (bs->file, + file_offset, + merged_sector + trailing_length, +VHDX_LOG_SECTOR_SIZE - trailing_length); +if (ret <0 ) { + goto exit; + } +memcpy (merged_sector, data_tmp, trailing_length); + bytes_written = trailing_length; + sector_write = merged_sector; + }else { + bytes_written =VHDX_LOG_SECTOR_SIZE ; + sector_write = data_tmp; + } + +/* populate the raw sector data into the proper structures, +* as well as update the descriptor, and convert to proper +* endianness */ +vhdx_log_raw_to_le_sector (new_desc, data_sector, sector_write, + s->log.sequence); + + data_tmp += bytes_written; + data_sector++; + new_desc++; + file_offset +=VHDX_LOG_SECTOR_SIZE ; + } + +/* checksum covers entire entry, from the log header through the +* last data sector */ +vhdx_update_checksum (buffer, total_length,4 ); +cpu_to_le32s ((uint32_t *)(buffer +4 )); + +/* now write to the log */ +vhdx_log_write_sectors (bs, &s->log, §ors_written, buffer, + desc_sectors + sectors); +if (ret <0 ) { + goto exit; + } + +if (sectors_written != desc_sectors + sectors) { +/* instead of failing, we could flush the log here */ + ret = -EINVAL ; + goto exit; + } + + s->log.sequence++; +/* write new tail */ + s->log.tail = s->log.write; + +exit: +qemu_vfree (buffer); +qemu_vfree (merged_sector); +return ret; +} + +/* Perform a log write, and then immediately flush the entire log */ +int vhdx_log_write_and_flush (BlockDriverState *bs,BDRVVHDXState *s, +void *data,uint32_t length,uint64_t offset) +{ +int ret =0 ; +VHDXLogSequence logs = { .valid = true, + .count =1 , + .hdr = {0 } }; + + + ret =vhdx_log_write (bs, s, data, length, offset); +if (ret <0 ) { + goto exit; + } + logs.log = s->log; + + ret =vhdx_log_flush (bs, s, &logs); +if (ret <0 ) { + goto exit; + } + + s->log = logs.log; + +exit: +return ret; +} +block/vhdx.h +3
-0@@ -393,6 +393,9 @@ bool vhdx_checksum_is_valid(uint8_t *buf, size_t size, int crc_offset);int vhdx_parse_log (BlockDriverState *bs,BDRVVHDXState *s); +int vhdx_log_write_and_flush (BlockDriverState *bs,BDRVVHDXState *s, +void *data,uint32_t length,uint64_t offset); +static inline void leguid_to_cpus (MSGUID *guid) {le32_to_cpus (&guid->data1);Signature · 1 line
1.8.1.4
Signature · 1 line
Fam
Re: [Qemu-devel] [PATCH 9/9] block: vhdx write support
In reply to Jeff CodyQuoted 32 lines · Jeff Cody
This adds support for writing to VHDX image files, using coroutines. Writes into the BAT table goes through the VHDX log. Currently, BAT table writes occur when expanding a dynamic VHDX file, and allocating a new BAT entry.
Signed-off-by Jeff Cody <[email protected]>--- block/vhdx.c | 149 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 147 insertions(+), 2 deletions(-)block/vhdx.c +6
-1@@ -831,7 +831,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags)vhdx_update_headers (bs, s, false,NULL ); } -/* TODO: differencing files, write */ +/* TODO: differencing files */ return 0 ; fail:@@ -963,7 +963,45 @@ exit:return ret; } +/* +* Allocate a new payload block at the end of the file. +* +* Allocation will happen at 1MB alignment inside the file +*
block/vhdx.c exit: · line 980
Quoted 10 lines
block/vhdx.c +9
-0+* Returns the file offset start of the new payload block +*/ +static int vhdx_allocate_block (BlockDriverState *bs,BDRVVHDXState *s, +uint64_t *new_offset) +{ + *new_offset =bdrv_getlength (bs->file); +/* per the spec, the address for a block is in units of 1MB */ +if (*new_offset % (1024 *1024 )) { + *new_offset = ((*new_offset >>20 ) +1 ) <<20 ;/* round up to 1MB */
You can use ROUND_UP() macro here:
*new_offset = ROUND_UP(*new_offset, 1 << 20);
Fam
Re: [Qemu-devel] [PATCH 4/9] block: vhdx - log support struct and defines
In reply to Fam ZhengQuoted 27 lines · Fam Zheng
Quoted 26 lines · Jeff Cody
This adds some magic number defines, and internal structure definitions for VHDX log replay support.
Signed-off-by Jeff Cody <[email protected]>--- block/vhdx.h | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-)block/vhdx.h +3
-0@@ -151,7 +151,10 @@ typedef struct QEMU_PACKED VHDXRegionTableEntry {/* ---- LOG ENTRY STRUCTURES ---- */ +#define VHDX_LOG_MIN_SIZE (1024*1024) +#define VHDX_LOG_SECTOR_SIZE 4096#define VHDX_LOG_HDR_SIZE 64 +#define VHDX_LOG_SIGNATURE 0x65676f6c typedef struct QEMU_PACKED VHDXLogEntryHeader {uint32_t signature;/* "loge" in ASCII */ uint32_t checksum;/* CRC-32C hash of the 64KB table */ @@ -174,7 +177,8 @@ typedef struct QEMU_PACKED VHDXLogEntryHeader { } VHDXLogEntryHeader;
Quoted 8 lines
block/vhdx.h typedef struct QEMU_PACKED VHDXLogEntryHeader { · line 181
Quoted 4 lines
block/vhdx.h +2
-1#define VHDX_LOG_DESC_SIZE 32 - +#define VHDX_LOG_DESC_SIGNATURE 0x63736564 +#define VHDX_LOG_ZERO_SIGNATURE 0x6f72657aAre these macros really used? I see "desc" and "zero" used to compare signatures.
