freebsd-dev/sys/dev/esp/ncr53c9x.c

3041 lines
80 KiB
C
Raw Normal View History

/*-
* Copyright (c) 2004 Scott Long
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
*/
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
/* $NetBSD: ncr53c9x.c,v 1.114 2005/02/27 00:27:02 perry Exp $ */
/*-
* Copyright (c) 1998, 2002 The NetBSD Foundation, Inc.
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by Charles M. Hannum.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the NetBSD
* Foundation, Inc. and its contributors.
* 4. Neither the name of The NetBSD Foundation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
* TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
/*-
* Copyright (c) 1994 Peter Galbavy
* Copyright (c) 1995 Paul Kranenburg
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by Peter Galbavy
* 4. The name of the author may not be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
* WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT,
* INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
* ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
/*
* Based on aic6360 by Jarle Greipsland
*
* Acknowledgements: Many of the algorithms used in this driver are
* inspired by the work of Julian Elischer (julian@FreeBSD.org) and
* Charles Hannum (mycroft@duality.gnu.ai.mit.edu). Thanks a million!
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/bus.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <sys/resource.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/queue.h>
#include <sys/time.h>
#include <sys/callout.h>
#include <cam/cam.h>
#include <cam/cam_ccb.h>
#include <cam/cam_debug.h>
#include <cam/cam_sim.h>
#include <cam/cam_xpt_sim.h>
#include <cam/scsi/scsi_all.h>
#include <cam/scsi/scsi_message.h>
#include <dev/esp/ncr53c9xreg.h>
#include <dev/esp/ncr53c9xvar.h>
int ncr53c9x_debug = NCR_SHOWMISC /*|NCR_SHOWPHASE|NCR_SHOWTRAC|NCR_SHOWCMDS*/;
#ifdef DEBUG
int ncr53c9x_notag = 0;
#endif
static void ncr53c9x_select(struct ncr53c9x_softc *, struct ncr53c9x_ecb *);
static int ncr53c9x_reselect(struct ncr53c9x_softc *, int, int, int);
static void ncr53c9x_scsi_reset(struct ncr53c9x_softc *);
static void ncr53c9x_poll(struct cam_sim *);
static void ncr53c9x_sched(struct ncr53c9x_softc *);
static void ncr53c9x_done(struct ncr53c9x_softc *, struct ncr53c9x_ecb *);
static void ncr53c9x_msgin(struct ncr53c9x_softc *);
static void ncr53c9x_msgout(struct ncr53c9x_softc *);
static void ncr53c9x_timeout(void *arg);
static void ncr53c9x_watch(void *arg);
static void ncr53c9x_abort(struct ncr53c9x_softc *, struct ncr53c9x_ecb *);
static void ncr53c9x_dequeue(struct ncr53c9x_softc *,
struct ncr53c9x_ecb *);
static void ncr53c9x_sense(struct ncr53c9x_softc *, struct ncr53c9x_ecb *);
static void ncr53c9x_free_ecb(struct ncr53c9x_softc *,
struct ncr53c9x_ecb *);
static void ncr53c9x_wrfifo(struct ncr53c9x_softc *, u_char *, int);
static int ncr53c9x_rdfifo(struct ncr53c9x_softc *, int);
static struct ncr53c9x_ecb *ncr53c9x_get_ecb(struct ncr53c9x_softc *);
static struct ncr53c9x_linfo *ncr53c9x_lunsearch(struct ncr53c9x_tinfo *,
int64_t lun);
static __inline void ncr53c9x_readregs(struct ncr53c9x_softc *);
static __inline int ncr53c9x_stp2cpb(struct ncr53c9x_softc *, int);
static __inline void ncr53c9x_setsync(struct ncr53c9x_softc *,
struct ncr53c9x_tinfo *);
#define NCR_RDFIFO_START 0
#define NCR_RDFIFO_CONTINUE 1
#define NCR_SET_COUNT(sc, size) do { \
NCR_WRITE_REG((sc), NCR_TCL, (size)); \
NCR_WRITE_REG((sc), NCR_TCM, (size) >> 8); \
if ((sc->sc_cfg2 & NCRCFG2_FE) || \
(sc->sc_rev == NCR_VARIANT_FAS366)) { \
NCR_WRITE_REG((sc), NCR_TCH, (size) >> 16); \
} \
if (sc->sc_rev == NCR_VARIANT_FAS366) { \
NCR_WRITE_REG(sc, NCR_RCH, 0); \
} \
} while (0)
#ifndef mstohz
#define mstohz(ms) \
(((ms) < 0x20000) ? \
((ms +0u) / 1000u) * hz : \
((ms +0u) * hz) /1000u)
#endif
/*
* Names for the NCR53c9x variants, corresponding to the variant tags
* in ncr53c9xvar.h.
*/
static const char *ncr53c9x_variant_names[] = {
"ESP100",
"ESP100A",
"ESP200",
"NCR53C94",
"NCR53C96",
"ESP406",
"FAS408",
"FAS216",
"AM53C974",
"FAS366/HME",
"NCR53C90 (86C01)",
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
"FAS100A",
"FAS236",
};
/*
* Search linked list for LUN info by LUN id.
*/
static struct ncr53c9x_linfo *
ncr53c9x_lunsearch(struct ncr53c9x_tinfo *ti, int64_t lun)
{
struct ncr53c9x_linfo *li;
LIST_FOREACH(li, &ti->luns, link)
if (li->lun == lun)
return (li);
return (NULL);
}
/*
* Attach this instance, and then all the sub-devices.
*/
int
ncr53c9x_attach(struct ncr53c9x_softc *sc)
{
struct cam_devq *devq;
struct cam_sim *sim;
struct cam_path *path;
struct ncr53c9x_ecb *ecb;
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
int error, i;
mtx_init(&sc->sc_lock, "ncr", "ncr53c9x lock", MTX_DEF);
/*
* Note, the front-end has set us up to print the chip variation.
*/
if (sc->sc_rev >= NCR_VARIANT_MAX) {
device_printf(sc->sc_dev, "unknown variant %d, devices not "
"attached\n", sc->sc_rev);
return (EINVAL);
}
device_printf(sc->sc_dev, "%s, %dMHz, SCSI ID %d\n",
ncr53c9x_variant_names[sc->sc_rev], sc->sc_freq, sc->sc_id);
sc->sc_ntarg = (sc->sc_rev == NCR_VARIANT_FAS366) ? 16 : 8;
/*
* Allocate SCSI message buffers.
* Front-ends can override allocation to avoid alignment
* handling in the DMA engines. Note that that ncr53c9x_msgout()
* can request a 1 byte DMA transfer.
*/
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
if (sc->sc_omess == NULL) {
sc->sc_omess_self = 1;
sc->sc_omess = malloc(NCR_MAX_MSG_LEN, M_DEVBUF, M_NOWAIT);
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
if (sc->sc_omess == NULL) {
device_printf(sc->sc_dev,
"cannot allocate MSGOUT buffer\n");
return (ENOMEM);
}
} else
sc->sc_omess_self = 0;
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
if (sc->sc_imess == NULL) {
sc->sc_imess_self = 1;
sc->sc_imess = malloc(NCR_MAX_MSG_LEN + 1, M_DEVBUF, M_NOWAIT);
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
if (sc->sc_imess == NULL) {
device_printf(sc->sc_dev,
"cannot allocate MSGIN buffer\n");
error = ENOMEM;
goto fail_omess;
}
} else
sc->sc_imess_self = 0;
sc->sc_tinfo = malloc(sc->sc_ntarg * sizeof(sc->sc_tinfo[0]),
M_DEVBUF, M_NOWAIT | M_ZERO);
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
if (sc->sc_tinfo == NULL) {
device_printf(sc->sc_dev,
"cannot allocate target info buffer\n");
error = ENOMEM;
goto fail_imess;
}
callout_init(&sc->sc_watchdog, 0);
/*
* Treat NCR53C90 with the 86C01 DMA chip exactly as ESP100
* from now on.
*/
if (sc->sc_rev == NCR_VARIANT_NCR53C90_86C01)
sc->sc_rev = NCR_VARIANT_ESP100;
sc->sc_ccf = FREQTOCCF(sc->sc_freq);
/* The value *must not* be == 1. Make it 2 */
if (sc->sc_ccf == 1)
sc->sc_ccf = 2;
/*
* The recommended timeout is 250ms. This register is loaded
* with a value calculated as follows, from the docs:
*
* (timout period) x (CLK frequency)
* reg = -------------------------------------
* 8192 x (Clock Conversion Factor)
*
* Since CCF has a linear relation to CLK, this generally computes
* to the constant of 153.
*/
sc->sc_timeout = ((250 * 1000) * sc->sc_freq) / (8192 * sc->sc_ccf);
/* CCF register only has 3 bits; 0 is actually 8 */
sc->sc_ccf &= 7;
/*
* Register with CAM
*/
devq = cam_simq_alloc(sc->sc_ntarg);
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
if (devq == NULL) {
device_printf(sc->sc_dev, "cannot allocate device queue\n");
error = ENOMEM;
goto fail_tinfo;
}
sim = cam_sim_alloc(ncr53c9x_action, ncr53c9x_poll, "esp", sc,
device_get_unit(sc->sc_dev), 1,
NCR_TAG_DEPTH, devq);
if (sim == NULL) {
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
device_printf(sc->sc_dev, "cannot allocate SIM entry\n");
error = ENOMEM;
goto fail_devq;
}
if (xpt_bus_register(sim, 0) != CAM_SUCCESS) {
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
device_printf(sc->sc_dev, "cannot register bus\n");
error = EIO;
goto fail_sim;
}
if (xpt_create_path(&path, NULL, cam_sim_path(sim),
CAM_TARGET_WILDCARD, CAM_LUN_WILDCARD)
!= CAM_REQ_CMP) {
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
device_printf(sc->sc_dev, "cannot create path\n");
error = EIO;
goto fail_bus;
}
sc->sc_sim = sim;
sc->sc_path = path;
/* Reset state & bus */
#if 0
sc->sc_cfflags = sc->sc_dev.dv_cfdata->cf_flags;
#endif
sc->sc_state = 0;
ncr53c9x_init(sc, 1);
TAILQ_INIT(&sc->free_list);
if ((sc->ecb_array = malloc(sizeof(struct ncr53c9x_ecb) * NCR_TAG_DEPTH,
M_DEVBUF, M_NOWAIT|M_ZERO)) == NULL) {
device_printf(sc->sc_dev, "cannot allocate ECB array\n");
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
error = ENOMEM;
goto fail_path;
}
for (i = 0; i < NCR_TAG_DEPTH; i++) {
ecb = &sc->ecb_array[i];
ecb->sc = sc;
ecb->tag_id = i;
TAILQ_INSERT_HEAD(&sc->free_list, ecb, free_links);
}
callout_reset(&sc->sc_watchdog, 60*hz, ncr53c9x_watch, sc);
return (0);
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
fail_path:
xpt_free_path(path);
fail_bus:
xpt_bus_deregister(cam_sim_path(sim));
fail_sim:
cam_sim_free(sim, TRUE);
fail_devq:
cam_simq_free(devq);
fail_tinfo:
free(sc->sc_tinfo, M_DEVBUF);
fail_imess:
if (sc->sc_imess_self)
free(sc->sc_imess, M_DEVBUF);
fail_omess:
if (sc->sc_omess_self)
free(sc->sc_omess, M_DEVBUF);
return (error);
}
int
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
ncr53c9x_detach(struct ncr53c9x_softc *sc)
{
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
callout_drain(&sc->sc_watchdog);
mtx_lock(&sc->sc_lock);
ncr53c9x_init(sc, 1);
mtx_unlock(&sc->sc_lock);
xpt_free_path(sc->sc_path);
xpt_bus_deregister(cam_sim_path(sc->sc_sim));
cam_sim_free(sc->sc_sim, TRUE);
free(sc->ecb_array, M_DEVBUF);
free(sc->sc_tinfo, M_DEVBUF);
if (sc->sc_imess_self)
free(sc->sc_imess, M_DEVBUF);
if (sc->sc_omess_self)
free(sc->sc_omess, M_DEVBUF);
mtx_destroy(&sc->sc_lock);
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
return (0);
}
/*
* This is the generic ncr53c9x reset function. It does not reset the SCSI bus,
* only this controller, but kills any on-going commands, and also stops
* and resets the DMA.
*
* After reset, registers are loaded with the defaults from the attach
* routine above.
