d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
62e3af5225
freebsd-nq/sys/sparc64/sbus/lsi64854.c

728 lines
19 KiB
C
Raw Normal View History

Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/*-
* 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.
*
*/
Sync licenses and the corresponding RCS IDs with NetBSD, mainly switching the licenses of Matthew R. Green and the TNF to 2-clause. Obtained from: NetBSD
2011-03-12 14:33:32 +00:00
/* $NetBSD: lsi64854.c,v 1.33 2008/04/28 20:23:50 martin Exp $ */
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/*-
* Copyright (c) 1998 The NetBSD Foundation, Inc.
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by Paul Kranenburg.
*
* 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 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.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/bus.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/mutex.h>
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
#include <sys/rman.h>
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
#include <machine/bus.h>
#include <cam/cam.h>
#include <cam/cam_ccb.h>
#include <cam/scsi/scsi_all.h>
Fix paths after repocopies done by scottl Reviewed by: marius OK'ed by: scottl
2004-11-10 14:09:52 +00:00
#include <sparc64/sbus/lsi64854reg.h>
#include <sparc64/sbus/lsi64854var.h>
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
#include <dev/esp/ncr53c9xreg.h>
#include <dev/esp/ncr53c9xvar.h>
#ifdef DEBUG
#define LDB_SCSI 1
#define LDB_ENET 2
#define LDB_PP 4
#define LDB_ANY 0xff
int lsi64854debug = 0;
#define DPRINTF(a,x) do { if (lsi64854debug & (a)) printf x ; } while (0)
#else
#define DPRINTF(a,x)
#endif
#define MAX_DMA_SZ (16*1024*1024)
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
static void lsi64854_reset(struct lsi64854_softc *);
static void lsi64854_map_scsi(void *, bus_dma_segment_t *, int, int);
static int lsi64854_setup(struct lsi64854_softc *, caddr_t *, size_t *,
int, size_t *);
static int lsi64854_scsi_intr(void *);
static int lsi64854_enet_intr(void *);
static int lsi64854_setup_pp(struct lsi64854_softc *, caddr_t *, size_t *,
int, size_t *);
static int lsi64854_pp_intr(void *);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/*
* Finish attaching this DMA device.
* Front-end must fill in these fields:
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
* sc_res
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
* sc_burst
* sc_channel (one of SCSI, ENET, PP)
* sc_client (one of SCSI, ENET, PP `soft_c' pointers)
*/
- 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
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
lsi64854_attach(struct lsi64854_softc *sc)
{
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
bus_dma_lock_t *lockfunc;
struct ncr53c9x_softc *nsc;
void *lockfuncarg;
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
uint32_t csr;
- 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;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
lockfunc = NULL;
lockfuncarg = NULL;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
switch (sc->sc_channel) {
case L64854_CHANNEL_SCSI:
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
nsc = sc->sc_client;
if (NCR_LOCK_INITIALIZED(nsc) == 0) {
device_printf(sc->sc_dev, "mutex not initialized\n");
return (ENXIO);
}
lockfunc = busdma_lock_mutex;
lockfuncarg = &nsc->sc_lock;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
sc->intr = lsi64854_scsi_intr;
sc->setup = lsi64854_setup;
break;
case L64854_CHANNEL_ENET:
sc->intr = lsi64854_enet_intr;
break;
case L64854_CHANNEL_PP:
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
sc->intr = lsi64854_pp_intr;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
sc->setup = lsi64854_setup_pp;
break;
default:
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
device_printf(sc->sc_dev, "unknown channel\n");
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
sc->reset = lsi64854_reset;
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
if (sc->setup != NULL) {
error = bus_dma_tag_create(
sc->sc_parent_dmat, /* parent */
1, 0, /* alignment, boundary */
BUS_SPACE_MAXADDR, /* lowaddr */
BUS_SPACE_MAXADDR, /* highaddr */
