lastdragon/linux-kernel
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Resguardo inicial

Browse Source
This commit is contained in:
lastdragon
2026-09-06 12:53:21 -06:00
parent 662efe3c2a
commit ad2088d178
94853 changed files with 43200404 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
linux-7.2.3/arch/x86/pci/Makefile
+25
View File
@@ -0,0 +1,25 @@
# SPDX-License-Identifier: GPL-2.0
obj-y := i386.o init.o
obj-$(CONFIG_PCI_BIOS) += pcbios.o
obj-$(CONFIG_PCI_MMCONFIG) += mmconfig_$(BITS).o direct.o mmconfig-shared.o
obj-$(CONFIG_PCI_DIRECT) += direct.o
obj-$(CONFIG_PCI_OLPC) += olpc.o
obj-$(CONFIG_PCI_XEN) += xen.o
obj-y += fixup.o
obj-$(CONFIG_ACPI) += acpi.o
obj-y += legacy.o irq.o
obj-$(CONFIG_X86_INTEL_CE) += ce4100.o
obj-$(CONFIG_X86_INTEL_MID) += intel_mid.o
obj-$(CONFIG_X86_NUMACHIP) += numachip.o
obj-y += common.o early.o
obj-y += bus_numa.o
obj-$(CONFIG_AMD_NB) += amd_bus.o
obj-$(CONFIG_PCI_CNB20LE_QUIRK) += broadcom_bus.o
ccflags-$(CONFIG_PCI_DEBUG) += -DDEBUG
linux-7.2.3/arch/x86/pci/acpi.c
+631
View File
@@ -0,0 +1,631 @@
// SPDX-License-Identifier: GPL-2.0
#define pr_fmt(fmt) "PCI: " fmt
#include <linux/pci.h>
#include <linux/acpi.h>
#include <linux/init.h>
#include <linux/irq.h>
#include <linux/dmi.h>
#include <linux/slab.h>
#include <linux/pci-acpi.h>
#include <asm/numa.h>
#include <asm/pci_x86.h>
struct pci_root_info {
struct acpi_pci_root_info common;
struct pci_sysdata sd;
#ifdef CONFIG_PCI_MMCONFIG
bool mcfg_added;
u8 start_bus;
u8 end_bus;
#endif
};
bool pci_use_e820 = true;
static bool pci_use_crs = true;
static bool pci_ignore_seg;
static int __init set_use_crs(const struct dmi_system_id *id)
{
pci_use_crs = true;
return 0;
}
static int __init set_nouse_crs(const struct dmi_system_id *id)
{
pci_use_crs = false;
return 0;
}
static int __init set_ignore_seg(const struct dmi_system_id *id)
{
pr_info("%s detected: ignoring ACPI _SEG\n", id->ident);
pci_ignore_seg = true;
return 0;
}
static int __init set_no_e820(const struct dmi_system_id *id)
{
pr_info("%s detected: not clipping E820 regions from _CRS\n",
id->ident);
pci_use_e820 = false;
return 0;
}
static const struct dmi_system_id pci_crs_quirks[] __initconst = {
/* http://bugzilla.kernel.org/show_bug.cgi?id=14183 */
{
.callback = set_use_crs,
.ident = "IBM System x3800",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
DMI_MATCH(DMI_PRODUCT_NAME, "x3800"),
},
},
/* https://bugzilla.kernel.org/show_bug.cgi?id=16007 */
/* 2006 AMD HT/VIA system with two host bridges */
{
.callback = set_use_crs,
.ident = "ASRock ALiveSATA2-GLAN",
.matches = {
DMI_MATCH(DMI_PRODUCT_NAME, "ALiveSATA2-GLAN"),
},
},
/* https://bugzilla.kernel.org/show_bug.cgi?id=30552 */
/* 2006 AMD HT/VIA system with two host bridges */
{
.callback = set_use_crs,
.ident = "ASUS M2V-MX SE",
.matches = {
DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
DMI_MATCH(DMI_BOARD_NAME, "M2V-MX SE"),
DMI_MATCH(DMI_BIOS_VENDOR, "American Megatrends Inc."),
},
},
/* https://bugzilla.kernel.org/show_bug.cgi?id=42619 */
{
.callback = set_use_crs,
.ident = "MSI MS-7253",
.matches = {
DMI_MATCH(DMI_BOARD_VENDOR, "MICRO-STAR INTERNATIONAL CO., LTD"),
DMI_MATCH(DMI_BOARD_NAME, "MS-7253"),
DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies, LTD"),
},
},
/* https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/931368 */
/* https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/1033299 */
{
.callback = set_use_crs,
.ident = "Foxconn K8M890-8237A",
.matches = {
DMI_MATCH(DMI_BOARD_VENDOR, "Foxconn"),
DMI_MATCH(DMI_BOARD_NAME, "K8M890-8237A"),
DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies, LTD"),
},
},
/* Now for the blacklist.. */
/* https://bugzilla.redhat.com/show_bug.cgi?id=769657 */
{
.callback = set_nouse_crs,
.ident = "Dell Studio 1557",
.matches = {
DMI_MATCH(DMI_BOARD_VENDOR, "Dell Inc."),
DMI_MATCH(DMI_PRODUCT_NAME, "Studio 1557"),
DMI_MATCH(DMI_BIOS_VERSION, "A09"),
},
},
/* https://bugzilla.redhat.com/show_bug.cgi?id=769657 */
{
.callback = set_nouse_crs,
.ident = "Thinkpad SL510",
.matches = {
DMI_MATCH(DMI_BOARD_VENDOR, "LENOVO"),
DMI_MATCH(DMI_BOARD_NAME, "2847DFG"),
DMI_MATCH(DMI_BIOS_VERSION, "6JET85WW (1.43 )"),
},
},
/* https://bugzilla.kernel.org/show_bug.cgi?id=42606 */
{
.callback = set_nouse_crs,
.ident = "Supermicro X8DTH",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Supermicro"),
DMI_MATCH(DMI_PRODUCT_NAME, "X8DTH-i/6/iF/6F"),
DMI_MATCH(DMI_BIOS_VERSION, "2.0a"),
},
},
/* https://bugzilla.kernel.org/show_bug.cgi?id=15362 */
{
.callback = set_ignore_seg,
.ident = "HP xw9300",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
DMI_MATCH(DMI_PRODUCT_NAME, "HP xw9300 Workstation"),
},
},
/*
* Many Lenovo models with "IIL" in their DMI_PRODUCT_VERSION have
* an E820 reserved region that covers the entire 32-bit host
* bridge memory window from _CRS. Using the E820 region to clip
* _CRS means no space is available for hot-added or uninitialized
* PCI devices. This typically breaks I2C controllers for touchpads
* and hot-added Thunderbolt devices. See the commit log for
* models known to require this quirk and related bug reports.
*/
{
.callback = set_no_e820,
.ident = "Lenovo *IIL* product version",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
DMI_MATCH(DMI_PRODUCT_VERSION, "IIL"),
},
},
/*
* The Acer Spin 5 (SP513-54N) has the same E820 reservation covering
* the entire _CRS 32-bit window issue as the Lenovo *IIL* models.
* See https://bugs.launchpad.net/bugs/1884232
*/
{
.callback = set_no_e820,
.ident = "Acer Spin 5 (SP513-54N)",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Acer"),
DMI_MATCH(DMI_PRODUCT_NAME, "Spin SP513-54N"),
},
},
/*
* Clevo X170KM-G barebones have the same E820 reservation covering
* the entire _CRS 32-bit window issue as the Lenovo *IIL* models.
* See https://bugzilla.kernel.org/show_bug.cgi?id=214259
*/
{
.callback = set_no_e820,
.ident = "Clevo X170KM-G Barebone",
.matches = {
DMI_MATCH(DMI_BOARD_NAME, "X170KM-G"),
},
},
{}
};
void __init pci_acpi_crs_quirks(void)
{
int year = dmi_get_bios_year();
if (year >= 0 && year < 2008 && iomem_resource.end <= 0xffffffff)
pci_use_crs = false;
/*
* Some firmware includes unusable space (host bridge registers,
* hidden PCI device BARs, etc) in PCI host bridge _CRS. This is a
* firmware defect, and 4dc2287c1805 ("x86: avoid E820 regions when
* allocating address space") has clipped out the unusable space in
* the past.
