On Sat, Apr 23, 2022 at 03:24:03PM +0200, Christophe JAILLET wrote:
> platform_device_add_data() duplicates the memory it is passed. So we can
> free some memory to save a few bytes that would remain unused otherwise.
>
> Signed-off-by: Christophe JAILLET <christophe.jaillet@xxxxxxxxxx>
> ---
> arch/mips/sgi-ip27/ip27-xtalk.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/arch/mips/sgi-ip27/ip27-xtalk.c b/arch/mips/sgi-ip27/ip27-xtalk.c
> index 000ede156bdc..e762886d1dda 100644
> --- a/arch/mips/sgi-ip27/ip27-xtalk.c
> +++ b/arch/mips/sgi-ip27/ip27-xtalk.c
> @@ -53,6 +53,8 @@ static void bridge_platform_create(nasid_t nasid, int widget, int masterwid)
> }
> platform_device_add_resources(pdev, &w1_res, 1);
> platform_device_add_data(pdev, wd, sizeof(*wd));
> + /* platform_device_add_data() duplicates the data */
> + kfree(wd);
> platform_device_add(pdev);
>
> bd = kzalloc(sizeof(*bd), GFP_KERNEL);
> @@ -83,6 +85,8 @@ static void bridge_platform_create(nasid_t nasid, int widget, int masterwid)
> bd->io_offset = offset;
>
> platform_device_add_data(pdev, bd, sizeof(*bd));
> + /* platform_device_add_data() duplicates the data */
> + kfree(bd);
> platform_device_add(pdev);
> pr_info("xtalk:n%d/%x bridge widget\n", nasid, widget);
> return;
> --
> 2.32.0
applied to mips-next.
Thomas.
--
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea. [ RFC1925, 2.3 ]