*/
void
ncr53c9x_reset(struct ncr53c9x_softc *sc)
{
/* reset DMA first */
NCRDMA_RESET(sc);
/* reset SCSI chip */
NCRCMD(sc, NCRCMD_RSTCHIP);
NCRCMD(sc, NCRCMD_NOP);
DELAY(500);
/* do these backwards, and fall through */
switch (sc->sc_rev) {
case NCR_VARIANT_ESP406:
case NCR_VARIANT_FAS408:
NCR_WRITE_REG(sc, NCR_CFG5, sc->sc_cfg5 | NCRCFG5_SINT);
NCR_WRITE_REG(sc, NCR_CFG4, sc->sc_cfg4);
case NCR_VARIANT_AM53C974:
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
case NCR_VARIANT_FAS100A:
case NCR_VARIANT_FAS216:
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
case NCR_VARIANT_FAS236:
case NCR_VARIANT_NCR53C94:
case NCR_VARIANT_NCR53C96:
case NCR_VARIANT_ESP200:
sc->sc_features |= NCR_F_HASCFG3;
NCR_WRITE_REG(sc, NCR_CFG3, sc->sc_cfg3);
case NCR_VARIANT_ESP100A:
sc->sc_features |= NCR_F_SELATN3;
NCR_WRITE_REG(sc, NCR_CFG2, sc->sc_cfg2);
case NCR_VARIANT_ESP100:
NCR_WRITE_REG(sc, NCR_CFG1, sc->sc_cfg1);
NCR_WRITE_REG(sc, NCR_CCF, sc->sc_ccf);
NCR_WRITE_REG(sc, NCR_SYNCOFF, 0);
NCR_WRITE_REG(sc, NCR_TIMEOUT, sc->sc_timeout);
break;
case NCR_VARIANT_FAS366:
sc->sc_features |=
NCR_F_HASCFG3 | NCR_F_FASTSCSI | NCR_F_SELATN3;
sc->sc_cfg3 = NCRFASCFG3_FASTCLK | NCRFASCFG3_OBAUTO;
sc->sc_cfg3_fscsi = NCRFASCFG3_FASTSCSI;
NCR_WRITE_REG(sc, NCR_CFG3, sc->sc_cfg3);
sc->sc_cfg2 = 0; /* NCRCFG2_HMEFE| NCRCFG2_HME32 */
NCR_WRITE_REG(sc, NCR_CFG2, sc->sc_cfg2);
NCR_WRITE_REG(sc, NCR_CFG1, sc->sc_cfg1);
NCR_WRITE_REG(sc, NCR_CCF, sc->sc_ccf);
NCR_WRITE_REG(sc, NCR_SYNCOFF, 0);
NCR_WRITE_REG(sc, NCR_TIMEOUT, sc->sc_timeout);
break;
default:
device_printf(sc->sc_dev, "unknown revision code, "
"assuming ESP100\n");
NCR_WRITE_REG(sc, NCR_CFG1, sc->sc_cfg1);
NCR_WRITE_REG(sc, NCR_CCF, sc->sc_ccf);
NCR_WRITE_REG(sc, NCR_SYNCOFF, 0);
NCR_WRITE_REG(sc, NCR_TIMEOUT, sc->sc_timeout);
}
if (sc->sc_rev == NCR_VARIANT_AM53C974)
NCR_WRITE_REG(sc, NCR_AMDCFG4, sc->sc_cfg4);
#if 0
device_printf(sc->sc_dev, "ncr53c9x_reset: revision %d\n",
sc->sc_rev);
device_printf(sc->sc_dev, "ncr53c9x_reset: cfg1 0x%x, cfg2 0x%x, "
"cfg3 0x%x, ccf 0x%x, timeout 0x%x\n",
sc->sc_cfg1, sc->sc_cfg2, sc->sc_cfg3, sc->sc_ccf, sc->sc_timeout);
#endif
}
/*
* Reset the SCSI bus, but not the chip
*/
static void
ncr53c9x_scsi_reset(struct ncr53c9x_softc *sc)
{
(*sc->sc_glue->gl_dma_stop)(sc);
NCR_MISC(("%s: resetting SCSI bus\n", device_get_nameunit(sc->sc_dev)));
NCRCMD(sc, NCRCMD_RSTSCSI);
DELAY(250000); /* Give the bus a fighting chance to settle */
}
/*
* Initialize ncr53c9x state machine
*/
void
ncr53c9x_init(struct ncr53c9x_softc *sc, int doreset)
{
struct ncr53c9x_ecb *ecb;
struct ncr53c9x_linfo *li;
int i, r;
NCR_MISC(("[NCR_INIT(%d) %d] ", doreset, sc->sc_state));
if (sc->sc_state == 0) {
/* First time through; initialize. */
TAILQ_INIT(&sc->ready_list);
sc->sc_nexus = NULL;
memset(sc->sc_tinfo, 0, sizeof(sc->sc_tinfo));
for (r = 0; r < sc->sc_ntarg; r++) {
LIST_INIT(&sc->sc_tinfo[r].luns);
}
} else {
/* Cancel any active commands. */
sc->sc_state = NCR_CLEANING;
sc->sc_msgify = 0;
if ((ecb = sc->sc_nexus) != NULL) {
ecb->ccb->ccb_h.status = CAM_CMD_TIMEOUT;
ncr53c9x_done(sc, ecb);
}
/* Cancel outstanding disconnected commands on each LUN */
for (r = 0; r < sc->sc_ntarg; r++) {
LIST_FOREACH(li, &sc->sc_tinfo[r].luns, link) {
if ((ecb = li->untagged) != NULL) {
li->untagged = NULL;
/*
* XXXXXXX
*
* Should we terminate a command
* that never reached the disk?
*/
li->busy = 0;
ecb->ccb->ccb_h.status =
CAM_CMD_TIMEOUT;
ncr53c9x_done(sc, ecb);
}
for (i = 0; i < 256; i++)
if ((ecb = li->queued[i])) {
li->queued[i] = NULL;
ecb->ccb->ccb_h.status =
CAM_CMD_TIMEOUT;
ncr53c9x_done(sc, ecb);
}
li->used = 0;
}
}
}
/*
* reset the chip to a known state
*/
ncr53c9x_reset(sc);
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
sc->sc_flags = 0;
sc->sc_msgpriq = sc->sc_msgout = sc->sc_msgoutq = 0;
sc->sc_phase = sc->sc_prevphase = INVALID_PHASE;
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
for (r = 0; r < sc->sc_ntarg; r++) {
struct ncr53c9x_tinfo *ti = &sc->sc_tinfo[r];
/* XXX - config flags per target: low bits: no reselect; high bits: no synch */
ti->flags = ((sc->sc_minsync && !(sc->sc_cfflags & (1<<((r&7)+8))))
? 0 : T_SYNCHOFF) |
((sc->sc_cfflags & (1<<(r&7))) ? T_RSELECTOFF : 0);
#ifdef DEBUG
if (ncr53c9x_notag)
ti->flags &= ~T_TAG;
#endif
ti->period = sc->sc_minsync;
ti->offset = 0;
ti->cfg3 = 0;
}
if (doreset) {
sc->sc_state = NCR_SBR;
NCRCMD(sc, NCRCMD_RSTSCSI);
} else {
sc->sc_state = NCR_IDLE;
ncr53c9x_sched(sc);
}
}
/*
* Read the NCR registers, and save their contents for later use.
* NCR_STAT, NCR_STEP & NCR_INTR are mostly zeroed out when reading
* NCR_INTR - so make sure it is the last read.
*
* I think that (from reading the docs) most bits in these registers
* only make sense when he DMA CSR has an interrupt showing. Call only
* if an interrupt is pending.
*/
static __inline void
ncr53c9x_readregs(struct ncr53c9x_softc *sc)
{
sc->sc_espstat = NCR_READ_REG(sc, NCR_STAT);
/* Only the stepo bits are of interest */
sc->sc_espstep = NCR_READ_REG(sc, NCR_STEP) & NCRSTEP_MASK;
if (sc->sc_rev == NCR_VARIANT_FAS366)
sc->sc_espstat2 = NCR_READ_REG(sc, NCR_STAT2);
sc->sc_espintr = NCR_READ_REG(sc, NCR_INTR);
if (sc->sc_glue->gl_clear_latched_intr != NULL)
(*sc->sc_glue->gl_clear_latched_intr)(sc);
/*
* Determine the SCSI bus phase, return either a real SCSI bus phase
* or some pseudo phase we use to detect certain exceptions.
*/
sc->sc_phase = (sc->sc_espintr & NCRINTR_DIS) ?
/* Disconnected */ BUSFREE_PHASE : sc->sc_espstat & NCRSTAT_PHASE;
NCR_INTS(("regs[intr=%02x,stat=%02x,step=%02x,stat2=%02x] ",
sc->sc_espintr, sc->sc_espstat, sc->sc_espstep, sc->sc_espstat2));
}
/*
* Convert Synchronous Transfer Period to chip register Clock Per Byte value.
*/
static __inline int
ncr53c9x_stp2cpb(struct ncr53c9x_softc *sc, int period)
{
int v;
v = (sc->sc_freq * period) / 250;
if (ncr53c9x_cpb2stp(sc, v) < period)
/* Correct round-down error */
v++;
return (v);
}
static __inline void
ncr53c9x_setsync(struct ncr53c9x_softc *sc, struct ncr53c9x_tinfo *ti)
{
u_char syncoff, synctp;
u_char cfg3 = sc->sc_cfg3 | ti->cfg3;
if (ti->flags & T_SYNCMODE) {
syncoff = ti->offset;
synctp = ncr53c9x_stp2cpb(sc, ti->period);
if (sc->sc_features & NCR_F_FASTSCSI) {
/*
* If the period is 200ns or less (ti->period <= 50),
* put the chip in Fast SCSI mode.
*/
if (ti->period <= 50)
/*
* There are (at least) 4 variations of the
* configuration 3 register. The drive attach
* routine sets the appropriate bit to put the
* chip into Fast SCSI mode so that it doesn't
* have to be figured out here each time.
*/
cfg3 |= sc->sc_cfg3_fscsi;
}
/*
* Am53c974 requires different SYNCTP values when the
* FSCSI bit is off.
*/
if (sc->sc_rev == NCR_VARIANT_AM53C974 &&
(cfg3 & NCRAMDCFG3_FSCSI) == 0)
synctp--;
} else {
syncoff = 0;
synctp = 0;
}
if (sc->sc_features & NCR_F_HASCFG3)
NCR_WRITE_REG(sc, NCR_CFG3, cfg3);
NCR_WRITE_REG(sc, NCR_SYNCOFF, syncoff);
NCR_WRITE_REG(sc, NCR_SYNCTP, synctp);
}
/*
* Send a command to a target, set the driver state to NCR_SELECTING
* and let the caller take care of the rest.
*
* Keeping this as a function allows me to say that this may be done
* by DMA instead of programmed I/O soon.
*/
static void
ncr53c9x_select(struct ncr53c9x_softc *sc, struct ncr53c9x_ecb *ecb)
{
int target = ecb->ccb->ccb_h.target_id;
int lun = ecb->ccb->ccb_h.target_lun;
struct ncr53c9x_tinfo *ti;
int tiflags;
u_char *cmd;
int clen;
int selatn3, selatns;
size_t dmasize;
NCR_TRACE(("[ncr53c9x_select(t%d,l%d,cmd:%x,tag:%x,%x)] ",
target, lun, ecb->cmd.cmd.opcode, ecb->tag[0], ecb->tag[1]));
ti = &sc->sc_tinfo[target];
tiflags = ti->flags;
sc->sc_state = NCR_SELECTING;
/*
* Schedule the timeout now, the first time we will go away
* expecting to come back due to an interrupt, because it is
* always possible that the interrupt may never happen.
*/
ecb->ccb->ccb_h.timeout_ch =
timeout(ncr53c9x_timeout, ecb, mstohz(ecb->timeout));
/*
* The docs say the target register is never reset, and I
* can't think of a better place to set it
*/
if (sc->sc_rev == NCR_VARIANT_FAS366) {
NCRCMD(sc, NCRCMD_FLUSH);
NCR_WRITE_REG(sc, NCR_SELID, target | NCR_BUSID_HME);
} else {
NCR_WRITE_REG(sc, NCR_SELID, target);
}
ncr53c9x_setsync(sc, ti);
if ((ecb->flags & ECB_SENSE) != 0) {
/*
* For REQUEST SENSE, we should not send an IDENTIFY or
* otherwise mangle the target. There should be no MESSAGE IN
* phase.
*/
if (sc->sc_features & NCR_F_DMASELECT) {
/* setup DMA transfer for command */
dmasize = clen = ecb->clen;
sc->sc_cmdlen = clen;
sc->sc_cmdp = (caddr_t)&ecb->cmd.cmd;
/* Program the SCSI counter */
NCR_SET_COUNT(sc, dmasize);
if (sc->sc_rev != NCR_VARIANT_FAS366)
NCRCMD(sc, NCRCMD_NOP|NCRCMD_DMA);
/* And get the targets attention */
NCRCMD(sc, NCRCMD_SELNATN | NCRCMD_DMA);
NCRDMA_SETUP(sc, &sc->sc_cmdp, &sc->sc_cmdlen, 0,
&dmasize);
NCRDMA_GO(sc);
} else {
ncr53c9x_wrfifo(sc, (u_char *)&ecb->cmd.cmd, ecb->clen);
NCRCMD(sc, NCRCMD_SELNATN);
}
return;
}
selatn3 = selatns = 0;
if (ecb->tag[0] != 0) {
if (sc->sc_features & NCR_F_SELATN3)
/* use SELATN3 to send tag messages */
selatn3 = 1;
else
/* We don't have SELATN3; use SELATNS to send tags */
selatns = 1;
}
if (ti->flags & T_NEGOTIATE) {
/* We have to use SELATNS to send sync/wide messages */
selatn3 = 0;
selatns = 1;
}
cmd = (u_char *)&ecb->cmd.cmd;
if (selatn3) {
/* We'll use tags with SELATN3 */
clen = ecb->clen + 3;
cmd -= 3;
cmd[0] = MSG_IDENTIFY(lun, 1); /* msg[0] */
cmd[1] = ecb->tag[0]; /* msg[1] */
cmd[2] = ecb->tag[1]; /* msg[2] */
} else {
/* We don't have tags, or will send messages with SELATNS */
clen = ecb->clen + 1;
cmd -= 1;
cmd[0] = MSG_IDENTIFY(lun, (tiflags & T_RSELECTOFF) == 0);
}
if ((sc->sc_features & NCR_F_DMASELECT) && !selatns) {
/* setup DMA transfer for command */
dmasize = clen;
sc->sc_cmdlen = clen;
sc->sc_cmdp = cmd;
/* Program the SCSI counter */
NCR_SET_COUNT(sc, dmasize);
/* load the count in */
/* if (sc->sc_rev != NCR_VARIANT_FAS366) */
NCRCMD(sc, NCRCMD_NOP|NCRCMD_DMA);
/* And get the targets attention */
if (selatn3) {
sc->sc_msgout = SEND_TAG;
sc->sc_flags |= NCR_ATN;
NCRCMD(sc, NCRCMD_SELATN3 | NCRCMD_DMA);
} else
NCRCMD(sc, NCRCMD_SELATN | NCRCMD_DMA);
NCRDMA_SETUP(sc, &sc->sc_cmdp, &sc->sc_cmdlen, 0, &dmasize);
NCRDMA_GO(sc);
return;
}
/*
* Who am I? This is where we tell the target that we are
* happy for it to disconnect etc.
