Quoting Dmitry Baryshkov (2021-04-06 16:06:06)
> devm_clk_hw_register_fixed_factor_release(), the release function for
> the devm_clk_hw_register_fixed_factor(), calls
> clk_hw_unregister_fixed_factor(), which will kfree() the clock. However
> after that the devres functions will also kfree the allocated data,
> resulting in double free/memory corruption. Just call
> clk_hw_unregister() instead, leaving kfree() to devres code.
> 
> Reported-by: Rob Clark <robdcl...@chromium.org>
> Cc: Daniel Palmer <dan...@0x0f.com>
> Signed-off-by: Dmitry Baryshkov <dmitry.barysh...@linaro.org>
> ---

Applied to clk-fixes. I also squashed this in to cleanup that ugly cast.

---8<----

diff --git a/drivers/clk/clk-fixed-factor.c b/drivers/clk/clk-fixed-factor.c
index 390c16f321a6..4e4b6d367612 100644
--- a/drivers/clk/clk-fixed-factor.c
+++ b/drivers/clk/clk-fixed-factor.c
@@ -66,12 +66,14 @@ EXPORT_SYMBOL_GPL(clk_fixed_factor_ops);
 
 static void devm_clk_hw_register_fixed_factor_release(struct device *dev, void 
*res)
 {
+       struct clk_fixed_factor *fix = res;
+
        /*
         * We can not use clk_hw_unregister_fixed_factor, since it will kfree()
         * the hw, resulting in double free. Just unregister the hw and let
         * devres code kfree() it.
         */
-       clk_hw_unregister(&((struct clk_fixed_factor *)res)->hw);
+       clk_hw_unregister(&fix->hw);
 }
 
 static struct clk_hw *
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

Reply via email to