freebsd-nq/sys/pci/amdpm.c

673 lines
16 KiB
C
Raw Normal View History

/*-
* Copyright (c) 2000 Matthew C. Forman
*
* Based (heavily) on alpm.c which is:
*
* Copyright (c) 1998, 1999 Nicolas Souchu
* 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.
*/
/*
* Power management function/SMBus function support for the AMD 756 chip.
*/
2003-06-11 06:34:30 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/bus.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/module.h>
#include <sys/mutex.h>
#include <sys/systm.h>
#include <machine/bus.h>
#include <machine/resource.h>
#include <sys/rman.h>
#include <dev/pci/pcivar.h>
#include <dev/pci/pcireg.h>
#include <dev/smbus/smbconf.h>
#include "smbus_if.h"
#define AMDPM_DEBUG(x) if (amdpm_debug) (x)
#ifdef DEBUG
static int amdpm_debug = 1;
#else
static int amdpm_debug = 0;
#endif
#define AMDPM_VENDORID_AMD 0x1022
#define AMDPM_DEVICEID_AMD756PM 0x740b
#define AMDPM_DEVICEID_AMD766PM 0x7413
#define AMDPM_DEVICEID_AMD768PM 0x7443
Fix PCI ID of the AMD-8111 System Management controller so it matches SMBus 1.0 and not SMBus 2.0. AMD-8111 hub (datasheet is publically available) implements both SMBus 2.0 (a separate PCI device) and SMBus 1.0 (a subfunction of the System Management Controller device with the base I/O address is accessible through the CSR 0x58). This driver only supports AMD-756 SMBus 1.0 compatible devices. With the patched sysutils/xmbmon port (to also fix PCI ID and to enable smb(4) support), I now get: pciconf: none0@pci0:7:2: class=0x0c0500 card=0x746a1022 chip=0x746a1022 rev=0x02 hdr=0x00 vendor = 'Advanced Micro Devices (AMD)' device = 'AMD-8111 SMBus 2.0 Controller' class = serial bus subclass = SMBus amdpm0@pci0:7:3: class=0x068000 card=0x746b1022 chip=0x746b1022 rev=0x05 hdr=0x00 vendor = 'Advanced Micro Devices (AMD)' device = 'AMD-8111 ACPI System Management Controller' class = bridge dmesg: amdpm0: <AMD 756/766/768/8111 Power Management Controller> port 0x10e0-0x10ff at device 7.3 on pci0 smbus0: <System Management Bus> on amdpm0 # mbmon -A -d Summary of Detection: * SMB monitor(s)[ioctl:AMD8111]: ** Winbond Chip W83627HF/THF/THF-A found at slave address: 0x50. ** Analog Dev. Chip ADM1027 found at slave address: 0x5C. * ISA monitor(s): ** Winbond Chip W83627HF/THF/THF-A found. I think the confusion comes from the fact that nobody really tried SMBus with xmbmon :-), since sysutils/xmbmon port doesn't come with SMBus support enabled, neither in FreeBSD 4, nor in later versions, so mbmon(1) was just showing the values from the Winbond sensors accessible through the ISA I/O method (mbmon -I), for me anyway. On my test machine, the amdpm(4) didn't even attach due to I/O port allocation failure (who knows what the hell it read from CSR 0x58 of the SMBus 2.0 device :-), which isn't in the CSR space). I've also checked that lm_sensors.org uses correct PCI ID for SMBus 1.0 of AMD-8111: i2c-amd756.c: {PCI_VENDOR_ID_AMD, 0x746B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AMD8111 }, This driver is analogous to our amdpm.c which supports SMBus 1.0 AMD-756 and compatible devices, including SMBus 1.0 on AMD-8111. i2c-amd8111.c: { 0x1022, 0x746a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, This driver is analogous to nForce-2/3/4, i2c-nforce2.c, which supports SMBus 2.0, and which our amdpm.c does NOT support (SMBus 2.0 uses a different, ACPI-unified, API to talk to SMBus). At least I know for sure it doesn't work with my nForce3. :-) (The xmbmon port will be fixed to correct the PCI ID too and to enable the smb(4) support.)
