Block fixes for stable-0.13
The following changes since commit 72230c523bda18d4dd2f7d16f96cc59e8fbbd6c9:
Update version for 0.13.0-rc1 (2010-08-31 08:19:23 -0500)
are available in the git repository at:
git://repo.or.cz/qemu/kevin.git for-stable-0.13
Bernhard Kohl (6):
scsi-disk: fix the mode data length field returned by the MODE SENSE command
scsi-disk: fix the mode data header returned by the MODE SENSE(10) command
scsi-disk: respect the page control (PC) field in the MODE SENSE command
scsi-disk: fix the block descriptor returned by the MODE SENSE command
scsi-disk: return CHECK CONDITION for unknown page codes in the MODE SENSE command
scsi-disk: fix the check of the DBD bit in the MODE SENSE command
Kevin Wolf (7):
raw-posix: Don't use file name for host_cdrom detection on Linux qemu-img convert: Use cache=unsafe for output image
block: Fix BDRV_O_CACHE_MASK
qcow2: Remove unnecessary flush after L2 write
vvfat: Fix segfault on write to read-only disk
vvfat: Fix double free for opening the image rw
vvfat: Use cache=unsafe
block.h | 2 +- block/qcow2-cluster.c | 16 ++++++-- block/raw-posix.c | 3 -- block/vvfat.c | 26 ++++++++++---- hw/scsi-disk.c | 90 ++++++++++++++++++++++++++++++++++++++---------- qemu-img.c | 3 +- 6 files changed, 105 insertions(+), 35 deletions(-)
PATCH 1/13 scsi-disk: fix the mode data length field returned by the MODE SENSE command
From: Bernhard Kohl <[email protected]>
The MODE DATA LENGTH field indicates the length in bytes of the following data that is available to be transferred. The mode data length does not include the number of bytes in the MODE DATA LENGTH field.
--- hw/scsi-disk.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
hw/scsi-disk.c +1 -1
@@ -652,7 +652,7 @@ static int scsi_disk_emulate_mode_sense(SCSIRequest *req, uint8_t *outbuf) } buflen = p - outbuf; - outbuf[0 ] = buflen -4 ; + outbuf[0 ] = buflen -1 ;if (buflen > req->cmd.xfer) buflen = req->cmd.xfer;return buflen;
Signature · 1 line
1.7.2.2
PATCH 2/13 scsi-disk: fix the mode data header returned by the MODE SENSE(10) command
In reply to Kevin WolfFrom: Bernhard Kohl <[email protected]>
The header for the MODE SENSE(10) command is 8 bytes long.
--- hw/scsi-disk.c | 38 ++++++++++++++++++++++++++++++++------ 1 files changed, 32 insertions(+), 6 deletions(-)
hw/scsi-disk.c +32 -6
@@ -606,6 +606,7 @@ static int scsi_disk_emulate_mode_sense(SCSIRequest *req, uint8_t *outbuf)uint64_t nb_sectors;int page, dbd, buflen;uint8_t *p; +uint8_t dev_specific_param; dbd = req->cmd.buf[1 ] &0x8 ; page = req->cmd.buf[2 ] &0x3f ;
@@ -613,16 +614,31 @@ static int scsi_disk_emulate_mode_sense(SCSIRequest *req, uint8_t *outbuf)memset (outbuf,0 , req->cmd.xfer); p = outbuf; - p[1 ] =0 ;/* Default media type. */ - p[3 ] =0 ;/* Block descriptor length. */ if (bdrv_is_read_only (s->bs)) { - p[2 ] =0x80 ;/* Readonly. */ + dev_specific_param =0x80 ;/* Readonly. */ + }else { + dev_specific_param =0x00 ; + } + +if (req->cmd.buf[0 ] ==MODE_SENSE ) { + p[1 ] =0 ;/* Default media type. */ + p[2 ] = dev_specific_param; + p[3 ] =0 ;/* Block descriptor length. */ + p +=4 ; + }else {/* MODE_SENSE_10 */ + p[2 ] =0 ;/* Default media type. */ + p[3 ] = dev_specific_param; + p[6 ] = p[7 ] =0 ;/* Block descriptor length. */ + p +=8 ; } - p +=4 ;bdrv_get_geometry (s->bs, &nb_sectors);if ((~dbd) & nb_sectors) { - outbuf[3 ] =8 ;/* Block descriptor length */ +if (req->cmd.buf[0 ] ==MODE_SENSE ) { + outbuf[3 ] =8 ;/* Block descriptor length */ + }else {/* MODE_SENSE_10 */ + outbuf[7 ] =8 ;/* Block descriptor length */ + } nb_sectors /= s->cluster_size; nb_sectors--;if (nb_sectors >0xffffff )
