From: Eric Blake <eblake@redhat.com>
To: Chunyan Liu <cyliu@suse.com>, qemu-devel@nongnu.org
Cc: stefanha@redhat.com
Subject: Re: [Qemu-devel] [PATCH v23 08/32] add convert functions between QEMUOptionParameter to QemuOpts
Date: Tue, 25 Mar 2014 15:35:21 -0600 [thread overview]
Message-ID: <5331F699.3010502@redhat.com> (raw)
In-Reply-To: <1395396763-26081-9-git-send-email-cyliu@suse.com>
[-- Attachment #1: Type: text/plain, Size: 4064 bytes --]
On 03/21/2014 04:12 AM, Chunyan Liu wrote:
> Add two temp convert functions between QEMUOptionParameter to QemuOpts,
s/convert/conversion/ here and in subject
> so that next patch can use it. It will simplify later patch for easier
> review. And will be finally removed after all backend drivers switch to
> QemuOpts.
>
> Signed-off-by: Chunyan Liu <cyliu@suse.com>
> ---
> +++ b/include/qemu/option.h
> @@ -103,6 +103,11 @@ typedef struct QemuOptDesc {
> } QemuOptDesc;
>
> struct QemuOptsList {
> + /* FIXME: Temp used for QEMUOptionParamter->QemuOpts conversion to
> + * indicate free memory. Will remove after all drivers switch to QemuOpts.
> + */
> + bool mallocd;
Spelling looks odd; I might have used 'allocated'. But as it gets
deleted later, I don't care.
> + num_opts = count_option_parameters(list);
> + opts = g_malloc0(sizeof(QemuOptsList) +
> + (num_opts + 1) * sizeof(QemuOptDesc));
[1]...
> + while (list && list->name) {
> + opts->desc[i].name = g_strdup(list->name);
> + opts->desc[i].help = g_strdup(list->help);
> + switch (list->type) {
> + case OPT_FLAG:
> + opts->desc[i].type = QEMU_OPT_BOOL;
> + opts->desc[i].def_value_str =
> + g_strdup(list->value.n ? "on" : "off");
This always sets def_value_str...
> + break;
> +
> + case OPT_NUMBER:
> + opts->desc[i].type = QEMU_OPT_NUMBER;
> + if (list->value.n) {
> + opts->desc[i].def_value_str =
> + g_strdup_printf("%" PRIu64, list->value.n);
> + }
...whereas this only sets def_value_str for non-zero values. But can't
0 be a valid setting? Is this mismatch in what gets converted going to
bite us? Should you be paying attention to list->assigned instead or in
addition to just checking for non-zero values?
> + break;
> +
> + case OPT_SIZE:
> + opts->desc[i].type = QEMU_OPT_SIZE;
> + if (list->value.n) {
> + opts->desc[i].def_value_str =
> + g_strdup_printf("%" PRIu64, list->value.n);
> + }
Same question for 0 values.
> + break;
> +
> + case OPT_STRING:
> + opts->desc[i].type = QEMU_OPT_STRING;
> + opts->desc[i].def_value_str = g_strdup(list->value.s);
> + break;
> + }
> +
> + i++;
> + list++;
> + opts->desc[i].name = NULL;
...[1] This assignment is dead code, because you used malloc0 which
guarantees that desc[i].name is already NULL.
> +/* convert QemuOpts to QEMUOptionParamter
s/Paramter/Parameter/
> + * Note: result QEMUOptionParameter has shorter lifetime than
> + * input QemuOpts.
> + * FIXME: this function will be removed after all drivers
> + * switch to QemuOpts
> + */
> +QEMUOptionParameter *opts_to_params(QemuOpts *opts)
> +{
> + num_opts = count_opts_list(opts->list);
> + dest = g_malloc0((num_opts + 1) * sizeof(QEMUOptionParameter));
> +
> + i++;
> + desc++;
> + dest[i].name = NULL;
Another dead assignment.
> + }
> +
> + return dest;
> +}
> +
> +void qemu_opts_free(QemuOptsList *list)
> +{
> + /* List members point to new malloced space and need to free.
> + * FIXME:
> + * Introduced for QEMUOptionParamter->QemuOpts conversion.