*/
/* Now get the command into the FIFO */
ncr53c9x_wrfifo(sc, cmd, clen);
/* And get the targets attention */
if (selatns) {
NCR_MSGS(("SELATNS \n"));
/* Arbitrate, select and stop after IDENTIFY message */
NCRCMD(sc, NCRCMD_SELATNS);
} else if (selatn3) {
sc->sc_msgout = SEND_TAG;
sc->sc_flags |= NCR_ATN;
NCRCMD(sc, NCRCMD_SELATN3);
} else
NCRCMD(sc, NCRCMD_SELATN);
}
static void
ncr53c9x_free_ecb(struct ncr53c9x_softc *sc, struct ncr53c9x_ecb *ecb)
{
ecb->flags = 0;
TAILQ_INSERT_TAIL(&sc->free_list, ecb, free_links);
return;
}
static struct ncr53c9x_ecb *
ncr53c9x_get_ecb(struct ncr53c9x_softc *sc)
{
struct ncr53c9x_ecb *ecb;
ecb = TAILQ_FIRST(&sc->free_list);
if (ecb) {
if (ecb->flags != 0)
panic("ecb flags not cleared\n");
TAILQ_REMOVE(&sc->free_list, ecb, free_links);
ecb->flags = ECB_ALLOC;
bzero(&ecb->ccb, sizeof(struct ncr53c9x_ecb) -
offsetof(struct ncr53c9x_ecb, ccb));
}
return (ecb);
}
/*
* DRIVER FUNCTIONS CALLABLE FROM HIGHER LEVEL DRIVERS:
*/
/*
* Start a SCSI-command
* This function is called by the higher level SCSI-driver to queue/run
* SCSI-commands.
*/
void
ncr53c9x_action(struct cam_sim *sim, union ccb *ccb)
{
struct ncr53c9x_softc *sc;
struct ncr53c9x_ecb *ecb;
NCR_TRACE(("[ncr53c9x_action %d]", ccb->ccb_h.func_code));
sc = cam_sim_softc(sim);
mtx_lock(&sc->sc_lock);
switch (ccb->ccb_h.func_code) {
case XPT_RESET_BUS:
ncr53c9x_scsi_reset(sc);
ccb->ccb_h.status = CAM_REQ_CMP;
mtx_unlock(&sc->sc_lock);
xpt_done(ccb);
return;
case XPT_CALC_GEOMETRY:
mtx_unlock(&sc->sc_lock);
cam_calc_geometry(&ccb->ccg, sc->sc_extended_geom);
xpt_done(ccb);
return;
case XPT_PATH_INQ:
{
struct ccb_pathinq *cpi = &ccb->cpi;
cpi->version_num = 1;
cpi->hba_inquiry = PI_SDTR_ABLE|PI_TAG_ABLE;
cpi->hba_inquiry |=
(sc->sc_rev == NCR_VARIANT_FAS366) ? PI_WIDE_16 : 0;
cpi->target_sprt = 0;
cpi->hba_misc = 0;
cpi->hba_eng_cnt = 0;
cpi->max_target = sc->sc_ntarg - 1;
cpi->max_lun = 8;
cpi->initiator_id = sc->sc_id;
cpi->bus_id = 0;
cpi->base_transfer_speed = 3300;
strncpy(cpi->sim_vid, "FreeBSD", SIM_IDLEN);
strncpy(cpi->hba_vid, "Sun", HBA_IDLEN);
strncpy(cpi->dev_name, cam_sim_name(sim), DEV_IDLEN);
cpi->unit_number = cam_sim_unit(sim);
cpi->transport = XPORT_SPI;
cpi->transport_version = 2;
cpi->protocol = PROTO_SCSI;
cpi->protocol_version = SCSI_REV_2;
ccb->ccb_h.status = CAM_REQ_CMP;
mtx_unlock(&sc->sc_lock);
xpt_done(ccb);
return;
}
case XPT_GET_TRAN_SETTINGS:
{
struct ccb_trans_settings *cts = &ccb->cts;
struct ncr53c9x_tinfo *ti = &sc->sc_tinfo[ccb->ccb_h.target_id];
struct ccb_trans_settings_scsi *scsi =
&cts->proto_specific.scsi;
struct ccb_trans_settings_spi *spi =
&cts->xport_specific.spi;
cts->protocol = PROTO_SCSI;
cts->protocol_version = SCSI_REV_2;
cts->transport = XPORT_SPI;
cts->transport_version = 2;
if (cts->type == CTS_TYPE_CURRENT_SETTINGS) {
spi->sync_period = ti->period;
spi->sync_offset = ti->offset;
spi->bus_width = ti->width;
if ((ti->flags & T_TAG) != 0) {
spi->flags |= CTS_SPI_FLAGS_DISC_ENB;
scsi->flags |= CTS_SCSI_FLAGS_TAG_ENB;
} else {
spi->flags &= ~CTS_SPI_FLAGS_DISC_ENB;
scsi->flags &= ~CTS_SCSI_FLAGS_TAG_ENB;
}
} else {
spi->sync_period = sc->sc_maxsync;
spi->sync_offset = sc->sc_maxoffset;
spi->bus_width = sc->sc_maxwidth;
spi->flags |= CTS_SPI_FLAGS_DISC_ENB;
scsi->flags |= CTS_SCSI_FLAGS_TAG_ENB;
}
spi->valid =
CTS_SPI_VALID_BUS_WIDTH |
CTS_SPI_VALID_SYNC_RATE |
CTS_SPI_VALID_SYNC_OFFSET |
CTS_SPI_VALID_DISC;
scsi->valid = CTS_SCSI_VALID_TQ;
ccb->ccb_h.status = CAM_REQ_CMP;
mtx_unlock(&sc->sc_lock);
xpt_done(ccb);
return;
}
case XPT_ABORT:
printf("XPT_ABORT called\n");
ccb->ccb_h.status = CAM_FUNC_NOTAVAIL;
mtx_unlock(&sc->sc_lock);
xpt_done(ccb);
return;
case XPT_TERM_IO:
printf("XPT_TERM_IO called\n");
ccb->ccb_h.status = CAM_FUNC_NOTAVAIL;
mtx_unlock(&sc->sc_lock);
xpt_done(ccb);
return;
case XPT_RESET_DEV:
printf("XPT_RESET_DEV called\n");
case XPT_SCSI_IO:
{
struct ccb_scsiio *csio;
if (ccb->ccb_h.target_id < 0 ||
ccb->ccb_h.target_id >= sc->sc_ntarg) {
ccb->ccb_h.status = CAM_PATH_INVALID;
mtx_unlock(&sc->sc_lock);
xpt_done(ccb);
return;
}
/* Get an ECB to use. */
ecb = ncr53c9x_get_ecb(sc);
/*
* This should never happen as we track resources
* in the mid-layer.
*/
if (ecb == NULL) {
xpt_freeze_simq(sim, 1);
ccb->ccb_h.status = CAM_REQUEUE_REQ;
printf("unable to allocate ecb\n");
mtx_unlock(&sc->sc_lock);
xpt_done(ccb);
return;
}
/* Initialize ecb */
ecb->ccb = ccb;
ecb->timeout = ccb->ccb_h.timeout;
if (ccb->ccb_h.func_code == XPT_RESET_DEV) {
ecb->flags |= ECB_RESET;
ecb->clen = 0;
ecb->dleft = 0;
} else {
csio = &ccb->csio;
if ((ccb->ccb_h.flags & CAM_CDB_POINTER) != 0)
bcopy(csio->cdb_io.cdb_ptr, &ecb->cmd.cmd,
csio->cdb_len);
else
bcopy(csio->cdb_io.cdb_bytes, &ecb->cmd.cmd,
csio->cdb_len);
ecb->clen = csio->cdb_len;
ecb->daddr = csio->data_ptr;
ecb->dleft = csio->dxfer_len;
}
ecb->stat = 0;
TAILQ_INSERT_TAIL(&sc->ready_list, ecb, chain);
ecb->flags |= ECB_READY;
if (sc->sc_state == NCR_IDLE)
ncr53c9x_sched(sc);
break;
}
case XPT_SET_TRAN_SETTINGS:
{
struct ccb_trans_settings *cts = &ccb->cts;
int target = ccb->ccb_h.target_id;
struct ncr53c9x_tinfo *ti = &sc->sc_tinfo[target];
struct ccb_trans_settings_scsi *scsi =
&cts->proto_specific.scsi;
struct ccb_trans_settings_spi *spi =
&cts->xport_specific.spi;
if ((scsi->valid & CTS_SCSI_VALID_TQ) != 0) {
if ((sc->sc_cfflags & (1<<((target & 7) + 16))) == 0 &&
(scsi->flags & CTS_SCSI_FLAGS_TAG_ENB)) {
NCR_MISC(("%s: target %d: tagged queuing\n",
device_get_nameunit(sc->sc_dev), target));
ti->flags |= T_TAG;
} else
ti->flags &= ~T_TAG;
}
if ((spi->valid & CTS_SPI_VALID_BUS_WIDTH) != 0) {
if (spi->bus_width != 0) {
NCR_MISC(("%s: target %d: wide negotiation\n",
device_get_nameunit(sc->sc_dev), target));
if (sc->sc_rev == NCR_VARIANT_FAS366) {
ti->flags |= T_WIDE;
ti->width = 1;
}
} else {
ti->flags &= ~T_WIDE;
ti->width = 0;
}
ti->flags |= T_NEGOTIATE;
}
if ((spi->valid & CTS_SPI_VALID_SYNC_RATE) != 0) {
NCR_MISC(("%s: target %d: sync period negotiation\n",
device_get_nameunit(sc->sc_dev), target));
ti->flags |= T_NEGOTIATE;
ti->period = spi->sync_period;
}
if ((spi->valid & CTS_SPI_VALID_SYNC_OFFSET) != 0) {
NCR_MISC(("%s: target %d: sync offset negotiation\n",
device_get_nameunit(sc->sc_dev), target));
ti->flags |= T_NEGOTIATE;
ti->offset = spi->sync_offset;
}
mtx_unlock(&sc->sc_lock);
ccb->ccb_h.status = CAM_REQ_CMP;
xpt_done(ccb);
return;
}
default:
device_printf(sc->sc_dev, "Unhandled function code %d\n",
ccb->ccb_h.func_code);
ccb->ccb_h.status = CAM_PROVIDE_FAIL;
mtx_unlock(&sc->sc_lock);
xpt_done(ccb);
return;
}
mtx_unlock(&sc->sc_lock);
}
/*
* Used when interrupt driven I/O is not allowed, e.g. during boot.
*/
static void
ncr53c9x_poll(struct cam_sim *sim)
{
struct ncr53c9x_softc *sc;
NCR_TRACE(("[ncr53c9x_poll] "));
sc = cam_sim_softc(sim);
if (NCRDMA_ISINTR(sc)) {
ncr53c9x_intr(sc);
}
}
/*
* LOW LEVEL SCSI UTILITIES
*/
/*
* Schedule a scsi operation. This has now been pulled out of the interrupt
* handler so that we may call it from ncr53c9x_scsipi_request and
* ncr53c9x_done. This may save us an unnecessary interrupt just to get
* things going. Should only be called when state == NCR_IDLE and at bio pl.
*/
static void
ncr53c9x_sched(struct ncr53c9x_softc *sc)
{
struct ncr53c9x_ecb *ecb;
struct ncr53c9x_tinfo *ti;
struct ncr53c9x_linfo *li;
int lun;
int tag;
NCR_TRACE(("[ncr53c9x_sched] "));
if (sc->sc_state != NCR_IDLE)
panic("ncr53c9x_sched: not IDLE (state=%d)", sc->sc_state);
/*
* Find first ecb in ready queue that is for a target/lunit
* combinations that is not busy.