@@ -652,7 +668,17 @@ static int scsi_disk_emulate_mode_sense(SCSIRequest *req, uint8_t *outbuf) } buflen = p - outbuf; - outbuf[0 ] = buflen -1 ; +/* +* The mode data length field specifies the length in bytes of the +* following data that is available to be transferred. The mode data +* length does not include itself. +*/ +if (req->cmd.buf[0 ] ==MODE_SENSE ) { + outbuf[0 ] = buflen -1 ; + }else {/* MODE_SENSE_10 */ + outbuf[0 ] = ((buflen -2 ) >>8 ) &0xff ; + outbuf[1 ] = (buflen -2 ) &0xff ; + }if (buflen > req->cmd.xfer) buflen = req->cmd.xfer;return buflen;
Signature · 1 line
1.7.2.2
PATCH 3/13 scsi-disk: respect the page control (PC) field in the MODE SENSE command
In reply to Kevin WolfFrom: Bernhard Kohl <[email protected]>
The page control (PC) field defines the type of mode parameter values to be returned in the mode pages:
PC=0 : Current values
PC=1 : Changeable values
PC=2 : Default values
PC=3 : Saved values
The current implementation always returns the same type of parameters. This is OK for Current and Default values as we don't support changes to be done by the MODE SELECT command.
For Saved values the following applies (implemented by this patch):
"A PC field value of 3h requests that the target return the saved values of the mode parameters. Implementation of saved page parameters is optional. Mode parameters not supported by the target shall be set to zero. If saved values are not implemented, the command shall be terminated with CHECK CONDITION status, the sense key set to ILLEGAL REQUEST and the additional sense code set to
SAVING PARAMETERS NOT SUPPORTED."
For Changeable values the following applies (implemented by this patch):
"A PC field value of 1h requests that the target return a mask denoting those mode parameters that are changeable. In the mask, the fields of the mode parameters that are changeable shall be set to all one bits and the fields of the mode parameters that are non-changeable (i.e. defined by the target) shall be set to all zero bits."
In newer versions of the SCSI-2 spec the following clause was added. "If the logical unit does not implement changeable parameters mode pages and the device server receives a MODE SENSE command with 01b in the PC field, then the command shall be terminated with CHECK CONDITION status, with the sense key set to ILLEGAL REQUEST, and the additional sense code set to INVALID FIELD IN CDB."
This was not yet included in the SCSI-2 Working Drafts from 1986-1993. I assume that the variant to return CHECK CONDITION for PC=1 is not widely implemented by real devices. I have a legacy OS which fails, if MODE_SENSE returns non GOOD for PC=1. So for highest compatibility I implemented the former variant with this patch.
The last Working Draft X3T9.2 Rev. 10L 7-SEP-93 can be found here:
http://ldkelley.com/SCSI2/SCSI2/SCSI2-08.html#8.2.10
In mode_sense_page() this patch also avoids multiple hard coded definitions of the same mode page length. Instead I use the varable p[1]. In fact the returned length of the mode pages 4 and 5 were wrong (2 bytes less).
