From: Markus Armbruster <armbru@redhat.com>
To: Fabiano Rosas <farosas@suse.de>
Cc: qemu-devel@nongnu.org, "Marc-André Lureau" <marcandre.lureau@redhat.com>
Subject: Re: [PATCH] qemu-option: Cope with a missing QemuOptsList
Date: Tue, 15 Sep 2026 16:50:06 +0200 [thread overview]
Message-ID: <87wlsmfu1t.fsf@pond.sub.org> (raw)
In-Reply-To: <20260915134412.4187164-1-farosas@suse.de> (Fabiano Rosas's message of "Tue, 15 Sep 2026 10:44:12 -0300")
Fabiano Rosas <farosas@suse.de> writes:
> It's a common pattern in vl.c to chain the qemu_find_opts() output
> into the first argument of qemu_opts_parse_noisily():
>
> opts = qemu_opts_parse_noisily(qemu_find_opts("spice"), optarg, false);
> if (!opts) {
> exit(1);
> }
>
> In cases such as spice that have the group defined in the module file,
> it's possible to reach qemu_opts_parse_noisily() with a NULL QemuOptsList if
> the module is not present in the host filesystem.
>
> $ ../configure --enable-modules --enable-spice
> $ make
> $ mv qemu-bundle/usr/local/lib64/qemu/ui-spice-core.so{,.not}
> $ ./qemu-system-x86_64 -spice a
> qemu-system-x86_64: -spice a: There is no option group 'spice'
> Segmentation fault (core dumped)
>
> Return NULL from qemu_opts_parse_noisily() if there is no list.
>
> Signed-off-by: Fabiano Rosas <farosas@suse.de>
> ---
> util/qemu-option.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/util/qemu-option.c b/util/qemu-option.c
> index 9fbf425f86..0ec12d252c 100644
> --- a/util/qemu-option.c
> +++ b/util/qemu-option.c
> @@ -945,6 +945,10 @@ QemuOpts *qemu_opts_parse_noisily(QemuOptsList *list, const char *params,
> QemuOpts *opts;
> bool help_wanted = false;
>
> + if (!list) {
> + return NULL;
> + }
> +
> opts = opts_parse(list, params, permit_abbrev, true,
> opts_accepts_any(list) ? NULL : &help_wanted,
> &err);
Before the patch, qemu_opts_parse_noisily() either
* Succeeds and returns non-null
* Fails, reports an error, and returns null
* Prints help and returns null
Your patch adds a fourth case:
* Fails silently and returns null
I dislike this case. Functions should either always print something
whent they fail, or never.
Are all callers prepared for silent failure?
The pattern you quoted in the commit message is, because
qemu_find_opts() reports an error.
Here's a cleaner solution for this pattern. Replace
opts = qemu_opts_parse_noisily(qemu_find_opts(...), ...)
by a call of a new helper function that does
list = qemu_find_opts(...);
if (!list) {
return NULL;
}
return qemu_opts_parse_noisily(list, ...);
Thoughts?
next prev parent reply other threads:[~2026-09-15 14:51 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-15 13:44 [PATCH] qemu-option: Cope with a missing QemuOptsList Fabiano Rosas
2026-09-15 14:20 ` Marc-André Lureau
2026-09-15 14:50 ` Markus Armbruster [this message]
2026-09-18 22:45 ` Fabiano Rosas
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=87wlsmfu1t.fsf@pond.sub.org \
--to=armbru@redhat.com \
--cc=farosas@suse.de \
--cc=marcandre.lureau@redhat.com \
--cc=qemu-devel@nongnu.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.