On 2024/6/14 18:33, Adrian Hunter wrote:
On 13/06/24 04:43, Chen Wang wrote:
From: Chen Wang <unicorn_wang@xxxxxxxxxxx>
[......]
static void cv18xx_sdhci_reset(struct sdhci_host *host, u8 mask)
{
struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
@@ -1230,46 +1270,15 @@ static int dwcmshc_probe(struct platform_device *pdev)
host->mmc_host_ops.execute_tuning = dwcmshc_execute_tuning;
if (pltfm_data == &sdhci_dwcmshc_rk35xx_pdata) {
- rk_priv = devm_kzalloc(&pdev->dev, sizeof(struct rk35xx_priv), GFP_KERNEL);
- if (!rk_priv) {
- err = -ENOMEM;
- goto err_clk;
- }
-
- if (of_device_is_compatible(pdev->dev.of_node, "rockchip,rk3588-dwcmshc"))
- rk_priv->devtype = DWCMSHC_RK3588;
- else
- rk_priv->devtype = DWCMSHC_RK3568;
-
- priv->priv = rk_priv;
-
- err = rk35xx_init(host, priv);
+ err = rk35xx_init(&pdev->dev, host, priv);
rk_priv is used further on, but it is not assigned anymore.
You are right. It seems unnecessary to provide this patch separately
just to extract the init function, and it also violates the principle of
atomicity of patch modification. I will merge this patch with the next one.
[......]
