On Mon, 17 May 2021 16:23:56 +1000
David Gibson <david@gibson.dropbear.id.au> wrote:

> On Sat, May 15, 2021 at 01:07:59PM +0530, Vaibhav Jain wrote:
[...]
> > +    rc = (result == MEMTX_OK) ?
> > +        scm_perf_check_rr_buffer(perfstats, addr, size, &num_stats) :
> > +        H_PRIVILEGE;
> 
> This is a bit cryptic.  Just deal with the memtx error first, then run
> the buffer validation.  Actually, you can unify the exit paths for
> these and the success case by using a goto label near the end which
> has the g_free() and return rc.
> 

It seems all the g_free() calls could even be avoided by
converting perfstats to g_autofree.