Page MenuHomeFreeBSD
Differential D1433

paravirt: Add detection of VMware using new hypervisor interface
AbandonedPublic
Actions

Authored by bryanv on Jan 4 2015, 5:56 AM.
Tags
None
Referenced Files
Unknown Object (File)
Fri, Apr 18, 4:18 AM
Unknown Object (File)
Fri, Apr 18, 2:29 AM
Unknown Object (File)
Fri, Apr 18, 12:28 AM
Unknown Object (File)
Thu, Apr 17, 7:37 PM
Unknown Object (File)
Mar 3 2025, 4:11 PM
Unknown Object (File)
Jan 8 2025, 2:49 AM
Unknown Object (File)
Nov 28 2024, 2:21 AM
Unknown Object (File)
Nov 28 2024, 2:21 AM
View All 96 Files
Subscribers
rwindz0_gmail.com

Details

Reviewers
None
Summary

While here, move the VMware TSC frequency code into vmware.c. We may to later want to generalize this by adding a callback to either cpu_ops or init_ops.

Diff Detail

Lint
Lint Skipped
Unit
Tests Skipped

Event Timeline

bryanv updated this revision to Diff 2986.Jan 4 2015, 5:56 AM
bryanv retitled this revision from to paravirt: Add detection of VMware using new hypervisor interface.
bryanv updated this object.
bryanv edited the test plan for this revision. (Show Details)
bryanv set the repository for this revision to rS FreeBSD src repository - subversion.
rwindz0_gmail.com added a subscriber: rwindz0_gmail.com.Nov 21 2016, 3:22 PM
bryanv abandoned this revision.Jul 18 2018, 3:03 PM

Revision Contents
Changeset List

PathSize
sys/
conf/
1 line
1 line
x86/
include/
2 lines
x86/
19 lines
84 lines

Diff 2986

View Options

sys/conf/files.amd64

  • This file was added.
/*-
* Copyright (c) 2022 Takanori Watanabe
*
* 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 "opt_acpi.h"
#include <sys/param.h>
andrewUnsubmitted
Done
Inline Actions

Is opt_evdev.h needed?

andrew: Is `opt_evdev.h` needed?
#include <sys/bus.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <sys/module.h>
#include <sys/rman.h>
#include <contrib/dev/acpica/include/acpi.h>
#include <contrib/dev/acpica/include/accommon.h>
#include <dev/acpica/acpivar.h>
/* Hooks for the ACPI CA debugging infrastructure */
#define _COMPONENT ACPI_GED
ACPI_MODULE_NAME("GED")
static MALLOC_DEFINE(M_ACPIGED, "acpiged", "ACPI Generic event data");
struct acpi_ged_events {
device_t dev;
hrsUnsubmitted
Done
Inline Actions

I suggest s/events/event/ because this structure is for a single event.

hrs: I suggest s/events/event/ because this structure is for a single event.
struct resource *r;
int rid;
void *cookie;
};
struct acpi_ged_softc {
int numevts;
struct acpi_ged_events *evts;
};
static int acpi_ged_probe(device_t dev);
static int acpi_ged_attach(device_t dev);
static int acpi_ged_detach(device_t dev);
static char *ged_ids[] = { "ACPI0013", NULL };
static device_method_t acpi_ged_methods[] = {
/* Device interface */
DEVMETHOD(device_probe, acpi_ged_probe),
DEVMETHOD(device_attach, acpi_ged_attach),
DEVMETHOD(device_detach, acpi_ged_detach), DEVMETHOD_END
};
hrsUnsubmitted
Done
Inline Actions

DEVMETHOD_END should be on a separate line.

hrs: DEVMETHOD_END should be on a separate line.
static driver_t acpi_ged_driver = {
"acpi_ged",
acpi_ged_methods,
sizeof(struct acpi_ged_softc),
};
DRIVER_MODULE(acpi_ged, acpi, acpi_ged_driver, 0, 0);
MODULE_DEPEND(acpi_ged, acpi, 1, 1, 1);
static void
acpi_ged_evt(void *arg)
{
struct acpi_ged_events *evt = arg;
device_t dev = evt->dev;
ACPI_OBJECT_LIST args;
ACPI_OBJECT arg1;
unsigned long irq ;
#ifdef INTRNG
struct intr_map_data_acpi *ima;
ima = rman_get_virtual(evt->r);
irq = ima->irq;
#else
irq = rman_get_start(evt->r);
#endif
hrsUnsubmitted
Done
Inline Actions

I am not sure if this matters, but according to the specification[1], a _CRS can have _Exx or _Lxx method, instead of _EVT, for the interrupt-signaled event with an event number <= 255.