2005-12-16 15:03:16 +00:00
#define AMDPM_DEVICEID_AMD8111PM 0x746B
/* nVidia nForce chipset */
#define AMDPM_VENDORID_NVIDIA 0x10de
#define AMDPM_DEVICEID_NF_SMB 0x01b4
/* PCI Configuration space registers */
#define AMDPCI_PMBASE 0x58
#define NFPCI_PMBASE 0x14
#define AMDPCI_GEN_CONFIG_PM 0x41
#define AMDPCI_PMIOEN (1<<7)
#define AMDPCI_SCIINT_CONFIG_PM 0x42
#define AMDPCI_SCISEL_IRQ11 11
#define AMDPCI_REVID 0x08
/*
* I/O registers.
* Base address programmed via AMDPCI_PMBASE.
*/
#define AMDSMB_GLOBAL_STATUS (0x00)
#define AMDSMB_GS_TO_STS (1<<5)
#define AMDSMB_GS_HCYC_STS (1<<4)
#define AMDSMB_GS_HST_STS (1<<3)
#define AMDSMB_GS_PRERR_STS (1<<2)
#define AMDSMB_GS_COL_STS (1<<1)
#define AMDSMB_GS_ABRT_STS (1<<0)
#define AMDSMB_GS_CLEAR_STS (AMDSMB_GS_TO_STS|AMDSMB_GS_HCYC_STS|AMDSMB_GS_PRERR_STS|AMDSMB_GS_COL_STS|AMDSMB_GS_ABRT_STS)
#define AMDSMB_GLOBAL_ENABLE (0x02)
#define AMDSMB_GE_ABORT (1<<5)
#define AMDSMB_GE_HCYC_EN (1<<4)
#define AMDSMB_GE_HOST_STC (1<<3)
#define AMDSMB_GE_CYC_QUICK 0
#define AMDSMB_GE_CYC_BYTE 1
#define AMDSMB_GE_CYC_BDATA 2
#define AMDSMB_GE_CYC_WDATA 3
#define AMDSMB_GE_CYC_PROCCALL 4
#define AMDSMB_GE_CYC_BLOCK 5
#define LSB 0x1 /* XXX: Better name: Read/Write? */
#define AMDSMB_HSTADDR (0x04)
#define AMDSMB_HSTDATA (0x06)
#define AMDSMB_HSTCMD (0x08)
#define AMDSMB_HSTDFIFO (0x09)
#define AMDSMB_HSLVDATA (0x0A)
#define AMDSMB_HSLVDA (0x0C)
#define AMDSMB_HSLVDDR (0x0E)
#define AMDSMB_SNPADDR (0x0F)
struct amdpm_softc {
int base;
int rid;
struct resource *res;
bus_space_tag_t smbst;
bus_space_handle_t smbsh;
device_t smbus;
struct mtx lock;
};
#define AMDPM_LOCK(amdpm) mtx_lock(&(amdpm)->lock)
#define AMDPM_UNLOCK(amdpm) mtx_unlock(&(amdpm)->lock)
#define AMDPM_LOCK_ASSERT(amdpm) mtx_assert(&(amdpm)->lock, MA_OWNED)
#define AMDPM_SMBINB(amdpm,register) \
(bus_space_read_1(amdpm->smbst, amdpm->smbsh, register))
#define AMDPM_SMBOUTB(amdpm,register,value) \
(bus_space_write_1(amdpm->smbst, amdpm->smbsh, register, value))
#define AMDPM_SMBINW(amdpm,register) \
(bus_space_read_2(amdpm->smbst, amdpm->smbsh, register))
#define AMDPM_SMBOUTW(amdpm,register,value) \
(bus_space_write_2(amdpm->smbst, amdpm->smbsh, register, value))
static int amdpm_detach(device_t dev);
static int
amdpm_probe(device_t dev)
{
u_long base;
u_int16_t vid;
u_int16_t did;
vid = pci_get_vendor(dev);
did = pci_get_device(dev);
if ((vid == AMDPM_VENDORID_AMD) &&
((did == AMDPM_DEVICEID_AMD756PM) ||
(did == AMDPM_DEVICEID_AMD766PM) ||
2004-04-20 13:36:44 +00:00
(did == AMDPM_DEVICEID_AMD768PM) ||
(did == AMDPM_DEVICEID_AMD8111PM))) {
device_set_desc(dev, "AMD 756/766/768/8111 Power Management Controller");
2003-09-06 14:04:30 +00:00
/*
2003-09-06 14:04:30 +00:00
* We have to do this, since the BIOS won't give us the
* resource info (not mine, anyway).
