Re: [PATCH] HID: intel-ish-hid: fixes incorrect error handling

On Fri, 7 Dec 2018, Benjamin Tissoires wrote:

quoted

The memory chunk allocated by hid_allocate_device() should be released
by hid_destroy_device(), not kfree().

Fixes: 0b28cb4bcb1("HID: intel-ish-hid: ISH HID client driver")
Signed-off-by: Pan Bian <redacted>
---
 drivers/hid/intel-ish-hid/ishtp-hid.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/hid/intel-ish-hid/ishtp-hid.c b/drivers/hid/intel-ish-hid/ishtp-hid.c
index cd23903..e918d78 100644
--- a/drivers/hid/intel-ish-hid/ishtp-hid.c
+++ b/drivers/hid/intel-ish-hid/ishtp-hid.c

@@ -222,7 +222,7 @@ int ishtp_hid_probe(unsigned int cur_hid_dev,
 err_hid_device:
        kfree(hid_data);
 err_hid_data:
-       kfree(hid);
+       hid_destroy_device(hid);

Looks good to me. Srinivas, any comments?
FWIW:
Reviewed-by: Benjamin Tissoires <redacted>

Jiri, do you think we should send this one as 4.20/fixes material or
wait for 4.21?

Given that this has been around since 4.9, I wouldn't be devastated if it 
lands only in next merge window. So I'd just put it to 4.20/fixes and wait 
for other more serious trigger for sending that to Linus eventually.

Thanks,

-- 
Jiri Kosina
SUSE Labs