Also, in the case that there is no _Exx/_Lxx/_EVT method, acpi_ged_evt() is called, and it fails. Registering such event handlers does not make sense.

Considering them, it looks better for me to get the ACPI handle for the method, put it into "struct acpi_ged_event" during acpi_ged_attach(), and call it upon an interrupt. If we cannot locate the ACPI handle, we should ignore it with displaying a warning.

[1] https://uefi.org/htmlspecs/ACPI_Spec_6_4_html/05_ACPI_Software_Programming_Model/ACPI_Software_Programming_Model.html#crs-object-for-interrupt-signaled-events

hrs: I am not sure if this matters, but according to the specification[1], a _CRS can have _Exx or…
arg1.Type = ACPI_TYPE_INTEGER;
arg1.Integer.Value = irq;
args.Count = 1;
args.Pointer = &arg1;
AcpiEvaluateObject(acpi_get_handle(dev), "_EVT", &args, NULL);
}
static void
acpi_ged_intr(void *arg)
{
AcpiOsExecute(OSL_GPE_HANDLER, acpi_ged_evt, arg);
}
static int
acpi_ged_probe(device_t dev)
{
int rv;
if (acpi_disabled("ged"))
return (ENXIO);
rv = ACPI_ID_PROBE(device_get_parent(dev), dev, ged_ids, NULL);
if (rv > 0)
return (ENXIO);
device_set_desc(dev, "Generic Event Device");
return (rv);
}
static int
acpi_ged_attach(device_t dev)
{
struct acpi_ged_softc *sc = device_get_softc(dev);
struct resource_list *rl;
struct resource_list_entry *rle;
int i;
andrewUnsubmitted
Done
Inline Actions

No need to zero this, the softc will be zero when allocated.

andrew: No need to zero this, the softc will be zero when allocated.
ACPI_FUNCTION_TRACE((char *)(uintptr_t) __func__);
rl = BUS_GET_RESOURCE_LIST(device_get_parent(dev), dev);
STAILQ_FOREACH (rle, rl, link) {
if (rle->type == SYS_RES_IRQ) {
sc->numevts++;
}
}
sc->evts = mallocarray(sc->numevts, sizeof(*sc->evts), M_ACPIGED,
hrsUnsubmitted
Done
Inline Actions

I wonder why we need to allocate an IRQ with no RF_SHAREABLE (or without checking whether it can be RF_SHAREABLE). Is there any reason?

hrs: I wonder why we need to allocate an IRQ with no RF_SHAREABLE (or without checking whether it…
M_WAITOK | M_ZERO);
for (i = 0; i < sc->numevts; i++) {
sc->evts[i].dev = dev;
sc->evts[i].rid = i;
sc->evts[i].r = bus_alloc_resource_any(dev, SYS_RES_IRQ,
&sc->evts[i].rid, RF_ACTIVE);
if (sc->evts[i].r == NULL) {
device_printf(dev, "Cannot alloc %dth irq\n", i);
continue;
}
if (bus_setup_intr(dev, sc->evts[i].r,
INTR_TYPE_MISC | INTR_MPSAFE, NULL, acpi_ged_intr,
&sc->evts[i], &sc->evts[i].cookie) != 0) {
device_printf(dev, "Failed to setup intr %d\n", i);
}
}
return_VALUE(0);
}
static int
acpi_ged_detach(device_t dev)
{
struct acpi_ged_softc *sc = device_get_softc(dev);
int i;
for (i = 0; i < sc->numevts; i++) {
if (sc->evts[i].cookie) {
bus_teardown_intr(dev, sc->evts[i].r,
sc->evts[i].cookie);
}
if (sc->evts[i].r) {
bus_release_resource(dev, SYS_RES_IRQ, sc->evts[i].rid,
sc->evts[i].r);
}
}
free(sc->evts, M_ACPIGED);
return 0;
}
cyUnsubmitted
Not Done
Inline Actions