NULL, NULL, /* filter, filterarg */
MAX_DMA_SZ, /* maxsize */
1, /* nsegments */
MAX_DMA_SZ, /* maxsegsize */
BUS_DMA_ALLOCNOW, /* flags */
lockfunc, lockfuncarg, /* lockfunc, lockfuncarg */
&sc->sc_buffer_dmat);
if (error != 0) {
device_printf(sc->sc_dev,
"cannot allocate buffer DMA tag\n");
return (error);
}
error = bus_dmamap_create(sc->sc_buffer_dmat, 0,
&sc->sc_dmamap);
if (error != 0) {
device_printf(sc->sc_dev, "DMA map create failed\n");
bus_dma_tag_destroy(sc->sc_buffer_dmat);
return (error);
}
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
csr = L64854_GCSR(sc);
sc->sc_rev = csr & L64854_DEVID;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
if (sc->sc_rev == DMAREV_HME)
- 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);
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
device_printf(sc->sc_dev, "DMA rev. ");
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
switch (sc->sc_rev) {
case DMAREV_0:
printf("0");
break;
case DMAREV_ESC:
- 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
printf("ESC");
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
break;
case DMAREV_1:
printf("1");
break;
case DMAREV_PLUS:
printf("1+");
break;
case DMAREV_2:
printf("2");
break;
default:
printf("unknown (0x%x)", sc->sc_rev);
}
DPRINTF(LDB_ANY, (", burst 0x%x, csr 0x%x", sc->sc_burst, csr));
printf("\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
return (0);
}
int
lsi64854_detach(struct lsi64854_softc *sc)
{
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
if (sc->setup != NULL) {
bus_dmamap_sync(sc->sc_buffer_dmat, sc->sc_dmamap,
(L64854_GCSR(sc) & L64854_WRITE) != 0 ?
BUS_DMASYNC_PREREAD : BUS_DMASYNC_PREWRITE);
- 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
bus_dmamap_unload(sc->sc_buffer_dmat, sc->sc_dmamap);
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
bus_dmamap_destroy(sc->sc_buffer_dmat, sc->sc_dmamap);
bus_dma_tag_destroy(sc->sc_buffer_dmat);
}
- 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);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
/*
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
* DMAWAIT waits while condition is true.
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
*/
#define DMAWAIT(SC, COND, MSG, DONTPANIC) do if (COND) { \
int count = 500000; \
while ((COND) && --count > 0) DELAY(1); \
if (count == 0) { \
printf("%s: line %d: CSR = 0x%lx\n", __FILE__, __LINE__, \
(u_long)L64854_GCSR(SC)); \
if (DONTPANIC) \
printf(MSG); \
else \
panic(MSG); \
} \
} while (0)
#define DMA_DRAIN(sc, dontpanic) do { \
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
uint32_t csr; \
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/* \
* DMA rev0 & rev1: we are not allowed to touch the DMA "flush" \
* and "drain" bits while it is still thinking about a \
* request. \
* other revs: D_ESC_R_PEND bit reads as 0 \
*/ \
DMAWAIT(sc, L64854_GCSR(sc) & D_ESC_R_PEND, "R_PEND", dontpanic);\
if (sc->sc_rev != DMAREV_HME) { \
/* \
* Select drain bit based on revision \
* also clears errors and D_TC flag \
*/ \
csr = L64854_GCSR(sc); \
if (sc->sc_rev == DMAREV_1 || sc->sc_rev == DMAREV_0) \
csr |= D_ESC_DRAIN; \
else \
csr |= L64854_INVALIDATE; \
\
L64854_SCSR(sc,csr); \
} \
/* \
* Wait for draining to finish \
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
* rev0 & rev1 call this PACKCNT \
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
*/ \
DMAWAIT(sc, L64854_GCSR(sc) & L64854_DRAINING, "DRAINING", dontpanic);\
} while(0)
#define DMA_FLUSH(sc, dontpanic) do { \
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
uint32_t csr; \
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/* \
* DMA rev0 & rev1: we are not allowed to touch the DMA "flush" \
* and "drain" bits while it is still thinking about a \
* request. \
* other revs: D_ESC_R_PEND bit reads as 0 \
*/ \
DMAWAIT(sc, L64854_GCSR(sc) & D_ESC_R_PEND, "R_PEND", dontpanic);\
csr = L64854_GCSR(sc); \
csr &= ~(L64854_WRITE|L64854_EN_DMA); /* no-ops on ENET */ \
csr |= L64854_INVALIDATE; /* XXX FAS ? */ \
L64854_SCSR(sc,csr); \
} while(0)
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
static void
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
lsi64854_reset(struct lsi64854_softc *sc)