*
* But other firmware supplies E820 reserved regions that cover
* entire _CRS windows, so clipping throws away the entire window,
* leaving none for hot-added or uninitialized devices. These E820
* entries are probably *not* a firmware defect, so disable the
* clipping by default for post-2022 machines.
*
* We already have quirks to disable clipping for pre-2023
* machines, and we'll likely need quirks to *enable* clipping for
* post-2022 machines that incorrectly include unusable space in
* _CRS.
*/
if (year >= 2023)
pci_use_e820 = false;
dmi_check_system(pci_crs_quirks);
/*
* If the user specifies "pci=use_crs" or "pci=nocrs" explicitly, that
* takes precedence over anything we figured out above.
*/
if (pci_probe & PCI_ROOT_NO_CRS)
pci_use_crs = false;
else if (pci_probe & PCI_USE__CRS)
pci_use_crs = true;
pr_info("%s host bridge windows from ACPI; if necessary, use \"pci=%s\" and report a bug\n",
pci_use_crs ? "Using" : "Ignoring",
pci_use_crs ? "nocrs" : "use_crs");
/* "pci=use_e820"/"pci=no_e820" on the kernel cmdline takes precedence */
if (pci_probe & PCI_NO_E820)
pci_use_e820 = false;
else if (pci_probe & PCI_USE_E820)
pci_use_e820 = true;
pr_info("%s E820 reservations for host bridge windows\n",
pci_use_e820 ? "Using" : "Ignoring");
if (pci_probe & (PCI_NO_E820 | PCI_USE_E820))
pr_info("Please notify linux-pci@vger.kernel.org so future kernels can do this automatically\n");
}
/*
* Check if pdev is part of a PCIe switch that is directly below the
* specified bridge.
*/
static bool pcie_switch_directly_under(struct pci_dev *bridge,
struct pci_dev *pdev)
{
struct pci_dev *parent = pci_upstream_bridge(pdev);
/* If the device doesn't have a parent, it's not under anything */
if (!parent)
return false;
/*
* If the device has a PCIe type, check if it is below the
* corresponding PCIe switch components (if applicable). Then check
* if its upstream port is directly beneath the specified bridge.
*/
switch (pci_pcie_type(pdev)) {
case PCI_EXP_TYPE_UPSTREAM:
return parent == bridge;
case PCI_EXP_TYPE_DOWNSTREAM:
if (pci_pcie_type(parent) != PCI_EXP_TYPE_UPSTREAM)
return false;
parent = pci_upstream_bridge(parent);
return parent == bridge;
case PCI_EXP_TYPE_ENDPOINT:
if (pci_pcie_type(parent) != PCI_EXP_TYPE_DOWNSTREAM)
return false;
parent = pci_upstream_bridge(parent);
if (!parent || pci_pcie_type(parent) != PCI_EXP_TYPE_UPSTREAM)
return false;
parent = pci_upstream_bridge(parent);
return parent == bridge;
}
return false;
}
static bool pcie_has_usb4_host_interface(struct pci_dev *pdev)
{
struct fwnode_handle *fwnode;
/*
* For USB4, the tunneled PCIe Root or Downstream Ports are marked
* with the "usb4-host-interface" ACPI property, so we look for
* that first. This should cover most cases.
*/
fwnode = fwnode_find_reference(dev_fwnode(&pdev->dev),
"usb4-host-interface", 0);
if (!IS_ERR(fwnode)) {
fwnode_handle_put(fwnode);
return true;
}
/*
* Any integrated Thunderbolt 3/4 PCIe Root Ports from Intel
* before Alder Lake do not have the "usb4-host-interface"
* property so we use their PCI IDs instead. All these are
* tunneled. This list is not expected to grow.
