drivers/kitprog: workaround KitProg firmware bug of missing ZLP

KitProg firmware does not send a zero length packet at the end of the bulk-in
transmission of a length divisible by a bulk packet size. This is inconsistent
with the USB specification and results in jtag_libusb_bulk_read()
waits forever when a transmission of specific size is received.

Limit bulk read size to expected number of bytes for problematic tranfer sizes.
Use 1 second timeout as the last resort.

Change-Id: Ice80306424afd76e9fbc6851911ffd5109c84501
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/4426
Tested-by: jenkins
Reviewed-by: Bohdan Tymkiv <bhdt@cypress.com>
riscv-compliance-dev
Tomas Vanek 2018-02-21 22:19:58 +01:00
parent a28dea0fe4
commit 7829bb701f
1 changed files with 13 additions and 3 deletions

View File

@ -741,12 +741,22 @@ static int kitprog_swd_run_queue(void)
break;
}
/* We use the maximum buffer size here because the KitProg sometimes
* doesn't like bulk reads of fewer than 62 bytes. (?!?!)
/* KitProg firmware does not send a zero length packet
* after the bulk-in transmission of a length divisible by bulk packet
* size (64 bytes) as required by the USB specification.
* Therefore libusb would wait for continuation of transmission.
* Workaround: Limit bulk read size to expected number of bytes
* for problematic tranfer sizes. Otherwise use the maximum buffer
* size here because the KitProg sometimes doesn't like bulk reads
* of fewer than 62 bytes. (?!?!)
*/
size_t read_count_workaround = SWD_MAX_BUFFER_LENGTH;
if (read_count % 64 == 0)
read_count_workaround = read_count;
ret = jtag_libusb_bulk_read(kitprog_handle->usb_handle,
BULK_EP_IN | LIBUSB_ENDPOINT_IN, (char *)buffer,
SWD_MAX_BUFFER_LENGTH, 0);
read_count_workaround, 1000);
if (ret > 0) {
/* Handle garbage data by offsetting the initial read index */
if ((unsigned int)ret > read_count)