{
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
uint32_t csr;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
DMA_FLUSH(sc, 1);
csr = L64854_GCSR(sc);
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
DPRINTF(LDB_ANY, ("%s: csr 0x%x\n", __func__, csr));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
if (sc->sc_dmasize != 0) {
bus_dmamap_sync(sc->sc_buffer_dmat, sc->sc_dmamap,
(csr & D_WRITE) != 0 ? BUS_DMASYNC_PREREAD :
BUS_DMASYNC_PREWRITE);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
bus_dmamap_unload(sc->sc_buffer_dmat, sc->sc_dmamap);
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
}
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
if (sc->sc_rev == DMAREV_HME)
L64854_SCSR(sc, csr | D_HW_RESET_FAS366);
csr |= L64854_RESET; /* reset DMA */
L64854_SCSR(sc, csr);
DELAY(200); /* > 10 Sbus clocks(?) */
/*DMAWAIT1(sc); why was this here? */
csr = L64854_GCSR(sc);
csr &= ~L64854_RESET; /* de-assert reset line */
L64854_SCSR(sc, csr);
DELAY(5); /* allow a few ticks to settle */
csr = L64854_GCSR(sc);
csr |= L64854_INT_EN; /* enable interrupts */
if (sc->sc_rev > DMAREV_1 && sc->sc_channel == L64854_CHANNEL_SCSI) {
if (sc->sc_rev == DMAREV_HME)
csr |= D_TWO_CYCLE;
else
csr |= D_FASTER;
}
/* Set burst */
switch (sc->sc_rev) {
case DMAREV_HME:
case DMAREV_2:
csr &= ~L64854_BURST_SIZE;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
if (sc->sc_burst == 32)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr |= L64854_BURST_32;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
else if (sc->sc_burst == 16)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr |= L64854_BURST_16;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
else
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr |= L64854_BURST_0;
break;
case DMAREV_ESC:
csr |= D_ESC_AUTODRAIN; /* Auto-drain */
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
if (sc->sc_burst == 32)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr &= ~D_ESC_BURST;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
else
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr |= D_ESC_BURST;
break;
default:
break;
}
L64854_SCSR(sc, csr);
if (sc->sc_rev == DMAREV_HME) {
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
bus_write_4(sc->sc_res, L64854_REG_ADDR, 0);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
sc->sc_dmactl = csr;
}
sc->sc_active = 0;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
DPRINTF(LDB_ANY, ("%s: done, csr 0x%x\n", __func__, csr));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
static void
lsi64854_map_scsi(void *arg, bus_dma_segment_t *segs, int nseg, int error)
{
struct lsi64854_softc *sc;
sc = (struct lsi64854_softc *)arg;
if (nseg != 1)
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
panic("%s: cannot map %d segments\n", __func__, nseg);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
bus_dmamap_sync(sc->sc_buffer_dmat, sc->sc_dmamap,
sc->sc_datain ? BUS_DMASYNC_PREREAD : BUS_DMASYNC_PREWRITE);
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
bus_write_4(sc->sc_res, L64854_REG_ADDR, segs[0].ds_addr);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
#define DMAMAX(a) (MAX_DMA_SZ - ((a) & (MAX_DMA_SZ - 1)))
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/*
* setup a DMA transfer
*/
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
static int
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
lsi64854_setup(struct lsi64854_softc *sc, caddr_t *addr, size_t *len,
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
int datain, size_t *dmasize)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
{
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
long bcnt;
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
uint32_t csr;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
DMA_FLUSH(sc, 0);
#if 0
DMACSR(sc) &= ~D_INT_EN;
#endif
sc->sc_dmaaddr = addr;
sc->sc_dmalen = len;
sc->sc_datain = datain;
/*
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
* The rules say we cannot transfer more than the limit
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
* of this DMA chip (64k for old and 16Mb for new),
* and we cannot cross a 16Mb boundary.
