On Mon, Jul 04, 2022 at 02:23:38PM +0100, Bruce Richardson wrote:
> On Sun, Jul 03, 2022 at 01:22:42PM +0100, Kevin Laatz wrote:
> > During PCI device close, any allocated memory needs to be free'd.
> > Currently, one of the free's is being called on an incorrect idxd_dmadev
> > struct member, namely 'batch_idx_ring', causing a memleak from the
> > pointer that should have been free'd.
>
> I think you need to explain that the two rings are beside each other in
> memory and we need to free using the pointer to the start of the block,
> rather than the pointer to the middle of it.
>
> > This patch fixes this memleak by calling free on the correct pointer.
> >
> > Fixes: 9449330a8458 ("dma/idxd: create dmadev instances on PCI probe")
> > Cc: sta...@dpdk.org
> > Cc: bruce.richard...@intel.com
> >
> > Signed-off-by: Kevin Laatz <kevin.la...@intel.com>
>
> With more explanation in the commit log
>
Correction (obviously!)
Acked-by: Bruce Richardson <bruce.richard...@intel.com>
