From: Stefan Hajnoczi <stefanha@gmail.com>
To: Fam Zheng <famz@redhat.com>
Cc: kwolf@redhat.com, pl@kamp.de, qemu-devel@nongnu.org, stefanha@redhat.com
Subject: Re: [Qemu-devel] [PATCH] vmdk: Allow vmdk_create to work with protocol
Date: Thu, 19 Dec 2013 14:12:32 +0100 [thread overview]
Message-ID: <20131219131232.GC2537@stefanha-thinkpad.redhat.com> (raw)
In-Reply-To: <1387281600-2111-1-git-send-email-famz@redhat.com>
On Tue, Dec 17, 2013 at 08:00:00PM +0800, Fam Zheng wrote:
> @@ -1511,48 +1521,55 @@ static int vmdk_create_extent(const char *filename, int64_t filesize,
> header.check_bytes[3] = 0xa;
>
> /* write all the data */
> - ret = qemu_write_full(fd, &magic, sizeof(magic));
> - if (ret != sizeof(magic)) {
> - ret = -errno;
> + ret = bdrv_pwrite(bs, 0, &magic, sizeof(magic));
> + if (ret < 0) {
> + error_set(errp, QERR_IO_ERROR);
> goto exit;
> }
> - ret = qemu_write_full(fd, &header, sizeof(header));
> + ret = bdrv_pwrite(bs, sizeof(magic), &header, sizeof(header));
> if (ret != sizeof(header)) {
> + error_set(errp, QERR_IO_ERROR);
> ret = -errno;
This line should be deleted.
Also, I noticed you changed ret != sizeof(magic) to ret < 0 for the
magic number bdrv_pwrite() but did not change the condition for the
header write. Please keep the error handling condition consistent.
> goto exit;
> }
>
> - ret = ftruncate(fd, le64_to_cpu(header.grain_offset) << 9);
> + ret = bdrv_truncate(bs, (le64_to_cpu(header.grain_offset)) << 9);
Why add parentheses around le64_to_cpu()?
> if (ret < 0) {
> - ret = -errno;
> - goto exit;
> + error_setg(errp, "Could not truncate file");
goto exit?
> }
>
> /* write grain directory */
> - lseek(fd, le64_to_cpu(header.rgd_offset) << 9, SEEK_SET);
> - for (i = 0, tmp = le64_to_cpu(header.rgd_offset) + gd_size;
> + gd_buf_size = gd_sectors * BDRV_SECTOR_SIZE * sizeof(*gd_buf);
> + gd_buf = g_malloc0(gd_buf_size);
> + for (i = 0, tmp = le64_to_cpu(header.rgd_offset) + gd_sectors;
> i < gt_count; i++, tmp += gt_size) {
> - ret = qemu_write_full(fd, &tmp, sizeof(tmp));
> - if (ret != sizeof(tmp)) {
> - ret = -errno;
> - goto exit;
> - }
Was this old code not endian-safe? It appears to be writing native
endian values. The new code is different.
> @@ -1771,33 +1791,34 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
> total_size / (int64_t)(63 * number_heads * 512),
> number_heads,
> adapter_type);
> - if (split || flat) {
> - fd = qemu_open(filename,
> - O_WRONLY | O_CREAT | O_TRUNC | O_BINARY | O_LARGEFILE,
> - 0644);
> + desc_len = strlen(desc);
> + /* the descriptor offset = 0x200 */
> + if (!split && !flat) {
> + desc_offset = 0x200;
> } else {
> - fd = qemu_open(filename,
> - O_WRONLY | O_BINARY | O_LARGEFILE,
> - 0644);
> + ret = bdrv_create_file(filename, options, &local_err);
Missing error handling if bdrv_create_file() fails.
> }
> - if (fd < 0) {
> - ret = -errno;
> + ret = bdrv_file_open(&new_bs, filename, NULL, BDRV_O_RDWR, &local_err);
> + if (ret < 0) {
> + error_setg_errno(errp, -ret, "Could not write description");
> goto exit;
> }
> - /* the descriptor offset = 0x200 */
> - if (!split && !flat && 0x200 != lseek(fd, 0x200, SEEK_SET)) {
> - ret = -errno;
> - goto close_exit;
> + ret = bdrv_pwrite(new_bs, desc_offset, desc, desc_len);
> + if (ret < 0) {
> + error_setg_errno(errp, -ret, "Could not write description");
goto close_exit?
> }
> - ret = qemu_write_full(fd, desc, strlen(desc));
> - if (ret != strlen(desc)) {
> - ret = -errno;
> - goto close_exit;
> + /* bdrv_pwrite write padding zeros to align to sector, we don't need that
> + * for description file */
> + if (desc_offset == 0) {
> + ret = bdrv_truncate(new_bs, desc_offset + desc_len);
We know desc_offset == 0, so desc_offset (0) + desc_len is really just
desc_len.
next prev parent reply other threads:[~2013-12-19 13:13 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-12-17 12:00 [Qemu-devel] [PATCH] vmdk: Allow vmdk_create to work with protocol Fam Zheng
2013-12-19 13:12 ` Stefan Hajnoczi [this message]
2013-12-20 1:48 ` Fam Zheng
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20131219131232.GC2537@stefanha-thinkpad.redhat.com \
--to=stefanha@gmail.com \
--cc=famz@redhat.com \
--cc=kwolf@redhat.com \
--cc=pl@kamp.de \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).