Should evt be evts here?

cy: Should evt be evts here?
View Options

sys/conf/files.i386

  • This file was added.
/*-
* Copyright (c) 2022 Takanori Watanabe
*
* 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 "opt_acpi.h"
#include <sys/param.h>
andrewUnsubmitted
Done
Inline Actions

Is opt_evdev.h needed?

andrew: Is `opt_evdev.h` needed?
#include <sys/bus.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <sys/module.h>
#include <sys/rman.h>
#include <contrib/dev/acpica/include/acpi.h>
#include <contrib/dev/acpica/include/accommon.h>
#include <dev/acpica/acpivar.h>
/* Hooks for the ACPI CA debugging infrastructure */
#define _COMPONENT ACPI_GED
ACPI_MODULE_NAME("GED")
static MALLOC_DEFINE(M_ACPIGED, "acpiged", "ACPI Generic event data");
struct acpi_ged_events {
device_t dev;
hrsUnsubmitted
Done
Inline Actions

I suggest s/events/event/ because this structure is for a single event.

hrs: I suggest s/events/event/ because this structure is for a single event.
struct resource *r;
int rid;
void *cookie;
};
struct acpi_ged_softc {
int numevts;
struct acpi_ged_events *evts;
};
static int acpi_ged_probe(device_t dev);
static int acpi_ged_attach(device_t dev);
static int acpi_ged_detach(device_t dev);
static char *ged_ids[] = { "ACPI0013", NULL };
static device_method_t acpi_ged_methods[] = {
/* Device interface */
DEVMETHOD(device_probe, acpi_ged_probe),
DEVMETHOD(device_attach, acpi_ged_attach),
DEVMETHOD(device_detach, acpi_ged_detach), DEVMETHOD_END
};
hrsUnsubmitted
Done
Inline Actions

DEVMETHOD_END should be on a separate line.

hrs: DEVMETHOD_END should be on a separate line.
static driver_t acpi_ged_driver = {
"acpi_ged",
acpi_ged_methods,
sizeof(struct acpi_ged_softc),
};
DRIVER_MODULE(acpi_ged, acpi, acpi_ged_driver, 0, 0);
MODULE_DEPEND(acpi_ged, acpi, 1, 1, 1);
static void
acpi_ged_evt(void *arg)
{
struct acpi_ged_events *evt = arg;
device_t dev = evt->dev;
ACPI_OBJECT_LIST args;
ACPI_OBJECT arg1;
unsigned long irq ;
#ifdef INTRNG
struct intr_map_data_acpi *ima;
ima = rman_get_virtual(evt->r);
irq = ima->irq;
#else
irq = rman_get_start(evt->r);
#endif
hrsUnsubmitted
Done
Inline Actions

I am not sure if this matters, but according to the specification[1], a _CRS can have _Exx or _Lxx method, instead of _EVT, for the interrupt-signaled event with an event number <= 255.

Also, in the case that there is no _Exx/_Lxx/_EVT method, acpi_ged_evt() is called, and it fails. Registering such event handlers does not make sense.

Considering them, it looks better for me to get the ACPI handle for the method, put it into "struct acpi_ged_event" during acpi_ged_attach(), and call it upon an interrupt. If we cannot locate the ACPI handle, we should ignore it with displaying a warning.

[1] https://uefi.org/htmlspecs/ACPI_Spec_6_4_html/05_ACPI_Software_Programming_Model/ACPI_Software_Programming_Model.html#crs-object-for-interrupt-signaled-events

