Hi Zhang,
On Sun, 29 Jun 2025 at 17:57, Zhang Shurong <zhang_shurong@xxxxxxxxxxx> wrote:
> pm_runtime_get_sync will increment pm usage counter
> even it failed. Forgetting to putting operation will
> result in reference leak here. We fix it by replacing
> it with pm_runtime_resume_and_get to keep usage counter
> balanced.
>
> Fixes: 87244fe5abdf ("dmaengine: rcar-dmac: Add Renesas R-Car Gen2 DMA Controller (DMAC) driver")
> Signed-off-by: Zhang Shurong <zhang_shurong@xxxxxxxxxxx>
Thanks for your patch!
Reviewed-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
> --- a/drivers/dma/sh/rcar-dmac.c
> +++ b/drivers/dma/sh/rcar-dmac.c
> @@ -1068,7 +1068,7 @@ static int rcar_dmac_alloc_chan_resources(struct dma_chan *chan)
> if (ret < 0)
> return -ENOMEM;
>
> - return pm_runtime_get_sync(chan->device->dev);
> + return pm_runtime_resume_and_get(chan->device->dev);
Note that there are other issues with this function: in case of failure,
none of the memory allocated before is freed. Probably the original
author assumed none of this can really fail.
> }
>
> static void rcar_dmac_free_chan_resources(struct dma_chan *chan)
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