*/
base = pci_read_config(dev, AMDPCI_PMBASE, 4);
base &= 0xff00;
bus_set_resource(dev, SYS_RES_IOPORT, AMDPCI_PMBASE,
base+0xe0, 32);
2005-02-24 21:32:56 +00:00
return (BUS_PROBE_DEFAULT);
}
if ((vid == AMDPM_VENDORID_NVIDIA) &&
(did == AMDPM_DEVICEID_NF_SMB)) {
device_set_desc(dev, "nForce SMBus Controller");
/*
* We have to do this, since the BIOS won't give us the
* resource info (not mine, anyway).
*/
base = pci_read_config(dev, NFPCI_PMBASE, 4);
base &= 0xff00;
bus_set_resource(dev, SYS_RES_IOPORT, NFPCI_PMBASE,
base, 32);
return (BUS_PROBE_DEFAULT);
}
return ENXIO;
}
static int
amdpm_attach(device_t dev)
{
struct amdpm_softc *amdpm_sc = device_get_softc(dev);
u_char val_b;
/* Enable I/O block access */
val_b = pci_read_config(dev, AMDPCI_GEN_CONFIG_PM, 1);
pci_write_config(dev, AMDPCI_GEN_CONFIG_PM, val_b | AMDPCI_PMIOEN, 1);
/* Allocate I/O space */
if (pci_get_vendor(dev) == AMDPM_VENDORID_AMD)
amdpm_sc->rid = AMDPCI_PMBASE;
else
amdpm_sc->rid = NFPCI_PMBASE;
amdpm_sc->res = bus_alloc_resource_any(dev, SYS_RES_IOPORT,
&amdpm_sc->rid, RF_ACTIVE);
if (amdpm_sc->res == NULL) {
device_printf(dev, "could not map i/o space\n");
return (ENXIO);
}
amdpm_sc->smbst = rman_get_bustag(amdpm_sc->res);
amdpm_sc->smbsh = rman_get_bushandle(amdpm_sc->res);
mtx_init(&amdpm_sc->lock, device_get_nameunit(dev), "amdpm", MTX_DEF);
/* Allocate a new smbus device */
amdpm_sc->smbus = device_add_child(dev, "smbus", -1);
if (!amdpm_sc->smbus) {
amdpm_detach(dev);
2003-09-06 14:04:30 +00:00
return (EINVAL);
}
bus_generic_attach(dev);
return (0);
}
static int
amdpm_detach(device_t dev)
{
struct amdpm_softc *amdpm_sc = device_get_softc(dev);
if (amdpm_sc->smbus) {
2003-09-06 14:04:30 +00:00
device_delete_child(dev, amdpm_sc->smbus);
amdpm_sc->smbus = NULL;
}
mtx_destroy(&amdpm_sc->lock);
if (amdpm_sc->res)
2003-09-06 14:04:30 +00:00
bus_release_resource(dev, SYS_RES_IOPORT, amdpm_sc->rid,
amdpm_sc->res);
return (0);
}
static int
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
amdpm_callback(device_t dev, int index, void *data)
{
int error = 0;
switch (index) {
case SMB_REQUEST_BUS:
case SMB_RELEASE_BUS:
break;
default:
error = EINVAL;
}
return (error);
}
static int
amdpm_clear(struct amdpm_softc *sc)
{
AMDPM_LOCK_ASSERT(sc);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_STATUS, AMDSMB_GS_CLEAR_STS);
DELAY(10);
return (0);
}
#if 0
static int
amdpm_abort(struct amdpm_softc *sc)
{
u_short l;
l = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_ENABLE);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_ENABLE, l | AMDSMB_GE_ABORT);
return (0);
}
#endif
static int
amdpm_idle(struct amdpm_softc *sc)
{
u_short sts;
AMDPM_LOCK_ASSERT(sc);
sts = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_STATUS);
AMDPM_DEBUG(printf("amdpm: busy? STS=0x%x\n", sts));
return (~(sts & AMDSMB_GS_HST_STS));
}
/*
* Poll the SMBus controller
*/
static int
amdpm_wait(struct amdpm_softc *sc)
{
int count = 10000;
u_short sts = 0;
int error;
AMDPM_LOCK_ASSERT(sc);
/* Wait for command to complete (SMBus controller is idle) */
while(count--) {
DELAY(10);