hrs: I am not sure if this matters, but according to the specification[1], a _CRS can have _Exx or…
arg1.Type = ACPI_TYPE_INTEGER;
arg1.Integer.Value = irq;
args.Count = 1;
args.Pointer = &arg1;
AcpiEvaluateObject(acpi_get_handle(dev), "_EVT", &args, NULL);
}
static void
acpi_ged_intr(void *arg)
{
AcpiOsExecute(OSL_GPE_HANDLER, acpi_ged_evt, arg);
}
static int
acpi_ged_probe(device_t dev)
{
int rv;
if (acpi_disabled("ged"))
return (ENXIO);
rv = ACPI_ID_PROBE(device_get_parent(dev), dev, ged_ids, NULL);
if (rv > 0)
return (ENXIO);
device_set_desc(dev, "Generic Event Device");
return (rv);
}
static int
acpi_ged_attach(device_t dev)
{
struct acpi_ged_softc *sc = device_get_softc(dev);
struct resource_list *rl;
struct resource_list_entry *rle;
int i;
andrewUnsubmitted
Done
Inline Actions

No need to zero this, the softc will be zero when allocated.

andrew: No need to zero this, the softc will be zero when allocated.
ACPI_FUNCTION_TRACE((char *)(uintptr_t) __func__);
rl = BUS_GET_RESOURCE_LIST(device_get_parent(dev), dev);
STAILQ_FOREACH (rle, rl, link) {
if (rle->type == SYS_RES_IRQ) {
sc->numevts++;
}
}
sc->evts = mallocarray(sc->numevts, sizeof(*sc->evts), M_ACPIGED,
hrsUnsubmitted
Done
Inline Actions

I wonder why we need to allocate an IRQ with no RF_SHAREABLE (or without checking whether it can be RF_SHAREABLE). Is there any reason?

hrs: I wonder why we need to allocate an IRQ with no RF_SHAREABLE (or without checking whether it…
M_WAITOK | M_ZERO);
for (i = 0; i < sc->numevts; i++) {
sc->evts[i].dev = dev;
sc->evts[i].rid = i;
sc->evts[i].r = bus_alloc_resource_any(dev, SYS_RES_IRQ,
&sc->evts[i].rid, RF_ACTIVE);
if (sc->evts[i].r == NULL) {
device_printf(dev, "Cannot alloc %dth irq\n", i);
continue;
}
if (bus_setup_intr(dev, sc->evts[i].r,
INTR_TYPE_MISC | INTR_MPSAFE, NULL, acpi_ged_intr,
&sc->evts[i], &sc->evts[i].cookie) != 0) {
device_printf(dev, "Failed to setup intr %d\n", i);
}
}
return_VALUE(0);
}
static int
acpi_ged_detach(device_t dev)
{
struct acpi_ged_softc *sc = device_get_softc(dev);
int i;
for (i = 0; i < sc->numevts; i++) {
if (sc->evts[i].cookie) {
bus_teardown_intr(dev, sc->evts[i].r,
sc->evts[i].cookie);
}
if (sc->evts[i].r) {
bus_release_resource(dev, SYS_RES_IRQ, sc->evts[i].rid,
sc->evts[i].r);
}
}
free(sc->evts, M_ACPIGED);
return 0;
}
cyUnsubmitted
Not Done
Inline Actions

Should evt be evts here?

cy: Should evt be evts here?
View Options

sys/x86/include/vmware.h

  • This file was added.
/*-
* Copyright (c) 2022 Takanori Watanabe
*
* 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 "opt_acpi.h"
#include <sys/param.h>
andrewUnsubmitted
Done
Inline Actions

Is opt_evdev.h needed?

andrew: Is `opt_evdev.h` needed?
#include <sys/bus.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <sys/module.h>
#include <sys/rman.h>
#include <contrib/dev/acpica/include/acpi.h>
#include <contrib/dev/acpica/include/accommon.h>
#include <dev/acpica/acpivar.h>
/* Hooks for the ACPI CA debugging infrastructure */
#define _COMPONENT ACPI_GED
ACPI_MODULE_NAME("GED")
static MALLOC_DEFINE(M_ACPIGED, "acpiged", "ACPI Generic event data");
struct acpi_ged_events {
device_t dev;
hrsUnsubmitted
Done
Inline Actions

I suggest s/events/event/ because this structure is for a single event.