*/
*dmasize = sc->sc_dmasize =
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
ulmin(*dmasize, DMAMAX((size_t)*sc->sc_dmaaddr));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
DPRINTF(LDB_ANY, ("%s: dmasize=%ld\n", __func__, (long)sc->sc_dmasize));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/*
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
* XXX what length?
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
*/
if (sc->sc_rev == DMAREV_HME) {
L64854_SCSR(sc, sc->sc_dmactl | L64854_RESET);
L64854_SCSR(sc, sc->sc_dmactl);
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
bus_write_4(sc->sc_res, L64854_REG_CNT, *dmasize);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
/* Program the DMA address */
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
if (sc->sc_dmasize != 0)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
if (bus_dmamap_load(sc->sc_buffer_dmat, sc->sc_dmamap,
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
*sc->sc_dmaaddr, sc->sc_dmasize, lsi64854_map_scsi, sc, 0))
panic("%s: cannot allocate DVMA address", __func__);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
if (sc->sc_rev == DMAREV_ESC) {
/* DMA ESC chip bug work-around */
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
bcnt = sc->sc_dmasize;
if (((bcnt + (long)*sc->sc_dmaaddr) & PAGE_MASK_8K) != 0)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
bcnt = roundup(bcnt, PAGE_SIZE_8K);
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
bus_write_4(sc->sc_res, L64854_REG_CNT, bcnt);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
/* Setup DMA control register */
csr = L64854_GCSR(sc);
if (datain)
csr |= L64854_WRITE;
else
csr &= ~L64854_WRITE;
csr |= L64854_INT_EN;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
if (sc->sc_rev == DMAREV_HME)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr |= (D_DSBL_SCSI_DRN | D_EN_DMA);
L64854_SCSR(sc, csr);
return (0);
}
/*
* Pseudo (chained) interrupt from the esp driver to kick the
* current running DMA transfer. Called from ncr53c9x_intr()
* for now.
*
* return 1 if it was a DMA continue.
*/
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
static int
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
lsi64854_scsi_intr(void *arg)
{
struct lsi64854_softc *sc = arg;
struct ncr53c9x_softc *nsc = sc->sc_client;
int trans, resid;
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
uint32_t csr;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr = L64854_GCSR(sc);
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
DPRINTF(LDB_SCSI, ("%s: addr 0x%x, csr %b\n", __func__,
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
bus_read_4(sc->sc_res, L64854_REG_ADDR), csr, DDMACSR_BITS));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
if (csr & (D_ERR_PEND|D_SLAVE_ERR)) {
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
device_printf(sc->sc_dev, "error: csr=%b\n", csr, DDMACSR_BITS);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr &= ~D_EN_DMA; /* Stop DMA */
/* Invalidate the queue; SLAVE_ERR bit is write-to-clear */
csr |= D_INVALIDATE|D_SLAVE_ERR;
L64854_SCSR(sc, csr);
return (-1);
}
/* This is an "assertion" :) */
if (sc->sc_active == 0)
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
panic("%s: DMA wasn't active", __func__);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
DMA_DRAIN(sc, 0);
/* DMA has stopped */
csr &= ~D_EN_DMA;
L64854_SCSR(sc, csr);
sc->sc_active = 0;
if (sc->sc_dmasize == 0) {
/* A "Transfer Pad" operation completed */
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
DPRINTF(LDB_SCSI, ("%s: discarded %d bytes (tcl=%d, tcm=%d)\n",
__func__, NCR_READ_REG(nsc, NCR_TCL) |
(NCR_READ_REG(nsc, NCR_TCM) << 8),
NCR_READ_REG(nsc, NCR_TCL), NCR_READ_REG(nsc, NCR_TCM)));
return (0);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
resid = 0;
/*
* If a transfer onto the SCSI bus gets interrupted by the device
* (e.g. for a SAVEPOINTER message), the data in the FIFO counts
* as residual since the NCR53C9X counter registers get decremented
* as bytes are clocked into the FIFO.