*/
if (pdev->vendor == PCI_VENDOR_ID_INTEL) {
switch (pdev->device) {
/* Ice Lake Thunderbolt 3 PCIe Root Ports */
case 0x8a1d:
case 0x8a1f:
case 0x8a21:
case 0x8a23:
/* Tiger Lake-LP Thunderbolt 4 PCIe Root Ports */
case 0x9a23:
case 0x9a25:
case 0x9a27:
case 0x9a29:
/* Tiger Lake-H Thunderbolt 4 PCIe Root Ports */
case 0x9a2b:
case 0x9a2d:
case 0x9a2f:
case 0x9a31:
return true;
}
}
return false;
}
bool arch_pci_dev_is_removable(struct pci_dev *pdev)
{
struct pci_dev *parent, *root;
/* pdev without a parent or Root Port is never tunneled */
parent = pci_upstream_bridge(pdev);
if (!parent)
return false;
root = pcie_find_root_port(pdev);
if (!root)
return false;
/* Internal PCIe devices are not tunneled */
if (!root->external_facing)
return false;
/* Anything directly behind a "usb4-host-interface" is tunneled */
if (pcie_has_usb4_host_interface(parent))
return true;
/*
* Check if this is a discrete Thunderbolt/USB4 controller that is
* directly behind the non-USB4 PCIe Root Port marked as
* "ExternalFacingPort". Those are not behind a PCIe tunnel.
*/
if (pcie_switch_directly_under(root, pdev))
return false;
/* PCIe devices after the discrete chip are tunneled */
return true;
}
#ifdef CONFIG_PCI_MMCONFIG
static int check_segment(u16 seg, struct device *dev, char *estr)
{
if (seg) {
dev_err(dev, "%s can't access configuration space under this host bridge\n",
estr);
return -EIO;
}
/*
* Failure in adding MMCFG information is not fatal,
* just can't access extended configuration space of
* devices under this host bridge.
*/
dev_warn(dev, "%s can't access extended configuration space under this bridge\n",
estr);
return 0;
}
static int setup_mcfg_map(struct acpi_pci_root_info *ci)
{
int result, seg;
struct pci_root_info *info;
struct acpi_pci_root *root = ci->root;
struct device *dev = &ci->bridge->dev;
info = container_of(ci, struct pci_root_info, common);
info->start_bus = (u8)root->secondary.start;
info->end_bus = (u8)root->secondary.end;
info->mcfg_added = false;
seg = info->sd.domain;
dev_dbg(dev, "%s(%04x %pR ECAM %pa)\n", __func__, seg,
&root->secondary, &root->mcfg_addr);
/* return success if MMCFG is not in use */
if (raw_pci_ext_ops && raw_pci_ext_ops != &pci_mmcfg)
return 0;
if (!(pci_probe & PCI_PROBE_MMCONF))
return check_segment(seg, dev, "MMCONFIG is disabled,");
result = pci_mmconfig_insert(dev, seg, info->start_bus, info->end_bus,
root->mcfg_addr);
if (result == 0) {
/* enable MMCFG if it hasn't been enabled yet */
if (raw_pci_ext_ops == NULL)
raw_pci_ext_ops = &pci_mmcfg;
info->mcfg_added = true;
} else if (result != -EEXIST)
return check_segment(seg, dev,
"fail to add MMCONFIG information,");
return 0;
}
static void teardown_mcfg_map(struct acpi_pci_root_info *ci)
{
struct pci_root_info *info;
info = container_of(ci, struct pci_root_info, common);
if (info->mcfg_added) {
pci_mmconfig_delete(info->sd.domain,
info->start_bus, info->end_bus);
info->mcfg_added = false;
}
}
#else
static int setup_mcfg_map(struct acpi_pci_root_info *ci)
{
return 0;
}
static void teardown_mcfg_map(struct acpi_pci_root_info *ci)
{
}
#endif
static int pci_acpi_root_get_node(struct acpi_pci_root *root)
{
int busnum = root->secondary.start;
struct acpi_device *device = root->device;
int node = acpi_get_node(device->handle);
if (node == NUMA_NO_NODE) {
node = x86_pci_root_bus_node(busnum);
if (node != 0 && node != NUMA_NO_NODE)
dev_info(&device->dev, FW_BUG "no _PXM; falling back to node %d from hardware (may be inconsistent with ACPI node numbers)\n",
node);
}
if (node != NUMA_NO_NODE && !node_online(node))
node = NUMA_NO_NODE;
return node;
}
static int pci_acpi_root_init_info(struct acpi_pci_root_info *ci)
{
return setup_mcfg_map(ci);
}
static void pci_acpi_root_release_info(struct acpi_pci_root_info *ci)
{
teardown_mcfg_map(ci);
kfree(container_of(ci, struct pci_root_info, common));
}
/*
* An IO port or MMIO resource assigned to a PCI host bridge may be
* consumed by the host bridge itself or available to its child
* bus/devices. The ACPI specification defines a bit (Producer/Consumer)
* to tell whether the resource is consumed by the host bridge itself,
* but firmware hasn't used that bit consistently, so we can't rely on it.