sts = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_STATUS);
if (!(sts & AMDSMB_GS_HST_STS))
break;
}
AMDPM_DEBUG(printf("amdpm: STS=0x%x (count=%d)\n", sts, count));
error = SMB_ENOERR;
if (!count)
error |= SMB_ETIMEOUT;
if (sts & AMDSMB_GS_ABRT_STS)
error |= SMB_EABORT;
if (sts & AMDSMB_GS_COL_STS)
error |= SMB_ENOACK;
if (sts & AMDSMB_GS_PRERR_STS)
error |= SMB_EBUSERR;
if (error != SMB_ENOERR)
amdpm_clear(sc);
return (error);
}
static int
amdpm_quick(device_t dev, u_char slave, int how)
{
struct amdpm_softc *sc = (struct amdpm_softc *)device_get_softc(dev);
int error;
u_short l;
AMDPM_LOCK(sc);
amdpm_clear(sc);
if (!amdpm_idle(sc)) {
AMDPM_UNLOCK(sc);
return (EBUSY);
}
switch (how) {
case SMB_QWRITE:
AMDPM_DEBUG(printf("amdpm: QWRITE to 0x%x", slave));
AMDPM_SMBOUTW(sc, AMDSMB_HSTADDR, slave & ~LSB);
break;
case SMB_QREAD:
AMDPM_DEBUG(printf("amdpm: QREAD to 0x%x", slave));
AMDPM_SMBOUTW(sc, AMDSMB_HSTADDR, slave | LSB);
break;
default:
panic("%s: unknown QUICK command (%x)!", __func__, how);
}
l = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_ENABLE);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_ENABLE, (l & 0xfff8) | AMDSMB_GE_CYC_QUICK | AMDSMB_GE_HOST_STC);
error = amdpm_wait(sc);
AMDPM_DEBUG(printf(", error=0x%x\n", error));
AMDPM_UNLOCK(sc);
return (error);
}
static int
amdpm_sendb(device_t dev, u_char slave, char byte)
{
struct amdpm_softc *sc = (struct amdpm_softc *)device_get_softc(dev);
int error;
u_short l;
AMDPM_LOCK(sc);
amdpm_clear(sc);
if (!amdpm_idle(sc)) {
AMDPM_UNLOCK(sc);
return (SMB_EBUSY);
}
AMDPM_SMBOUTW(sc, AMDSMB_HSTADDR, slave & ~LSB);
AMDPM_SMBOUTW(sc, AMDSMB_HSTDATA, byte);
l = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_ENABLE);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_ENABLE, (l & 0xfff8) | AMDSMB_GE_CYC_BYTE | AMDSMB_GE_HOST_STC);
error = amdpm_wait(sc);
AMDPM_DEBUG(printf("amdpm: SENDB to 0x%x, byte=0x%x, error=0x%x\n", slave, byte, error));
AMDPM_UNLOCK(sc);
return (error);
}
static int
amdpm_recvb(device_t dev, u_char slave, char *byte)
{
struct amdpm_softc *sc = (struct amdpm_softc *)device_get_softc(dev);
int error;
u_short l;
AMDPM_LOCK(sc);
amdpm_clear(sc);
if (!amdpm_idle(sc)) {
AMDPM_UNLOCK(sc);
return (SMB_EBUSY);
}
AMDPM_SMBOUTW(sc, AMDSMB_HSTADDR, slave | LSB);
l = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_ENABLE);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_ENABLE, (l & 0xfff8) | AMDSMB_GE_CYC_BYTE | AMDSMB_GE_HOST_STC);
if ((error = amdpm_wait(sc)) == SMB_ENOERR)
*byte = AMDPM_SMBINW(sc, AMDSMB_HSTDATA);
AMDPM_DEBUG(printf("amdpm: RECVB from 0x%x, byte=0x%x, error=0x%x\n", slave, *byte, error));
AMDPM_UNLOCK(sc);
return (error);
}
static int
amdpm_writeb(device_t dev, u_char slave, char cmd, char byte)
{
struct amdpm_softc *sc = (struct amdpm_softc *)device_get_softc(dev);
int error;
u_short l;
AMDPM_LOCK(sc);
amdpm_clear(sc);
if (!amdpm_idle(sc)) {
AMDPM_UNLOCK(sc);
return (SMB_EBUSY);
}
AMDPM_SMBOUTW(sc, AMDSMB_HSTADDR, slave & ~LSB);
AMDPM_SMBOUTW(sc, AMDSMB_HSTDATA, byte);
AMDPM_SMBOUTB(sc, AMDSMB_HSTCMD, cmd);
l = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_ENABLE);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_ENABLE, (l & 0xfff8) | AMDSMB_GE_CYC_BDATA | AMDSMB_GE_HOST_STC);
error = amdpm_wait(sc);
AMDPM_DEBUG(printf("amdpm: WRITEB to 0x%x, cmd=0x%x, byte=0x%x, error=0x%x\n", slave, cmd, byte, error));
AMDPM_UNLOCK(sc);
return (error);
}
static int
amdpm_readb(device_t dev, u_char slave, char cmd, char *byte)
{
struct amdpm_softc *sc = (struct amdpm_softc *)device_get_softc(dev);
int error;
u_short l;
AMDPM_LOCK(sc);
amdpm_clear(sc);
if (!amdpm_idle(sc)) {
AMDPM_UNLOCK(sc);
return (SMB_EBUSY);
}
AMDPM_SMBOUTW(sc, AMDSMB_HSTADDR, slave | LSB);
AMDPM_SMBOUTB(sc, AMDSMB_HSTCMD, cmd);
l = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_ENABLE);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_ENABLE, (l & 0xfff8) | AMDSMB_GE_CYC_BDATA | AMDSMB_GE_HOST_STC);
if ((error = amdpm_wait(sc)) == SMB_ENOERR)
*byte = AMDPM_SMBINW(sc, AMDSMB_HSTDATA);
AMDPM_DEBUG(printf("amdpm: READB from 0x%x, cmd=0x%x, byte=0x%x, error=0x%x\n", slave, cmd, *byte, error));
AMDPM_UNLOCK(sc);
return (error);
}
static int
amdpm_writew(device_t dev, u_char slave, char cmd, short word)
{
struct amdpm_softc *sc = (struct amdpm_softc *)device_get_softc(dev);
int error;
u_short l;
AMDPM_LOCK(sc);
amdpm_clear(sc);
if (!amdpm_idle(sc)) {
AMDPM_UNLOCK(sc);
return (SMB_EBUSY);
}
AMDPM_SMBOUTW(sc, AMDSMB_HSTADDR, slave & ~LSB);
AMDPM_SMBOUTW(sc, AMDSMB_HSTDATA, word);
AMDPM_SMBOUTB(sc, AMDSMB_HSTCMD, cmd);
l = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_ENABLE);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_ENABLE, (l & 0xfff8) | AMDSMB_GE_CYC_WDATA | AMDSMB_GE_HOST_STC);
error = amdpm_wait(sc);
AMDPM_DEBUG(printf("amdpm: WRITEW to 0x%x, cmd=0x%x, word=0x%x, error=0x%x\n", slave, cmd, word, error));
AMDPM_UNLOCK(sc);
return (error);
}
static int
amdpm_readw(device_t dev, u_char slave, char cmd, short *word)
{
struct amdpm_softc *sc = (struct amdpm_softc *)device_get_softc(dev);
int error;
u_short l;
AMDPM_LOCK(sc);
amdpm_clear(sc);
if (!amdpm_idle(sc)) {
AMDPM_UNLOCK(sc);
return (SMB_EBUSY);
}
AMDPM_SMBOUTW(sc, AMDSMB_HSTADDR, slave | LSB);
AMDPM_SMBOUTB(sc, AMDSMB_HSTCMD, cmd);
l = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_ENABLE);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_ENABLE, (l & 0xfff8) | AMDSMB_GE_CYC_WDATA | AMDSMB_GE_HOST_STC);
if ((error = amdpm_wait(sc)) == SMB_ENOERR)
*word = AMDPM_SMBINW(sc, AMDSMB_HSTDATA);
AMDPM_DEBUG(printf("amdpm: READW from 0x%x, cmd=0x%x, word=0x%x, error=0x%x\n", slave, cmd, *word, error));
AMDPM_UNLOCK(sc);
return (error);
}
static int
amdpm_bwrite(device_t dev, u_char slave, char cmd, u_char count, char *buf)
{
struct amdpm_softc *sc = (struct amdpm_softc *)device_get_softc(dev);
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
u_char i;
int error;
u_short l;
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
if (count < 1 || count > 32)
return (SMB_EINVAL);
AMDPM_LOCK(sc);
amdpm_clear(sc);
if (!amdpm_idle(sc)) {
AMDPM_UNLOCK(sc);
return (SMB_EBUSY);
}
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
