On 2012年07月05日 03:00, Alan Stern wrote:
On Wed, 4 Jul 2012, Lan Tianyu wrote:
Since port may power off, usb device descriptor can't be get
at that time and usb_get_configuration() should return error.
Signed-off-by: Lan Tianyu<tianyu.lan@xxxxxxxxx>
---
drivers/usb/core/config.c | 4 +---
1 files changed, 1 insertions(+), 3 deletions(-)
diff --git a/drivers/usb/core/config.c b/drivers/usb/core/config.c
index f4bdd0c..072e8a2 100644
--- a/drivers/usb/core/config.c
+++ b/drivers/usb/core/config.c
@@ -702,9 +702,7 @@ int usb_get_configuration(struct usb_device *dev)
if (result< 0) {
dev_err(ddev, "unable to read config index %d "
"descriptor/%s: %d\n", cfgno, "start", result);
- dev_err(ddev, "chopping to %d config(s)\n", cfgno);
- dev->descriptor.bNumConfigurations = cfgno;
- break;
+ goto err;
I really would prefer it if you goto err only when result != -EPIPE.
Otherwise, do the same thing as the existing code.
Oh sorry. I miss your comment at your last mail. I get it. But return error is
-71 (EPROTO)
How about this?
dev_err(ddev, "unable to read config index %d "
"descriptor/%s: %d\n", cfgno, "start", result)
if (result == -EPROTO)
goto err;
dev_err(ddev, "chopping to %d config(s)\n", cfgno);
dev->descriptor.bNumConfigurations = cfgno;
break;
Why did you suggest "result != -EPIPE"?
And if we use it, the judgement should be before dev_err("chopping to %d
config(s)\n"...).
Becasue the error info is not necessary. Right?
Think about what happens if a USB device has a mistake in its device
descriptor and sets bNumConfigurations higher than the actual number
of configs.
Alan Stern
--
Best Regards
Tianyu Lan
linux kernel enabling team
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