*/
for (ecb = TAILQ_FIRST(&sc->ready_list); ecb != NULL;
ecb = TAILQ_NEXT(ecb, chain)) {
ti = &sc->sc_tinfo[ecb->ccb->ccb_h.target_id];
lun = ecb->ccb->ccb_h.target_lun;
/* Select type of tag for this command */
if ((ti->flags & (T_RSELECTOFF)) != 0)
tag = 0;
else if ((ti->flags & (T_TAG)) == 0)
tag = 0;
else if ((ecb->flags & ECB_SENSE) != 0)
tag = 0;
else if ((ecb->ccb->ccb_h.flags & CAM_TAG_ACTION_VALID) == 0)
tag = 0;
else if (ecb->ccb->csio.tag_action == CAM_TAG_ACTION_NONE)
tag = 0;
else
tag = ecb->ccb->csio.tag_action;
li = TINFO_LUN(ti, lun);
if (li == NULL) {
/* Initialize LUN info and add to list. */
if ((li = malloc(sizeof(*li),
M_DEVBUF, M_NOWAIT|M_ZERO)) == NULL) {
continue;
}
li->lun = lun;
LIST_INSERT_HEAD(&ti->luns, li, link);
if (lun < NCR_NLUN)
ti->lun[lun] = li;
}
li->last_used = time_second;
if (tag == 0) {
/* Try to issue this as an un-tagged command */
if (li->untagged == NULL)
li->untagged = ecb;
}
if (li->untagged != NULL) {
tag = 0;
if ((li->busy != 1) && li->used == 0) {
/* We need to issue this untagged command now */
ecb = li->untagged;
} else {
/* Not ready yet */
continue;
}
}
ecb->tag[0] = tag;
if (tag != 0) {
li->queued[ecb->tag_id] = ecb;
ecb->tag[1] = ecb->tag_id;
li->used++;
}
if (li->untagged != NULL && (li->busy != 1)) {
li->busy = 1;
TAILQ_REMOVE(&sc->ready_list, ecb, chain);
ecb->flags &= ~ECB_READY;
sc->sc_nexus = ecb;
ncr53c9x_select(sc, ecb);
break;
}
if (li->untagged == NULL && tag != 0) {
TAILQ_REMOVE(&sc->ready_list, ecb, chain);
ecb->flags &= ~ECB_READY;
sc->sc_nexus = ecb;
ncr53c9x_select(sc, ecb);
break;
} else
NCR_TRACE(("%d:%d busy\n",
ecb->ccb->ccb_h.target_id,
ecb->ccb->ccb_h.target_lun));
}
}
static void
ncr53c9x_sense(struct ncr53c9x_softc *sc, struct ncr53c9x_ecb *ecb)
{
union ccb *ccb = ecb->ccb;
struct ncr53c9x_tinfo *ti;
struct scsi_request_sense *ss = (void *)&ecb->cmd.cmd;
struct ncr53c9x_linfo *li;
int lun;
NCR_TRACE(("requesting sense "));
lun = ccb->ccb_h.target_lun;
ti = &sc->sc_tinfo[ccb->ccb_h.target_id];
/* Next, setup a request sense command block */
memset(ss, 0, sizeof(*ss));
ss->opcode = REQUEST_SENSE;
ss->byte2 = ccb->ccb_h.target_lun << SCSI_CMD_LUN_SHIFT;
ss->length = sizeof(struct scsi_sense_data);
ecb->clen = sizeof(*ss);
ecb->daddr = (char *)&ecb->ccb->csio.sense_data;
ecb->dleft = sizeof(struct scsi_sense_data);
ecb->flags |= ECB_SENSE;
ecb->timeout = NCR_SENSE_TIMEOUT;
ti->senses++;
li = TINFO_LUN(ti, lun);
if (li->busy)
li->busy = 0;
ncr53c9x_dequeue(sc, ecb);
li->untagged = ecb; /* must be executed first to fix C/A */
li->busy = 2;
if (ecb == sc->sc_nexus) {
ncr53c9x_select(sc, ecb);
} else {
TAILQ_INSERT_HEAD(&sc->ready_list, ecb, chain);
ecb->flags |= ECB_READY;
if (sc->sc_state == NCR_IDLE)
ncr53c9x_sched(sc);
}
}
/*
* POST PROCESSING OF SCSI_CMD (usually current)
*/
static void
ncr53c9x_done(struct ncr53c9x_softc *sc, struct ncr53c9x_ecb *ecb)
{
union ccb *ccb = ecb->ccb;
struct ncr53c9x_tinfo *ti;
struct ncr53c9x_linfo *li;
int lun;
NCR_TRACE(("[ncr53c9x_done(status:%x)] ", ccb->ccb_h.status));
ti = &sc->sc_tinfo[ccb->ccb_h.target_id];
lun = ccb->ccb_h.target_lun;
li = TINFO_LUN(ti, lun);
untimeout(ncr53c9x_timeout, ecb, ccb->ccb_h.timeout_ch);
/*
* Now, if we've come here with no error code, i.e. we've kept the
* initial XS_NOERROR, and the status code signals that we should
* check sense, we'll need to set up a request sense cmd block and
* push the command back into the ready queue *before* any other
* commands for this target/lunit, else we lose the sense info.
* We don't support chk sense conditions for the request sense cmd.
*/
if (ccb->ccb_h.status == CAM_REQ_CMP) {
if ((ecb->flags & ECB_ABORT) != 0) {
ccb->ccb_h.status = CAM_CMD_TIMEOUT;
} else if ((ecb->flags & ECB_SENSE) != 0 &&
(ecb->stat != SCSI_STATUS_CHECK_COND)) {
ccb->ccb_h.status = CAM_AUTOSNS_VALID;
} else if (ecb->stat == SCSI_STATUS_CHECK_COND) {
if ((ecb->flags & ECB_SENSE) != 0)
ccb->ccb_h.status = CAM_AUTOSENSE_FAIL;
else {
/* First, save the return values */
ccb->csio.resid = ecb->dleft;
ncr53c9x_sense(sc, ecb);
return;
}
} else {
ccb->csio.resid = ecb->dleft;
}
#if 0
if (xs->status == SCSI_QUEUE_FULL || xs->status == XS_BUSY)
xs->error = XS_BUSY;
#endif
}
#ifdef NCR53C9X_DEBUG
if (ncr53c9x_debug & NCR_SHOWTRAC) {
if (ccb->csio.resid != 0)
printf("resid=%d ", ccb->csio.resid);
#if 0
if (xs->error == XS_SENSE)
printf("sense=0x%02x\n",
xs->sense.scsi_sense.error_code);
else
printf("error=%d\n", xs->error);
#endif
}
#endif
/*
* Remove the ECB from whatever queue it's on.
*/
ncr53c9x_dequeue(sc, ecb);
if (ecb == sc->sc_nexus) {
sc->sc_nexus = NULL;
if (sc->sc_state != NCR_CLEANING) {
sc->sc_state = NCR_IDLE;
ncr53c9x_sched(sc);
}
}
if (ccb->ccb_h.status == CAM_SEL_TIMEOUT) {
/* Selection timeout -- discard this LUN if empty */
if (li->untagged == NULL && li->used == 0) {
if (lun < NCR_NLUN)
ti->lun[lun] = NULL;
LIST_REMOVE(li, link);
free(li, M_DEVBUF);
}
}
ncr53c9x_free_ecb(sc, ecb);
ti->cmds++;
xpt_done(ccb);
}
static void
ncr53c9x_dequeue(struct ncr53c9x_softc *sc, struct ncr53c9x_ecb *ecb)
{
struct ncr53c9x_tinfo *ti;
struct ncr53c9x_linfo *li;
int64_t lun;
ti = &sc->sc_tinfo[ecb->ccb->ccb_h.target_id];
lun = ecb->ccb->ccb_h.target_lun;
li = TINFO_LUN(ti, lun);
#ifdef DIAGNOSTIC
if (li == NULL || li->lun != lun)
panic("ncr53c9x_dequeue: lun %qx for ecb %p does not exist",
(long long) lun, ecb);
#endif
if (li->untagged == ecb) {
li->busy = 0;
li->untagged = NULL;
}
if (ecb->tag[0] && li->queued[ecb->tag[1]] != NULL) {
#ifdef DIAGNOSTIC
if (li->queued[ecb->tag[1]] != NULL &&
(li->queued[ecb->tag[1]] != ecb))
panic("ncr53c9x_dequeue: slot %d for lun %qx has %p "
"instead of ecb %p\n", ecb->tag[1],
(long long) lun,
li->queued[ecb->tag[1]], ecb);
#endif
li->queued[ecb->tag[1]] = NULL;
li->used--;
}
if ((ecb->flags & ECB_READY) != 0) {
ecb->flags &= ~ECB_READY;
TAILQ_REMOVE(&sc->ready_list, ecb, chain);
}
}
/*
* INTERRUPT/PROTOCOL ENGINE
*/
/*
* Schedule an outgoing message by prioritizing it, and asserting
* attention on the bus. We can only do this when we are the initiator
* else there will be an illegal command interrupt.
*/
#define ncr53c9x_sched_msgout(m) \
do { \
NCR_MSGS(("ncr53c9x_sched_msgout %x %d", m, __LINE__)); \
NCRCMD(sc, NCRCMD_SETATN); \
sc->sc_flags |= NCR_ATN; \
sc->sc_msgpriq |= (m); \
} while (0)
static void
ncr53c9x_flushfifo(struct ncr53c9x_softc *sc)
{
NCR_TRACE(("[flushfifo] "));
NCRCMD(sc, NCRCMD_FLUSH);
if (sc->sc_phase == COMMAND_PHASE ||
sc->sc_phase == MESSAGE_OUT_PHASE)
DELAY(2);
}
static int
ncr53c9x_rdfifo(struct ncr53c9x_softc *sc, int how)
{
int i, n;
u_char *buf;
switch(how) {
case NCR_RDFIFO_START:
buf = sc->sc_imess;
sc->sc_imlen = 0;
break;
case NCR_RDFIFO_CONTINUE:
buf = sc->sc_imess + sc->sc_imlen;
break;
default:
panic("ncr53c9x_rdfifo: bad flag");
break;
}
/*
* XXX buffer (sc_imess) size for message
*/
n = NCR_READ_REG(sc, NCR_FFLAG) & NCRFIFO_FF;
if (sc->sc_rev == NCR_VARIANT_FAS366) {
n *= 2;
for (i = 0; i < n; i++)
buf[i] = NCR_READ_REG(sc, NCR_FIFO);
if (sc->sc_espstat2 & NCRFAS_STAT2_ISHUTTLE) {
NCR_WRITE_REG(sc, NCR_FIFO, 0);
buf[i++] = NCR_READ_REG(sc, NCR_FIFO);
NCR_READ_REG(sc, NCR_FIFO);
ncr53c9x_flushfifo(sc);
}
} else {
for (i = 0; i < n; i++)
buf[i] = NCR_READ_REG(sc, NCR_FIFO);
}
sc->sc_imlen += i;
#if 0
#ifdef NCR53C9X_DEBUG
{
int j;
NCR_TRACE(("\n[rdfifo %s (%d):",
(how == NCR_RDFIFO_START) ? "start" : "cont",
(int)sc->sc_imlen));
if (ncr53c9x_debug & NCR_SHOWTRAC) {
for (j = 0; j < sc->sc_imlen; j++)
printf(" %02x", sc->sc_imess[j]);
printf("]\n");
}
}
#endif
#endif
return sc->sc_imlen;
}
static void
ncr53c9x_wrfifo(struct ncr53c9x_softc *sc, u_char *p, int len)
{
int i;
#ifdef NCR53C9X_DEBUG
NCR_MSGS(("[wrfifo(%d):", len));
if (ncr53c9x_debug & NCR_SHOWMSGS) {
for (i = 0; i < len; i++)
printf(" %02x", p[i]);
printf("]\n");
}
#endif
for (i = 0; i < len; i++) {
NCR_WRITE_REG(sc, NCR_FIFO, p[i]);
if (sc->sc_rev == NCR_VARIANT_FAS366)
NCR_WRITE_REG(sc, NCR_FIFO, 0);
}
}
static int
ncr53c9x_reselect(struct ncr53c9x_softc *sc, int message, int tagtype,
int tagid)
{
u_char selid, target, lun;
struct ncr53c9x_ecb *ecb = NULL;
struct ncr53c9x_tinfo *ti;
struct ncr53c9x_linfo *li;
if (sc->sc_rev == NCR_VARIANT_FAS366) {
target = sc->sc_selid;
} else {
/*
* The SCSI chip made a snapshot of the data bus
* while the reselection was being negotiated.
* This enables us to determine which target did
* the reselect.
*/
selid = sc->sc_selid & ~(1 << sc->sc_id);
if (selid & (selid - 1)) {
device_printf(sc->sc_dev, "reselect with invalid "
"selid %02x; sending DEVICE RESET\n", selid);
goto reset;
}
target = ffs(selid) - 1;
}
lun = message & 0x07;
/*
* Search wait queue for disconnected cmd
* The list should be short, so I haven't bothered with
* any more sophisticated structures than a simple
* singly linked list.
*/
ti = &sc->sc_tinfo[target];
li = TINFO_LUN(ti, lun);
/*
* We can get as far as the LUN with the IDENTIFY
* message. Check to see if we're running an
* un-tagged command. Otherwise ack the IDENTIFY
* and wait for a tag message.
