-
Attached is a patchset with all of the changes I have made to dcfldd to date.
Additions:
Allows for partial write of block during conv=sync if at the end of input.
Allows limit= to limit the count of input, rather than count=.
Closes all popened processes. (Uses internal popen call, rather than original popen2.)
Allows sizeprobe= to manually specify...
2008-04-15 23:29:07 UTC by dloveall
-
I was having problems with sending data to a stream via the of:= method. Tracked it down to the process not having a chance to finish. The patch below is a stab at implementing the pclose2() method for streamed output. This already required keeping a handle around for the stream, rather than just the fileno.
At them moment, I'm not using the functionality to send hashes to multiple outputs.
2008-02-24 05:12:42 UTC by dloveall
-
The patch below allows the user to specify an arbitrary limit to the size of the input in bytes. The following commands generate the same value:
dcfldd if=/dev/sda bs=1234 count=1 | md5sum
dcfldd if=/dev/sda bs=1 count=1234 | md5sum
dcfldd if=/dev/sda bs=1024 limit=1234 | md5sum
This allows for aligned block reads, as required. In particular, this is useful for verifying the hash of a...
2008-01-31 04:00:44 UTC by dloveall
-
In the event that the input is from a stream which has a known size, it would be nice to provide a method to provide the size to dcfldd for time estimation purposes. This patch implements this as an extension to the sizeprobe= option.
In the event that sizeprobe=XXX is provided and XXX is neither if or of, it attempts to parse the value as a number in the same method as bs, etc (allowing...
2008-01-31 00:24:12 UTC by dloveall
-
Segmented dd images are recognized natively by OS X, if named in the format:
test.dmg
test.002.dmgpart
test.003.dmgpart
...
This is a patch that adds support for this naming structure. It is called using splitformat=MAC. If the number of file segments exceeds 999, the file names will have additional characters. I am unaware how OS X treats this.
Dave.
2008-01-30 21:32:21 UTC by dloveall
-
Patch for this:
diff -ur dcfldd-1.3.4.orig/split.c dcfldd-1.3.4/split.c
--- dcfldd-1.3.4.orig/split.c 2005-05-13 13:52:06.000000000 -0500
+++ dcfldd-1.3.4/split.c 2008-01-30 14:40:24.000000000 -0600
@@ -107,6 +107,7 @@
if (fd < 0)
syscall_error(fname);
+ close(split->currfd);
split->currfd = fd;
split->curr_bytes = 0;.
2008-01-30 20:46:56 UTC by dloveall
-
Currently, if a read is done of a hard drive using sync and a large block size that is not an integer multiple of the drive, the input and output files will end up as different sizes. As such, it makes things like hashing a bit more complicated. The patch attached below attempts to provide a fix by utilizing the size probe functionality.
Scenario:
dcfldd if=/dev/zero of=test bs=1024...
2008-01-30 20:20:38 UTC by dloveall
-
jessekornblum committed revision 41 to the dcfldd SVN repository, changing 3 files.
2007-12-04 11:28:14 UTC by jessekornblum
-
jessekornblum committed revision 40 to the dcfldd SVN repository, changing 4 files.
2007-10-28 16:04:46 UTC by jessekornblum
-
harbourn committed revision 39 to the dcfldd SVN repository, changing 2 files.
2007-09-29 06:51:59 UTC by harbourn