AMDPM_SMBOUTW(sc, AMDSMB_HSTADDR, slave & ~LSB);
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
/*
* Do we have to reset the internal 32-byte buffer?
* Can't see how to do this from the data sheet.
*/
AMDPM_SMBOUTW(sc, AMDSMB_HSTDATA, count);
/* Fill the 32-byte internal buffer */
for (i = 0; i < count; i++) {
AMDPM_SMBOUTB(sc, AMDSMB_HSTDFIFO, buf[i]);
DELAY(2);
}
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
AMDPM_SMBOUTB(sc, AMDSMB_HSTCMD, cmd);
l = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_ENABLE);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_ENABLE,
(l & 0xfff8) | AMDSMB_GE_CYC_BLOCK | AMDSMB_GE_HOST_STC);
error = amdpm_wait(sc);
AMDPM_DEBUG(printf("amdpm: WRITEBLK to 0x%x, count=0x%x, cmd=0x%x, error=0x%x", slave, count, cmd, error));
AMDPM_UNLOCK(sc);
return (error);
}
static int
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
amdpm_bread(device_t dev, u_char slave, char cmd, u_char *count, char *buf)
{
struct amdpm_softc *sc = (struct amdpm_softc *)device_get_softc(dev);
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
u_char data, len, i;
int error;
u_short l;
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
if (*count < 1 || *count > 32)
return (SMB_EINVAL);
AMDPM_LOCK(sc);
amdpm_clear(sc);
if (!amdpm_idle(sc)) {
AMDPM_UNLOCK(sc);
return (SMB_EBUSY);
}
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
AMDPM_SMBOUTW(sc, AMDSMB_HSTADDR, slave | LSB);
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
AMDPM_SMBOUTB(sc, AMDSMB_HSTCMD, cmd);
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
l = AMDPM_SMBINW(sc, AMDSMB_GLOBAL_ENABLE);
AMDPM_SMBOUTW(sc, AMDSMB_GLOBAL_ENABLE,
(l & 0xfff8) | AMDSMB_GE_CYC_BLOCK | AMDSMB_GE_HOST_STC);
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
if ((error = amdpm_wait(sc)) != SMB_ENOERR)
goto error;
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
len = AMDPM_SMBINW(sc, AMDSMB_HSTDATA);
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
/* Read the 32-byte internal buffer */
for (i = 0; i < len; i++) {
data = AMDPM_SMBINB(sc, AMDSMB_HSTDFIFO);
if (i < *count)
buf[i] = data;
DELAY(2);
}
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
*count = len;
error:
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
AMDPM_DEBUG(printf("amdpm: READBLK to 0x%x, count=0x%x, cmd=0x%x, error=0x%x", slave, *count, cmd, error));
AMDPM_UNLOCK(sc);
return (error);
}
static devclass_t amdpm_devclass;
static device_method_t amdpm_methods[] = {
/* Device interface */
DEVMETHOD(device_probe, amdpm_probe),
DEVMETHOD(device_attach, amdpm_attach),
DEVMETHOD(device_detach, amdpm_detach),
/* SMBus interface */
DEVMETHOD(smbus_callback, amdpm_callback),
DEVMETHOD(smbus_quick, amdpm_quick),
DEVMETHOD(smbus_sendb, amdpm_sendb),
DEVMETHOD(smbus_recvb, amdpm_recvb),
DEVMETHOD(smbus_writeb, amdpm_writeb),
DEVMETHOD(smbus_readb, amdpm_readb),
DEVMETHOD(smbus_writew, amdpm_writew),
DEVMETHOD(smbus_readw, amdpm_readw),
DEVMETHOD(smbus_bwrite, amdpm_bwrite),
DEVMETHOD(smbus_bread, amdpm_bread),
{ 0, 0 }
};
static driver_t amdpm_driver = {
"amdpm",
amdpm_methods,