*/
if (li != NULL) {
if (li->untagged != NULL && li->busy)
ecb = li->untagged;
else if (tagtype != MSG_SIMPLE_Q_TAG) {
/* Wait for tag to come by */
sc->sc_state = NCR_IDENTIFIED;
return (0);
} else if (tagtype)
ecb = li->queued[tagid];
}
if (ecb == NULL) {
device_printf(sc->sc_dev, "reselect from target %d lun %d "
"tag %x:%x with no nexus; sending ABORT\n",
target, lun, tagtype, tagid);
goto abort;
}
/* Make this nexus active again. */
sc->sc_state = NCR_CONNECTED;
sc->sc_nexus = ecb;
ncr53c9x_setsync(sc, ti);
if (ecb->flags & ECB_RESET)
ncr53c9x_sched_msgout(SEND_DEV_RESET);
else if (ecb->flags & ECB_ABORT)
ncr53c9x_sched_msgout(SEND_ABORT);
/* Do an implicit RESTORE POINTERS. */
sc->sc_dp = ecb->daddr;
sc->sc_dleft = ecb->dleft;
return (0);
reset:
ncr53c9x_sched_msgout(SEND_DEV_RESET);
return (1);
abort:
ncr53c9x_sched_msgout(SEND_ABORT);
return (1);
}
/* From NetBSD. These should go into CAM at some point */
#define MSG_ISEXTENDED(m) ((m) == MSG_EXTENDED)
#define MSG_IS1BYTE(m) \
((!MSG_ISEXTENDED(m) && (m) < 0x20) || MSG_ISIDENTIFY(m))
#define MSG_IS2BYTE(m) (((m) & 0xf0) == 0x20)
static inline int
__verify_msg_format(u_char *p, int len)
{
if (len == 1 && MSG_IS1BYTE(p[0]))
return 1;
if (len == 2 && MSG_IS2BYTE(p[0]))
return 1;
if (len >= 3 && MSG_ISEXTENDED(p[0]) &&
len == p[1] + 2)
return 1;
return 0;
}
/*
* Get an incoming message as initiator.
*
* The SCSI bus must already be in MESSAGE_IN_PHASE and there is a
* byte in the FIFO
*/
static void
ncr53c9x_msgin(struct ncr53c9x_softc *sc)
{
NCR_TRACE(("[ncr53c9x_msgin(curmsglen:%ld)] ", (long)sc->sc_imlen));
if (sc->sc_imlen == 0) {
device_printf(sc->sc_dev, "msgin: no msg byte available\n");
return;
}
/*
* Prepare for a new message. A message should (according
* to the SCSI standard) be transmitted in one single
* MESSAGE_IN_PHASE. If we have been in some other phase,
* then this is a new message.
*/
if (sc->sc_prevphase != MESSAGE_IN_PHASE &&
sc->sc_state != NCR_RESELECTED) {
device_printf(sc->sc_dev, "phase change, dropping message, "
"prev %d, state %d\n", sc->sc_prevphase, sc->sc_state);
sc->sc_flags &= ~NCR_DROP_MSGI;
sc->sc_imlen = 0;
}
/*
* If we're going to reject the message, don't bother storing
* the incoming bytes. But still, we need to ACK them.
*/
if ((sc->sc_flags & NCR_DROP_MSGI) != 0) {
NCRCMD(sc, NCRCMD_MSGOK);
printf("<dropping msg byte %x>", sc->sc_imess[sc->sc_imlen]);
return;
}
if (sc->sc_imlen >= NCR_MAX_MSG_LEN) {
ncr53c9x_sched_msgout(SEND_REJECT);
sc->sc_flags |= NCR_DROP_MSGI;
} else {
u_char *pb;
int plen;
switch (sc->sc_state) {
/*
* if received message is the first of reselection
* then first byte is selid, and then message
*/
case NCR_RESELECTED:
pb = sc->sc_imess + 1;
plen = sc->sc_imlen - 1;
break;
default:
pb = sc->sc_imess;
plen = sc->sc_imlen;
break;
}
if (__verify_msg_format(pb, plen))
goto gotit;
}
/* Ack what we have so far */
NCRCMD(sc, NCRCMD_MSGOK);
return;
gotit:
NCR_MSGS(("gotmsg(%x) state %d", sc->sc_imess[0], sc->sc_state));
/* We got a complete message, flush the imess, */
/* XXX nobody uses imlen below */
sc->sc_imlen = 0;
/*
* Now we should have a complete message (1 byte, 2 byte
* and moderately long extended messages). We only handle
* extended messages which total length is shorter than
* NCR_MAX_MSG_LEN. Longer messages will be amputated.
*/
switch (sc->sc_state) {
struct ncr53c9x_ecb *ecb;
struct ncr53c9x_tinfo *ti;
struct ncr53c9x_linfo *li;
int lun;
case NCR_CONNECTED:
ecb = sc->sc_nexus;
ti = &sc->sc_tinfo[ecb->ccb->ccb_h.target_id];
switch (sc->sc_imess[0]) {
case MSG_CMDCOMPLETE:
NCR_MSGS(("cmdcomplete "));
if (sc->sc_dleft < 0) {
xpt_print_path(ecb->ccb->ccb_h.path);
printf("got %ld extra bytes\n",
-(long)sc->sc_dleft);
sc->sc_dleft = 0;
}
ecb->dleft = (ecb->flags & ECB_TENTATIVE_DONE) ?
0 : sc->sc_dleft;
if ((ecb->flags & ECB_SENSE) == 0)
ecb->ccb->csio.resid = ecb->dleft;
sc->sc_state = NCR_CMDCOMPLETE;
break;
case MSG_MESSAGE_REJECT:
NCR_MSGS(("msg reject (msgout=%x) ", sc->sc_msgout));
switch (sc->sc_msgout) {
case SEND_TAG:
/*
* Target does not like tagged queuing.
* - Flush the command queue
* - Disable tagged queuing for the target
* - Dequeue ecb from the queued array.
*/
device_printf(sc->sc_dev, "tagged queuing "
"rejected: target %d\n",
ecb->ccb->ccb_h.target_id);
NCR_MSGS(("(rejected sent tag)"));
NCRCMD(sc, NCRCMD_FLUSH);
DELAY(1);
ti->flags &= ~T_TAG;
lun = ecb->ccb->ccb_h.target_lun;
li = TINFO_LUN(ti, lun);
if (ecb->tag[0] &&
li->queued[ecb->tag[1]] != NULL) {
li->queued[ecb->tag[1]] = NULL;
li->used--;
}
ecb->tag[0] = ecb->tag[1] = 0;
li->untagged = ecb;
li->busy = 1;
break;
case SEND_SDTR:
device_printf(sc->sc_dev, "sync transfer "
"rejected: target %d\n",
ecb->ccb->ccb_h.target_id);
sc->sc_flags &= ~NCR_SYNCHNEGO;
ti->flags &= ~(T_NEGOTIATE | T_SYNCMODE);
ncr53c9x_setsync(sc, ti);
break;
case SEND_WDTR:
device_printf(sc->sc_dev, "wide transfer "
"rejected: target %d\n",
ecb->ccb->ccb_h.target_id);
ti->flags &= ~(T_WIDE | T_WDTRSENT);
ti->width = 0;
break;
case SEND_INIT_DET_ERR:
goto abort;
}
break;
case MSG_NOOP:
NCR_MSGS(("noop "));
break;
case MSG_HEAD_OF_Q_TAG:
case MSG_SIMPLE_Q_TAG:
case MSG_ORDERED_Q_TAG:
NCR_MSGS(("TAG %x:%x",
sc->sc_imess[0], sc->sc_imess[1]));
break;
case MSG_DISCONNECT:
NCR_MSGS(("disconnect "));
ti->dconns++;
sc->sc_state = NCR_DISCONNECT;
/*
* Mark the fact that all bytes have moved. The
* target may not bother to do a SAVE POINTERS
* at this stage. This flag will set the residual
* count to zero on MSG COMPLETE.
*/
if (sc->sc_dleft == 0)
ecb->flags |= ECB_TENTATIVE_DONE;
break;
case MSG_SAVEDATAPOINTER:
NCR_MSGS(("save datapointer "));
ecb->daddr = sc->sc_dp;
ecb->dleft = sc->sc_dleft;
break;
case MSG_RESTOREPOINTERS:
NCR_MSGS(("restore datapointer "));
sc->sc_dp = ecb->daddr;
sc->sc_dleft = ecb->dleft;
break;
case MSG_EXTENDED:
NCR_MSGS(("extended(%x) ", sc->sc_imess[2]));
switch (sc->sc_imess[2]) {
case MSG_EXT_SDTR:
NCR_MSGS(("SDTR period %d, offset %d ",
sc->sc_imess[3], sc->sc_imess[4]));
if (sc->sc_imess[1] != 3)
goto reject;
ti->period = sc->sc_imess[3];
ti->offset = sc->sc_imess[4];
ti->flags &= ~T_NEGOTIATE;
if (sc->sc_minsync == 0 ||
ti->offset == 0 ||
ti->period > 124) {
#if 0
#ifdef NCR53C9X_DEBUG
xpt_print_path(ecb->ccb->ccb_h.path);
printf("async mode\n");
#endif
#endif
ti->flags &= ~T_SYNCMODE;
if ((sc->sc_flags&NCR_SYNCHNEGO) == 0) {
/*
* target initiated negotiation
*/
ti->offset = 0;
ncr53c9x_sched_msgout(
SEND_SDTR);
}
} else {
int p;
p = ncr53c9x_stp2cpb(sc, ti->period);
ti->period = ncr53c9x_cpb2stp(sc, p);
if ((sc->sc_flags&NCR_SYNCHNEGO) == 0) {
/*
* target initiated negotiation
*/
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
if (ti->period < sc->sc_minsync)
ti->period =
sc->sc_minsync;
if (ti->offset > 15)
ti->offset = 15;
ti->flags &= ~T_SYNCMODE;
ncr53c9x_sched_msgout(
SEND_SDTR);
} else {
/* we are sync */
ti->flags |= T_SYNCMODE;
}
}
sc->sc_flags &= ~NCR_SYNCHNEGO;
ncr53c9x_setsync(sc, ti);
break;
case MSG_EXT_WDTR:
#ifdef NCR53C9X_DEBUG
device_printf(sc->sc_dev, "wide mode %d\n",
sc->sc_imess[3]);
#endif
if (sc->sc_imess[3] == 1) {
ti->cfg3 |= NCRFASCFG3_EWIDE;
ncr53c9x_setsync(sc, ti);
} else
ti->width = 0;
/*
* Device started width negotiation.
*/
if (!(ti->flags & T_WDTRSENT))
ncr53c9x_sched_msgout(SEND_WDTR);
ti->flags &= ~(T_WIDE | T_WDTRSENT);
break;
default:
xpt_print_path(ecb->ccb->ccb_h.path);
printf("unrecognized MESSAGE EXTENDED;"
" sending REJECT\n");
goto reject;
}
break;
default:
NCR_MSGS(("ident "));
xpt_print_path(ecb->ccb->ccb_h.path);
printf("unrecognized MESSAGE; sending REJECT\n");
reject:
ncr53c9x_sched_msgout(SEND_REJECT);
break;
}
break;
case NCR_IDENTIFIED:
/*
* IDENTIFY message was received and queue tag is expected now
*/
if ((sc->sc_imess[0] != MSG_SIMPLE_Q_TAG) ||
(sc->sc_msgify == 0)) {
device_printf(sc->sc_dev, "TAG reselect without "
"IDENTIFY; MSG %x; sending DEVICE RESET\n",
sc->sc_imess[0]);
goto reset;
}
(void) ncr53c9x_reselect(sc, sc->sc_msgify,
sc->sc_imess[0], sc->sc_imess[1]);
break;
case NCR_RESELECTED:
if (MSG_ISIDENTIFY(sc->sc_imess[1])) {
sc->sc_msgify = sc->sc_imess[1];
} else {
device_printf(sc->sc_dev, "reselect without IDENTIFY;"
" MSG %x; sending DEVICE RESET\n", sc->sc_imess[1]);
goto reset;
}
(void) ncr53c9x_reselect(sc, sc->sc_msgify, 0, 0);
break;
default:
device_printf(sc->sc_dev, "unexpected MESSAGE IN; "
"sending DEVICE RESET\n");
reset:
ncr53c9x_sched_msgout(SEND_DEV_RESET);
break;
abort:
ncr53c9x_sched_msgout(SEND_ABORT);
break;
}
/* if we have more messages to send set ATN */
if (sc->sc_msgpriq)
NCRCMD(sc, NCRCMD_SETATN);
/* Ack last message byte */
NCRCMD(sc, NCRCMD_MSGOK);
/* Done, reset message pointer. */
sc->sc_flags &= ~NCR_DROP_MSGI;
sc->sc_imlen = 0;
}
/*
* Send the highest priority, scheduled message
*/
static void
ncr53c9x_msgout(struct ncr53c9x_softc *sc)
{
struct ncr53c9x_tinfo *ti;
struct ncr53c9x_ecb *ecb;
size_t size;
NCR_TRACE(("[ncr53c9x_msgout(priq:%x, prevphase:%x)]",
sc->sc_msgpriq, sc->sc_prevphase));
/*
* XXX - the NCR_ATN flag is not in sync with the actual ATN
* condition on the SCSI bus. The 53c9x chip
* automatically turns off ATN before sending the
* message byte. (See also the comment below in the
* default case when picking out a message to send.)
