path: root/sgp_dd.8

.TH SGP_DD "8" "July 2006" "sg3_utils-1.22" SG3_UTILS
.SH NAME
sgp_dd \- copies data to and from files and devices. Specialised for
devices that understand the SCSI command set.
.SH SYNOPSIS
.B sgp_dd
[\fIbs=<n>\fR] [\fIcount=<n>\fR] [\fIibs=<n>\fR] [\fIif=<ifile>\fR]
[\fIiflag=<flags>\fR] [\fIobs=<n>\fR] [\fIof=<ofile>\fR]
[\fIoflag=<flags>\fR] [\fIseek=<n>\fR] [\fIskip=<n>\fR]
[\fI--help\fR] [\fI--version\fR]
.PP
[\fIbpt=<n>\fR] [\fIcdbsz=6|10|12|16\fR] [\fIdeb=<n>\fR] [\fIdio=0|1\fR]
[\fIsync=0|1\fR] [\fIthr=<n>\fR] [\fItime=0|1\fR] [\fIverbose=<n>\fR]
.SH DESCRIPTION
.\" Add any additional description here
.PP
Copy data to and from any files. Specialised for "files" that are
Linux SCSI generic (sg) and raw devices. Similar syntax and semantics to 
.B dd(1) 
but does not perform any conversions. Uses POSIX threads to increase
the amount of parallelism. This improves speed in some cases.
.PP
The first group in the synopsis above are "standard" Unix
.B dd(1)
arguments. The second group are extra arguments added by this utility.
Both groups are defined below.
.TP
bpt=BLOCKS
each IO transaction will be made using this number of blocks (or less if
near the end of count). Default is 128 for block sizes less that 2048
bytes, otherwise the default is 32. So for bs=512 the reads and writes
will each convey 64 KiB of data by default (less if near the end of the
transfer or memory restrictions). When cd/dvd drives are accessed, the
block size is typically 2048 bytes and bpt defaults to 32 which again
implies 64 KiB transfers.
.TP
bs=BYTES
this 
.B must 
be the block size of the physical device. Note that this differs from
.B dd(1) 
which permits 'bs' to be an integral multiple of the actual device block
size. Default is 512 which is usually correct for disks but incorrect for
cdroms (which normally have 2048 byte blocks).
.TP
cdbsz=6 | 10 | 12 | 16
size of SCSI READ and/or WRITE commands issued on sg device names.
Default is 10 byte SCSI command blocks (unless calculations indicate
that a 4 byte block number may be exceeded, in which case it defaults
to 16 byte SCSI commands).
.TP
count=BLOCKS
copy this number of blocks from 'if' to 'of'. Default is the minimum (
of 'if' and 'of') number of blocks that sg devices return from READ
CAPACITY SCSI commands or that block devices (or their partitions) report.
Normal files are not probed for their size. If 'skip'
or 'seek' are given and the count is deduced (i.e. not explicitly given)
then that count is scaled back so that the copy will not overrun the
device. If the file name is a block device partition and count is not given
then the size of the partition rather than the size of the whole device is
used. If count is not given and cannot be deduced then an error message
is issued and no copy takes place.
.TP
deb=NUM
outputs debug information. If NUM is 0 (default) then none and as NUM
increases so does the amount of debug (max debug output when NUM is 9)
.TP
dio=0 | 1
default is 0 which selects indirect IO. Value of 1 attempts direct
IO which, if not available, falls back to indirect IO and notes this
at completion. If direct IO is selected and /proc/scsi/sg/allow_dio
has the value of 0 then a warning is issued (and indirect IO is performed)
.TP
ibs=BYTES
if given must be the same value as 'bs'.
.TP
if=FILE
read from FILE instead of stdin. A file name of - is taken to be stdin.
Starts reading at the beginning of FILE unless 'skip=<n>' is given.
.TP
iflag=FLAGS
where FLAGS is a comma separated list of one or more flags outlined below.
These flags are associated with <ifile> and are ignored when <ifile> is
stdin.
.TP
obs=BYTES
if given must be the same as 'bs'.
.TP
of=FILE
write to FILE instead of stdout. A file name of - is taken to be stdout.
If FILE is /dev/null then no actual writes are performed. If FILE is .
