freebsd-nq/sys/dev/ipmi/ipmi_acpi.c

215 lines
5.7 KiB
C
Raw Normal View History

Update the ipmi(4) driver: - Split out the communication protocols into their own files and use a couple of function pointers in the softc that the commuication protocols setup in their own attach routine. - Add support for the SSIF interface (talking to IPMI over SMBus). - Add an ACPI attachment. - Add a PCI attachment that attaches to devices with the IPMI interface subclass. - Split the ISA attachment out into its own file: ipmi_isa.c. - Change the code to probe the SMBIOS table for an IPMI entry to just use pmap_mapbios() to map the table in rather than trying to setup a fake resource on an isa device and then activating the resource to map in the table. - Make bus attachments leaner by adding attach functions for each communication interface (ipmi_kcs_attach(), ipmi_smic_attach(), etc.) that setup per-interface data. - Formalize the model used by the driver to handle requests by adding an explicit struct ipmi_request object that holds the state of a given request and reply for the entire lifetime of the request. By bundling the request into an object, it is easier to add retry logic to the various communication backends (as well as eventually support BT mode which uses a slightly different message format than KCS, SMIC, and SSIF). - Add a per-softc lock and remove D_NEEDGIANT as the driver is now MPSAFE. - Add 32-bit compatibility ioctl shims so you can use a 32-bit ipmitool on FreeBSD/amd64. - Add ipmi(4) to i386 and amd64 NOTES. Submitted by: ambrisko (large portions of 2 and 3) Sponsored by: IronPort Systems, Yahoo! MFC after: 6 days
2006-09-22 22:11:29 +00:00
/*-
* SPDX-License-Identifier: BSD-2-Clause-FreeBSD
*
Update the ipmi(4) driver: - Split out the communication protocols into their own files and use a couple of function pointers in the softc that the commuication protocols setup in their own attach routine. - Add support for the SSIF interface (talking to IPMI over SMBus). - Add an ACPI attachment. - Add a PCI attachment that attaches to devices with the IPMI interface subclass. - Split the ISA attachment out into its own file: ipmi_isa.c. - Change the code to probe the SMBIOS table for an IPMI entry to just use pmap_mapbios() to map the table in rather than trying to setup a fake resource on an isa device and then activating the resource to map in the table. - Make bus attachments leaner by adding attach functions for each communication interface (ipmi_kcs_attach(), ipmi_smic_attach(), etc.) that setup per-interface data. - Formalize the model used by the driver to handle requests by adding an explicit struct ipmi_request object that holds the state of a given request and reply for the entire lifetime of the request. By bundling the request into an object, it is easier to add retry logic to the various communication backends (as well as eventually support BT mode which uses a slightly different message format than KCS, SMIC, and SSIF). - Add a per-softc lock and remove D_NEEDGIANT as the driver is now MPSAFE. - Add 32-bit compatibility ioctl shims so you can use a 32-bit ipmitool on FreeBSD/amd64. - Add ipmi(4) to i386 and amd64 NOTES. Submitted by: ambrisko (large portions of 2 and 3) Sponsored by: IronPort Systems, Yahoo! MFC after: 6 days
2006-09-22 22:11:29 +00:00
* Copyright (c) 2006 IronPort Systems Inc. <ambrisko@ironport.com>
* 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.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/bus.h>
#include <sys/condvar.h>
#include <sys/kernel.h>
#include <sys/module.h>
#include <sys/rman.h>
#include <sys/selinfo.h>
2009-06-05 18:44:36 +00:00
#include <contrib/dev/acpica/include/acpi.h>