> + * Will remove after all drivers switch to QemuOpts.
> + */
> + if (list && list->mallocd) {
> + QemuOptDesc *desc = list->desc;
> + while (desc && desc->name) {
> + g_free((char *)desc->name);
> + g_free((char *)desc->help);
Are these casts still necessary, given patch 4?
> + g_free((char *)desc->def_value_str);
However, it looks like you are correct that this one is casting away const.
--
Eric Blake eblake redhat com +1-919-301-3266
Libvirt virtualization library http://libvirt.org
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 604 bytes --]
next prev parent reply other threads:[~2014-03-25 21:35 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-03-21 10:12 [Qemu-devel] [PATCH v23 00/32] replace QEMUOptionParameter with QemuOpts Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 01/32] move find_desc_by_name ahead for later calling Chunyan Liu
2014-03-21 23:16 ` Eric Blake
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 02/32] add def_value_str to QemuOptDesc Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 03/32] qapi: output def_value_str when query command line options Chunyan Liu
2014-03-21 23:27 ` Eric Blake
2014-03-24 3:18 ` Chun Yan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 04/32] change opt->name and opt->str from (const char *) to (char *) Chunyan Liu
2014-03-25 19:00 ` Leandro Dorileo
2014-03-25 19:23 ` Eric Blake
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 05/32] move qemu_opt_del ahead for later calling Chunyan Liu
2014-03-25 19:29 ` Eric Blake
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 06/32] add qemu_opt_get_*_del functions for replace work Chunyan Liu
2014-03-25 20:33 ` Eric Blake
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 07/32] add qemu_opts_print_help to replace print_option_help Chunyan Liu
2014-03-25 19:07 ` Leandro Dorileo
2014-03-25 20:43 ` Eric Blake
2014-03-26 2:58 ` Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 08/32] add convert functions between QEMUOptionParameter to QemuOpts Chunyan Liu
2014-03-25 21:35 ` Eric Blake [this message]
2014-03-26 3:26 ` Chunyan Liu
2014-03-26 6:30 ` Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 09/32] add qemu_opts_append to repalce append_option_parameters Chunyan Liu
2014-03-25 19:13 ` Leandro Dorileo
2014-03-25 21:40 ` Eric Blake
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 10/32] check NULL input for qemu_opts_del Chunyan Liu
2014-03-25 21:41 ` Eric Blake
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 11/32] qemu_opts_print: change fprintf stderr to printf Chunyan Liu
2014-03-25 20:10 ` Leandro Dorileo
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 12/32] qcow2.c: remove 'assigned' check in amend Chunyan Liu
2014-03-25 19:25 ` Leandro Dorileo
2014-03-26 7:37 ` Chunyan Liu
2014-03-27 7:27 ` Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 13/32] change block layer to support both QemuOpts and QEMUOptionParamter Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 14/32] vvfat.c: handle cross_driver's create_options and create_opts Chunyan Liu
2014-03-25 19:17 ` Leandro Dorileo
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 15/32] cow.c: replace QEMUOptionParameter with QemuOpts Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 16/32] gluster.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 17/32] iscsi.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 18/32] qcow.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 19/32] qcow2.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 20/32] qed.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 21/32] raw-posix.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 22/32] raw-win32.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 23/32] raw_bsd.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 24/32] rbd.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 25/32] sheepdog.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 26/32] ssh.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 27/32] vdi.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 28/32] vhdx.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 29/32] vmdk.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 30/32] vpc.c: " Chunyan Liu
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 31/32] cleanup QEMUOptionParameter Chunyan Liu
2014-03-25 18:02 ` Leandro Dorileo
2014-03-21 10:12 ` [Qemu-devel] [PATCH v23 32/32] cleanup tmp 'mallocd' member from QemuOptsList Chunyan Liu
2014-03-25 18:09 ` [Qemu-devel] [PATCH v23 00/32] replace QEMUOptionParameter with QemuOpts Leandro Dorileo
2014-03-25 20:22 ` Leandro Dorileo
2014-03-26 15:18 ` Stefan Hajnoczi
2014-03-27 7:20 ` Chunyan Liu
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=5331F699.3010502@redhat.com \
--to=eblake@redhat.com \
--cc=cyliu@suse.com \
--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).