On Sun, 25 Jan 2015 10:33:41 +0300 Boris Baykov dev@borisbaykov.com wrote:
Added MPSSE reset according to FTDI Application Notes AN114.
Signed-off-by: Boris Baykov dev@borisbaykov.com
diff -U 5 -N ./flashrom/ft2232_spi.c ./flashrom.ftdi_mpsse_init/ft2232_spi.c --- ./flashrom/ft2232_spi.c 2015-01-25 10:16:05.372174400 +0300 +++ ./flashrom.ftdi_mpsse_init/ft2232_spi.c 2015-01-25 10:17:50.000000000 +0300 @@ -75,10 +75,11 @@ {0}, };
#define DEFAULT_DIVISOR 2
+#define BITMODE_BITBANG_RESET 0 #define BITMODE_BITBANG_NORMAL 1 #define BITMODE_BITBANG_SPI 2
/* The variables cs_bits and pindir store the values for the "set data bits low byte" MPSSE command that
- sets the initial state and the direction of the I/O pins. The pin offsets are as follows:
@@ -352,10 +353,14 @@
if (ftdi_write_data_set_chunksize(ftdic, 256)) { msg_perr("Unable to set chunk size (%s).\n", ftdi_get_error_string(ftdic)); }
if (ftdi_set_bitmode(ftdic, 0x00, BITMODE_BITBANG_RESET) < 0) {
msg_perr("Unable to reset MPSSE controller (%s).\n", ftdi_get_error_string(ftdic));
}
if (ftdi_set_bitmode(ftdic, 0x00, BITMODE_BITBANG_SPI) < 0) { msg_perr("Unable to set bitmode to SPI (%s).\n", ftdi_get_error_string(ftdic)); } if (clock_5x) {
Yes, that's probably a good idea, thank you. For *H devices we should also disable three-phase data clocking (0x8D) and adaptive clocking (0x97). It's probably overkill but it shouldn't hurt either I guess (and it is recommended by the docs). That said, I don't want to commit that before the imminent 0.9.8 release. I will commit it (a refined version with the additional disables) shortly after release.