--- hw/scsi-disk.c | 45 +++++++++++++++++++++++++++++++++++---------- 1 files changed, 35 insertions(+), 10 deletions(-)
hw/scsi-disk.c +35 -10
@@ -485,16 +485,26 @@ static int scsi_disk_emulate_inquiry(SCSIRequest *req, uint8_t *outbuf)return buflen; } -static int mode_sense_page (SCSIRequest *req,int page,uint8_t *p) +static int mode_sense_page (SCSIRequest *req,int page,uint8_t *p, +int page_control) {SCSIDiskState *s =DO_UPCAST (SCSIDiskState, qdev, req->dev);BlockDriverState *bdrv = s->bs;int cylinders, heads, secs; +/* +* If Changeable Values are requested, a mask denoting those mode parameters +* that are changeable shall be returned. As we currently don't support +* parameter changes via MODE_SELECT all bits are returned set to zero. +* The buffer was already menset to zero by the caller of this function. +*/ switch (page) {case 4 :/* Rigid disk device geometry page. */ p[0 ] =4 ; p[1 ] =0x16 ; +if (page_control ==1 ) {/* Changeable Values */ +return p[1 ] +2 ; + }/* if a geometry hint is available, use it */ bdrv_get_geometry_hint (bdrv, &cylinders, &heads, &secs); p[2 ] = (cylinders >>16 ) &0xff ;
@@ -519,11 +529,14 @@ static int mode_sense_page(SCSIRequest *req, int page, uint8_t *p)/* Medium rotation rate [rpm], 5400 rpm */ p[20 ] = (5400 >>8 ) &0xff ; p[21 ] =5400 &0xff ; -return 0x16 ; +return p[1 ] +2 ;case 5 :/* Flexible disk device geometry page. */ p[0 ] =5 ; p[1 ] =0x1e ; +if (page_control ==1 ) {/* Changeable Values */ +return p[1 ] +2 ; + }/* Transfer rate [kbit/s], 5Mbit/s */ p[2 ] =5000 >>8 ; p[3 ] =5000 &0xff ;
@@ -555,21 +568,27 @@ static int mode_sense_page(SCSIRequest *req, int page, uint8_t *p)/* Medium rotation rate [rpm], 5400 rpm */ p[28 ] = (5400 >>8 ) &0xff ; p[29 ] =5400 &0xff ; -return 0x1e ; +return p[1 ] +2 ;case 8 :/* Caching page. */ p[0 ] =8 ; p[1 ] =0x12 ; +if (page_control ==1 ) {/* Changeable Values */ +return p[1 ] +2 ; + }if (bdrv_enable_write_cache (s->bs)) { p[2 ] =4 ;/* WCE */ } -return 20 ; +return p[1 ] +2 ;case 0x2a :/* CD Capabilities and Mechanical Status page. */ if (bdrv_get_type_hint (bdrv) !=BDRV_TYPE_CDROM )return 0 ; p[0 ] =0x2a ; p[1 ] =0x14 ; +if (page_control ==1 ) {/* Changeable Values */ +return p[1 ] +2 ; + } p[2 ] =3 ;// CD-R & CD-RW read p[3 ] =0 ;// Writing not supported p[4 ] =0x7f ;/* Audio, composite, digital out,
@@ -593,7 +612,7 @@ static int mode_sense_page(SCSIRequest *req, int page, uint8_t *p) p[19 ] = (16 *176 ) &0xff ; p[20 ] = (16 *176 ) >>8 ;// 16x write speed current p[21 ] = (16 *176 ) &0xff ; -return 22 ; +return p[1 ] +2 ;default :return 0 ;
@@ -604,13 +623,15 @@ static int scsi_disk_emulate_mode_sense(SCSIRequest *req, uint8_t *outbuf) {SCSIDiskState *s =DO_UPCAST (SCSIDiskState, qdev, req->dev);uint64_t nb_sectors; -int page, dbd, buflen; +int page, dbd, buflen, page_control;uint8_t *p;uint8_t dev_specific_param; dbd = req->cmd.buf[1 ] &0x8 ; page = req->cmd.buf[2 ] &0x3f ; -DPRINTF ("Mode Sense (page %d, len %zd)\n" , page, req->cmd.xfer); + page_control = (req->cmd.buf[2 ] &0xc0 ) >>6 ; +DPRINTF ("Mode Sense(%d) (page %d, len %d, page_control %d)\n" , + (req->cmd.buf[0 ] ==MODE_SENSE ) ?6 :10 , page, len, page_control);memset (outbuf,0 , req->cmd.xfer); p = outbuf;
@@ -654,16 +675,20 @@ static int scsi_disk_emulate_mode_sense(SCSIRequest *req, uint8_t *outbuf) p +=8 ; } +if (page_control ==3 ) {/* Saved Values */ +return -1 ;/* ILLEGAL_REQUEST */ + } +switch (page) {case 0x04 :case 0x05 :case 0x08 :case 0x2a : - p +=mode_sense_page (req, page, p); + p +=mode_sense_page (req, page, p, page_control);break ;case 0x3f : - p +=mode_sense_page (req,0x08 , p); - p +=mode_sense_page (req,0x2a , p); + p +=mode_sense_page (req,0x08 , p, page_control); + p +=mode_sense_page (req,0x2a , p, page_control);break ; }
Signature · 1 line
1.7.2.2
PATCH 4/13 scsi-disk: fix the block descriptor returned by the MODE SENSE command
In reply to Kevin WolfFrom: Bernhard Kohl <[email protected]>
The block descriptor contains the number of blocks, not the highest LBA. Real hard disks return 0 if the number of blocks exceed the maximum 0xFFFFFF.
