On Tue, Jun 27, 2017 at 04:18:42PM +0100, Ian Abbott wrote:
> The global spinlock `l3_remap_lock` is reinitialized every time the
> "probe" function `alt_fpga_bridge_probe()` is called. It should only be
> initialized once. Use `DEFINE_SPINLOCK()` to initialize it statically.
>
> Fixes: e5f8efa5c8bf ("ARM: socfpga: fpga bridge driver support")
> Cc: <stable@xxxxxxxxxxxxxxx> # 4.10+
> Signed-off-by: Ian Abbott <abbotti@xxxxxxxxx>
Reviewed-By: Moritz Fischer <mdf@xxxxxxxxxx>
Thanks for catching that,
Moritz
> ---
> drivers/fpga/altera-hps2fpga.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/drivers/fpga/altera-hps2fpga.c b/drivers/fpga/altera-hps2fpga.c
> index 3066b805f2d0..08c0ecb7d109 100644
> --- a/drivers/fpga/altera-hps2fpga.c
> +++ b/drivers/fpga/altera-hps2fpga.c
> @@ -66,7 +66,7 @@ static int alt_hps2fpga_enable_show(struct fpga_bridge *bridge)
>
> /* The L3 REMAP register is write only, so keep a cached value. */
> static unsigned int l3_remap_shadow;
> -static spinlock_t l3_remap_lock;
> +static DEFINE_SPINLOCK(l3_remap_lock);
>
> static int _alt_hps2fpga_enable_set(struct altera_hps2fpga_data *priv,
> bool enable)
> @@ -171,8 +171,6 @@ static int alt_fpga_bridge_probe(struct platform_device *pdev)
> return -EBUSY;
> }
>
> - spin_lock_init(&l3_remap_lock);
> -
> if (!of_property_read_u32(dev->of_node, "bridge-enable", &enable)) {
> if (enable > 1) {
> dev_warn(dev, "invalid bridge-enable %u > 1\n", enable);
> --
> 2.11.0
>
Attachment:
signature.asc
Description: PGP signature