(period) then it is treated the same way as /dev/null (this is a
shorthand notation). Starts (over-)writing at the beginning of FILE
unless either 'seek=<n>' or 'oflag=append' is given (it is not permitted
to give both).
.TP
oflag=FLAGS
where FLAGS is a comma separated list of one or more flags outlined below.
These flags are associated with <ofile> and are ignored when <ofile>
is /dev/null, . (period), or stdout.
.TP
seek=BLOCKS
start writing BLOCKS bs-sized blocks from the start of the output file.
Default is block 0 (i.e. start of file).
.TP
skip=BLOCKS
start reading BLOCKS bs-sized blocks from the start of input file.
Default is block 0 (i.e. start of file).
.TP
sync=0 | 1
when 1, does SYNCHRONIZE CACHE command on 'of' at the end of the transfer.
Only active when 'of' is a sg device file name
.TP
thr=NUM
this is the number or worker threads (default 4) that attempt to
copy in parallel. Minimum is 0 and maximum is 16
.TP
time=0 | 1
when 1, the transfer is timed and throughput calculation is
performed, outputting the results (to stderr) at completion. When
0 (default) no timing is performed.
.TP
verbose=NUM
increase verbosity. Same as 'deb=NUM'. Added for compatibility with
sg_dd and sgm_dd.
.TP
--help
outputs usage message and exits
.TP
--version
outputs version number information and exits
.SH FLAGS
Here is a list of flags and their meanings:
.TP
append
causes the O_APPEND flag to be added to the open of <ofile>. For normal
files this will lead to data appended to the end of any existing data.
Cannot be used together with the 'seek=<n>' option as they conflict.
The default action of this utility is to overwrite any existing data
from the beginning of the file or, if 'seek=<n>' is given, starting at
block <n>. Note that attempting to "append" to a device file will
usually be ignored or may cause an error to be reported.
.TP
coe
continue on error. When given with 'iflag=', an error that is detected
in a single SCSI command (typically 'bpt' blocks) is noted (by an error
message sent to stderr), then zeroes are substituted into the buffer
for the corresponding write operation and the copy continues. Note that the 
.B sg_dd
utility is more sophisticated in such error situations when 'iflag=coe'.
When given with 'oflag=', any error reported by a SCSI WRITE command is
reported to stderr and the copy continues (as if nothing went wrong).
.TP
direct
causes the O_DIRECT flag to be added to the open of <ifile> and/or <ofile>.
This flag requires some memory alignment on IO. Hence user memory buffers
are aligned to the page size. Has no effect on sg, normal or raw files.
.TP
dpo
set the DPO bit (disable page out) in READ and WRITE SCSI commands. Not
supported for 6 byte cdb variants of READ and WRITE. Indicates that
data is unlikely to be required to stay in device (e.g. disk) cache.
May speed media copy and/or cause a media copy to have less impact
on other device users.
.TP
dsync
causes the O_SYNC flag to be added to the open of <ifile> and/or <ofile>.
The "d" is prepended to lower confusion with the 'sync=0|1' option which
has another action (i.e. a synchronisation to media at the end of the
transfer).
.TP
excl
causes the O_EXCL flag to be added to the open of <ifile> and/or <ofile>.
.TP
fua
causes the FUA (force unit access) bit to be set in READ and/or WRITE
SCSI commands. This only has effect with sg devices. The 6 byte variants
of the READ and WRITE SCSI commands do not support the FUA bit.
Only active for sg device file names.
.SH RETIRED OPTIONS
Here are some retired options that are still present:
.TP
coe=0 | 1
continue on error is 0 (off) by default. When it is 1, it is
equivalent to 'iflag=coe oflag=coe' described in the FLAGS section
above.  Similar to 'conv=noerror' in
.B dd(1)
utility. Default is 0 which implies stop on error. More advanced
coe=1 processing on reads is performed by the sg_dd utility.
.TP
fua=0 | 1 | 2 | 3
force unit access bit. When 3, fua is set on both 'if' and 'of'; when 2, fua
is set on 'if'; when 1, fua is set on 'of'; when 0 (default), fua is cleared
on both. See the 'fua' flag.
.SH NOTES
A raw device must be bound to a block device prior to using sgp_dd.
See
.B raw(8)
for more information about binding raw devices. To be safe, the sg device
mapping to SCSI block devices should be checked with 'cat /proc/scsi/scsi'
before use.