hrs: I suggest s/events/event/ because this structure is for a single event.
struct resource *r;
int rid;
void *cookie;
};
struct acpi_ged_softc {
int numevts;
struct acpi_ged_events *evts;
};
static int acpi_ged_probe(device_t dev);
static int acpi_ged_attach(device_t dev);
static int acpi_ged_detach(device_t dev);
static char *ged_ids[] = { "ACPI0013", NULL };
static device_method_t acpi_ged_methods[] = {
/* Device interface */
DEVMETHOD(device_probe, acpi_ged_probe),
DEVMETHOD(device_attach, acpi_ged_attach),
DEVMETHOD(device_detach, acpi_ged_detach), DEVMETHOD_END
};
hrsUnsubmitted
Done
Inline Actions

DEVMETHOD_END should be on a separate line.

hrs: DEVMETHOD_END should be on a separate line.
static driver_t acpi_ged_driver = {
"acpi_ged",
acpi_ged_methods,
sizeof(struct acpi_ged_softc),
};
DRIVER_MODULE(acpi_ged, acpi, acpi_ged_driver, 0, 0);
MODULE_DEPEND(acpi_ged, acpi, 1, 1, 1);
static void
acpi_ged_evt(void *arg)
{
struct acpi_ged_events *evt = arg;
device_t dev = evt->dev;
ACPI_OBJECT_LIST args;
ACPI_OBJECT arg1;
unsigned long irq ;
#ifdef INTRNG
struct intr_map_data_acpi *ima;
ima = rman_get_virtual(evt->r);
irq = ima->irq;
#else
irq = rman_get_start(evt->r);
#endif
hrsUnsubmitted
Done
Inline Actions

I am not sure if this matters, but according to the specification[1], a _CRS can have _Exx or _Lxx method, instead of _EVT, for the interrupt-signaled event with an event number <= 255.

Also, in the case that there is no _Exx/_Lxx/_EVT method, acpi_ged_evt() is called, and it fails. Registering such event handlers does not make sense.

Considering them, it looks better for me to get the ACPI handle for the method, put it into "struct acpi_ged_event" during acpi_ged_attach(), and call it upon an interrupt. If we cannot locate the ACPI handle, we should ignore it with displaying a warning.

[1] https://uefi.org/htmlspecs/ACPI_Spec_6_4_html/05_ACPI_Software_Programming_Model/ACPI_Software_Programming_Model.html#crs-object-for-interrupt-signaled-events

hrs: I am not sure if this matters, but according to the specification[1], a _CRS can have _Exx or…
arg1.Type = ACPI_TYPE_INTEGER;
arg1.Integer.Value = irq;
args.Count = 1;
args.Pointer = &arg1;
AcpiEvaluateObject(acpi_get_handle(dev), "_EVT", &args, NULL);
}
static void
acpi_ged_intr(void *arg)
{
AcpiOsExecute(OSL_GPE_HANDLER, acpi_ged_evt, arg);
}
static int
acpi_ged_probe(device_t dev)
{
int rv;
if (acpi_disabled("ged"))
return (ENXIO);
rv = ACPI_ID_PROBE(device_get_parent(dev), dev, ged_ids, NULL);
if (rv > 0)
return (ENXIO);
device_set_desc(dev, "Generic Event Device");
return (rv);
}
static int
acpi_ged_attach(device_t dev)
{
struct acpi_ged_softc *sc = device_get_softc(dev);
struct resource_list *rl;
struct resource_list_entry *rle;
int i;
andrewUnsubmitted
Done
Inline Actions

No need to zero this, the softc will be zero when allocated.

andrew: No need to zero this, the softc will be zero when allocated.
ACPI_FUNCTION_TRACE((char *)(uintptr_t) __func__);
rl = BUS_GET_RESOURCE_LIST(device_get_parent(dev), dev);
STAILQ_FOREACH (rle, rl, link) {
if (rle->type == SYS_RES_IRQ) {
sc->numevts++;
}
}
sc->evts = mallocarray(sc->numevts, sizeof(*sc->evts), M_ACPIGED,
hrsUnsubmitted
Done
Inline Actions

I wonder why we need to allocate an IRQ with no RF_SHAREABLE (or without checking whether it can be RF_SHAREABLE). Is there any reason?