*/
if (sc->sc_flags & NCR_ATN) {
if (sc->sc_prevphase != MESSAGE_OUT_PHASE) {
new:
NCRCMD(sc, NCRCMD_FLUSH);
/* DELAY(1); */
sc->sc_msgoutq = 0;
sc->sc_omlen = 0;
}
} else {
if (sc->sc_prevphase == MESSAGE_OUT_PHASE) {
ncr53c9x_sched_msgout(sc->sc_msgoutq);
goto new;
} else {
device_printf(sc->sc_dev, "at line %d: unexpected "
"MESSAGE OUT phase\n", __LINE__);
}
}
if (sc->sc_omlen == 0) {
/* Pick up highest priority message */
sc->sc_msgout = sc->sc_msgpriq & -sc->sc_msgpriq;
sc->sc_msgoutq |= sc->sc_msgout;
sc->sc_msgpriq &= ~sc->sc_msgout;
sc->sc_omlen = 1; /* "Default" message len */
switch (sc->sc_msgout) {
case SEND_SDTR:
ecb = sc->sc_nexus;
ti = &sc->sc_tinfo[ecb->ccb->ccb_h.target_id];
sc->sc_omess[0] = MSG_EXTENDED;
sc->sc_omess[1] = MSG_EXT_SDTR_LEN;
sc->sc_omess[2] = MSG_EXT_SDTR;
sc->sc_omess[3] = ti->period;
sc->sc_omess[4] = ti->offset;
sc->sc_omlen = 5;
if ((sc->sc_flags & NCR_SYNCHNEGO) == 0) {
ti->flags |= T_SYNCMODE;
ncr53c9x_setsync(sc, ti);
}
break;
case SEND_WDTR:
ecb = sc->sc_nexus;
ti = &sc->sc_tinfo[ecb->ccb->ccb_h.target_id];
sc->sc_omess[0] = MSG_EXTENDED;
sc->sc_omess[1] = MSG_EXT_WDTR_LEN;
sc->sc_omess[2] = MSG_EXT_WDTR;
sc->sc_omess[3] = ti->width;
sc->sc_omlen = 4;
break;
case SEND_IDENTIFY:
if (sc->sc_state != NCR_CONNECTED) {
device_printf(sc->sc_dev, "at line %d: no "
"nexus\n", __LINE__);
}
ecb = sc->sc_nexus;
sc->sc_omess[0] =
MSG_IDENTIFY(ecb->ccb->ccb_h.target_lun, 0);
break;
case SEND_TAG:
if (sc->sc_state != NCR_CONNECTED) {
device_printf(sc->sc_dev, "at line %d: no "
"nexus\n", __LINE__);
}
ecb = sc->sc_nexus;
sc->sc_omess[0] = ecb->tag[0];
sc->sc_omess[1] = ecb->tag[1];
sc->sc_omlen = 2;
break;
case SEND_DEV_RESET:
sc->sc_flags |= NCR_ABORTING;
sc->sc_omess[0] = MSG_BUS_DEV_RESET;
ecb = sc->sc_nexus;
ti = &sc->sc_tinfo[ecb->ccb->ccb_h.target_id];
ti->flags &= ~T_SYNCMODE;
if ((ti->flags & T_SYNCHOFF) == 0)
/* We can re-start sync negotiation */
ti->flags |= T_NEGOTIATE;
break;
case SEND_PARITY_ERROR:
sc->sc_omess[0] = MSG_PARITY_ERROR;
break;
case SEND_ABORT:
sc->sc_flags |= NCR_ABORTING;
sc->sc_omess[0] = MSG_ABORT;
break;
case SEND_INIT_DET_ERR:
sc->sc_omess[0] = MSG_INITIATOR_DET_ERR;
break;
case SEND_REJECT:
sc->sc_omess[0] = MSG_MESSAGE_REJECT;
break;
default:
/*
* We normally do not get here, since the chip
* automatically turns off ATN before the last
* byte of a message is sent to the target.
* However, if the target rejects our (multi-byte)
* message early by switching to MSG IN phase
* ATN remains on, so the target may return to
* MSG OUT phase. If there are no scheduled messages
* left we send a NO-OP.
*
* XXX - Note that this leaves no useful purpose for
* the NCR_ATN flag.
*/
sc->sc_flags &= ~NCR_ATN;
sc->sc_omess[0] = MSG_NOOP;
break;
}
sc->sc_omp = sc->sc_omess;
}
#ifdef DEBUG
if (ncr53c9x_debug & NCR_SHOWMSGS) {
int i;
NCR_MSGS(("<msgout:"));
for (i = 0; i < sc->sc_omlen; i++)
NCR_MSGS((" %02x", sc->sc_omess[i]));
NCR_MSGS(("> "));
}
#endif
if (sc->sc_rev == NCR_VARIANT_FAS366) {
/*
* XXX fifo size
*/
ncr53c9x_flushfifo(sc);
ncr53c9x_wrfifo(sc, sc->sc_omp, sc->sc_omlen);
NCRCMD(sc, NCRCMD_TRANS);
} else {
/* (re)send the message */
size = min(sc->sc_omlen, sc->sc_maxxfer);
NCRDMA_SETUP(sc, &sc->sc_omp, &sc->sc_omlen, 0, &size);
/* Program the SCSI counter */
NCR_SET_COUNT(sc, size);
/* Load the count in and start the message-out transfer */
NCRCMD(sc, NCRCMD_NOP|NCRCMD_DMA);
NCRCMD(sc, NCRCMD_TRANS|NCRCMD_DMA);
NCRDMA_GO(sc);
}
}
/*
* This is the most critical part of the driver, and has to know
* how to deal with *all* error conditions and phases from the SCSI
* bus. If there are no errors and the DMA was active, then call the
* DMA pseudo-interrupt handler. If this returns 1, then that was it
* and we can return from here without further processing.
*
* Most of this needs verifying.
*/
void
ncr53c9x_intr(void *arg)
{
struct ncr53c9x_softc *sc = arg;
struct ncr53c9x_ecb *ecb;
struct ncr53c9x_tinfo *ti;
size_t size;
int nfifo;
NCR_INTS(("[ncr53c9x_intr: state %d]", sc->sc_state));
if (!NCRDMA_ISINTR(sc))
return;
mtx_lock(&sc->sc_lock);
again:
/* and what do the registers say... */
ncr53c9x_readregs(sc);
/*
* At the moment, only a SCSI Bus Reset or Illegal
* Command are classed as errors. A disconnect is a
* valid condition, and we let the code check is the
* "NCR_BUSFREE_OK" flag was set before declaring it
* and error.
*
* Also, the status register tells us about "Gross
* Errors" and "Parity errors". Only the Gross Error
* is really bad, and the parity errors are dealt
* with later
*
* TODO
* If there are too many parity error, go to slow
* cable mode ?
*/
/* SCSI Reset */
if ((sc->sc_espintr & NCRINTR_SBR) != 0) {
if ((NCR_READ_REG(sc, NCR_FFLAG) & NCRFIFO_FF) != 0) {
NCRCMD(sc, NCRCMD_FLUSH);
DELAY(1);
}
if (sc->sc_state != NCR_SBR) {
device_printf(sc->sc_dev, "SCSI bus reset\n");
ncr53c9x_init(sc, 0); /* Restart everything */
goto out;
}
#if 0
/*XXX*/ printf("<expected bus reset: "
"[intr %x, stat %x, step %d]>\n",
sc->sc_espintr, sc->sc_espstat, sc->sc_espstep);
#endif
if (sc->sc_nexus != NULL)
panic("%s: nexus in reset state",
device_get_nameunit(sc->sc_dev));
goto sched;
}
ecb = sc->sc_nexus;
#define NCRINTR_ERR (NCRINTR_SBR|NCRINTR_ILL)
if (sc->sc_espintr & NCRINTR_ERR ||
sc->sc_espstat & NCRSTAT_GE) {
if ((sc->sc_espstat & NCRSTAT_GE) != 0) {
/* Gross Error; no target ? */
if (NCR_READ_REG(sc, NCR_FFLAG) & NCRFIFO_FF) {
NCRCMD(sc, NCRCMD_FLUSH);
DELAY(1);
}
if (sc->sc_state == NCR_CONNECTED ||
sc->sc_state == NCR_SELECTING) {
ecb->ccb->ccb_h.status = CAM_SEL_TIMEOUT;
ncr53c9x_done(sc, ecb);
}
goto out;
}
if ((sc->sc_espintr & NCRINTR_ILL) != 0) {
if ((sc->sc_flags & NCR_EXPECT_ILLCMD) != 0) {
/*
* Eat away "Illegal command" interrupt
* on a ESP100 caused by a re-selection
* while we were trying to select
* another target.
*/
#ifdef DEBUG
device_printf(sc->sc_dev, "ESP100 work-around "
"activated\n");
#endif
sc->sc_flags &= ~NCR_EXPECT_ILLCMD;
goto out;
}
/* illegal command, out of sync ? */
device_printf(sc->sc_dev, "illegal command: 0x%x "
"(state %d, phase %x, prevphase %x)\n",
sc->sc_lastcmd,
sc->sc_state, sc->sc_phase, sc->sc_prevphase);
if (NCR_READ_REG(sc, NCR_FFLAG) & NCRFIFO_FF) {
NCRCMD(sc, NCRCMD_FLUSH);
DELAY(1);
}
ncr53c9x_init(sc, 1); /* Restart everything */
goto out;
}
}
sc->sc_flags &= ~NCR_EXPECT_ILLCMD;
/*
* Call if DMA is active.
*
* If DMA_INTR returns true, then maybe go 'round the loop
* again in case there is no more DMA queued, but a phase
* change is expected.
*/
if (NCRDMA_ISACTIVE(sc)) {
int r = NCRDMA_INTR(sc);
if (r == -1) {
device_printf(sc->sc_dev, "DMA error; resetting\n");
ncr53c9x_init(sc, 1);
goto out;
}
/* If DMA active here, then go back to work... */
if (NCRDMA_ISACTIVE(sc))
goto out;
if ((sc->sc_espstat & NCRSTAT_TC) == 0) {
/*
* DMA not completed. If we can not find a
* acceptable explanation, print a diagnostic.
*/
if (sc->sc_state == NCR_SELECTING)
/*
* This can happen if we are reselected
* while using DMA to select a target.
*/
/*void*/;
else if (sc->sc_prevphase == MESSAGE_OUT_PHASE) {
/*
* Our (multi-byte) message (eg SDTR) was
* interrupted by the target to send
* a MSG REJECT.
* Print diagnostic if current phase
* is not MESSAGE IN.
*/
if (sc->sc_phase != MESSAGE_IN_PHASE)
device_printf(sc->sc_dev,"!TC on MSGOUT"
" [intr %x, stat %x, step %d]"
" prevphase %x, resid %lx\n",
sc->sc_espintr,
sc->sc_espstat,
sc->sc_espstep,
sc->sc_prevphase,
(u_long)sc->sc_omlen);
} else if (sc->sc_dleft == 0) {
/*
* The DMA operation was started for
* a DATA transfer. Print a diagnostic
* if the DMA counter and TC bit
* appear to be out of sync.
*
* XXX This is fatal and usually means that
* the DMA engine is hopelessly out of
* sync with reality. A disk is likely
* getting spammed at this point.
*/
device_printf(sc->sc_dev, "!TC on DATA XFER"
" [intr %x, stat %x, step %d]"
" prevphase %x, resid %x\n",
sc->sc_espintr,
sc->sc_espstat,
sc->sc_espstep,
sc->sc_prevphase,
ecb ? ecb->dleft : -1);
panic("esp: unrecoverable DMA error");
}
}
}
/*
* Check for less serious errors.
*/
if ((sc->sc_espstat & NCRSTAT_PE) != 0) {
device_printf(sc->sc_dev, "SCSI bus parity error\n");
if (sc->sc_prevphase == MESSAGE_IN_PHASE)
ncr53c9x_sched_msgout(SEND_PARITY_ERROR);
else
ncr53c9x_sched_msgout(SEND_INIT_DET_ERR);
}
if ((sc->sc_espintr & NCRINTR_DIS) != 0) {
sc->sc_msgify = 0;
NCR_INTS(("<DISC [intr %x, stat %x, step %d]>",
sc->sc_espintr,sc->sc_espstat,sc->sc_espstep));
if (NCR_READ_REG(sc, NCR_FFLAG) & NCRFIFO_FF) {
NCRCMD(sc, NCRCMD_FLUSH);
/* DELAY(1); */
}
/*
* This command must (apparently) be issued within
* 250mS of a disconnect. So here you are...
*/
NCRCMD(sc, NCRCMD_ENSEL);
switch (sc->sc_state) {
case NCR_RESELECTED:
goto sched;
case NCR_SELECTING:
{
struct ncr53c9x_linfo *li;
ecb->ccb->ccb_h.status = CAM_SEL_TIMEOUT;
/* Selection timeout -- discard all LUNs if empty */
ti = &sc->sc_tinfo[ecb->ccb->ccb_h.target_id];
li = LIST_FIRST(&ti->luns);
while (li != NULL) {
if (li->untagged == NULL && li->used == 0) {
if (li->lun < NCR_NLUN)
ti->lun[li->lun] = NULL;
LIST_REMOVE(li, link);
free(li, M_DEVBUF);
/*
* Restart the search at the beginning
*/
li = LIST_FIRST(&ti->luns);
continue;
}
li = LIST_NEXT(li, link);
}
goto finish;
}
case NCR_CONNECTED:
if ((sc->sc_flags & NCR_SYNCHNEGO) != 0) {
#ifdef NCR53C9X_DEBUG
if (ecb != NULL)
xpt_print_path(ecb->ccb->ccb_h.path);
printf("sync nego not completed!\n");
#endif
ti = &sc->sc_tinfo[ecb->ccb->ccb_h.target_id];
sc->sc_flags &= ~NCR_SYNCHNEGO;
ti->flags &= ~(T_NEGOTIATE | T_SYNCMODE);
}
/* it may be OK to disconnect */
if ((sc->sc_flags & NCR_ABORTING) == 0) {
/*
* Section 5.1.1 of the SCSI 2 spec
* suggests issuing a REQUEST SENSE
* following an unexpected disconnect.