sizeof(struct amdpm_softc),
};
DRIVER_MODULE(amdpm, pci, amdpm_driver, amdpm_devclass, 0, 0);
Minor overhaul of SMBus support: - Change smbus_callback() to pass a void * rather than caddr_t. - Change smbus_bread() to pass a pointer to the count and have it be an in/out parameter. The input is the size of the buffer (same as before), but on return it will contain the actual amount of data read back from the bus. Note that this value may be larger than the input value. It is up to the caller to treat this as an error if desired. - Change the SMB_BREAD ioctl to write out the updated struct smbcmd which will contain the actual number of bytes read in the 'count' field. To preserve the previous ABI, the old ioctl value is mapped to SMB_OLD_BREAD which doesn't copy the updated smbcmd back out to userland. I doubt anyone actually used the old BREAD anyway as it was rediculous to do a bulk-read but not tell the using program how much data was actually read. - Make the smbus driver and devclass public in the smbus module and push all the DRIVER_MODULE()'s for attaching the smbus driver to various foosmb drivers out into the foosmb modules. This makes all the foosmb logic centralized and allows new foosmb modules to be self-contained w/o having to hack smbus.c everytime a new smbus driver is added. - Add a new SMB_EINVAL error bit and use it in place of EINVAL to return an error for bad arguments (such as invalid counts for bread and bwrite). - Map SMB bus error bits to EIO in smbus_error(). - Make the smbus driver call bus_generic_probe() and require child drivers such as smb(4) to create device_t's via identify routines. Previously, smbus just created one anonymous device during attach, and if you had multiple drivers that could attach it was just random chance as to which driver got to probe for the sole device_t first. - Add a mutex to the smbus(4) softc and use it in place of dummy splhigh() to protect the 'owner' field and perform necessary synchronization for smbus_request_bus() and smbus_release_bus(). - Change the bread() and bwrite() methods of alpm(4), amdpm(4), and viapm(4) to only perform a single transaction and not try to use a loop of multiple transactions for a large request. The framing and commands to use for a large transaction depend on the upper-layer protocol (such as SSIF for IPMI over SMBus) from what I can tell, and the smb(4) driver never allowed bulk read/writes of more than 32-bytes anyway. The other smb drivers only performed single transactions. - Fix buffer overflows in the bread() methods of ichsmb(4), alpm(4), amdpm(4), amdsmb(4), intpm(4), and nfsmb(4). - Use SMB_xxx errors in viapm(4). - Destroy ichsmb(4)'s mutex after bus_generic_detach() to avoid problems from child devices making smb upcalls that would use the mutex during their detach methods. MFC after: 1 week Reviewed by: jmg (mostly)
2006-09-11 20:52:41 +00:00
DRIVER_MODULE(smbus, amdpm, smbus_driver, smbus_devclass, 0, 0);
MODULE_DEPEND(amdpm, pci, 1, 1, 1);
MODULE_DEPEND(amdpm, smbus, SMBUS_MINVER, SMBUS_PREFVER, SMBUS_MAXVER);
MODULE_VERSION(amdpm, 1);