SCSI-Spec:
http://ldkelley.com/SCSI2/SCSI2/SCSI2-08.html#8.3.3
"The number of blocks field specifies the number of logical blocks on the medium to which the density code and block length fields apply. A value of zero indicates that all of the remaining logical blocks of the logical unit shall have the medium characteristics specified."
--- hw/scsi-disk.c | 3 +-- 1 files changed, 1 insertions(+), 2 deletions(-)
hw/scsi-disk.c +1 -2
@@ -661,9 +661,8 @@ static int scsi_disk_emulate_mode_sense(SCSIRequest *req, uint8_t *outbuf) outbuf[7 ] =8 ;/* Block descriptor length */ } nb_sectors /= s->cluster_size; - nb_sectors--;if (nb_sectors >0xffffff ) - nb_sectors =0xffffff ; + nb_sectors =0 ; p[0 ] =0 ;/* media density code */ p[1 ] = (nb_sectors >>16 ) &0xff ; p[2 ] = (nb_sectors >>8 ) &0xff ;
Signature · 1 line
1.7.2.2
PATCH 5/13 scsi-disk: return CHECK CONDITION for unknown page codes in the MODE SENSE command
In reply to Kevin WolfFrom: Bernhard Kohl <[email protected]>
SCSI-Spec:
http://ldkelley.com/SCSI2/SCSI2/SCSI2-08.html#8.2.10
"An initiator may request any one or all of the supported mode pages from a target. If an initiator issues a MODE SENSE command with a page code value not implemented by the target, the target shall return CHECK CONDITION status and shall set the sense key to ILLEGAL REQUEST and the additional sense code to INVALID FIELD IN CDB."
--- hw/scsi-disk.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)
hw/scsi-disk.c +2 -0
@@ -689,6 +689,8 @@ static int scsi_disk_emulate_mode_sense(SCSIRequest *req, uint8_t *outbuf) p +=mode_sense_page (req,0x08 , p, page_control); p +=mode_sense_page (req,0x2a , p, page_control);break ; +default : +return -1 ;/* ILLEGAL_REQUEST */ } buflen = p - outbuf;
Signature · 1 line
1.7.2.2
PATCH 6/13 scsi-disk: fix the check of the DBD bit in the MODE SENSE command
In reply to Kevin WolfFrom: Bernhard Kohl <[email protected]>
The DBD bit does not work as expected.
SCSI-Spec:
http://ldkelley.com/SCSI2/SCSI2/SCSI2-08.html#8.2.10
"A disable block descriptors (DBD) bit of zero indicates that the target may return zero or more block descriptors in the returned MODE SENSE data (see 8.3.3), at the target's discretion. A DBD bit of one specifies that the target shall not return any block descriptors in the returned MODE SENSE data."
--- hw/scsi-disk.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
hw/scsi-disk.c +1 -1
@@ -654,7 +654,7 @@ static int scsi_disk_emulate_mode_sense(SCSIRequest *req, uint8_t *outbuf) }bdrv_get_geometry (s->bs, &nb_sectors); -if ((~dbd) & nb_sectors) { +if (!dbd && nb_sectors) {if (req->cmd.buf[0 ] ==MODE_SENSE ) { outbuf[3 ] =8 ;/* Block descriptor length */ }else {/* MODE_SENSE_10 */
Signature · 1 line
1.7.2.2
PATCH 7/13 raw-posix: Don't use file name for host_cdrom detection on Linux
In reply to Kevin WolfOn Linux, we have code to detect CD-ROMs using an ioctl. We shouldn't lose anything but false positives by removing the check for a /dev/cd* path.
