On 8/2/24 13:22, Andrew Davis wrote:
This is simpler and removes the need to assume the id length to be 1
digit, which then removes a W=1 compile warning about the same.

Signed-off-by: Andrew Davis <a...@ti.com>
Acked/Tested-by: Hari Nagalla <hnaga...@ti.com>
---
  drivers/remoteproc/keystone_remoteproc.c | 6 +-----
  1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/remoteproc/keystone_remoteproc.c 
b/drivers/remoteproc/keystone_remoteproc.c
index 7e57b90bcaf85..81b179e269a1e 100644
--- a/drivers/remoteproc/keystone_remoteproc.c
+++ b/drivers/remoteproc/keystone_remoteproc.c
@@ -366,8 +366,6 @@ static int keystone_rproc_probe(struct platform_device 
*pdev)
        struct rproc *rproc;
        int dsp_id;
        char *fw_name = NULL;
-       char *template = "keystone-dsp%d-fw";
-       int name_len = 0;
        int ret = 0;
if (!np) {
@@ -382,11 +380,9 @@ static int keystone_rproc_probe(struct platform_device 
*pdev)
        }
/* construct a custom default fw name - subject to change in future */
-       name_len = strlen(template); /* assuming a single digit alias */
-       fw_name = devm_kzalloc(dev, name_len, GFP_KERNEL);
+       fw_name = devm_kasprintf(dev, GFP_KERNEL, "keystone-dsp%d-fw", dsp_id);
        if (!fw_name)
                return -ENOMEM;
-       snprintf(fw_name, name_len, template, dsp_id);
rproc = rproc_alloc(dev, dev_name(dev), &keystone_rproc_ops, fw_name,
                            sizeof(*ksproc));


Reply via email to