Re: [PATCH v2] imx-drm: Fix probe failure

[Fabio Estevam]

Hi Sascha,

On Thu, Aug 29, 2013 at 3:37 PM, Sascha Hauer <s.hauer@xxxxxxxxxxxxxx> wrote:
> It's not about the module refcount. It's only that a on a drm device
> which is in use you better do not add/remove components.
>
> Again: The imx-drm driver does *not* do hotplugging. It will *not* add
> or remove components when the device is opened. The code you remove
> exactly makes (or made before Daniels patch) that sure.

I am not sure what would be the correct fix for this.

What about this?

--- a/drivers/staging/imx-drm/imx-drm-core.c
+++ b/drivers/staging/imx-drm/imx-drm-core.c
@@ -438,6 +438,8 @@ static int imx_drm_driver_load(struct drm_device
*drm, unsigned long flags)
                ret = -EINVAL;

        ret = 0;
+
+       imxdrm->references = -1;

 err_init:
        mutex_unlock(&imxdrm->mutex);
@@ -485,7 +487,7 @@ int imx_drm_add_crtc(struct drm_crtc *crtc,

        mutex_lock(&imxdrm->mutex);

-       if (imxdrm->references) {
+       if (imxdrm->references > 0) {
                ret = -EBUSY;
                goto err_busy;
        }
@@ -564,7 +566,7 @@ int imx_drm_add_encoder(struct drm_encoder *encoder,

        mutex_lock(&imxdrm->mutex);

-       if (imxdrm->references) {
+       if (imxdrm->references > 0) {
                ret = -EBUSY;
                goto err_busy;
        }
@@ -709,7 +711,7 @@ int imx_drm_add_connector(struct drm_connector *connector,

        mutex_lock(&imxdrm->mutex);

-       if (imxdrm->references) {
+       if (imxdrm->references > 0) {
                ret = -EBUSY;
                goto err_busy;
        }
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/