Re: [Qemu-devel] [Qemu-stable] [PATCH 09/14] blockdev: Plug memory leak in drive_init()

[Markus Armbruster]

Benoît Canet <address@hidden> writes:

> The Tuesday 27 May 2014 à 21:44:15 (+0200), Markus Armbruster wrote :
>> Benoît Canet <address@hidden> writes:
>> 
>> > The Tuesday 27 May 2014 à 21:11:45 (+0200), Markus Armbruster wrote :
>> >> Benoît Canet <address@hidden> writes:
>> >> 
>> >> > The Tuesday 27 May 2014 à 18:13:12 (+0200), Markus Armbruster wrote :
>> >> >> Benoît Canet <address@hidden> writes:
>> >> >> 
>> >> >> > The Monday 26 May 2014 à 19:37:10 (+0200), Markus Armbruster wrote :
>> >> >> >> Introduced in commit f298d07.  Spotted by Coverity.
>> >> >> >> 
>> >> >> >> Signed-off-by: Markus Armbruster <address@hidden>
>> >> >> >> ---
>> >> >> >>  blockdev.c | 2 ++
>> >> >> >>  1 file changed, 2 insertions(+)
>> >> >> >> 
>> >> >> >> diff --git a/blockdev.c b/blockdev.c
>> >> >> >> index 6460c70..7ec7d79 100644
>> >> >> >> --- a/blockdev.c
>> >> >> >> +++ b/blockdev.c
>> >> >> >> @@ -941,6 +941,7 @@ DriveInfo *drive_init(QemuOpts *all_opts, 
>> >> >> >> BlockInterfaceType block_default_type)
>> >> >> >>  
>> >> >> >>      /* Actual block device init: Functionality shared with 
>> >> >> >> blockdev-add */
>> >> >> >>      dinfo = blockdev_init(filename, bs_opts, &local_err);
>> >> >> >> +    bs_opts = NULL;
>> >> >
>> >> > What is the purpose of this line ? I though it was to avoid double 
>> >> > unref.
>> >> 
>> >> Before this patch, bs_opts gets leaked on any path from its qdict_new()
>> >> that doesn't go through blockdev_init().
>> >> 
>> >> The new line below frees it, but without the line above, it would free
>> >> it a second time on paths that go through blockdev_init().
>> >> 
>> >> Clear now?
>> >
>> > Clear from the start it fixes a potential double free.
>> > "This commits seems to fix two thing a leak and a double free."
>> 
>> Well, before the patch, the leak exists, but there is no double-free.
>> 
>> The patch fixes only one thing: the leak.  It takes care not to break
>> things by freeing when it shouldn't.
>> 
>> Do you still think the commit message should be amended?  How?
>
> Maybe just says it also take care not to introduce a double free.

Adding this paragraph:

    bs_opts is leaked on all paths from its qdev_new() that don't got
    through blockdev_init().  Add the missing QDECREF(), and zap bs_opts
    after blockdev_init(), so the new QDECREF() does nothing when we go
    through blockdev_init().

Thanks!