They are used in the log write patch (when creating new log sectors). Right now, we only create data sectors (nothing uses the zero sectors yet), so only the _DESC_SIGNATURE is used.
Thanks,
Jeff
Re: [Qemu-devel] [PATCH 7/9] block: vhdx - log parsing, replay, and flush support
In reply to Fam ZhengQuoted 129 lines · Fam Zheng
Quoted 128 lines · Jeff Cody
This adds support for VHDX v0 logs, as specified in Microsoft's VHDX Specification Format v1.00:
https://www.microsoft.com/en-us/download/details.aspx?id=34750The following support is added:
* Log parsing, and validation - validate that an existing log is correct.
* Log search - search through an existing log, to find any valid sequence of entries.
* Log replay and flush - replay an existing log, and flush/clear the log when complete.
The VHDX log is a circular buffer, with elements (sectors) of 4KB.
A log entry is a variably-length number of sectors, that is comprised of a header and 'descriptors', that describe each sector.
A log may contain multiple entries, know as a log sequence. In a log sequence, each log entry immediately follows the previous entry, with an incrementing sequence number. There can only ever be one active and valid sequence in the log.
Each log entry must match the file log GUID in order to be valid (along with other criteria). Once we have flushed all valid log entries, we marked the file log GUID to be zero, which indicates a buffer with no valid entries.
Signed-off-by Jeff Cody <[email protected]>--- block/Makefile.objs | 2 +- block/vhdx-log.c | 734 ++++++++++++++++++++++++++++++++++++++++++++++++++++ block/vhdx.c | 44 +--- block/vhdx.h | 7 +- 4 files changed, 743 insertions(+), 44 deletions(-) create mode 100644 block/vhdx-log.cblock/Makefile.objs +1
-1@@ -2,7 +2,7 @@ block-obj-y += raw.o cow.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o block-obj-y += qed-check.o -block-obj-$(CONFIG_VHDX) += vhdx.o vhdx-endian.o +block-obj-$(CONFIG_VHDX) += vhdx.o vhdx-endian.o vhdx-log.o block-obj-y += parallels.o blkdebug.o blkverify.o block-obj-y += snapshot.o qapi.o block-obj-$(CONFIG_WIN32) += raw-win32.o win32-aio.oblock/vhdx-log.c new file +70
-0@@ -0,0 +1,734 @@ +/* +* Block driver for Hyper-V VHDX Images +* +* Copyright (c) 2013 Red Hat, Inc., +* +* Authors: +* Jeff Cody <[email protected]> +* +* This is based on the "VHDX Format Specification v1.00", published 8/25/2012 +* by Microsoft: +* https://www.microsoft.com/en-us/download/details.aspx?id=34750 +* +* This file covers the functionality of the metadata log writing, parsing, and +* replay. +* +* This work is licensed under the terms of the GNU LGPL, version 2 or later. +* See the COPYING.LIB file in the top-level directory. +* +*/ +#include "qemu-common.h" +#include "block/block_int.h" +#include "qemu/module.h" +#include "block/vhdx.h" + + +typedef struct VHDXLogSequence { +bool valid; +uint32_t count; +VHDXLogEntries log; +VHDXLogEntryHeader hdr; +}VHDXLogSequence ; + +typedef struct VHDXLogDescEntries { +VHDXLogEntryHeader hdr; +VHDXLogDescriptor desc[]; +}VHDXLogDescEntries ; + + +/* Returns true if the GUID is zero */ +static bool vhdx_log_guid_is_zero (MSGUID *guid) +{ +int i; +int ret =0 ; + +/* If either the log guid, or log length is zero, +* then a replay log is not present */ +for (i =0 ; i <sizeof (MSGUID ); i++) { + ret |= ((uint8_t *) guid)[i]; + } + +return ret ==0 ; +} + +/* The log located on the disk is circular buffer containing +* sectors of 4096 bytes each. +* +* It is assumed for the read/write functions below that the +* circular buffer scheme uses a 'one sector open' to indicate +* the buffer is full. Given the validation methods used for each +* sector, this method should be compatible with other methods that +* do not waste a sector. +*/ + + +/* Allow peeking at the hdr entry at the beginning of the current +* read index, without advancing the read index */ +static int vhdx_log_peek_hdr (BlockDriverState *bs,VHDXLogEntries *log, +VHDXLogEntryHeader *hdr) +{ +int ret =0 ;
Quoted 9 lines
block/vhdx-log.c line 76
Quoted 6 lines
block/vhdx-log.c new file +6
-0+uint64_t offset; +uint32_t read; + +assert (hdr !=NULL ); + +/* peek is only support on sector boundaries */ s/support/supported/
Thanks
Quoted 135 lines
Quoted 135 lines
block/vhdx-log.c new file +135
