Please, igrnore. I'll send updated patchset today. On Thu, Apr 28, 2016 at 10:14:52AM +0300, Ed Bartosh wrote: > Used truncate instead of dd to create wic image for the > following reasons: > - truncate syntax is much more clear > - dd requires additional calculations of the image size > in blocks > - the way dd was used in the code is not entirely correct. > It was still writing one block to the file, which made it not > 100% sparse. > > Signed-off-by: Ed Bartosh <[email protected]> > --- > scripts/lib/wic/utils/fs_related.py | 12 +++--------- > 1 file changed, 3 insertions(+), 9 deletions(-) > > diff --git a/scripts/lib/wic/utils/fs_related.py > b/scripts/lib/wic/utils/fs_related.py > index 2e74461..2658dcf 100644 > --- a/scripts/lib/wic/utils/fs_related.py > +++ b/scripts/lib/wic/utils/fs_related.py > @@ -71,14 +71,8 @@ class DiskImage(Disk): > def create(self): > if self.device is not None: > return > - > - blocks = self.size / 1024 > - if self.size - blocks * 1024: > - blocks += 1 > - > - # create disk image > - dd_cmd = "dd if=/dev/zero of=%s bs=1024 seek=%d count=1" % \ > - (self.image_file, blocks) > - exec_cmd(dd_cmd) > + # create sparse disk image > + cmd = "truncate %s -s %s" % (self.image_file, self.size) > + exec_cmd(cmd) > > self.device = self.image_file > -- > 2.1.4 >
-- -- Regards, Ed -- _______________________________________________ Openembedded-core mailing list [email protected] http://lists.openembedded.org/mailman/listinfo/openembedded-core