*/
if (!(csr & D_WRITE) &&
(resid = (NCR_READ_REG(nsc, NCR_FFLAG) & NCRFIFO_FF)) != 0) {
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
DPRINTF(LDB_SCSI, ("%s: empty esp FIFO of %d ", __func__,
resid));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
if (nsc->sc_rev == NCR_VARIANT_FAS366 &&
(NCR_READ_REG(nsc, NCR_CFG3) & NCRFASCFG3_EWIDE))
resid <<= 1;
}
if ((nsc->sc_espstat & NCRSTAT_TC) == 0) {
/*
* `Terminal count' is off, so read the residue
* out of the NCR53C9X counter registers.
*/
resid += (NCR_READ_REG(nsc, NCR_TCL) |
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
(NCR_READ_REG(nsc, NCR_TCM) << 8) |
((nsc->sc_cfg2 & NCRCFG2_FE) ?
(NCR_READ_REG(nsc, NCR_TCH) << 16) : 0));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
if (resid == 0 && sc->sc_dmasize == 65536 &&
(nsc->sc_cfg2 & NCRCFG2_FE) == 0)
/* A transfer of 64K is encoded as `TCL=TCM=0' */
resid = 65536;
}
trans = sc->sc_dmasize - resid;
Correct spelling in comments. Submitted by: brucec
2011-04-22 09:31:40 +00:00
if (trans < 0) { /* transferred < 0? */
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
#if 0
/*
* This situation can happen in perfectly normal operation
* if the ESP is reselected while using DMA to select
* another target. As such, don't print the warning.
*/
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
device_printf(sc->sc_dev, "xfer (%d) > req (%d)\n", trans,
sc->sc_dmasize);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
#endif
trans = sc->sc_dmasize;
}
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
DPRINTF(LDB_SCSI, ("%s: tcl=%d, tcm=%d, tch=%d; trans=%d, resid=%d\n",
__func__, NCR_READ_REG(nsc, NCR_TCL), NCR_READ_REG(nsc, NCR_TCM),
(nsc->sc_cfg2 & NCRCFG2_FE) ? NCR_READ_REG(nsc, NCR_TCH) : 0,
trans, resid));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
if (sc->sc_dmasize != 0) {
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
bus_dmamap_sync(sc->sc_buffer_dmat, sc->sc_dmamap,
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
(csr & D_WRITE) != 0 ? BUS_DMASYNC_POSTREAD :
BUS_DMASYNC_POSTWRITE);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
bus_dmamap_unload(sc->sc_buffer_dmat, sc->sc_dmamap);
}
*sc->sc_dmalen -= trans;
*sc->sc_dmaaddr += trans;
#if 0 /* this is not normal operation just yet */
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
if (*sc->sc_dmalen == 0 || nsc->sc_phase != nsc->sc_prevphase)
return (0);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/* and again */
dma_start(sc, sc->sc_dmaaddr, sc->sc_dmalen, DMACSR(sc) & D_WRITE);
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
return (1);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
#endif
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
return (0);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
/*
* Pseudo (chained) interrupt to le driver to handle DMA errors.