* Some devices go into a contingent
* allegiance condition when
* disconnecting, and this is necessary
* to clean up their state.
*/
device_printf(sc->sc_dev, "unexpected "
"disconnect [state %d, intr %x, stat %x, "
"phase(c %x, p %x)]; ", sc->sc_state,
sc->sc_espintr, sc->sc_espstat,
sc->sc_phase, sc->sc_prevphase);
/*
* XXX This will cause a chip reset and will
* prevent us from finding out the real
* problem with the device. However, it's
* neccessary until a way can be found to
* safely cancel the DMA that is in
* progress.
*/
if (1 || (ecb->flags & ECB_SENSE) != 0) {
printf("resetting\n");
goto reset;
}
printf("sending REQUEST SENSE\n");
untimeout(ncr53c9x_timeout, ecb,
ecb->ccb->ccb_h.timeout_ch);
ncr53c9x_sense(sc, ecb);
goto out;
}
ecb->ccb->ccb_h.status = CAM_CMD_TIMEOUT;
goto finish;
case NCR_DISCONNECT:
sc->sc_nexus = NULL;
goto sched;
case NCR_CMDCOMPLETE:
ecb->ccb->ccb_h.status = CAM_REQ_CMP;
goto finish;
}
}
switch (sc->sc_state) {
case NCR_SBR:
device_printf(sc->sc_dev, "waiting for Bus Reset to happen\n");
goto out;
case NCR_RESELECTED:
/*
* we must be continuing a message ?
*/
device_printf(sc->sc_dev, "unhandled reselect continuation, "
"state %d, intr %02x\n", sc->sc_state, sc->sc_espintr);
ncr53c9x_init(sc, 1);
goto out;
break;
case NCR_IDENTIFIED:
ecb = sc->sc_nexus;
if (sc->sc_phase != MESSAGE_IN_PHASE) {
int i = (NCR_READ_REG(sc, NCR_FFLAG) & NCRFIFO_FF);
/*
* Things are seriously screwed up.
* Pull the brakes, i.e. reset
*/
device_printf(sc->sc_dev, "target didn't send tag: %d "
"bytes in fifo\n", i);
/* Drain and display fifo */
while (i-- > 0)
printf("[%d] ", NCR_READ_REG(sc, NCR_FIFO));
ncr53c9x_init(sc, 1);
goto out;
} else
goto msgin;
case NCR_IDLE:
case NCR_SELECTING:
ecb = sc->sc_nexus;
if (sc->sc_espintr & NCRINTR_RESEL) {
sc->sc_msgpriq = sc->sc_msgout = sc->sc_msgoutq = 0;
sc->sc_flags = 0;
/*
* If we're trying to select a
* target ourselves, push our command
* back into the ready list.
*/
if (sc->sc_state == NCR_SELECTING) {
NCR_INTS(("backoff selector "));
untimeout(ncr53c9x_timeout, ecb,
ecb->ccb->ccb_h.timeout_ch);
ncr53c9x_dequeue(sc, ecb);
TAILQ_INSERT_HEAD(&sc->ready_list, ecb, chain);
ecb->flags |= ECB_READY;
ecb = sc->sc_nexus = NULL;
}
sc->sc_state = NCR_RESELECTED;
if (sc->sc_phase != MESSAGE_IN_PHASE) {
/*
* Things are seriously screwed up.
* Pull the brakes, i.e. reset
*/
device_printf(sc->sc_dev, "target didn't "
"identify\n");
ncr53c9x_init(sc, 1);
goto out;
}
/*
* The C90 only inhibits FIFO writes until reselection
* is complete instead of waiting until the interrupt
* status register has been read. So, if the reselect
* happens while we were entering command bytes (for
* another target) some of those bytes can appear in
* the FIFO here, after the interrupt is taken.
*
* To remedy this situation, pull the Selection ID
* and Identify message from the FIFO directly, and
* ignore any extraneous fifo contents. Also, set
* a flag that allows one Illegal Command Interrupt
* to occur which the chip also generates as a result
* of writing to the FIFO during a reselect.
*/
if (sc->sc_rev == NCR_VARIANT_ESP100) {
nfifo = NCR_READ_REG(sc, NCR_FFLAG) & NCRFIFO_FF;
sc->sc_imess[0] = NCR_READ_REG(sc, NCR_FIFO);
sc->sc_imess[1] = NCR_READ_REG(sc, NCR_FIFO);
sc->sc_imlen = 2;
if (nfifo != 2) {
/* Flush the rest */
NCRCMD(sc, NCRCMD_FLUSH);
}
sc->sc_flags |= NCR_EXPECT_ILLCMD;
if (nfifo > 2)
nfifo = 2; /* We fixed it.. */
} else
nfifo = ncr53c9x_rdfifo(sc, NCR_RDFIFO_START);
if (nfifo != 2) {
device_printf(sc->sc_dev, "RESELECT: %d bytes "
"in FIFO! [intr %x, stat %x, step %d, "
"prevphase %x]\n",
nfifo,
sc->sc_espintr,
sc->sc_espstat,
sc->sc_espstep,
sc->sc_prevphase);
ncr53c9x_init(sc, 1);
goto out;
}
sc->sc_selid = sc->sc_imess[0];
NCR_INTS(("selid=%02x ", sc->sc_selid));
/* Handle identify message */
ncr53c9x_msgin(sc);
if (sc->sc_state != NCR_CONNECTED &&
sc->sc_state != NCR_IDENTIFIED) {
/* IDENTIFY fail?! */
device_printf(sc->sc_dev, "identify failed, "
"state %d, intr %02x\n", sc->sc_state,
sc->sc_espintr);
ncr53c9x_init(sc, 1);
goto out;
}
goto shortcut; /* ie. next phase expected soon */
}
#define NCRINTR_DONE (NCRINTR_FC|NCRINTR_BS)
if ((sc->sc_espintr & NCRINTR_DONE) == NCRINTR_DONE) {
/*
* Arbitration won; examine the `step' register
* to determine how far the selection could progress.
*/
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
if (ecb == NULL) {
/*
* When doing path inquiry during boot
* FAS100A trigger a stray interrupt which
* we just ignore instead of panicing.
*/
if (sc->sc_state == NCR_IDLE &&
sc->sc_espstep == 0)
goto out;
panic("ncr53c9x: no nexus");
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
}
ti = &sc->sc_tinfo[ecb->ccb->ccb_h.target_id];
switch (sc->sc_espstep) {
case 0:
/*
* The target did not respond with a
* message out phase - probably an old
* device that doesn't recognize ATN.
* Clear ATN and just continue, the
* target should be in the command
* phase.
* XXXX check for command phase?
*/
NCRCMD(sc, NCRCMD_RSTATN);
break;
case 1:
if ((ti->flags & T_NEGOTIATE) == 0 &&
ecb->tag[0] == 0) {
device_printf(sc->sc_dev, "step 1 & "
"!NEG\n");
goto reset;
}
if (sc->sc_phase != MESSAGE_OUT_PHASE) {
device_printf(sc->sc_dev, "!MSGOUT\n");
goto reset;
}
if (ti->flags & T_WIDE) {
ti->flags |= T_WDTRSENT;
ncr53c9x_sched_msgout(SEND_WDTR);
}
if (ti->flags & T_NEGOTIATE) {
/* Start negotiating */
sc->sc_flags |= NCR_SYNCHNEGO;
if (ecb->tag[0])
ncr53c9x_sched_msgout(
SEND_TAG|SEND_SDTR);
else
ncr53c9x_sched_msgout(
SEND_SDTR);
} else {
/* Could not do ATN3 so send TAG */
ncr53c9x_sched_msgout(SEND_TAG);
}
sc->sc_prevphase = MESSAGE_OUT_PHASE; /* XXXX */
break;
case 3:
/*
* Grr, this is supposed to mean
* "target left command phase prematurely".
* It seems to happen regularly when
* sync mode is on.
* Look at FIFO to see if command went out.
* (Timing problems?)
*/
if (sc->sc_features & NCR_F_DMASELECT) {
if (sc->sc_cmdlen == 0)
/* Hope for the best.. */
break;
} else if ((NCR_READ_REG(sc, NCR_FFLAG)
& NCRFIFO_FF) == 0) {
/* Hope for the best.. */
break;
}
printf("(%s:%d:%d): selection failed;"
" %d left in FIFO "
"[intr %x, stat %x, step %d]\n",
device_get_nameunit(sc->sc_dev),
ecb->ccb->ccb_h.target_id,
ecb->ccb->ccb_h.target_lun,
NCR_READ_REG(sc, NCR_FFLAG)
& NCRFIFO_FF,
sc->sc_espintr, sc->sc_espstat,
sc->sc_espstep);
NCRCMD(sc, NCRCMD_FLUSH);
ncr53c9x_sched_msgout(SEND_ABORT);
goto out;
case 2:
/* Select stuck at Command Phase */
NCRCMD(sc, NCRCMD_FLUSH);
break;
case 4:
if (sc->sc_features & NCR_F_DMASELECT &&
sc->sc_cmdlen != 0)
printf("(%s:%d:%d): select; "
"%lu left in DMA buffer "
"[intr %x, stat %x, step %d]\n",
device_get_nameunit(sc->sc_dev),
ecb->ccb->ccb_h.target_id,
ecb->ccb->ccb_h.target_lun,
(u_long)sc->sc_cmdlen,
sc->sc_espintr,
sc->sc_espstat,
sc->sc_espstep);
/* So far, everything went fine */
break;
}
sc->sc_prevphase = INVALID_PHASE; /* ?? */
/* Do an implicit RESTORE POINTERS. */
sc->sc_dp = ecb->daddr;
sc->sc_dleft = ecb->dleft;
sc->sc_state = NCR_CONNECTED;
break;
} else {
device_printf(sc->sc_dev, "unexpected status after "
"select: [intr %x, stat %x, step %x]\n",
sc->sc_espintr, sc->sc_espstat, sc->sc_espstep);
NCRCMD(sc, NCRCMD_FLUSH);
DELAY(1);
goto reset;
}
if (sc->sc_state == NCR_IDLE) {
device_printf(sc->sc_dev, "stray interrupt\n");
- Try to not leak resources in the attach functions of the esp(4) SBus front-end and the LSI64854 and NCR53C9x code in case one of these functions fails. Add detach functions to these parts and make esp(4) detachable. - Revert rev. 1.7 of esp_sbus.c, since rev. 1.34 of sbus.c the clockfreq IVAR defaults to the per-child values. - Merge ncr53c9x.c rev. 1.111 from NetBSD (partial): On reset, clear state flags and the msgout queue. In NetBSD code to notify the upper layer (i.e. CAM in FreeBSD) on reset was also added with this revision. This is believed to be not necessary in FreeBSD and was not merged. This makes ncr53c9x.c to be in sync with NetBSD up to rev. 1.114. - Conditionalize the LSI64854 support on sbus(4) only instead of sbus(4) and esp(4) as it's also required for the 'dma', 'espdma' and 'ledma' busses/devices as well as the 'SUNW,bpp' device (printer port) which all hang off of sbus(4). - Add a driver for the 'dma', 'espdma' and 'ledma' (pseudo-)busses/ devices. These busses and devices actually represent the LSI64854 DMA engines for the ESP SCSI and LANCE Ethernet controllers found on the SBus of Ultra 1 and SBus add-on cards. With 'espdma' and 'ledma' the 'esp' and 'le' devices hang off of the respective DMA bus instead of directly from the SBus. The 'dma' devices are either also used in this manner or on some add-on cards also as a companion device to an 'esp' device which also hangs off directly from the SBus. With the latter variant it's a bit tricky to glue the DMA engine to the core logic of the respective 'esp' device. With rev. 1.35 of sbus.c we are however guaranteed that such a 'dma' device is probed before the respective 'esp' device which simplifies things a lot. [1] - In the esp(4) SBus front-end read the part-unique ID code of Fast-SCSI capable chips the right way. This fixes erroneously detecting some chips as FAS366 when in fact they are not. Add explicit checks for the FAS100A, FAS216 and FAS236 variants instead treating all of these as ESP200. That way we can correctly set the respective Fast-SCSI config bits instead of driving them out of specs. This includes adding the FAS100A and FAS236 variants to the NCR53C9x core code. We probably still subsume some chip variants as ESP200 while in fact they are another variant which however shouldn't really matter as this will only happen when these chips are driven at 25MHz or less which implies not being able to run Fast-SCSI. [3] - Add a workaround to the NCR53C9x interrupt handler which ignores the stray interrupt generated by FAS100A when doing path inquiry during boot and which otherwiese would trigger a panic. - Add support for the 'esp' devices hanging off of a 'dma' or 'espdma' busses or which are companions of 'dma' devices to esp(4). In case of the variants that hang off of a DMA device this is a bit hackish as esp(4) then directly uses the softc of the respective parent to talk to the DMA engine. It might make sense to add an interface for this in order to implement this in a cleaner way however it's not yet clear how the requirements for the LANCE Ethernet controllers are and the hack works for now. [2] This effectively adds support for the onboard SCSI controller in Ultra 1 as well as most of the ESP-based SBus add-on cards to esp(4). With this the code for supporting the Performance Technologies SBS430 SBus SCSI add-on cards is also largely in place the remaining bits were however omitted as it's unclear from the NetBSD how to couple the DMA engine and the core logic together for these cards. Obtained from: OpenBSD [1] Obtained from: NetBSD [2] Clue from: BSD/OS [3] Reviewed by: scottl (earlier version) Tested with: FSBE/S add-on card (FAS236), SSHA add-on card (ESP100A), Ultra 1 (onboard FAS100A), Ultra 2 (onboard FAS366)
2005-05-19 14:51:10 +00:00
goto out;
}
break;
case NCR_CONNECTED:
if ((sc->sc_flags & NCR_ICCS) != 0) {
/* "Initiate Command Complete Steps" in progress */
u_char msg;
sc->sc_flags &= ~NCR_ICCS;
if (!(sc->sc_espintr & NCRINTR_DONE)) {
device_printf(sc->sc_dev, "ICCS: "
": [intr %x, stat %x, step %x]\n",
sc->sc_espintr, sc->sc_espstat,
sc->sc_espstep);
}
ncr53c9x_rdfifo(sc, NCR_RDFIFO_START);
if (sc->sc_imlen < 2)
device_printf(sc->sc_dev, "can't get status, "
"only %d bytes\n", (int)sc->sc_imlen);
ecb->stat = sc->sc_imess[sc->sc_imlen - 2];
msg = sc->sc_imess[sc->sc_imlen - 1];
NCR_PHASE(("<stat:(%x,%x)>", ecb->stat, msg));
if (msg == MSG_CMDCOMPLETE) {
ecb->dleft = (ecb->flags & ECB_TENTATIVE_DONE)
? 0 : sc->sc_dleft;
if ((ecb->flags & ECB_SENSE) == 0)
ecb->ccb->csio.resid = ecb->dleft;
sc->sc_state = NCR_CMDCOMPLETE;
} else
device_printf(sc->sc_dev, "STATUS_PHASE: "
"msg %d\n", msg);
sc->sc_imlen = 0;
NCRCMD(sc, NCRCMD_MSGOK);
goto shortcut; /* ie. wait for disconnect */
}
break;
default:
device_printf(sc->sc_dev, "invalid state: %d [intr %x, "
"phase(c %x, p %x)]\n", sc->sc_state,
sc->sc_espintr, sc->sc_phase, sc->sc_prevphase);
goto reset;
}
/*
* Driver is now in state NCR_CONNECTED, i.e. we
* have a current command working the SCSI bus.