*
* On x86 and IA64 platforms, all IO port and MMIO resources are assumed
* to be available to child bus/devices except one special case:
* IO port [0xCF8-0xCFF] is consumed by the host bridge itself
* to access PCI configuration space.
*
* So explicitly filter out PCI CFG IO ports[0xCF8-0xCFF].
*/
static bool resource_is_pcicfg_ioport(struct resource *res)
{
return (res->flags & IORESOURCE_IO) &&
res->start == 0xCF8 && res->end == 0xCFF;
}
static int pci_acpi_root_prepare_resources(struct acpi_pci_root_info *ci)
{
struct acpi_device *device = ci->bridge;
int busnum = ci->root->secondary.start;
struct resource_entry *entry, *tmp;
int status;
status = acpi_pci_probe_root_resources(ci);
if (pci_use_crs) {
resource_list_for_each_entry_safe(entry, tmp, &ci->resources)
if (resource_is_pcicfg_ioport(entry->res))
resource_list_destroy_entry(entry);
return status;
}
resource_list_for_each_entry_safe(entry, tmp, &ci->resources) {
dev_printk(KERN_DEBUG, &device->dev,
"host bridge window %pR (ignored)\n", entry->res);
resource_list_destroy_entry(entry);
}
x86_pci_root_bus_resources(busnum, &ci->resources);
return 0;
}
static struct acpi_pci_root_ops acpi_pci_root_ops = {
.pci_ops = &pci_root_ops,
.init_info = pci_acpi_root_init_info,
.release_info = pci_acpi_root_release_info,
.prepare_resources = pci_acpi_root_prepare_resources,
};
struct pci_bus *pci_acpi_scan_root(struct acpi_pci_root *root)
{
int domain = root->segment;
int busnum = root->secondary.start;
int node = pci_acpi_root_get_node(root);
struct pci_bus *bus;
if (pci_ignore_seg)
root->segment = domain = 0;
if (domain && !pci_domains_supported) {
pr_warn("pci_bus %04x:%02x: ignored (multiple domains not supported)\n",
domain, busnum);
return NULL;
}
bus = pci_find_bus(domain, busnum);
if (bus) {
/*
* If the desired bus has been scanned already, replace
* its bus->sysdata.
*/
struct pci_sysdata sd = {
.domain = domain,
.node = node,
.companion = root->device
};
memcpy(bus->sysdata, &sd, sizeof(sd));
} else {
struct pci_root_info *info;
info = kzalloc_obj(*info);
if (!info)
dev_err(&root->device->dev,
"pci_bus %04x:%02x: ignored (out of memory)\n",
domain, busnum);
else {
info->sd.domain = domain;
info->sd.node = node;
info->sd.companion = root->device;
bus = acpi_pci_root_create(root, &acpi_pci_root_ops,
&info->common, &info->sd);
}
}
/* After the PCI-E bus has been walked and all devices discovered,
* configure any settings of the fabric that might be necessary.