--- block/raw-posix.c | 3 --- 1 files changed, 0 insertions(+), 3 deletions(-)
block/raw-posix.c +0 -3
@@ -1154,9 +1154,6 @@ static int cdrom_probe_device(const char *filename)int fd, ret;int prio =0 ; -if (strstart (filename,"/dev/cd" ,NULL )) - prio =50 ; - fd =open (filename,O_RDONLY |O_NONBLOCK );if (fd <0 ) { goto out;
Signature · 1 line
1.7.2.2
PATCH 8/13 qemu-img convert: Use cache=unsafe for output image
In reply to Kevin WolfIf qemu-img crashes during the conversion, the user will throw away the broken output file anyway and start over. So no need to be too cautious.
--- qemu-img.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)
qemu-img.c +2 -1
@@ -783,7 +783,8 @@ static int img_convert(int argc, char **argv) goto out; } - out_bs =bdrv_new_open (out_filename, out_fmt,BDRV_O_FLAGS |BDRV_O_RDWR ); + out_bs =bdrv_new_open (out_filename, out_fmt, +BDRV_O_FLAGS |BDRV_O_RDWR |BDRV_O_NO_FLUSH );if (!out_bs) { ret =-1 ; goto out;
Signature · 1 line
1.7.2.2
PATCH 9/13 block: Fix BDRV_O_CACHE_MASK
In reply to Kevin WolfBDRV_O_CACHE_MASK should have been extended when cache=unsafe introduced a new flag BDRV_O_NO_FLUSH. There are currently no users that would change their behaviour because of this, but let's clean it up before things break.
--- block.h | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
block.h +1 -1
@@ -35,7 +35,7 @@ typedef struct QEMUSnapshotInfo {#define BDRV_O_NO_BACKING 0x0100/* don't open the backing file */ #define BDRV_O_NO_FLUSH 0x0200/* disable flushing on this disk */ -#define BDRV_O_CACHE_MASK (BDRV_O_NOCACHE | BDRV_O_CACHE_WB) +#define BDRV_O_CACHE_MASK (BDRV_O_NOCACHE | BDRV_O_CACHE_WB | BDRV_O_NO_FLUSH)#define BDRV_SECTOR_BITS 9#define BDRV_SECTOR_SIZE (1ULL << BDRV_SECTOR_BITS)
Signature · 1 line
1.7.2.2
PATCH 10/13 qcow2: Remove unnecessary flush after L2 write
In reply to Kevin WolfWhen a new cluster was allocated, we only need a flush after the write to the L2 table if it was a COW and we need to decrease the refcounts of the old clusters.
--- block/qcow2-cluster.c | 16 ++++++++++++---- 1 files changed, 12 insertions(+), 4 deletions(-)
block/qcow2-cluster.c +12 -4
@@ -655,7 +655,7 @@ static int write_l2_entries(BlockDriverState *bs, uint64_t *l2_table,int ret;BLKDBG_EVENT (bs->file,BLKDBG_L2_UPDATE ); - ret =bdrv_pwrite_sync (bs->file, l2_offset + start_offset, + ret =bdrv_pwrite (bs->file, l2_offset + start_offset, &l2_table[l2_start_index], len);if (ret <0 ) {return ret;
@@ -718,9 +718,17 @@ int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m) goto err; } -for (i =0 ; i < j; i++) -qcow2_free_any_clusters (bs, -be64_to_cpu (old_cluster[i]) & ~QCOW_OFLAG_COPIED ,1 ); +/* +* If this was a COW, we need to decrease the refcount of the old cluster. +* Also flush bs->file to get the right order for L2 and refcount update. +*/ +if (j !=0 ) { +bdrv_flush (bs->file); +for (i =0 ; i < j; i++) { +qcow2_free_any_clusters (bs, +be64_to_cpu (old_cluster[i]) & ~QCOW_OFLAG_COPIED ,1 ); + } + } ret =0 ; err:
Signature · 1 line
1.7.2.2
PATCH 11/13 vvfat: Fix segfault on write to read-only disk
In reply to Kevin WolfFrom: Kevin Wolf <[email protected]>
vvfat tries to set the readonly flag in its open function, but nowadays this is overwritted with the readonly=... command line option. Check in bdrv_write if the vvfat was opened read-only and return an error in this case.
