Lines Matching defs:phase
89 sym_print_addr(cmd, "illegal scsi phase (4/5).\n");
92 sym_print_addr(cmd, "ODD transfer in DATA OUT phase.\n");
95 sym_print_addr(cmd, "ODD transfer in DATA IN phase.\n");
935 printf("%s: handling phase mismatch from SCRIPTS.\n",
1811 * and/or hardware phase mismatch, since only such chips
1832 * If phase mismatch handled by scripts (895A/896/1010),
2283 * without the device having changed phase.
2287 * - A phase mismatch occurs before the MOV finished
2288 * and phase errors are to be handled by the C code.
2292 * - A phase mismatch occurs before the MOV finished and
2293 * phase errors are to be handled by SCRIPTS.
2294 * The chip will load the DSP with the phase mismatch
2305 int phase = cmd & 7;
2331 if ((cmd & 0xc0) || !(phase & 1) || !(sbcl & 0x8))
2343 np->msgout[0] = (phase == 7) ? M_PARITY : M_ID_ERROR;
2346 * If the old phase was DATA IN phase, we have to deal with
2352 if (phase == 1 || phase == 5) {
2359 /* No phase mismatch occurred */
2365 else if (phase == 7) /* We definitely cannot handle parity errors */
2366 #if 1 /* in message-in phase due to the relection */
2381 * chip exception handler for phase errors.
2418 * INPUT phase since the chip flushes everything before
2420 * For DATA IN phase, we will check for the SWIDE later.
2510 printf ("%s: SCSI phase error fixup: CCB already dequeued.\n",
2539 * If dt data phase, the MOVE instruction hasn't bit 4 of
2540 * the phase.
2551 * if old phase not dataphase, leave here.
2555 "phase change %x-%x %d@%08x resid=%d.\n",
2596 * fillin the phase mismatch context
2655 * Unexpected phase changes that occurs when the current phase
2656 * is not a DATA IN or DATA OUT phase are due to error conditions.
2668 * The code below does not care of the new phase and so
2670 * If the interrupted phase is COMMAND phase, we restart at
2686 case 2: /* COMMAND phase */
2690 case 3: /* STATUS phase */
2694 case 6: /* MSG OUT phase */
2721 case 7: /* MSG IN phase */
2756 * When an parity error is detected in input phase
2757 * and the device switches to msg-in phase inside a
2884 * A SCSI parity error (PAR) may be combined with a phase
3518 * to BUS FREE phase as we expected.
3771 * script address we want to return for the next data phase.
3808 * ofs < 0 there, and for the next data phase, we
3932 * If the target doesn't fetch the answer (no message out phase),
4409 * The device didn't go to MSG OUT phase after having
4414 "No MSG OUT phase after selection with ATN\n");
4417 * The device didn't switch to MSG IN phase after
4422 "No MSG IN phase after reselection\n");
4532 * The device switched to an illegal phase (4/5).