On Tue, Nov 20, 2012 at 4:30 AM, Egbert Eich <e...@suse.de> wrote:
> drm_get_edid() returns a pointer to an EDID block. The caller
> is responsible to free this pointer itself.
> Here the pointer gets assigned to the local variable raw_edid.
> Therefore it should be freed before the variable goes out of
> scope.
>
> Signed-off-by: Egbert Eich <e...@suse.de>
> ---
> drivers/gpu/drm/exynos/exynos_hdmi.c | 1 +
> 1 files changed, 1 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c
> b/drivers/gpu/drm/exynos/exynos_hdmi.c
> index 2c115f8..bc87bca 100644
> --- a/drivers/gpu/drm/exynos/exynos_hdmi.c
> +++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
> @@ -1293,6 +1293,7 @@ static int hdmi_get_edid(void *ctx, struct
> drm_connector *connector,
> DRM_DEBUG_KMS("%s : width[%d] x height[%d]\n",
> (hdata->dvi_mode ? "dvi monitor" : "hdmi monitor"),
> raw_edid->width_cm, raw_edid->height_cm);
> + kfree(raw_edid);
This will actually cause the memory to be freed twice.
The reason this happens is drm_get_edid attaches this to
connector->display_info.raw_edid, which is then freed in the
exynos_drm_connector function that gets the edid.
The whole thing is ugly, and needs to be revised. I've uploaded a
patch to refactor this against the chromium tree, but haven't yet
rebased against upstream. See
https://gerrit.chromium.org/gerrit/#/c/38406/
For now, please drop this patch.
Sean
> } else {
> return -ENODEV;
> }
> --
> 1.7.7
>
> _______________________________________________
> dri-devel mailing list
> dri-devel@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/dri-devel
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