-0+if (log->read %VHDX_LOG_SECTOR_SIZE ) { + ret = -EFAULT ; + goto exit; + } + + read = log->read; +/* we are guaranteed that a) log sectors are 4096 bytes, +* and b) the log length is a multiple of 1MB. So, there +* is always a round number of sectors in the buffer */ +if ((read +sizeof (VHDXLogEntryHeader)) > log->length) { + read =0 ; + } + +if (read == log->write) { + ret = -EINVAL ; + goto exit; + } + + offset = log->offset + read; + + ret =bdrv_pread (bs->file, offset, hdr,sizeof (VHDXLogEntryHeader)); +if (ret <0 ) { + goto exit; + } + +exit: +return ret; +} + +/* Index increment for log, based on sector boundaries */ +static int vhdx_log_inc_idx (uint32_t idx,uint64_t length) +{ + idx +=VHDX_LOG_SECTOR_SIZE ; +/* we are guaranteed that a) log sectors are 4096 bytes, +* and b) the log length is a multiple of 1MB. So, there +* is always a round number of sectors in the buffer */ +return idx >= length ?0 : idx; +} + + +/* Reset the log to empty */ +static void vhdx_log_reset (BlockDriverState *bs,BDRVVHDXState *s) +{ +MSGUID guid = {0 }; + s->log.read = s->log.write =0 ; +/* a log guid of 0 indicates an empty log to any parser of v0 +* VHDX logs */ +vhdx_update_headers (bs, s, false, &guid); +} + +/* Reads num_sectors from the log (all log sectors are 4096 bytes), +* into buffer 'buffer'. Upon return, *sectors_read will contain +* the number of sectors successfully read. +* +* It is assumed that 'buffer' is already allocated, and of sufficient +* size (i.e. >= 4096*num_sectors). +* +* If 'peek' is true, then the tail (read) pointer for the circular buffer is +* not modified. +* +* 0 is returned on success, -errno otherwise. */ +static int vhdx_log_read_sectors (BlockDriverState *bs,VHDXLogEntries *log, +uint32_t *sectors_read,void *buffer, +uint32_t num_sectors,bool peek) +{ +int ret =0 ; +uint64_t offset; +uint32_t read; + + read = log->read; + + *sectors_read =0 ; +while (num_sectors) { +if (read == log->write) { +/* empty */ +break ; + } + offset = log->offset + read; + + ret =bdrv_pread (bs->file, offset, buffer,VHDX_LOG_SECTOR_SIZE ); +if (ret <0 ) { + goto exit; + } + read =vhdx_log_inc_idx (read, log->length); + + *sectors_read = *sectors_read +1 ; + num_sectors--; + } + +exit: +if (!peek) { + log->read = read; + } +return ret; +} + +/* Validates a log entry header */ +static bool vhdx_log_hdr_is_valid (VHDXLogEntries *log,VHDXLogEntryHeader *hdr, +BDRVVHDXState *s) +{ +int valid = false; + +if (memcmp (&hdr->signature,"loge" ,4 )) { + goto exit; + } + +/* if the individual entry length is larger than the whole log +* buffer, that is obviously invalid */ +if (log->length < hdr->entry_length) { + goto exit; + } + +/* length of entire entry must be in units of 4KB (log sector size) */ +if (hdr->entry_length % (VHDX_LOG_SECTOR_SIZE )) { + goto exit; + } + +/* per spec, sequence # must be > 0 */ +if (hdr->sequence_number ==0 ) { + goto exit; + } + +/* log entries are only valid if they match the file-wide log guid +* found in the active header */ +if (!guid_eq (hdr->log_guid, s->headers[s->curr_header]->log_guid)) { + goto exit; + } + + valid = true; + +exit: +return valid; +} + +/*
Quoted 9 lines
block/vhdx-log.c line 217
Quoted 6 lines
block/vhdx-log.c new file +6
-0+* Given a log header, this will validate that the descriptors and the +* corresponding data sectors (if applicable) +* +* Validation consists of: +* 1. Making sure the sequence numbers matches the entry header +* 2. Verifying a valid signature ('zero' or desc' for descriptors) s/ desc'/ 'desc'/
Thanks
Quoted 7 lines
block/vhdx-log.c line 223
Quoted 6 lines
block/vhdx-log.c new file +6
-0+* 3. File offset field is a multiple of 4KB +* 4. If a data descriptor, the corresponding data sector +* has its signature ('data') and matching sequence number +* +* 'desc' is the data buffer containing the descriptor +* hdr is the log entry header
Quoted 5 lines
Please use gtkdoc format:
@desc: the data buffer ...
@hdr: the log entry header
Sure, I can do that.
Quoted 162 lines
Quoted 162 lines
block/vhdx-log.c new file +162
-0+* +* Returns true if valid +*/ +static bool vhdx_log_desc_is_valid (VHDXLogDescriptor *desc, +VHDXLogEntryHeader *hdr) +{ +bool ret = false; + +if (desc->sequence_number != hdr->sequence_number) { + goto exit; + } +if (desc->file_offset %VHDX_LOG_SECTOR_SIZE ) { + goto exit; + } + +if (!memcmp (&desc->signature,"zero" ,4 )) { +if (!desc->zero_length %VHDX_LOG_SECTOR_SIZE ) { +/* valid */ + ret = true; + } + }else if (!memcmp (&desc->signature,"desc" ,4 )) { +/* valid */ + ret = true; + } + +exit: +return ret; +} + + +/* Prior to sector data for a log entry, there is the header +* and the descriptors referenced in the header: +* +* [] = 4KB sector +* +* [ hdr, desc ][ desc ][ ... ][ data ][ ... ] +* +* The first sector in a log entry has a 64 byte header, and +* up to 126 32-byte descriptors. If more descriptors than +* 126 are required, then subsequent sectors can have up to 128 +* descriptors. Each sector is 4KB. Data follows the descriptor +* sectors. +* +* This will return the number of sectors needed to encompass +* the passed number of descriptors in desc_cnt. +* +* This will never return 0, even if desc_cnt is 0. +*/ +static