On 03/09/12 22:23, Andreas Färber wrote:
Am 03.09.2012 09:46, schrieb Brad Campbell:
Converting to an image with an output backing file would write out the
contents
of the source image whether or not it was already contained in the new backing
file. This commit ensures that the source file is checked against the new
backing file and output is only allocated if it obsoletes or extends data
already in the supplied backing file.
Signed-off-by: Brad Campbell <b...@fnarfbargle.com>
For a resend it would've been nice to get the typos out, I'm sure you
mean qemu-img. :) And it would be nice to have that marked as the topic
then so that it's easily recognized ("qemu-img: Bla bla").
To be honest, with a typo that big I would never have picked it up
myself no matter how many times I went over it. I'll take that
suggestion for the topic, thanks.
---
qemu-img.c | 74 ++++++++++++++++++++++++++++++++++++++++++++++-------------
1 files changed, 57 insertions(+), 17 deletions(-)
diff --git a/qemu-img.c b/qemu-img.c
index c8a70ff..3fb6cbf 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -629,11 +629,12 @@ static int img_convert(int argc, char **argv)
int progress = 0, flags;
const char *fmt, *out_fmt, *cache, *out_baseimg, *out_filename;
BlockDriver *drv, *proto_drv;
- BlockDriverState **bs = NULL, *out_bs = NULL;
+ BlockDriverState **bs = NULL, *out_bs = NULL, *out_bf = NULL;
int64_t total_sectors, nb_sectors, sector_num, bs_offset;
- uint64_t bs_sectors;
+ uint64_t bf_sectors, bs_sectors;
uint8_t * buf = NULL;
const uint8_t *buf1;
+ uint8_t * bf_buf = NULL;
Should be *bf_buf despite the * buf line above (checkpatch.pl sometimes
mixes up multiplication and declaration of pointer variables).
I'll correct and re-send.
BlockDriverInfo bdi;
QEMUOptionParameter *param = NULL, *create_options = NULL;
QEMUOptionParameter *out_baseimg_param;
[snip]
Note that the merge window for v1.3 is not yet open so you'll have to be
a bit patient before seeing your patch in qemu.git.
I'm not in any hurry, it's just that the mail config on the machine I
used git send-email from is so broken that the mail to Kevin bounced, so
I waited a week before I re-sent it just to avoid bombing the list. As
it turns out it's still broken anyway.
Thanks for the feedback.
Regards,
Brad