*/
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
static int
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
lsi64854_enet_intr(void *arg)
{
struct lsi64854_softc *sc = arg;
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
uint32_t csr;
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
int i, rv;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr = L64854_GCSR(sc);
/* If the DMA logic shows an interrupt, claim it */
rv = ((csr & E_INT_PEND) != 0) ? 1 : 0;
if (csr & (E_ERR_PEND|E_SLAVE_ERR)) {
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
device_printf(sc->sc_dev, "error: csr=%b\n", csr, EDMACSR_BITS);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr &= ~L64854_EN_DMA; /* Stop DMA */
/* Invalidate the queue; SLAVE_ERR bit is write-to-clear */
csr |= E_INVALIDATE|E_SLAVE_ERR;
L64854_SCSR(sc, csr);
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
/* Will be drained with the LE_C0_IDON interrupt. */
sc->sc_dodrain = 1;
return (-1);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
/* XXX - is this necessary with E_DSBL_WR_INVAL on? */
if (sc->sc_dodrain) {
i = 10;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr |= E_DRAIN;
L64854_SCSR(sc, csr);
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
while (i-- > 0 && (L64854_GCSR(sc) & E_DRAINING))
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
DELAY(1);
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
sc->sc_dodrain = 0;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
- 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 (rv);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
static void
lsi64854_map_pp(void *arg, bus_dma_segment_t *segs, int nsegs, int error)
{
struct lsi64854_softc *sc;
sc = (struct lsi64854_softc *)arg;
if (nsegs != 1)
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
panic("%s: cannot map %d segments\n", __func__, nsegs);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
bus_dmamap_sync(sc->sc_buffer_dmat, sc->sc_dmamap, sc->sc_datain ?
BUS_DMASYNC_PREREAD : BUS_DMASYNC_PREWRITE);
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
bus_write_4(sc->sc_res, L64854_REG_ADDR, segs[0].ds_addr);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
bus_write_4(sc->sc_res, L64854_REG_CNT, sc->sc_dmasize);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
/*
* setup a DMA transfer
*/
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
static int
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
lsi64854_setup_pp(struct lsi64854_softc *sc, caddr_t *addr, size_t *len,
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
int datain, size_t *dmasize)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
{
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
uint32_t csr;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
DMA_FLUSH(sc, 0);
sc->sc_dmaaddr = addr;
sc->sc_dmalen = len;
sc->sc_datain = datain;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
DPRINTF(LDB_PP, ("%s: pp start %ld@%p,%d\n", __func__,
(long)*sc->sc_dmalen, *sc->sc_dmaaddr, datain ? 1 : 0));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/*
* the rules say we cannot transfer more than the limit
* of this DMA chip (64k for old and 16Mb for new),
* and we cannot cross a 16Mb boundary.
*/
*dmasize = sc->sc_dmasize =
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
ulmin(*dmasize, DMAMAX((size_t)*sc->sc_dmaaddr));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
DPRINTF(LDB_PP, ("%s: dmasize=%ld\n", __func__, (long)sc->sc_dmasize));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/* Program the DMA address */
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
if (sc->sc_dmasize != 0)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
if (bus_dmamap_load(sc->sc_buffer_dmat, sc->sc_dmamap,
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
*sc->sc_dmaaddr, sc->sc_dmasize, lsi64854_map_pp, sc, 0))
panic("%s: pp cannot allocate DVMA address", __func__);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/* Setup DMA control register */
csr = L64854_GCSR(sc);
csr &= ~L64854_BURST_SIZE;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
if (sc->sc_burst == 32)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr |= L64854_BURST_32;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
else if (sc->sc_burst == 16)
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr |= L64854_BURST_16;
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
else
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr |= L64854_BURST_0;
csr |= P_EN_DMA|P_INT_EN|P_EN_CNT;
#if 0
/* This bit is read-only in PP csr register */
if (datain)
csr |= P_WRITE;
else
csr &= ~P_WRITE;
#endif
L64854_SCSR(sc, csr);
return (0);
}
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
/*
* Parallel port DMA interrupt.