Update the ipmi(4) driver: - Split out the communication protocols into their own files and use a couple of function pointers in the softc that the commuication protocols setup in their own attach routine. - Add support for the SSIF interface (talking to IPMI over SMBus). - Add an ACPI attachment. - Add a PCI attachment that attaches to devices with the IPMI interface subclass. - Split the ISA attachment out into its own file: ipmi_isa.c. - Change the code to probe the SMBIOS table for an IPMI entry to just use pmap_mapbios() to map the table in rather than trying to setup a fake resource on an isa device and then activating the resource to map in the table. - Make bus attachments leaner by adding attach functions for each communication interface (ipmi_kcs_attach(), ipmi_smic_attach(), etc.) that setup per-interface data. - Formalize the model used by the driver to handle requests by adding an explicit struct ipmi_request object that holds the state of a given request and reply for the entire lifetime of the request. By bundling the request into an object, it is easier to add retry logic to the various communication backends (as well as eventually support BT mode which uses a slightly different message format than KCS, SMIC, and SSIF). - Add a per-softc lock and remove D_NEEDGIANT as the driver is now MPSAFE. - Add 32-bit compatibility ioctl shims so you can use a 32-bit ipmitool on FreeBSD/amd64. - Add ipmi(4) to i386 and amd64 NOTES. Submitted by: ambrisko (large portions of 2 and 3) Sponsored by: IronPort Systems, Yahoo! MFC after: 6 days
2006-09-22 22:11:29 +00:00
#include <dev/acpica/acpivar.h>
/* Hooks for the ACPI CA debugging infrastructure */
#define _COMPONENT ACPI_BUTTON
ACPI_MODULE_NAME("IPMI")
#ifdef LOCAL_MODULE
#include <ipmi.h>
#include <ipmivars.h>
#else
#include <sys/ipmi.h>
#include <dev/ipmi/ipmivars.h>
#endif
static int ipmi_acpi_probe(device_t);
static int ipmi_acpi_attach(device_t);
int
ipmi_acpi_probe(device_t dev)
{
static char *ipmi_ids[] = {"IPI0001", NULL};
int rv;
Update the ipmi(4) driver: - Split out the communication protocols into their own files and use a couple of function pointers in the softc that the commuication protocols setup in their own attach routine. - Add support for the SSIF interface (talking to IPMI over SMBus). - Add an ACPI attachment. - Add a PCI attachment that attaches to devices with the IPMI interface subclass. - Split the ISA attachment out into its own file: ipmi_isa.c. - Change the code to probe the SMBIOS table for an IPMI entry to just use pmap_mapbios() to map the table in rather than trying to setup a fake resource on an isa device and then activating the resource to map in the table. - Make bus attachments leaner by adding attach functions for each communication interface (ipmi_kcs_attach(), ipmi_smic_attach(), etc.) that setup per-interface data. - Formalize the model used by the driver to handle requests by adding an explicit struct ipmi_request object that holds the state of a given request and reply for the entire lifetime of the request. By bundling the request into an object, it is easier to add retry logic to the various communication backends (as well as eventually support BT mode which uses a slightly different message format than KCS, SMIC, and SSIF). - Add a per-softc lock and remove D_NEEDGIANT as the driver is now MPSAFE. - Add 32-bit compatibility ioctl shims so you can use a 32-bit ipmitool on FreeBSD/amd64. - Add ipmi(4) to i386 and amd64 NOTES. Submitted by: ambrisko (large portions of 2 and 3) Sponsored by: IronPort Systems, Yahoo! MFC after: 6 days
2006-09-22 22:11:29 +00:00
if (ipmi_attached)
return (EBUSY);
if (acpi_disabled("ipmi"))
Update the ipmi(4) driver: - Split out the communication protocols into their own files and use a couple of function pointers in the softc that the commuication protocols setup in their own attach routine. - Add support for the SSIF interface (talking to IPMI over SMBus). - Add an ACPI attachment. - Add a PCI attachment that attaches to devices with the IPMI interface subclass. - Split the ISA attachment out into its own file: ipmi_isa.c. - Change the code to probe the SMBIOS table for an IPMI entry to just use pmap_mapbios() to map the table in rather than trying to setup a fake resource on an isa device and then activating the resource to map in the table. - Make bus attachments leaner by adding attach functions for each communication interface (ipmi_kcs_attach(), ipmi_smic_attach(), etc.) that setup per-interface data. - Formalize the model used by the driver to handle requests by adding an explicit struct ipmi_request object that holds the state of a given request and reply for the entire lifetime of the request. By bundling the request into an object, it is easier to add retry logic to the various communication backends (as well as eventually support BT mode which uses a slightly different message format than KCS, SMIC, and SSIF). - Add a per-softc lock and remove D_NEEDGIANT as the driver is now MPSAFE. - Add 32-bit compatibility ioctl shims so you can use a 32-bit ipmitool on FreeBSD/amd64. - Add ipmi(4) to i386 and amd64 NOTES. Submitted by: ambrisko (large portions of 2 and 3) Sponsored by: IronPort Systems, Yahoo! MFC after: 6 days
