On 08/03/2017 03:25 PM, Patrick DELAUNAY wrote:

Hi Marek,

Hi,

please stop top-posting.

Do think that this minor patch could be integrated in v2017.09 ?

http://patchwork.ozlabs.org/patch/762328/

Fix up the commit message, detail a bit more how to trigger the issue and use errno.h return value (-ENODEV ?) in the new code . Send V2 and I'll pick it.

Regards Patrick

...

-----Original Message----- From: Patrick DELAUNAY Sent: lundi 15 mai 2017 11:16 To: u-boot@lists.denx.de Cc: Yann GAUTIER yann.gautier@st.com; Patrick DELAUNAY patrick.delaunay@st.com; Michal Simek michal.simek@xilinx.com; Simon Glass sjg@chromium.org; Peng Fan van.freenix@gmail.com; Hans de Goede hdegoede@redhat.com Subject: [PATCH] cmd: usb: check usb_stor_curr_dev in usb dev with no parameter Importance: High

If 'usb start' does not find devices, the command 'usb dev' without parameter will lead to error= USB device -1: data abort A check on usb_stor_curr_dev avoid the issue

Signed-off-by: Patrick Delaunay patrick.delaunay@st.com

cmd/usb.c | 4 ++++ 1 file changed, 4 insertions(+)

diff --git a/cmd/usb.c b/cmd/usb.c index 4fa456e..bc90553 100644 --- a/cmd/usb.c +++ b/cmd/usb.c @@ -803,6 +803,10 @@ static int do_usb(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) printf("... is now current device\n"); return 0; } else {

• 	if (usb_stor_curr_dev < 0) {
  

• 		printf("no current device selected\n");
  

• 		return 1;
  

• 	}
  printf("\nUSB device %d: ", usb_stor_curr_dev);
  stor_dev = blk_get_devnum_by_type(IF_TYPE_USB,
  				  usb_stor_curr_dev);
  

-- 1.9.1