hrs: I wonder why we need to allocate an IRQ with no RF_SHAREABLE (or without checking whether it…
M_WAITOK | M_ZERO);
for (i = 0; i < sc->numevts; i++) {
sc->evts[i].dev = dev;
sc->evts[i].rid = i;
sc->evts[i].r = bus_alloc_resource_any(dev, SYS_RES_IRQ,
&sc->evts[i].rid, RF_ACTIVE);
if (sc->evts[i].r == NULL) {
device_printf(dev, "Cannot alloc %dth irq\n", i);
continue;
}
if (bus_setup_intr(dev, sc->evts[i].r,
INTR_TYPE_MISC | INTR_MPSAFE, NULL, acpi_ged_intr,
&sc->evts[i], &sc->evts[i].cookie) != 0) {
device_printf(dev, "Failed to setup intr %d\n", i);
}
}
return_VALUE(0);
}
static int
acpi_ged_detach(device_t dev)
{
struct acpi_ged_softc *sc = device_get_softc(dev);
int i;
for (i = 0; i < sc->numevts; i++) {
if (sc->evts[i].cookie) {
bus_teardown_intr(dev, sc->evts[i].r,
sc->evts[i].cookie);
}
if (sc->evts[i].r) {
bus_release_resource(dev, SYS_RES_IRQ, sc->evts[i].rid,
sc->evts[i].r);
}
}
free(sc->evts, M_ACPIGED);
return 0;
}
cyUnsubmitted
Not Done
Inline Actions

Should evt be evts here?

cy: Should evt be evts here?
View Options

sys/x86/x86/tsc.c

  • This file was added.
/*-
* Copyright (c) 2022 Takanori Watanabe
*
* 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 "opt_acpi.h"
#include <sys/param.h>
andrewUnsubmitted
Done
Inline Actions

Is opt_evdev.h needed?

andrew: Is `opt_evdev.h` needed?
#include <sys/bus.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <sys/module.h>
#include <sys/rman.h>
#include <contrib/dev/acpica/include/acpi.h>
#include <contrib/dev/acpica/include/accommon.h>
#include <dev/acpica/acpivar.h>
/* Hooks for the ACPI CA debugging infrastructure */
#define _COMPONENT ACPI_GED
ACPI_MODULE_NAME("GED")
static MALLOC_DEFINE(M_ACPIGED, "acpiged", "ACPI Generic event data");
struct acpi_ged_events {
device_t dev;
hrsUnsubmitted
Done
Inline Actions

I suggest s/events/event/ because this structure is for a single event.

hrs: I suggest s/events/event/ because this structure is for a single event.
struct resource *r;
int rid;
void *cookie;
};
struct acpi_ged_softc {
int numevts;
struct acpi_ged_events *evts;
};
static int acpi_ged_probe(device_t dev);
static int acpi_ged_attach(device_t dev);
static int acpi_ged_detach(device_t dev);
static char *ged_ids[] = { "ACPI0013", NULL };
static device_method_t acpi_ged_methods[] = {
/* Device interface */
DEVMETHOD(device_probe, acpi_ged_probe),
DEVMETHOD(device_attach, acpi_ged_attach),
DEVMETHOD(device_detach, acpi_ged_detach), DEVMETHOD_END
};
hrsUnsubmitted
Done
Inline Actions

DEVMETHOD_END should be on a separate line.

hrs: DEVMETHOD_END should be on a separate line.
static driver_t acpi_ged_driver = {
"acpi_ged",
acpi_ged_methods,
sizeof(struct acpi_ged_softc),
};
DRIVER_MODULE(acpi_ged, acpi, acpi_ged_driver, 0, 0);
MODULE_DEPEND(acpi_ged, acpi, 1, 1, 1);
static void
acpi_ged_evt(void *arg)
{
struct acpi_ged_events *evt = arg;
device_t dev = evt->dev;
ACPI_OBJECT_LIST args;
ACPI_OBJECT arg1;
unsigned long irq ;
#ifdef INTRNG
struct intr_map_data_acpi *ima;
ima = rman_get_virtual(evt->r);
irq = ima->irq;
#else
irq = rman_get_start(evt->r);
#endif
hrsUnsubmitted
Done
Inline Actions

I am not sure if this matters, but according to the specification[1], a _CRS can have _Exx or _Lxx method, instead of _EVT, for the interrupt-signaled event with an event number <= 255.