2006-09-22 22:11:29 +00:00
return (ENXIO);
rv = ACPI_ID_PROBE(device_get_parent(dev), dev, ipmi_ids, NULL);
if (rv <= 0)
device_set_desc(dev, "IPMI System Interface");
Update the ipmi(4) driver: - Split out the communication protocols into their own files and use a couple of function pointers in the softc that the commuication protocols setup in their own attach routine. - Add support for the SSIF interface (talking to IPMI over SMBus). - Add an ACPI attachment. - Add a PCI attachment that attaches to devices with the IPMI interface subclass. - Split the ISA attachment out into its own file: ipmi_isa.c. - Change the code to probe the SMBIOS table for an IPMI entry to just use pmap_mapbios() to map the table in rather than trying to setup a fake resource on an isa device and then activating the resource to map in the table. - Make bus attachments leaner by adding attach functions for each communication interface (ipmi_kcs_attach(), ipmi_smic_attach(), etc.) that setup per-interface data. - Formalize the model used by the driver to handle requests by adding an explicit struct ipmi_request object that holds the state of a given request and reply for the entire lifetime of the request. By bundling the request into an object, it is easier to add retry logic to the various communication backends (as well as eventually support BT mode which uses a slightly different message format than KCS, SMIC, and SSIF). - Add a per-softc lock and remove D_NEEDGIANT as the driver is now MPSAFE. - Add 32-bit compatibility ioctl shims so you can use a 32-bit ipmitool on FreeBSD/amd64. - Add ipmi(4) to i386 and amd64 NOTES. Submitted by: ambrisko (large portions of 2 and 3) Sponsored by: IronPort Systems, Yahoo! MFC after: 6 days
2006-09-22 22:11:29 +00:00
return (rv);
Update the ipmi(4) driver: - Split out the communication protocols into their own files and use a couple of function pointers in the softc that the commuication protocols setup in their own attach routine. - Add support for the SSIF interface (talking to IPMI over SMBus). - Add an ACPI attachment. - Add a PCI attachment that attaches to devices with the IPMI interface subclass. - Split the ISA attachment out into its own file: ipmi_isa.c. - Change the code to probe the SMBIOS table for an IPMI entry to just use pmap_mapbios() to map the table in rather than trying to setup a fake resource on an isa device and then activating the resource to map in the table. - Make bus attachments leaner by adding attach functions for each communication interface (ipmi_kcs_attach(), ipmi_smic_attach(), etc.) that setup per-interface data. - Formalize the model used by the driver to handle requests by adding an explicit struct ipmi_request object that holds the state of a given request and reply for the entire lifetime of the request. By bundling the request into an object, it is easier to add retry logic to the various communication backends (as well as eventually support BT mode which uses a slightly different message format than KCS, SMIC, and SSIF). - Add a per-softc lock and remove D_NEEDGIANT as the driver is now MPSAFE. - Add 32-bit compatibility ioctl shims so you can use a 32-bit ipmitool on FreeBSD/amd64. - Add ipmi(4) to i386 and amd64 NOTES. Submitted by: ambrisko (large portions of 2 and 3) Sponsored by: IronPort Systems, Yahoo! MFC after: 6 days
2006-09-22 22:11:29 +00:00
}
static int
ipmi_acpi_attach(device_t dev)
{
ACPI_HANDLE devh;
const char *mode;
struct ipmi_get_info info;
struct ipmi_softc *sc = device_get_softc(dev);
int count, error, flags, i, type;
int interface_type = 0, interface_version = 0;
error = 0;
devh = acpi_get_handle(dev);
if (ACPI_FAILURE(acpi_GetInteger(devh, "_IFT", &interface_type)))
return (ENXIO);
if (ACPI_FAILURE(acpi_GetInteger(devh, "_SRV", &interface_version)))
return (ENXIO);
switch (interface_type) {
case KCS_MODE:
count = 2;
mode = "KCS";
break;
case SMIC_MODE:
count = 3;
mode = "SMIC";