int vhdx_compute_desc_sectors (uint32_t desc_cnt) +{ +uint32_t desc_sectors; + + desc_cnt +=2 ;/* account for header in first sector */ + desc_sectors = desc_cnt /128 ; +if (desc_cnt %128 ) { + desc_sectors++; + } + +return desc_sectors; +} + + +/* Reads the log header, and subsequent descriptors (if any). This +* will allocate all the space for buffer, which must be NULL when +* passed into this function. Each descriptor will also be validated, +* and error returned if any are invalid. */ +static int vhdx_log_read_desc (BlockDriverState *bs,BDRVVHDXState *s, +VHDXLogEntries *log,VHDXLogDescEntries **buffer) +{ +int ret =0 ; +uint32_t desc_sectors; +uint32_t sectors_read; +VHDXLogEntryHeader hdr; +VHDXLogDescEntries *desc_entries =NULL ; +int i; + +assert (*buffer ==NULL ); + + ret =vhdx_log_peek_hdr (bs, log, &hdr); +if (ret <0 ) { + goto exit; + } +vhdx_log_entry_hdr_le_import (&hdr); +if (vhdx_log_hdr_is_valid (log, &hdr, s) == false) { + ret = -EINVAL ; + goto exit; + } + + desc_sectors =vhdx_compute_desc_sectors (hdr.descriptor_count); + desc_entries =qemu_blockalign (bs, desc_sectors *VHDX_LOG_SECTOR_SIZE ); + + ret =vhdx_log_read_sectors (bs, log, §ors_read, desc_entries, + desc_sectors, false); +if (ret <0 ) { + goto free_and_exit; + } +if (sectors_read != desc_sectors) { + ret = -EINVAL ; + goto free_and_exit; + } + +/* put in proper endianness, and validate each desc */ +for (i =0 ; i < hdr.descriptor_count; i++) { +vhdx_log_desc_le_import (&desc_entries->desc[i]); +if (vhdx_log_desc_is_valid (&desc_entries->desc[i], &hdr) == false) { + ret = -EINVAL ; + goto free_and_exit; + } + } + + *buffer = desc_entries; + goto exit; + +free_and_exit: +qemu_vfree (desc_entries); +exit: +return ret; +} + + +/* Flushes the descriptor described by desc to the VHDX image file. +* If the descriptor is a data descriptor, than 'data' must be non-NULL, +* and >= 4096 bytes (VHDX_LOG_SECTOR_SIZE), containing the data to be +* written. +* +* Verification is performed to make sure the sequence numbers of a data +* descriptor match the sequence number in the desc. +* +* For a zero descriptor, it may describe multiple sectors to fill with zeroes. +* In this case, it should be noted that zeroes are written to disk, and the +* image file is not extended as a sparse file. */ +static int vhdx_log_flush_desc (BlockDriverState *bs,VHDXLogDescriptor *desc, +VHDXLogDataSector *data) +{ +int ret =0 ; +uint64_t seq, file_offset; +uint32_t offset =0 ; +void *buffer =NULL ; +uint64_t count =1 ; +int i; + + buffer =qemu_blockalign (bs,VHDX_LOG_SECTOR_SIZE ); + +if (!memcmp (&desc->signature,"desc" ,4 )) { +/* data sector */ +if (data ==NULL ) { + ret = -EFAULT ; + goto exit; + } + +/* The sequence number of the data sector must match that +* in the descriptor */ + seq = data->sequence_high; + seq <<=32 ; + seq |= data->sequence_low &0xffffffff ; + +if (seq != desc->sequence_number) { + ret = -EINVAL ; + goto exit; + } + +/* Each data sector is in total 4096 bytes, however the first
Quoted 9 lines
block/vhdx-log.c line 391
Quoted 6 lines
block/vhdx-log.c new file +6
-0+* 8 bytes, and last 4 bytes, are located in the descriptor */ +memcpy (buffer, &desc->leading_bytes,sizeof (desc->leading_bytes)); + offset +=sizeof (desc->leading_bytes); + +memcpy (buffer+offset, data->data,4084 ); + offset +=4084 ;Could you use
sizeof(data->data)instead of 4084?
I suppose I should be consistent, and either use the explicit sizes, or use 'sizeof'. I was on the fence on whether it is better for these three variables to use sizeof() or explicit sizes.
Quoted 331 lines
Quoted 304 lines
block/vhdx-log.c new file +304
-0+ +memcpy (buffer+offset, &desc->trailing_bytes, +sizeof (desc->trailing_bytes)); + + }else if (!memcmp (&desc->signature,"zero" ,4 )) { +/* write 'count' sectors of sector */ +memset (buffer,0 ,VHDX_LOG_SECTOR_SIZE ); + count = desc->zero_length /VHDX_LOG_SECTOR_SIZE ; + } + + file_offset = desc->file_offset; + +/* count is only > 1 if we are writing zeroes */ +for (i =0 ; i < count; i++) { + ret =bdrv_pwrite_sync (bs->file, file_offset, buffer, +VHDX_LOG_SECTOR_SIZE ); +if (ret <0 ) { + goto exit; + } + file_offset +=VHDX_LOG_SECTOR_SIZE ; + } + +exit: +qemu_vfree (buffer); +return ret; +} + +/* Flush the entire log (as described by 'logs') to the VHDX image +* file, and then set the log to 'empty' status once complete. +* +* The log entries should be validate prior to flushing */ +static int vhdx_log_flush (BlockDriverState *bs,BDRVVHDXState *s, +VHDXLogSequence *logs) +{ +int ret =0 ; +int i; +uint32_t cnt, sectors_read; +uint64_t new_file_size; +void *data =NULL ; +VHDXLogDescEntries *desc_entries =NULL ; +VHDXLogEntryHeader hdr_tmp = {0 }; + + cnt = logs->count; + + data =qemu_blockalign (bs,VHDX_LOG_SECTOR_SIZE ); + +vhdx_user_visible_write (bs, s); + +/* each iteration represents one log sequence, which may span multiple +* sectors */ +while (cnt--) { + ret =vhdx_log_peek_hdr (bs, &logs->log, &hdr_tmp); +if (ret <0 ) { + goto exit; + } +/* if the log shows a FlushedFileOffset larger than our current file +* size, then that means the file has been truncated / corrupted, and +* we must refused to open it / use it */ +if (hdr_tmp.flushed_file_offset >bdrv_getlength (bs->file)) { + ret = -EINVAL ; + goto exit; + } + + ret =vhdx_log_read_desc (bs, s, &logs->log, &desc_entries); +if (ret <0 ) { + goto exit; + } + +for (i =0 ; i < desc_entries->hdr.descriptor_count; i++) { +if (!memcmp (&desc_entries->desc[i].signature,"desc" ,4 )) { +/* data sector, so read a sector to flush */ + ret =vhdx_log_read_sectors (bs, &logs->log, §ors_read, + data,1 , false); +if (ret <0 ) { + goto exit; + } +if (sectors_read !=1 ) { + ret = -EINVAL ; + goto exit; + } + } + + ret =vhdx_log_flush_desc (bs, &desc_entries->desc[i], data); +if (ret <0 ) { + goto exit; + } + } +if (bdrv_getlength (bs->file) < desc_entries->hdr.last_file_offset) { + new_file_size = desc_entries->hdr.last_file_offset; +if (new_file_size % (1024 *1024 )) { +/* round up to nearest 1MB boundary */ + new_file_size = ((new_file_size >>20 ) +1 ) <<20 ; +bdrv_truncate (bs->file, new_file_size); + } + } +qemu_vfree (desc_entries); + desc_entries =NULL ; + } + +/* once the log is fully flushed, indicate that we have an empty log +* now. This also sets the log guid to 0, to indicate an empty log */ +vhdx_log_reset (bs, s); + +exit: +qemu_vfree (data); +qemu_vfree (desc_entries); +return ret; +} + +static int vhdx_validate_log_entry (BlockDriverState *bs,BDRVVHDXState *s, +VHDXLogEntries *log,uint64_t seq, +bool *valid,VHDXLogEntryHeader *entry) +{ +int ret =0 ; +VHDXLogEntryHeader hdr; +void *buffer =NULL ; +uint32_t i, desc_sectors, total_sectors, crc; +uint32_t sectors_read =0 ; +VHDXLogDescEntries *desc_buffer =NULL ; + + *valid = false; + + ret =vhdx_log_peek_hdr (bs, log, &hdr); +if (ret <0 ) { + goto inc_and_exit; + } + +vhdx_log_entry_hdr_le_import (&hdr); + + +if (vhdx_log_hdr_is_valid (log, &hdr, s) == false) { + goto inc_and_exit; + } + +if (seq >0 ) { +if (hdr.sequence_number != seq +1 ) { + goto inc_and_exit; + } + } + + desc_sectors =vhdx_compute_desc_sectors (hdr.descriptor_count); + +/* Read desc sectors, and calculate log checksum */ + + total_sectors = hdr.entry_length /VHDX_LOG_SECTOR_SIZE ; + + +/* read_desc() will incrememnt the read idx */ + ret =vhdx_log_read_desc (bs, s, log, &desc_buffer); +if (ret <0 ) { + goto free_and_exit; + } + + crc =vhdx_checksum_calc (0xffffffff , (void *)desc_buffer, + desc_sectors *VHDX_LOG_SECTOR_SIZE ,4 ); + crc ^=0xffffffff ; + + buffer =qemu_blockalign (bs,VHDX_LOG_SECTOR_SIZE ); +if (total_sectors > desc_sectors) { +for (i =0 ; i < total_sectors - desc_sectors; i++) { + sectors_read =0 ; + ret =vhdx_log_read_sectors (bs, log, §ors_read, buffer, +1 , false); +if (ret <0 || sectors_read !=1 ) { + goto free_and_exit; + } + crc =vhdx_checksum_calc (crc, buffer,VHDX_LOG_SECTOR_SIZE ,-1 ); + crc ^=0xffffffff ; + } + } + crc ^=0xffffffff ; +if (crc != desc_buffer->hdr.checksum) { + goto free_and_exit; + } + + *valid = true; + *entry = hdr; + goto free_and_exit; + +inc_and_exit: + log->read =vhdx_log_inc_idx (log->read, log->length); + +free_and_exit: +qemu_vfree (buffer); +qemu_vfree (desc_buffer); +return ret; +} + +/* Search through the log circular buffer, and find the valid, active +* log sequence, if any exists +* */ +static int vhdx_log_search (BlockDriverState *bs,BDRVVHDXState *s, +VHDXLogSequence *logs) +{ +int ret =0 ; + +uint64_t curr_seq =0 ; +VHDXLogSequence candidate = {0 }; +VHDXLogSequence current = {0 }; + +uint32_t tail; +bool seq_valid = false; +VHDXLogEntryHeader hdr = {0 }; +VHDXLogEntries curr_log; + +memcpy (&curr_log, &s->log,sizeof (VHDXLogEntries)); + curr_log.write = curr_log.length;/* assume log is full */ + curr_log.read =0 ; + + +/* now we will go through the whole log sector by sector, until +* we find a valid, active log sequence, or reach the end of the +* log buffer */ +for (;;) { + tail = curr_log.read; + + curr_seq =0 ; +memset (¤t,0 ,sizeof (current)); + + ret =vhdx_validate_log_entry (bs, s, &curr_log, curr_seq, + &seq_valid, &hdr); +if (ret <0 ) { + goto exit; + } + +if (seq_valid) { + current.valid = true; + current.log = curr_log; + current.log.read = tail; + current.log.write = curr_log.read; + current.count =1 ; + current.hdr = hdr; + + +for (;;) { + ret =vhdx_validate_log_entry (bs, s, &curr_log, curr_seq, + &seq_valid, &hdr); +if (ret <0 ) { + goto exit; + } +if (seq_valid == false) { +break ; + } + current.log.write = curr_log.read; + current.count++; + + curr_seq = hdr.sequence_number; + } + } + +if (current.valid) { +if (candidate.valid == false || + current.hdr.sequence_number > candidate.hdr.sequence_number) { + candidate = current; + } + } + +if (curr_log.read < tail) { +break ; + } + } + + *logs = candidate; + +if (candidate.valid) { +/* this is the next sequence number, for writes */ + s->log.sequence = candidate.hdr.sequence_number +1 ; + } + + +exit: +return ret; +} + +/* Parse the replay log. Per the VHDX spec, if the log is present +* it must be replayed prior to opening the file, even read-only. +* +* If read-only, we must replay the log in RAM (or refuse to open +* a dirty VHDX file read-only */ +int vhdx_parse_log (BlockDriverState *bs,BDRVVHDXState *s) +{ +int ret =0 ; +VHDXHeader *hdr; +VHDXLogSequence logs = {0 }; + + hdr = s->headers[s->curr_header]; + +/* s->log.hdr is freed in vhdx_close() */ +if (s->log.hdr ==NULL ) { + s->log.hdr =qemu_blockalign (bs,sizeof (VHDXLogEntryHeader)); + } + + s->log.offset = hdr->log_offset; + s->log.length = hdr->log_length; + +if (s->log.offset <VHDX_LOG_MIN_SIZE || + s->log.offset %VHDX_LOG_MIN_SIZE ) { + ret = -EINVAL ; + goto exit; + } + +/* per spec, only log version of 0 is supported */ +if (hdr->log_version !=0 ) { + ret = -EINVAL ;block/vhdx-log.c line 705
Quoted 10 lines
block/vhdx-log.c new file +10
-0+ goto exit; + } + +/* If either the log guid, or log length is zero, +* then a replay log is not present */ +if (vhdx_log_guid_is_zero (&hdr->log_guid)) { + goto exit; + } + +Too many blank lines here.
Quoted 16 lines
block/vhdx-log.c new file +16
-0+ +if (hdr->log_length ==0 ) { + goto exit; + } + +if (hdr->log_length %VHDX_LOG_MIN_SIZE ) { + ret = -EINVAL ; + goto exit; + } + + +/* The log is present, we need to find if and where there is an active +* sequence of valid entries present in the log. */ + + ret =vhdx_log_search (bs, s, &logs); +if (ret <0 ) {
Quoted 9 lines
block/vhdx-log.c line 727
Quoted 6 lines
block/vhdx-log.c new file +6
-0+ goto exit; + } + +if (logs.valid) { +/* now flush the log */ + ret =vhdx_log_flush (bs, s, &logs);Does it flush log regardless of read-only open?
Yes - that will cause an error return to go up the stack once the write is attempted.
Alternatively, I could return error if the image file has a dirty log, and is opened read-only (in that case, could return -ENOTSUP).
Quoted 108 lines
Quoted 105 lines
block/vhdx-log.c new file +7
-0+ } + + +exit: +return ret; +} +block/vhdx.c +2
-42@@ -735,48 +735,6 @@ exit:return ret; } -/* Parse the replay log. Per the VHDX spec, if the log is present -* it must be replayed prior to opening the file, even read-only. -* -* If read-only, we must replay the log in RAM (or refuse to open -* a dirty VHDX file read-only */ -static int vhdx_parse_log (BlockDriverState *bs,BDRVVHDXState *s) -{ -int ret =0 ; -int i; -VHDXHeader *hdr; - - hdr = s->headers[s->curr_header]; - -/* either the log guid, or log length is zero, -* then a replay log is present */ -for (i =0 ; i <sizeof (hdr->log_guid.data4); i++) { - ret |= hdr->log_guid.data4[i]; - } -if (hdr->log_guid.data1 ==0 && - hdr->log_guid.data2 ==0 && - hdr->log_guid.data3 ==0 && - ret ==0 ) { - goto exit; - } - -/* per spec, only log version of 0 is supported */ -if (hdr->log_version !=0 ) { - ret = -EINVAL ; - goto exit; - } - -if (hdr->log_length ==0 ) { - goto exit; - } - -/* We currently do not support images with logs to replay */ - ret = -ENOTSUP ; - -exit: -return ret; -} -static int vhdx_open (BlockDriverState *bs,QDict *options,int flags) {@@ -789,6 +747,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags) s->bat =NULL ; s->first_visible_write = true; + s->log.write = s->log.read =0 ;qemu_co_mutex_init (&s->lock);@@ -1030,6 +989,7 @@ static void vhdx_close(BlockDriverState *bs)qemu_vfree (s->headers[1 ]);qemu_vfree (s->bat);qemu_vfree (s->parent_entries); +qemu_vfree (s->log.hdr); }static BlockDriver bdrv_vhdx = {block/vhdx.h +6
-1@@ -326,7 +326,11 @@ typedef struct VHDXMetadataEntries {typedef struct VHDXLogEntries {uint64_t offset;uint64_t length; -uint32_t head; +uint32_t write; +uint32_t read; +VHDXLogEntryHeader *hdr; +void *desc_buffer; +uint64_t sequence;uint32_t tail; }VHDXLogEntries ;@@ -387,6 +391,7 @@ uint32_t vhdx_checksum_calc(uint32_t crc, uint8_t *buf, size_t size,bool vhdx_checksum_is_valid (uint8_t *buf,size_t size,int crc_offset); +int vhdx_parse_log (BlockDriverState *bs,BDRVVHDXState *s);static inline void leguid_to_cpus (MSGUID *guid) {Signature · 1 line
1.8.1.4Signature · 1 line
Fam
Re: [Qemu-devel] [PATCH 8/9] block: vhdx - add log write support
In reply to Fam ZhengQuoted 134 lines · Fam Zheng
Quoted 133 lines · Jeff Cody
This adds support for writing to the VHDX log.
