static int hermes_init(struct hermes *hw)
{
u16 reg;
int err = 0;
int k;
hw->inten = 0x0;
hermes_write_regn(hw, INTEN, 0);
hermes_write_regn(hw, EVACK, 0xffff);
k = CMD_BUSY_TIMEOUT;
reg = hermes_read_regn(hw, CMD);
while (k && (reg & HERMES_CMD_BUSY)) {
if (reg == 0xffff)
return -ENODEV;
k--;
udelay(1);
reg = hermes_read_regn(hw, CMD);
}
reg = hermes_read_regn(hw, EVSTAT);
hermes_write_regn(hw, EVACK, reg);
err = hermes_doicmd_wait(hw, HERMES_CMD_INIT, 0, 0, 0, NULL);
return err;
}
int hermes_init(hermes_t *hw)
{
u16 reg;
int err = 0;
int k;
/* We don't want to be interrupted while resetting the chipset */
hw->inten = 0x0;
hermes_write_regn(hw, INTEN, 0);
hermes_write_regn(hw, EVACK, 0xffff);
/* Normally it's a "can't happen" for the command register to
be busy when we go to issue a command because we are
serializing all commands. However we want to have some
chance of resetting the card even if it gets into a stupid
state, so we actually wait to see if the command register
will unbusy itself here. */
k = CMD_BUSY_TIMEOUT;
reg = hermes_read_regn(hw, CMD);
while (k && (reg & HERMES_CMD_BUSY)) {
if (reg == 0xffff) /* Special case - the card has probably been removed,
so don't wait for the timeout */
return -ENODEV;
k--;
udelay(1);
reg = hermes_read_regn(hw, CMD);
}
/* No need to explicitly handle the timeout - if we've timed
out hermes_issue_cmd() will probably return -EBUSY below */
/* According to the documentation, EVSTAT may contain
obsolete event occurrence information. We have to acknowledge
it by writing EVACK. */
reg = hermes_read_regn(hw, EVSTAT);
hermes_write_regn(hw, EVACK, reg);
/* We don't use hermes_docmd_wait here, because the reset wipes
the magic constant in SWSUPPORT0 away, and it gets confused */
err = hermes_doicmd_wait(hw, HERMES_CMD_INIT, 0, 0, 0, NULL);
return err;
}