*/
if (sc->sc_state != NCR_CONNECTED || ecb == NULL) {
panic("ncr53c9x: no nexus");
}
switch (sc->sc_phase) {
case MESSAGE_OUT_PHASE:
NCR_PHASE(("MESSAGE_OUT_PHASE "));
ncr53c9x_msgout(sc);
sc->sc_prevphase = MESSAGE_OUT_PHASE;
break;
case MESSAGE_IN_PHASE:
msgin:
NCR_PHASE(("MESSAGE_IN_PHASE "));
if ((sc->sc_espintr & NCRINTR_BS) != 0) {
if ((sc->sc_rev != NCR_VARIANT_FAS366) ||
!(sc->sc_espstat2 & NCRFAS_STAT2_EMPTY)) {
NCRCMD(sc, NCRCMD_FLUSH);
}
sc->sc_flags |= NCR_WAITI;
NCRCMD(sc, NCRCMD_TRANS);
} else if ((sc->sc_espintr & NCRINTR_FC) != 0) {
if ((sc->sc_flags & NCR_WAITI) == 0) {
device_printf(sc->sc_dev, "MSGIN: unexpected "
"FC bit: [intr %x, stat %x, step %x]\n",
sc->sc_espintr, sc->sc_espstat,
sc->sc_espstep);
}
sc->sc_flags &= ~NCR_WAITI;
ncr53c9x_rdfifo(sc,
(sc->sc_prevphase == sc->sc_phase) ?
NCR_RDFIFO_CONTINUE : NCR_RDFIFO_START);
ncr53c9x_msgin(sc);
} else {
device_printf(sc->sc_dev, "MSGIN: weird bits: "
"[intr %x, stat %x, step %x]\n",
sc->sc_espintr, sc->sc_espstat, sc->sc_espstep);
}
sc->sc_prevphase = MESSAGE_IN_PHASE;
goto shortcut; /* i.e. expect data to be ready */
case COMMAND_PHASE:
/*
* Send the command block. Normally we don't see this
* phase because the SEL_ATN command takes care of
* all this. However, we end up here if either the
* target or we wanted to exchange some more messages
* first (e.g. to start negotiations).
*/
NCR_PHASE(("COMMAND_PHASE 0x%02x (%d) ",
ecb->cmd.cmd.opcode, ecb->clen));
if (NCR_READ_REG(sc, NCR_FFLAG) & NCRFIFO_FF) {
NCRCMD(sc, NCRCMD_FLUSH);
/* DELAY(1);*/
}
if (sc->sc_features & NCR_F_DMASELECT) {
/* setup DMA transfer for command */
size = ecb->clen;
sc->sc_cmdlen = size;
sc->sc_cmdp = (caddr_t)&ecb->cmd.cmd;
NCRDMA_SETUP(sc, &sc->sc_cmdp, &sc->sc_cmdlen,
0, &size);
/* Program the SCSI counter */
NCR_SET_COUNT(sc, size);
/* load the count in */
NCRCMD(sc, NCRCMD_NOP|NCRCMD_DMA);
/* start the command transfer */
NCRCMD(sc, NCRCMD_TRANS | NCRCMD_DMA);
NCRDMA_GO(sc);
} else {
ncr53c9x_wrfifo(sc, (u_char *)&ecb->cmd.cmd, ecb->clen);
NCRCMD(sc, NCRCMD_TRANS);
}
sc->sc_prevphase = COMMAND_PHASE;
break;
case DATA_OUT_PHASE:
NCR_PHASE(("DATA_OUT_PHASE [%ld] ",(long)sc->sc_dleft));
NCRCMD(sc, NCRCMD_FLUSH);
size = min(sc->sc_dleft, sc->sc_maxxfer);
NCRDMA_SETUP(sc, &sc->sc_dp, &sc->sc_dleft, 0, &size);
sc->sc_prevphase = DATA_OUT_PHASE;
goto setup_xfer;
case DATA_IN_PHASE:
NCR_PHASE(("DATA_IN_PHASE "));
if (sc->sc_rev == NCR_VARIANT_ESP100)
NCRCMD(sc, NCRCMD_FLUSH);
size = min(sc->sc_dleft, sc->sc_maxxfer);
NCRDMA_SETUP(sc, &sc->sc_dp, &sc->sc_dleft, 1, &size);
sc->sc_prevphase = DATA_IN_PHASE;
setup_xfer:
/* Target returned to data phase: wipe "done" memory */
ecb->flags &= ~ECB_TENTATIVE_DONE;
/* Program the SCSI counter */
NCR_SET_COUNT(sc, size);
/* load the count in */
NCRCMD(sc, NCRCMD_NOP|NCRCMD_DMA);
/*
* Note that if `size' is 0, we've already transceived
* all the bytes we want but we're still in DATA PHASE.
* Apparently, the device needs padding. Also, a
* transfer size of 0 means "maximum" to the chip
* DMA logic.
*/
NCRCMD(sc,
(size == 0 ? NCRCMD_TRPAD : NCRCMD_TRANS) | NCRCMD_DMA);
NCRDMA_GO(sc);
goto out;
case STATUS_PHASE:
NCR_PHASE(("STATUS_PHASE "));
sc->sc_flags |= NCR_ICCS;
NCRCMD(sc, NCRCMD_ICCS);
sc->sc_prevphase = STATUS_PHASE;
goto shortcut; /* i.e. expect status results soon */
case INVALID_PHASE:
break;
default:
device_printf(sc->sc_dev, "unexpected bus phase; resetting\n");
goto reset;
}
out:
mtx_unlock(&sc->sc_lock);
return;
reset:
ncr53c9x_init(sc, 1);
goto out;
finish:
ncr53c9x_done(sc, ecb);
goto out;
sched:
sc->sc_state = NCR_IDLE;
ncr53c9x_sched(sc);
goto out;
shortcut:
/*
* The idea is that many of the SCSI operations take very little
* time, and going away and getting interrupted is too high an
* overhead to pay. For example, selecting, sending a message
* and command and then doing some work can be done in one "pass".
*
* The delay is a heuristic. It is 2 when at 20MHz, 2 at 25MHz and 1
* at 40MHz. This needs testing.
*/
{
struct timeval wait, cur;
microtime(&wait);
wait.tv_usec += 50 / sc->sc_freq;
if (wait.tv_usec > 1000000) {
wait.tv_sec++;
wait.tv_usec -= 1000000;
}
do {
if (NCRDMA_ISINTR(sc))
goto again;
microtime(&cur);
} while (cur.tv_sec <= wait.tv_sec &&
cur.tv_usec <= wait.tv_usec);
}
goto out;
}
static void
ncr53c9x_abort(sc, ecb)
struct ncr53c9x_softc *sc;
struct ncr53c9x_ecb *ecb;
{
/* 2 secs for the abort */
ecb->timeout = NCR_ABORT_TIMEOUT;
ecb->flags |= ECB_ABORT;
if (ecb == sc->sc_nexus) {
/*
* If we're still selecting, the message will be scheduled
* after selection is complete.
*/
if (sc->sc_state == NCR_CONNECTED)
ncr53c9x_sched_msgout(SEND_ABORT);
/*
* Reschedule timeout.
*/
ecb->ccb->ccb_h.timeout_ch =
timeout(ncr53c9x_timeout, ecb, mstohz(ecb->timeout));
} else {
/*
* Just leave the command where it is.
* XXX - what choice do we have but to reset the SCSI
* eventually?
*/
if (sc->sc_state == NCR_IDLE)
ncr53c9x_sched(sc);
}
}
static void
ncr53c9x_timeout(void *arg)
{
struct ncr53c9x_ecb *ecb = arg;
union ccb *ccb = ecb->ccb;
struct ncr53c9x_softc *sc = ecb->sc;
struct ncr53c9x_tinfo *ti = &sc->sc_tinfo[ccb->ccb_h.target_id];
xpt_print_path(ccb->ccb_h.path);
device_printf(sc->sc_dev, "timed out [ecb %p (flags 0x%x, dleft %x, "
"stat %x)], <state %d, nexus %p, phase(l %x, c %x, p %x), "
"resid %lx, msg(q %x,o %x) %s>",
ecb, ecb->flags, ecb->dleft, ecb->stat,
sc->sc_state, sc->sc_nexus,
NCR_READ_REG(sc, NCR_STAT),
sc->sc_phase, sc->sc_prevphase,
(long)sc->sc_dleft, sc->sc_msgpriq, sc->sc_msgout,
NCRDMA_ISACTIVE(sc) ? "DMA active" : "");
#if defined(NCR53C9X_DEBUG) && NCR53C9X_DEBUG > 1
printf("TRACE: %s.", ecb->trace);
#endif
mtx_lock(&sc->sc_lock);
if (ecb->flags & ECB_ABORT) {
/* abort timed out */
printf(" AGAIN\n");
ncr53c9x_init(sc, 1);
} else {
/* abort the operation that has timed out */
printf("\n");
ccb->ccb_h.status = CAM_CMD_TIMEOUT;
ncr53c9x_abort(sc, ecb);
/* Disable sync mode if stuck in a data phase */
if (ecb == sc->sc_nexus &&
(ti->flags & T_SYNCMODE) != 0 &&
(sc->sc_phase & (MSGI|CDI)) == 0) {
/* XXX ASYNC CALLBACK! */
xpt_print_path(ccb->ccb_h.path);
printf("sync negotiation disabled\n");
sc->sc_cfflags |=
(1 << ((ccb->ccb_h.target_id & 7) + 8));
}
}
mtx_unlock(&sc->sc_lock);
}
static void
ncr53c9x_watch(void *arg)
{
struct ncr53c9x_softc *sc = (struct ncr53c9x_softc *)arg;
struct ncr53c9x_tinfo *ti;
struct ncr53c9x_linfo *li;
int t;
/* Delete any structures that have not been used in 10min. */
time_t old = time_second - (10 * 60);
mtx_lock(&sc->sc_lock);
for (t = 0; t < sc->sc_ntarg; t++) {
ti = &sc->sc_tinfo[t];
li = LIST_FIRST(&ti->luns);
while (li) {
if (li->last_used < old &&
li->untagged == NULL &&
li->used == 0) {
if (li->lun < NCR_NLUN)
ti->lun[li->lun] = NULL;
LIST_REMOVE(li, link);
free(li, M_DEVBUF);
/* Restart the search at the beginning */
li = LIST_FIRST(&ti->luns);
continue;
}
li = LIST_NEXT(li, link);
}
}
mtx_unlock(&sc->sc_lock);
callout_reset(&sc->sc_watchdog, 60 * hz, ncr53c9x_watch, sc);
}