*/
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
static int
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
lsi64854_pp_intr(void *arg)
{
struct lsi64854_softc *sc = arg;
int ret, trans, resid = 0;
Re-commit the following changes which were committed to these files at their old location in sys/dev/esp after they were repo-copied to sys/sparc64/sbus at rev. 1.1: sys/dev/esp/lsi64854.c rev. 1.2 sys/dev/esp/lsi64854var.h rev. 1.2 Add some style(9) touch ups; style(9) states that new code should follow these conventions and, well, this is a new driver. Tested on: i386, sparc64 Reviewed by: scottl
2005-04-17 12:45:20 +00:00
uint32_t csr;
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr = L64854_GCSR(sc);
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
DPRINTF(LDB_PP, ("%s: addr 0x%x, csr %b\n", __func__,
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
bus_read_4(sc->sc_res, L64854_REG_ADDR), csr, PDMACSR_BITS));
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
if (csr & (P_ERR_PEND|P_SLAVE_ERR)) {
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
resid = bus_read_4(sc->sc_res, L64854_REG_CNT);
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
device_printf(sc->sc_dev, "error: resid %d csr=%b\n", resid,
csr, PDMACSR_BITS);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
csr &= ~P_EN_DMA; /* Stop DMA */
/* Invalidate the queue; SLAVE_ERR bit is write-to-clear */
csr |= P_INVALIDATE|P_SLAVE_ERR;
L64854_SCSR(sc, csr);
o lsi64854_enet_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt and leave the reset of the DMA engine to the caller so we don't kill any state there. - Move the static 'dodrain' flag to struct lsi64854_softc as there can be more than one LSI64854 used for a LANCE in a system and reset it again once draining the E-cache is done so we don't keep draining the cache with every interrupt. - Remove calling sc->sc_intrchain(), we will call lsi64854_enet_intr() via sc->intr() in the interrupt handler of the LANCE driver and not use it in chained mode. o lsi64854_pp_intr(): - Like lsi64854_scsi_intr() return -1 in case there was a DMA error so the caller can distinguish it from a normal interrupt. o Remove the no longer used sc_intrchain* from struct lsi64854_softc. o Make lsi64854_reset(), lsi64854_setup*() and lsi64854_*_intr() static to lsi64854.c as we do and will only call them via the respective function pointers in struct lsi64854_softc. o While here fix style(9) bugs (variable definition inside a nested scope).
2006-01-31 12:50:02 +00:00
return (-1);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
ret = (csr & P_INT_PEND) != 0;
if (sc->sc_active != 0) {
DMA_DRAIN(sc, 0);
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
resid = bus_read_4(sc->sc_res, L64854_REG_CNT);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
}
/* DMA has stopped */
csr &= ~D_EN_DMA;
L64854_SCSR(sc, csr);
sc->sc_active = 0;
trans = sc->sc_dmasize - resid;
Correct spelling in comments. Submitted by: brucec
2011-04-22 09:31:40 +00:00
if (trans < 0) /* transferred < 0? */
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
trans = sc->sc_dmasize;
*sc->sc_dmalen -= trans;
*sc->sc_dmaaddr += trans;
o Move the MODULE_DEPEND() for cam(4) from the esp_sbus.c front-end to the ncr53c9x.c core where it actually belongs so future front-ends don't need to add it. o Use the correct OFW property when looking for the initiator ID of the SBus device. o Don't specify an alignment when creating the parent DMA tag for SUNW,fas; their DMA engine doesn't require an alignment constraint and it's no inherited by the child DMA tags anyway (which probably is a bug though). o Drop the superfluous sc_maxsync and use sc_minsync instead. The former apparently was added due to a confusion with the maximum frequency used in cam(4), which basically corresponds to the inverse of minimum sync period. o Merge ncr53c9x.c from NetBSD: 1.116: NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command in ncr53c9x_select(). 