d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
ff1625c61d
freebsd-dev/sys/dev/mc146818/mc146818.c

288 lines
7.6 KiB
C
Raw Normal View History

Start each of the license/copyright comments with /*-, minor shuffle of lines
2005-01-06 01:43:34 +00:00
/*-
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
* Copyright (c) 2003 Izumi Tsutsui. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The name of the author may not be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* from: NetBSD: mc146818.c,v 1.4 2003/11/24 06:20:40 tsutsui Exp
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
/*
* mc146818 and compatible time of day chip subroutines
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/bus.h>
#include <sys/clock.h>
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
#include <sys/lock.h>
#include <sys/mutex.h>
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
#include <machine/bus.h>
#include <dev/mc146818/mc146818reg.h>
#include <dev/mc146818/mc146818var.h>
#include "clock_if.h"
static u_int mc146818_def_getcent(device_t);
static void mc146818_def_setcent(device_t, u_int);
static u_int mc146818_def_read(device_t, u_int);
static void mc146818_def_write(device_t, u_int, u_int);
int
mc146818_attach(device_t dev)
{
struct mc146818_softc *sc;
sc = device_get_softc(dev);
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
if (mtx_initialized(&sc->sc_mtx) == 0) {
device_printf(dev, "%s: mutex not initialized\n", __func__);
return (ENXIO);
}
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
if (sc->sc_mcread == NULL)
sc->sc_mcread = mc146818_def_read;
if (sc->sc_mcwrite == NULL)
sc->sc_mcwrite = mc146818_def_write;
if (sc->sc_flag & MC146818_NO_CENT_ADJUST) {
/*
* Note that setting MC146818_NO_CENT_ADJUST means that
* the century has to be stored in NVRAM somewhere.
*/
if (sc->sc_getcent == NULL)
sc->sc_getcent = mc146818_def_getcent;
if (sc->sc_setcent == NULL)
sc->sc_setcent = mc146818_def_setcent;
}
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_lock_spin(&sc->sc_mtx);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
if (!(*sc->sc_mcread)(dev, MC_REGD) & MC_REGD_VRT) {
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_unlock_spin(&sc->sc_mtx);
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
device_printf(dev, "%s: battery low\n", __func__);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
return (ENXIO);
}
sc->sc_rega = MC_BASE_32_KHz;
(*sc->sc_mcwrite)(dev, MC_REGA, sc->sc_rega);
sc->sc_regb = 0;
sc->sc_regb |= (sc->sc_flag & MC146818_BCD) ? 0 : MC_REGB_BINARY;
sc->sc_regb |= (sc->sc_flag & MC146818_12HR) ? 0 : MC_REGB_24HR;
(*sc->sc_mcwrite)(dev, MC_REGB, sc->sc_regb);
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_unlock_spin(&sc->sc_mtx);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
clock_register(dev, 1000000); /* 1 second resolution. */
return (0);
}
/*
* Get time of day and convert it to a struct timespec.
* Return 0 on success, an error number otherwise.
*/
int
mc146818_gettime(device_t dev, struct timespec *ts)
{
struct mc146818_softc *sc;
struct clocktime ct;
int timeout, cent, year;
sc = device_get_softc(dev);
timeout = 1000000; /* XXX how long should we wait? */
/*
* If MC_REGA_UIP is 0 we have at least 244us before the next
* update. If it's 1 an update is imminent.
*/
for (;;) {
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_lock_spin(&sc->sc_mtx);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
if (!((*sc->sc_mcread)(dev, MC_REGA) & MC_REGA_UIP))
break;
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_unlock_spin(&sc->sc_mtx);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
if (--timeout < 0) {
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
device_printf(dev, "%s: timeout\n", __func__);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
return (EBUSY);
}
}
#define FROMREG(x) ((sc->sc_flag & MC146818_BCD) ? FROMBCD(x) : (x))
ct.nsec = 0;
ct.sec = FROMREG((*sc->sc_mcread)(dev, MC_SEC));
ct.min = FROMREG((*sc->sc_mcread)(dev, MC_MIN));
ct.hour = FROMREG((*sc->sc_mcread)(dev, MC_HOUR));
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
/* Map dow from 1 - 7 to 0 - 6. */
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
ct.dow = FROMREG((*sc->sc_mcread)(dev, MC_DOW)) - 1;
ct.day = FROMREG((*sc->sc_mcread)(dev, MC_DOM));
ct.mon = FROMREG((*sc->sc_mcread)(dev, MC_MONTH));
year = FROMREG((*sc->sc_mcread)(dev, MC_YEAR));
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
year += sc->sc_year0;
if (sc->sc_flag & MC146818_NO_CENT_ADJUST) {
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
cent = (*sc->sc_getcent)(dev);
year += cent * 100;
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
} else if (year < POSIX_BASE_YEAR)
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
year += 100;
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_unlock_spin(&sc->sc_mtx);
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
ct.year = year;
return (clock_ct_to_ts(&ct, ts));
}
#ifdef notyet
int
mc146818_getsecs(device_t dev, int *secp)
{
struct mc146818_softc *sc;
int sec, timeout;
sc = device_get_softc(dev);
timeout = 1000000; /* XXX how long should we wait? */
for (;;) {
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_lock_spin(&sc->sc_mtx);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
if (!((*sc->sc_mcread)(dev, MC_REGA) & MC_REGA_UIP)) {
sec = FROMREG((*sc->sc_mcread)(dev, MC_SEC));
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_unlock_spin(&sc->sc_mtx);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
break;
}
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_unlock_spin(&sc->sc_mtx);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
if (--timeout == 0) {
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
device_printf(dev, "%s: timeout\n", __func__);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
return (EBUSY);
}
}
#undef FROMREG
*secp = sec;
return (0);
}
#endif
/*
* Set the time of day clock based on the value of the struct timespec arg.