Without this check, vvfat tries to access the qcow bs, which is NULL without enabled write support.
--- block/vvfat.c | 5 +++++ 1 files changed, 5 insertions(+), 0 deletions(-)
block/vvfat.c +5 -0
@@ -2665,6 +2665,11 @@ static int vvfat_write(BlockDriverState *bs, int64_t sector_num,DLOG (checkpoint ()); +/* Check if we're operating in read-only mode */ +if (s->qcow ==NULL ) { +return -EACCES ; + } +vvfat_close_current_file (s);/*
Signature · 1 line
1.7.2.2
PATCH 12/13 vvfat: Fix double free for opening the image rw
In reply to Kevin WolfFrom: Kevin Wolf <[email protected]>
Allocation and deallocation of bs->opaque is not in the control of a block driver. Therefore it should not set bs->opaque to a data structure used by another bs, or closing the image will lead to a double free.
--- block/vvfat.c | 7 ++++--- 1 files changed, 4 insertions(+), 3 deletions(-)
block/vvfat.c +4 -3
@@ -2768,12 +2768,12 @@ static int vvfat_is_allocated(BlockDriverState *bs,static int write_target_commit (BlockDriverState *bs,int64_t sector_num,const uint8_t * buffer,int nb_sectors) { -BDRVVVFATState * s = bs->opaque; +BDRVVVFATState * s = *((BDRVVVFATState **) bs->opaque);return try_commit (s); }static void write_target_close (BlockDriverState *bs) { -BDRVVVFATState * s = bs->opaque; +BDRVVVFATState * s = *((BDRVVVFATState **) bs->opaque);bdrv_delete (s->qcow);free (s->qcow_filename); }
@@ -2816,7 +2816,8 @@ static int enable_write_target(BDRVVVFATState *s) s->bs->backing_hd =calloc (sizeof (BlockDriverState),1 ); s->bs->backing_hd->drv = &vvfat_write_target; - s->bs->backing_hd->opaque = s; + s->bs->backing_hd->opaque =qemu_malloc (sizeof (void *)); + *(void **)s->bs->backing_hd->opaque = s;return 0 ; }
Signature · 1 line
1.7.2.2
PATCH 13/13 vvfat: Use cache=unsafe
In reply to Kevin WolfFrom: Kevin Wolf <[email protected]>
The qcow file used for write support in vvfat is a temporary file, so we can use cache=unsafe there. Without this, write support is just too slow to be of any use.
--- block/vvfat.c | 14 ++++++++++---- 1 files changed, 10 insertions(+), 4 deletions(-)
block/vvfat.c +10 -4
@@ -2788,6 +2788,7 @@ static int enable_write_target(BDRVVVFATState *s) {BlockDriver *bdrv_qcow;QEMUOptionParameter *options; +int ret;int size =sector2cluster (s, s->sector_count); s->used_clusters =calloc (size,1 );
@@ -2803,11 +2804,16 @@ static int enable_write_target(BDRVVVFATState *s)if (bdrv_create (bdrv_qcow, s->qcow_filename, options) <0 )return -1 ; + s->qcow =bdrv_new ("" ); -if (s->qcow ==NULL || -bdrv_open (s->qcow, s->qcow_filename,BDRV_O_RDWR , bdrv_qcow) <0 ) - { -return -1 ; +if (s->qcow ==NULL ) { +return -1 ; + } + + ret =bdrv_open (s->qcow, s->qcow_filename, +BDRV_O_RDWR |BDRV_O_CACHE_WB |BDRV_O_NO_FLUSH , bdrv_qcow); +if (ret <0 ) { +return ret; }#ifndef _WIN32
Signature · 1 line
1.7.2.2