1.125: free allocated resources on detach. o Static'ize ncr53c9x_action(), ncr53c9x_init() and ncr53c9x_reset() as these are not required outside of ncr53c9x.c. o In ncr53c9x_attach() don't leak the device mutex in case attaching fails. o Register an asynchronous notification handler so in case cam(4) reports a lost device we can cancel outstanding commands and restore the default parameters for the target in question. o For FAS366 correctly support 16-bit target IDs and let it know that we use 32-bit transfers. o Overhaul the negotiation of transfer settings. This includes distinguishing between current and goal transfer settings of the target so we can renegotiate their goal settings when necessary and correcting the order in which tagged, wide and synchronous transfers are negotiated. o If we are requesting sense, force a renegotiation if we are currently using anything different from asynchronous at 8 bit as the target might have lost our transfer negotiations. o In case of an XPT_RESET_BUS just directly call ncr53c9x_init() instead of issuing a NCRCMD_RSTSCSI, which in turn will issue an interrupt that is treated as an unexpected SCSI bus reset by ncr53c9x_intr() and thus calls ncr53c9x_init(). Remove the now no longer used ncr53c9x_scsi_reset(). o Correct an off-by-one error when setting cpi->max_lun. o In replace printf(9) with device_printf(9) calls where appropriate and in ncr53c9x_action() remove some unnecessarily verbose messages. o In ncr53c9x_sched() use TAILQ_FOREACH() instead of reimplementing it and consolidate two tagging-related target info checks into one. o In ncr53c9x_done() set the CAM status to CAM_SCSI_STATUS_ERROR when appropriate, respect CAM_DIS_AUTOSENSE and teach it to return SCSI status information. o In ncr53c9x_dequeue() ensure the tags are cleared. o Use ulmin() instead of min() where appropriate. o In ncr53c9x_msgout() consistently use the reset label. o When we're interrupted during a data phase and the DMA engine is still active, don't panic but reset the core and the DMA engine as this should be sufficient. Also, the typical problem for triggering this was the lack of renegotiation when requesting sense. o Correctly handle DEVICE RESETs. o Adapt the locking of esp(4) to MPSAFE cam(4). This includes moving the calls of lsi64854_attach() to the bus front-ends so it can pass the esp(4) mutex to bus_dma_tag_create(9). o Change the LSI64854 driver to not create a DMA tag and map for the Ethernet channel as le(4) will handle these on its own as well as sync and unload the DMA maps for the SCSI and parallel port channel after a DMA transfer. o Cam(4)'ify some NetBSD-centric comments. o Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) and take advantage of rman_get_rid(9) in order to save some softc members. Reviewed by: scottl MFC after: 1 month
2008-09-08 20:20:44 +00:00
if (sc->sc_dmasize != 0) {
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
bus_dmamap_sync(sc->sc_buffer_dmat, sc->sc_dmamap,
Style and minor changes: - Merge lsi64854.c rev. 1.25 from NetBSD: nuke trailing whitespace. - Update NetBSD RCS IDs according to what was actually already merged. - Remove dv_name from the lsi64854_softc and use device_printf() instead. - Use __func__ instead of hardcoded function names in error messages. - Use ulmin() instead of min() for comparing the DMA sizes as the values involved actually are represented by 64bit unsigned instead of 32bit unsigned. As far as I can't tell this doesn't make a difference in practice though. - Some style(9) fixes (mainly indentation). - Remove unnecessary braces.
2005-04-17 17:41:32 +00:00
(csr & D_WRITE) != 0 ? BUS_DMASYNC_POSTREAD :
BUS_DMASYNC_POSTWRITE);
Port the NetBSD esp(4) driver. This only includes the sbus front-end, so its primary use is for the FEPS/FAS366 SCSI found in Sun Ultra 1e and 2 machines. Once the pci front-end is ported, this driver can replace the amd(4) driver. The code as-is is fairly stable. I've disabled tagged-queueing until I can figure out a corruption bug related to it. I'm importing it now so that people with these machines can (finally) stop netbooting and report bugs before 5.3.
2004-06-10 05:11:39 +00:00
bus_dmamap_unload(sc->sc_buffer_dmat, sc->sc_dmamap);
}
return (ret != 0);
}
Reference in New Issue Copy Permalink