On Thu, Feb 25, 2010 at 12:16:24AM -0800, David Brownell wrote:
> On Wednesday 24 February 2010, Mariano Alvira wrote:
> > + int channel;
>
> OK, I merged this one ... wrote a patch comment of my own, based
> on the e-mail discussions, since you didn't inlcude one.
Great, thanks!
>
>
> By the way ... I kind of suspect this will cause build breakage
> on systems configured to use FTDI's library, since INTERFACE_A
> and suchlike are not defined.
I don't think it will break much. If you are configured to use FTD2XX
it uses a different init function:
#if BUILD_FT2232_FTD2XX == 1
retval = ft2232_init_ftd2xx(ft2232_vid[i],
ft2232_pid[i],
more, &try_more);
#elif BUILD_FT2232_LIBFTDI == 1
retval = ft2232_init_libftdi(ft2232_vid[i],
ft2232_pid[i],
more, &try_more,
cur_layout->channel);
#endif
And to me it looks like that driver interprets the device string to
figure out what to open:
if (status != FT_OK) {
/* under Win32, the FTD2XX driver appends an "A" to
the end
* of the description, if we tried by the desc, then
* try by the alternate "A" description. */
if (openex_string == ft2232_device_desc) {
/* Try the alternate method. */
openex_string = ft2232_device_desc_A;
status = FT_OpenEx(openex_string,
openex_flags, &ftdih);
if (status == FT_OK) {
/* yea, the "alternate" method worked!
*/
} else {
/* drat, give the user a meaningfull
message.
* telling the use we tried *BOTH*
methods. */
LOG_WARNING("Unable to open FTDI
Device tried: '%s' and '%s'\n",
ft2232_device_desc,
ft2232_device_desc_A);
}
}
I suspect the "A" that gets appended indicates "INTERFACE_A", but I
have no way of knowing for sure so I left all that alone...
-Mar.
_______________________________________________
Openocd-development mailing list
Openocd-development@lists.berlios.de
https://lists.berlios.de/mailman/listinfo/openocd-development