For spec details, see VHDX Specification Format v1.00:
https://www.microsoft.com/en-us/download/details.aspx?id=34750There are a few limitations to this log support:
1.) There is no caching yet
2.) The log is flushed after each entryThe primary write interface,
vhdx_log_write_and_flush(), performs a log write followed by an immediate flush of the log.As each log entry sector is a minimum of 4KB, partial sector writes are filled in with data from the disk write destination.
If the current file log GUID is 0, a new GUID is generated and updated in the header.
Signed-off-by Jeff Cody <[email protected]>--- block/vhdx-log.c | 273 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ block/vhdx.h | 3 + 2 files changed, 276 insertions(+)block/vhdx-log.c +94
-0@@ -170,6 +170,53 @@ exit:return ret; } +/* Writes num_sectors to the log (all log sectors are 4096 bytes), +* from buffer 'buffer'. Upon return, *sectors_written will contain +* the number of sectors successfully written. +* +* It is assumed that 'buffer' is at least 4096*num_sectors large. +* +* 0 is returned on success, -errno otherwise */ +static int vhdx_log_write_sectors (BlockDriverState *bs,VHDXLogEntries *log, +uint32_t *sectors_written,void *buffer, +uint32_t num_sectors) +{ +int ret =0 ; +uint64_t offset; +uint32_t write; +void *buffer_tmp; +BDRVVHDXState *s = bs->opaque; + +vhdx_user_visible_write (bs, s); + + write = log->write; + + buffer_tmp = buffer; +while (num_sectors) { + + offset = log->offset + write; + write =vhdx_log_inc_idx (write, log->length); +if (write == log->read) { +/* full */ +break ; + } + ret =bdrv_pwrite_sync (bs->file, offset, buffer_tmp, +VHDX_LOG_SECTOR_SIZE ); +if (ret <0 ) { + goto exit; + } + buffer_tmp +=VHDX_LOG_SECTOR_SIZE ; + + log->write = write; + *sectors_written = *sectors_written +1 ; + num_sectors--; + } + +exit: +return ret; +} + +/* Validates a log entry header */ static bool vhdx_log_hdr_is_valid (VHDXLogEntries *log,VHDXLogEntryHeader *hdr,BDRVVHDXState *s)@@ -732,3 +779,229 @@ exit:return ret; } + + +static void vhdx_log_raw_to_le_sector (VHDXLogDescriptor *desc, +VHDXLogDataSector *sector,void *data, +uint64_t seq) +{ +memcpy (&desc->leading_bytes, data,8 ); + data +=8 ; +cpu_to_le64s (&desc->leading_bytes); +memcpy (sector->data, data,4084 ); + data +=4084 ; +memcpy (&desc->trailing_bytes, data,4 ); +cpu_to_le32s (&desc->trailing_bytes); + data +=4 ; + + sector->sequence_high = (uint32_t ) (seq >>32 ); + sector->sequence_low = (uint32_t ) (seq &0xffffffff ); + sector->data_signature =VHDX_LOG_DATA_SIGNATURE ; + +vhdx_log_desc_le_export (desc); +vhdx_log_data_le_export (sector); +} + + +static int vhdx_log_write (BlockDriverState *bs,BDRVVHDXState *s, +void *data,uint32_t length,uint64_t offset) +{ +int ret =0 ; +void *buffer =NULL ; +void *merged_sector =NULL ; +void *data_tmp, *sector_write; +unsigned int i; +int sector_offset; +uint32_t desc_sectors, sectors, total_length; +uint32_t sectors_written =0 ; +uint32_t aligned_length; +uint32_t leading_length =0 ; +uint32_t trailing_length =0 ; +uint32_t partial_sectors =0 ; +uint32_t bytes_written =0 ; +uint64_t file_offset; +VHDXHeader *header; +VHDXLogEntryHeader new_hdr; +VHDXLogDescriptor *new_desc =NULL ; +VHDXLogDataSector *data_sector =NULL ; +MSGUID new_guid = {0 }; +
Quoted 9 lines
block/vhdx-log.c exit: · line 834
Quoted 6 lines
block/vhdx-log.c +6
-0+ header = s->headers[s->curr_header]; + +/* need to have offset read data, and be on 4096 byte boundary */ + +if (length > header->log_length) { +/* no log present. we could create a log here instead of failing */ Does newly created vhdx have allocated log sectors?
I don't know of any way to make Hyper-V create a file without an allocated log area (I believe with the files I've generated, it allocates a 1MB log between the header and the BAT region).
The spec says that "LogLength" in the header should be a multiple of 1MB. And technically, 0 is a multiple of every number, so when parsing the header I don't fail out on a zero-lengthed log. In practice, I don't think Hyper-V creates files with zero-length logs, but I don't think the spec rules it out.
So we could either allocate a log in the file at this point, or fail.
Quoted 12 lines
block/vhdx-log.c exit: · line 845
Quoted 11 lines
block/vhdx-log.c +11
-0+ ret = -EINVAL ; + goto exit; + } + +if (vhdx_log_guid_is_zero (&header->log_guid)) { +vhdx_guid_generate (&new_guid); +vhdx_update_headers (bs, s, false, &new_guid); + }else { +/* currently, we require that the log be flushed after +* every write. */ + ret = -ENOTSUP ;
Quoted 2 lines
Can we make an assertion here?
I don't know if we should assert here - the VM could certainly continue on if this is not the primary drive.