break;
case BT_MODE:
device_printf(dev, "BT interface not supported\n");
return (ENXIO);
case SSIF_MODE:
if (ACPI_FAILURE(acpi_GetInteger(devh, "_ADR", &flags)))
return (ENXIO);
info.address = flags;
Update the ipmi(4) driver: - Split out the communication protocols into their own files and use a couple of function pointers in the softc that the commuication protocols setup in their own attach routine. - Add support for the SSIF interface (talking to IPMI over SMBus). - Add an ACPI attachment. - Add a PCI attachment that attaches to devices with the IPMI interface subclass. - Split the ISA attachment out into its own file: ipmi_isa.c. - Change the code to probe the SMBIOS table for an IPMI entry to just use pmap_mapbios() to map the table in rather than trying to setup a fake resource on an isa device and then activating the resource to map in the table. - Make bus attachments leaner by adding attach functions for each communication interface (ipmi_kcs_attach(), ipmi_smic_attach(), etc.) that setup per-interface data. - Formalize the model used by the driver to handle requests by adding an explicit struct ipmi_request object that holds the state of a given request and reply for the entire lifetime of the request. By bundling the request into an object, it is easier to add retry logic to the various communication backends (as well as eventually support BT mode which uses a slightly different message format than KCS, SMIC, and SSIF). - Add a per-softc lock and remove D_NEEDGIANT as the driver is now MPSAFE. - Add 32-bit compatibility ioctl shims so you can use a 32-bit ipmitool on FreeBSD/amd64. - Add ipmi(4) to i386 and amd64 NOTES. Submitted by: ambrisko (large portions of 2 and 3) Sponsored by: IronPort Systems, Yahoo! MFC after: 6 days
2006-09-22 22:11:29 +00:00
device_printf(dev, "SSIF interface not supported on ACPI\n");
return (0);
default:
return (ENXIO);
}
if (bus_get_resource(dev, SYS_RES_IOPORT, 0, NULL, NULL) == 0)
type = SYS_RES_IOPORT;
else if (bus_get_resource(dev, SYS_RES_MEMORY, 0, NULL, NULL) == 0)
type = SYS_RES_MEMORY;
else {
device_printf(dev, "unknown resource type\n");
return (ENXIO);
}
sc->ipmi_io_rid = 0;
sc->ipmi_io_res[0] = bus_alloc_resource_any(dev, type,
&sc->ipmi_io_rid, RF_ACTIVE);
sc->ipmi_io_type = type;
sc->ipmi_io_spacing = 1;
if (sc->ipmi_io_res[0] == NULL) {
device_printf(dev, "couldn't configure I/O resource\n");
return (ENXIO);
}
/* If we have multiple resources, allocate up to MAX_RES. */
for (i = 1; i < MAX_RES; i++) {
sc->ipmi_io_rid = i;
sc->ipmi_io_res[i] = bus_alloc_resource_any(dev, type,
&sc->ipmi_io_rid, RF_ACTIVE);
if (sc->ipmi_io_res[i] == NULL)
break;
}
sc->ipmi_io_rid = 0;
/* If we have multiple resources, make sure we have enough of them. */
if (sc->ipmi_io_res[1] != NULL && sc->ipmi_io_res[count - 1] == NULL) {
device_printf(dev, "too few I/O resources\n");
error = ENXIO;
goto bad;
}
device_printf(dev, "%s mode found at %s 0x%jx on %s\n",
mode, type == SYS_RES_IOPORT ? "io" : "mem",
(uintmax_t)rman_get_start(sc->ipmi_io_res[0]),
device_get_name(device_get_parent(dev)));
sc->ipmi_dev = dev;
/*
* Setup an interrupt if we have an interrupt resource. We
* don't support GPE interrupts via _GPE yet.
*/
sc->ipmi_irq_rid = 0;
sc->ipmi_irq_res = bus_alloc_resource_any(dev, SYS_RES_IRQ,
&sc->ipmi_irq_rid, RF_SHAREABLE | RF_ACTIVE);
/* Warn if _GPE exists. */
if (ACPI_SUCCESS(AcpiEvaluateObject(devh, "_GPE", NULL, NULL)))
device_printf(dev, "_GPE support not implemented\n");
/*
* We assume an alignment of 1 byte as currently the IPMI spec
* doesn't provide any way to determine the alignment via ACPI.
*/
switch (interface_type) {
case KCS_MODE:
error = ipmi_kcs_attach(sc);
if (error)
goto bad;
break;
case SMIC_MODE:
error = ipmi_smic_attach(sc);
if (error)
goto bad;
break;
}
error = ipmi_attach(dev);
if (error)
goto bad;
return (0);
bad:
ipmi_release_resources(dev);
return (error);
}
static device_method_t ipmi_methods[] = {
/* Device interface */
DEVMETHOD(device_probe, ipmi_acpi_probe),
DEVMETHOD(device_attach, ipmi_acpi_attach),
DEVMETHOD(device_detach, ipmi_detach),
{ 0, 0 }
};
static driver_t ipmi_acpi_driver = {
"ipmi",
ipmi_methods,
sizeof(struct ipmi_softc),
};
DRIVER_MODULE(ipmi_acpi, acpi, ipmi_acpi_driver, ipmi_devclass, 0, 0);
MODULE_DEPEND(ipmi_acpi, acpi, 1, 1, 1);