* Return 0 on success, an error number otherwise.
*/
int
mc146818_settime(device_t dev, struct timespec *ts)
{
struct mc146818_softc *sc;
struct clocktime ct;
int cent, year;
sc = device_get_softc(dev);
/* Accuracy is only one second. */
if (ts->tv_nsec >= 500000000)
ts->tv_sec++;
ts->tv_nsec = 0;
clock_ts_to_ct(ts, &ct);
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_lock_spin(&sc->sc_mtx);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
/* Disable RTC updates and interrupts (if enabled). */
(*sc->sc_mcwrite)(dev, MC_REGB,
((sc->sc_regb & (MC_REGB_BINARY | MC_REGB_24HR)) | MC_REGB_SET));
#define TOREG(x) ((sc->sc_flag & MC146818_BCD) ? TOBCD(x) : (x))
(*sc->sc_mcwrite)(dev, MC_SEC, TOREG(ct.sec));
(*sc->sc_mcwrite)(dev, MC_MIN, TOREG(ct.min));
(*sc->sc_mcwrite)(dev, MC_HOUR, TOREG(ct.hour));
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
/* Map dow from 0 - 6 to 1 - 7. */
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
(*sc->sc_mcwrite)(dev, MC_DOW, TOREG(ct.dow + 1));
(*sc->sc_mcwrite)(dev, MC_DOM, TOREG(ct.day));
(*sc->sc_mcwrite)(dev, MC_MONTH, TOREG(ct.mon));
year = ct.year - sc->sc_year0;
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
if (sc->sc_flag & MC146818_NO_CENT_ADJUST) {
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
cent = year / 100;
(*sc->sc_setcent)(dev, cent);
year -= cent * 100;
o mc146818(4): - Add locking. - Account for if the MC146818_NO_CENT_ADJUST flag is set we don't need to check wheter year < POSIX_BASE_YEAR. - Add some comments about mapping the day of week from the range the generic clock code uses to the range the chip uses and which I meant to add in the initial version. - Minor clean-up, use __func__ instead of hardcoded function names in error strings. o in the rtc(4) front-end additionally: - Don't leak resources in case mc146818_attach() fails. - Account for ebus(4) defaulting to SYS_RES_MEMORY for the memory resources since ebus.c rev. 1.22.
2005-05-19 21:20:42 +00:00
} else if (year > 99)
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
year -= 100;
(*sc->sc_mcwrite)(dev, MC_YEAR, TOREG(year));
/* Reenable RTC updates and interrupts. */
(*sc->sc_mcwrite)(dev, MC_REGB, sc->sc_regb);
After some input from bde@ and rereading the datasheet use a MTX_SPIN mutex instead of a MTX_DEF one in order to defer preemption while reading the date and time registers. If we don't manage to read them within the time slot where we are guaranteed that no updates occur we might actually read them during an update in which case the output is undefined.
2005-06-04 23:24:50 +00:00
mtx_unlock_spin(&sc->sc_mtx);
Add a driver back end for MC146818 and compatible clocks based on the respective NetBSD driver for use with the genclock interface. It's first use will be on sparc64 but it was also tested on alpha with a preliminary patch to switch alpha to use the genclock code together with this driver instead of the respective code in alpha/alpha/clock.c and the rather MD mcclock(4). Using it on i386 and amd64 won't be that hard but some changes/extensions to improve the genclock code in general should be done first, e.g. add locking and make it easier to access the NVRAM usually coupled with RTCs.
2004-11-17 16:37:25 +00:00
#undef TOREG
return (0);
}
#define MC_ADDR 0
#define MC_DATA 1
static u_int
mc146818_def_read(device_t dev, u_int reg)
{
struct mc146818_softc *sc;
sc = device_get_softc(dev);
bus_space_write_1(sc->sc_bst, sc->sc_bsh, MC_ADDR, reg);
return (bus_space_read_1(sc->sc_bst, sc->sc_bsh, MC_DATA));
}
static void
mc146818_def_write(device_t dev, u_int reg, u_int val)
{
struct mc146818_softc *sc;
sc = device_get_softc(dev);
bus_space_write_1(sc->sc_bst, sc->sc_bsh, MC_ADDR, reg);
bus_space_write_1(sc->sc_bst, sc->sc_bsh, MC_DATA, val);
}
/*
* Looks like it's common even across platforms to store the century at
* 0x32 in the NVRAM of the mc146818.
*/
#define MC_CENT 0x32
static u_int
mc146818_def_getcent(device_t dev)
{
struct mc146818_softc *sc;
sc = device_get_softc(dev);
return ((*sc->sc_mcread)(dev, MC_CENT));
}
static void
mc146818_def_setcent(device_t dev, u_int cent)
{
struct mc146818_softc *sc;
sc = device_get_softc(dev);
(*sc->sc_mcwrite)(dev, MC_CENT, cent);
}
Reference in New Issue Copy Permalink