Also, in the case that there is no _Exx/_Lxx/_EVT method, acpi_ged_evt() is called, and it fails. Registering such event handlers does not make sense.

Considering them, it looks better for me to get the ACPI handle for the method, put it into "struct acpi_ged_event" during acpi_ged_attach(), and call it upon an interrupt. If we cannot locate the ACPI handle, we should ignore it with displaying a warning.

[1] https://uefi.org/htmlspecs/ACPI_Spec_6_4_html/05_ACPI_Software_Programming_Model/ACPI_Software_Programming_Model.html#crs-object-for-interrupt-signaled-events

hrs: I am not sure if this matters, but according to the specification[1], a _CRS can have _Exx or…
arg1.Type = ACPI_TYPE_INTEGER;
arg1.Integer.Value = irq;
args.Count = 1;
args.Pointer = &arg1;
AcpiEvaluateObject(acpi_get_handle(dev), "_EVT", &args, NULL);
}
static void
acpi_ged_intr(void *arg)
{
AcpiOsExecute(OSL_GPE_HANDLER, acpi_ged_evt, arg);
}
static int
acpi_ged_probe(device_t dev)
{
int rv;
if (acpi_disabled("ged"))
return (ENXIO);
rv = ACPI_ID_PROBE(device_get_parent(dev), dev, ged_ids, NULL);
if (rv > 0)
return (ENXIO);
device_set_desc(dev, "Generic Event Device");
return (rv);
}
static int
acpi_ged_attach(device_t dev)
{
struct acpi_ged_softc *sc = device_get_softc(dev);
struct resource_list *rl;
struct resource_list_entry *rle;
int i;
andrewUnsubmitted
Done
Inline Actions

No need to zero this, the softc will be zero when allocated.

andrew: No need to zero this, the softc will be zero when allocated.
ACPI_FUNCTION_TRACE((char *)(uintptr_t) __func__);
rl = BUS_GET_RESOURCE_LIST(device_get_parent(dev), dev);
STAILQ_FOREACH (rle, rl, link) {
if (rle->type == SYS_RES_IRQ) {
sc->numevts++;
}
}
sc->evts = mallocarray(sc->numevts, sizeof(*sc->evts), M_ACPIGED,
hrsUnsubmitted
Done
Inline Actions

I wonder why we need to allocate an IRQ with no RF_SHAREABLE (or without checking whether it can be RF_SHAREABLE). Is there any reason?

hrs: I wonder why we need to allocate an IRQ with no RF_SHAREABLE (or without checking whether it…
M_WAITOK | M_ZERO);
for (i = 0; i < sc->numevts; i++) {
sc->evts[i].dev = dev;
sc->evts[i].rid = i;
sc->evts[i].r = bus_alloc_resource_any(dev, SYS_RES_IRQ,
&sc->evts[i].rid, RF_ACTIVE);
if (sc->evts[i].r == NULL) {
device_printf(dev, "Cannot alloc %dth irq\n", i);
continue;
}
if (bus_setup_intr(dev, sc->evts[i].r,
INTR_TYPE_MISC | INTR_MPSAFE, NULL, acpi_ged_intr,
&sc->evts[i], &sc->evts[i].cookie) != 0) {
device_printf(dev, "Failed to setup intr %d\n", i);
}
}
return_VALUE(0);
}
static int
acpi_ged_detach(device_t dev)
{
struct acpi_ged_softc *sc = device_get_softc(dev);
int i;
for (i = 0; i < sc->numevts; i++) {
if (sc->evts[i].cookie) {
bus_teardown_intr(dev, sc->evts[i].r,
sc->evts[i].cookie);
}
if (sc->evts[i].r) {
bus_release_resource(dev, SYS_RES_IRQ, sc->evts[i].rid,
sc->evts[i].r);
}
}
free(sc->evts, M_ACPIGED);
return 0;
}
cyUnsubmitted
Not Done
Inline Actions

Should evt be evts here?

cy: Should evt be evts here?
View Options

sys/x86/x86/vmware.c

Loading...