.PP
Raw device partition information can often be found with
.B fdisk(8)
[the "-ul" argument is useful in this respect].
.PP
BYTES and BLOCKS may be followed by one of these multiplicative suffixes:
c C *1; w W *2; b B *512; k K KiB *1,024; KB *1,000; m M MiB *1,048,576;
MB *1,000,000 . This pattern continues for "G", "T" and "P". The latter two
suffixes can only be used for count, skip and seek values). Also a suffix of
the form "x<n>" multiplies the leading number by <n>. These multiplicative
suffixes are compatible with GNU's dd command (since 2002) which claims
compliance with SI and with IEC 60027-2.
.PP
Alternatively numerical values can be given in hexadecimal preceded by
either "0x" or "0X". When hex numbers are given, multipliers cannot be
used.
.PP
The count, skip and seek parameters can take 64 bit values (i.e. very
big numbers). Other values are limited to what can fit in a signed
32 bit number.
.PP
Data usually gets to the user space in a 2 stage process: first the
SCSI adapter DMAs into kernel buffers and then the sg driver copies
this data into user memory (write operations reverse this sequence).
This is called "indirect IO" and there is a 'dio' option to select
"direct IO" which will DMA directly into user memory. Due to some
issues "direct IO" is disabled in the sg driver and needs a
configuration change to activate it.
.PP
All informative, warning and error output is sent to stderr so that
dd's output file can be stdout and remain unpolluted. If no options
are given, then the usage message is output and nothing else happens.
.PP
Why use sgp_dd? Because in some cases it is twice as fast as dd
(mainly with sg devices, raw devices give some improvement).
Another reason is that big copies fill the block device caches
which has a negative impact on other machine activity.
.SH SIGNALS
The signal handling has been borrowed from dd: SIGINT, SIGQUIT and
SIGPIPE output the number of remaining blocks to be transferred and
the records in + out counts; then they have their default action.
SIGUSR1 causes the same information to be output yet the copy continues.
All output caused by signals is sent to stderr.
.SH EXAMPLES
.PP
Looks quite similar in usage to dd:
.PP
   sgp_dd if=/dev/sg0 of=t bs=512 count=1MB
.PP
This will copy 1 million 512 byte blocks from the device associated with
/dev/sg0 (which should have 512 byte blocks) to a file called t.
Assuming /dev/sda and /dev/sg0 are the same device then the above is
equivalent to:
.PP
   dd if=/dev/sda of=t bs=512 count=1000000
.PP
although dd's speed may improve if bs was larger and count was
correspondingly scaled. Using a raw device to do something similar on a
ATA disk:
.PP
   raw /dev/raw/raw1 /dev/hda
.br
   sgp_dd if=/dev/raw/raw1 of=t bs=512 count=1MB
.PP
To copy a SCSI disk partition to an ATA disk partition:
.PP
   raw /dev/raw/raw2 /dev/hda3
.br
   sgp_dd if=/dev/sg0 skip=10123456 of=/dev/raw/raw2 bs=512
.PP
This assumes a valid partition is found on the SCSI disk at the given
skip block address (past the 5 GB point of that disk) and that
the partition goes to the end of the SCSI disk. An explicit count
is probably a safer option.
.PP
To do a fast copy from one SCSI disk to another one with similar
geometry (stepping over errors on the source disk):
.PP
   sgp_dd if=/dev/sg0 of=/dev/sg1 bs=512 coe=1
.SH EXIT STATUS
The exit status of sgp_dd is 0 when it is successful. Otherwise see
the sg3_utils(8) man page. Since this utility works at a higher level
than individual commands, and there are 'coe' and 'retries' flags,
individual SCSI command failures do not necessary cause the process
to exit.
.SH AUTHORS
Written by Doug Gilbert and Peter Allworth.
.SH "REPORTING BUGS"
Report bugs to <dgilbert at interlog dot com>.
.SH COPYRIGHT
Copyright \(co 2000-2006 Douglas Gilbert
.br
This software is distributed under the GPL version 2. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
.SH "SEE ALSO"
A simpler, non-threaded version of this utility but with more
advanced "continue on error" logic is called
.B sg_dd
and is also found in the sg3_utils package. The lmbench package contains
.B lmdd
which is also interesting.
.B raw(8), dd(1)