Quoted 183 lines
Quoted 180 lines
block/vhdx-log.c +162
-0+ } + +/* 0 is an invalid sequence number, but may also represent the first +* log write (or a wrapped seq) */ +if (s->log.sequence ==0 ) { + s->log.sequence =1 ; + } + + sector_offset = offset %VHDX_LOG_SECTOR_SIZE ; + file_offset = (offset /VHDX_LOG_SECTOR_SIZE ) *VHDX_LOG_SECTOR_SIZE ; + + aligned_length = length; + +/* add in the unaligned head and tail bytes */ +if (sector_offset) { + leading_length = (VHDX_LOG_SECTOR_SIZE - sector_offset); + leading_length = leading_length > length ? length : leading_length; + aligned_length -= leading_length; + partial_sectors++; + } + + sectors = aligned_length /VHDX_LOG_SECTOR_SIZE ; + trailing_length = aligned_length - (sectors *VHDX_LOG_SECTOR_SIZE ); +if (trailing_length) { + partial_sectors++; + } + + sectors += partial_sectors; + +/* sectors is now how many sectors the data itself takes, not +* including the header and descriptor metadata */ + + new_hdr = (VHDXLogEntryHeader ) { + .signature =VHDX_LOG_SIGNATURE , + .tail = s->log.tail, + .sequence_number = s->log.sequence, + .descriptor_count = sectors, + .reserved =0 , + .flushed_file_offset =bdrv_getlength (bs->file), + .last_file_offset =bdrv_getlength (bs->file), + }; + +memcpy (&new_hdr.log_guid, &header->log_guid,sizeof (MSGUID )); + + desc_sectors =vhdx_compute_desc_sectors (new_hdr.descriptor_count); + + total_length = (desc_sectors + sectors) *VHDX_LOG_SECTOR_SIZE ; + new_hdr.entry_length = total_length; + +vhdx_log_entry_hdr_le_export (&new_hdr); + + buffer =qemu_blockalign (bs, total_length); +memcpy (buffer, &new_hdr,sizeof (new_hdr)); + + new_desc = (VHDXLogDescriptor *) (buffer +sizeof (new_hdr)); + data_sector = buffer + (desc_sectors *VHDX_LOG_SECTOR_SIZE ); + data_tmp = data; + +/* All log sectors are 4KB, so for any partial sectors we must +* merge the data with preexisting data from the final file +* destination */ + merged_sector =qemu_blockalign (bs,VHDX_LOG_SECTOR_SIZE ); + +for (i =0 ; i < sectors; i++) { + new_desc->signature =VHDX_LOG_DESC_SIGNATURE ; + new_desc->sequence_number = s->log.sequence; + new_desc->file_offset = file_offset; + +if (i ==0 && leading_length) { +/* partial sector at the front of the buffer */ + ret =bdrv_pread (bs->file, file_offset, merged_sector, +VHDX_LOG_SECTOR_SIZE ); +if (ret <0 ) { + goto exit; + } +memcpy (merged_sector + sector_offset, data_tmp, leading_length); + bytes_written = leading_length; + sector_write = merged_sector; + }else if (i == sectors -1 && trailing_length) { +/* partial sector at the end of the buffer */ + ret =bdrv_pread (bs->file, + file_offset, + merged_sector + trailing_length, +VHDX_LOG_SECTOR_SIZE - trailing_length); +if (ret <0 ) { + goto exit; + } +memcpy (merged_sector, data_tmp, trailing_length); + bytes_written = trailing_length; + sector_write = merged_sector; + }else { + bytes_written =VHDX_LOG_SECTOR_SIZE ; + sector_write = data_tmp; + } + +/* populate the raw sector data into the proper structures, +* as well as update the descriptor, and convert to proper +* endianness */ +vhdx_log_raw_to_le_sector (new_desc, data_sector, sector_write, + s->log.sequence); + + data_tmp += bytes_written; + data_sector++; + new_desc++; + file_offset +=VHDX_LOG_SECTOR_SIZE ; + } + +/* checksum covers entire entry, from the log header through the +* last data sector */ +vhdx_update_checksum (buffer, total_length,4 ); +cpu_to_le32s ((uint32_t *)(buffer +4 )); + +/* now write to the log */ +vhdx_log_write_sectors (bs, &s->log, §ors_written, buffer, + desc_sectors + sectors); +if (ret <0 ) { + goto exit; + } + +if (sectors_written != desc_sectors + sectors) { +/* instead of failing, we could flush the log here */ + ret = -EINVAL ; + goto exit; + } + + s->log.sequence++; +/* write new tail */ + s->log.tail = s->log.write; + +exit: +qemu_vfree (buffer); +qemu_vfree (merged_sector); +return ret; +} + +/* Perform a log write, and then immediately flush the entire log */ +int vhdx_log_write_and_flush (BlockDriverState *bs,BDRVVHDXState *s, +void *data,uint32_t length,uint64_t offset) +{ +int ret =0 ; +VHDXLogSequence logs = { .valid = true, + .count =1 , + .hdr = {0 } }; + + + ret =vhdx_log_write (bs, s, data, length, offset); +if (ret <0 ) { + goto exit; + } + logs.log = s->log; + + ret =vhdx_log_flush (bs, s, &logs); +if (ret <0 ) { + goto exit; + } + + s->log = logs.log; + +exit: +return ret; +} +block/vhdx.h +3
-0@@ -393,6 +393,9 @@ bool vhdx_checksum_is_valid(uint8_t *buf, size_t size, int crc_offset);int vhdx_parse_log (BlockDriverState *bs,BDRVVHDXState *s); +int vhdx_log_write_and_flush (BlockDriverState *bs,BDRVVHDXState *s, +void *data,uint32_t length,uint64_t offset); +static inline void leguid_to_cpus (MSGUID *guid) {le32_to_cpus (&guid->data1);Signature · 1 line
1.8.1.4Signature · 1 line
Fam