From: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
To: "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>
Cc: Denis Lunev <den@virtuozzo.com>,
Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>,
"armbru@redhat.com" <armbru@redhat.com>,
"mdroth@linux.vnet.ibm.com" <mdroth@linux.vnet.ibm.com>
Subject: Re: [Qemu-devel] [PATCH v3] make check-unit: use after free in test-opts-visitor
Date: Tue, 20 Aug 2019 15:17:18 +0000 [thread overview]
Message-ID: <145da1b2-4a72-ee7e-515c-00b5f84ca8e6@virtuozzo.com> (raw)
In-Reply-To: <1565024586-387112-1-git-send-email-andrey.shinkevich@virtuozzo.com>
PINGING...
On 05/08/2019 20:03, Andrey Shinkevich wrote:
> In the struct OptsVisitor, the 'repeated_opts' member points to a list
> in the 'unprocessed_opts' hash table after the list has been destroyed.
> A subsequent call to visit_type_int() references the deleted list.
> It results in use-after-free issue reproduced by running the test case
> under the Valgrind: valgrind tests/test-opts-visitor.
> A new mode ListMode::LM_TRAVERSED is declared to mark the list
> traversal completed.
>
> Suggested-by: Markus Armbruster <armbru@redhat.com>
> Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
> ---
>
> v3:
> 01: The comment of the patch header was amended.
> 02: The change in spacing of 'ListMode' comment blocks was rolled back.
> 03: The 'repeated_opts' in opts_end_list() is now reset unconditionally
> as it was.
> 04: The 'name' in the error_setg() was removed as the pointer to the list
> name can be null (suggested by Markus).
>
> qapi/opts-visitor.c | 26 ++++++++++++++++++++++----
> 1 file changed, 22 insertions(+), 4 deletions(-)
>
> diff --git a/qapi/opts-visitor.c b/qapi/opts-visitor.c
> index 324b197..5fe0276 100644
> --- a/qapi/opts-visitor.c
> +++ b/qapi/opts-visitor.c
> @@ -24,7 +24,8 @@ enum ListMode
> {
> LM_NONE, /* not traversing a list of repeated options */
>
> - LM_IN_PROGRESS, /* opts_next_list() ready to be called.
> + LM_IN_PROGRESS, /*
> + * opts_next_list() ready to be called.
> *
> * Generating the next list link will consume the most
> * recently parsed QemuOpt instance of the repeated
> @@ -36,7 +37,8 @@ enum ListMode
> * LM_UNSIGNED_INTERVAL.
> */
>
> - LM_SIGNED_INTERVAL, /* opts_next_list() has been called.
> + LM_SIGNED_INTERVAL, /*
> + * opts_next_list() has been called.
> *
> * Generating the next list link will consume the most
> * recently stored element from the signed interval,
> @@ -48,7 +50,14 @@ enum ListMode
> * next element of the signed interval.
> */
>
> - LM_UNSIGNED_INTERVAL /* Same as above, only for an unsigned interval. */
> + LM_UNSIGNED_INTERVAL, /* Same as above, only for an unsigned interval. */
> +
> + LM_TRAVERSED /*
> + * opts_next_list() has been called.
> + *
> + * No more QemuOpt instance in the list.
> + * The traversal has been completed.
> + */
> };
>
> typedef enum ListMode ListMode;
> @@ -238,6 +247,8 @@ opts_next_list(Visitor *v, GenericList *tail, size_t size)
> OptsVisitor *ov = to_ov(v);
>
> switch (ov->list_mode) {
> + case LM_TRAVERSED:
> + return NULL;
> case LM_SIGNED_INTERVAL:
> case LM_UNSIGNED_INTERVAL:
> if (ov->list_mode == LM_SIGNED_INTERVAL) {
> @@ -258,6 +269,8 @@ opts_next_list(Visitor *v, GenericList *tail, size_t size)
> opt = g_queue_pop_head(ov->repeated_opts);
> if (g_queue_is_empty(ov->repeated_opts)) {
> g_hash_table_remove(ov->unprocessed_opts, opt->name);
> + ov->repeated_opts = NULL;
> + ov->list_mode = LM_TRAVERSED;
> return NULL;
> }
> break;
> @@ -289,7 +302,8 @@ opts_end_list(Visitor *v, void **obj)
>
> assert(ov->list_mode == LM_IN_PROGRESS ||
> ov->list_mode == LM_SIGNED_INTERVAL ||
> - ov->list_mode == LM_UNSIGNED_INTERVAL);
> + ov->list_mode == LM_UNSIGNED_INTERVAL ||
> + ov->list_mode == LM_TRAVERSED);
> ov->repeated_opts = NULL;
> ov->list_mode = LM_NONE;
> }
> @@ -306,6 +320,10 @@ lookup_scalar(const OptsVisitor *ov, const char *name, Error **errp)
> list = lookup_distinct(ov, name, errp);
> return list ? g_queue_peek_tail(list) : NULL;
> }
> + if (ov->list_mode == LM_TRAVERSED) {
> + error_setg(errp, "Fewer list elements than expected");
> + return NULL;
> + }
> assert(ov->list_mode == LM_IN_PROGRESS);
> return g_queue_peek_head(ov->repeated_opts);
> }
>
--
With the best regards,
Andrey Shinkevich
next prev parent reply other threads:[~2019-08-20 15:19 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-05 17:03 [Qemu-devel] [PATCH v3] make check-unit: use after free in test-opts-visitor Andrey Shinkevich
2019-08-20 15:17 ` Andrey Shinkevich [this message]
2019-08-21 11:25 ` Markus Armbruster
2019-08-21 11:55 ` Andrey Shinkevich
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=145da1b2-4a72-ee7e-515c-00b5f84ca8e6@virtuozzo.com \
--to=andrey.shinkevich@virtuozzo.com \
--cc=armbru@redhat.com \
--cc=den@virtuozzo.com \
--cc=mdroth@linux.vnet.ibm.com \
--cc=qemu-devel@nongnu.org \
--cc=vsementsov@virtuozzo.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).