*/
if (bus) {
struct pci_bus *child;
list_for_each_entry(child, &bus->children, node)
pcie_bus_configure_settings(child);
}
return bus;
}
int pcibios_root_bridge_prepare(struct pci_host_bridge *bridge)
{
/*
* We pass NULL as parent to pci_create_root_bus(), so if it is not NULL
* here, pci_create_root_bus() has been called by someone else and
* sysdata is likely to be different from what we expect. Let it go in
* that case.
*/
if (!bridge->dev.parent) {
struct pci_sysdata *sd = bridge->bus->sysdata;
ACPI_COMPANION_SET(&bridge->dev, sd->companion);
}
return 0;
}
int __init pci_acpi_init(void)
{
struct pci_dev *dev = NULL;
if (acpi_noirq)
return -ENODEV;
pr_info("Using ACPI for IRQ routing\n");
acpi_irq_penalty_init();
pcibios_enable_irq = acpi_pci_irq_enable;
pcibios_disable_irq = acpi_pci_irq_disable;
x86_init.pci.init_irq = x86_init_noop;
if (pci_routeirq) {
/*
* PCI IRQ routing is set up by pci_enable_device(), but we
* also do it here in case there are still broken drivers that
* don't use pci_enable_device().
*/
pr_info("Routing PCI interrupts for all devices because \"pci=routeirq\" specified\n");
for_each_pci_dev(dev)
acpi_pci_irq_enable(dev);
}
return 0;
}
linux-7.2.3/arch/x86/pci/amd_bus.c
+412
View File
@@ -0,0 +1,412 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/topology.h>
#include <linux/cpu.h>
#include <linux/range.h>
#include <asm/amd/nb.h>
#include <asm/pci_x86.h>
#include <asm/pci-direct.h>
#include "bus_numa.h"
#define AMD_NB_F0_NODE_ID 0x60
#define AMD_NB_F0_UNIT_ID 0x64
#define AMD_NB_F1_CONFIG_MAP_REG 0xe0
#define RANGE_NUM 16
#define AMD_NB_F1_CONFIG_MAP_RANGES 4
struct amd_hostbridge {
u32 bus;
u32 slot;
u32 device;
};
/*
* IMPORTANT NOTE:
* hb_probes[] and early_root_info_init() is in maintenance mode.
* It only supports K8, Fam10h, Fam11h, and Fam15h_00h-0fh .
* Future processor will rely on information in ACPI.
*/
static struct amd_hostbridge hb_probes[] __initdata = {
{ 0, 0x18, 0x1100 }, /* K8 */
{ 0, 0x18, 0x1200 }, /* Family10h */
{ 0xff, 0, 0x1200 }, /* Family10h */
{ 0, 0x18, 0x1300 }, /* Family11h */
{ 0, 0x18, 0x1600 }, /* Family15h */
};
static struct pci_root_info __init *find_pci_root_info(int node, int link)
{
struct pci_root_info *info;
/* find the position */
list_for_each_entry(info, &pci_root_infos, list)
if (info->node == node && info->link == link)
return info;
return NULL;
}
static inline resource_size_t cap_resource(u64 val)
{
if (val > RESOURCE_SIZE_MAX)
return RESOURCE_SIZE_MAX;
return val;
}
/**
* early_root_info_init()
* called before pcibios_scan_root and pci_scan_bus
* fills the mp_bus_to_cpumask array based according
* to the LDT Bus Number Registers found in the northbridge.
*/
static int __init early_root_info_init(void)
{
int i;
unsigned bus;
unsigned slot;
int node;
int link;
int def_node;
int def_link;
struct pci_root_info *info;
u32 reg;
u64 start;
u64 end;
struct range range[RANGE_NUM];
u64 val;
u32 address;
bool found;
struct resource fam10h_mmconf_res, *fam10h_mmconf;
u64 fam10h_mmconf_start;
u64 fam10h_mmconf_end;
if (!early_pci_allowed())
return -1;
found = false;
for (i = 0; i < ARRAY_SIZE(hb_probes); i++) {
u32 id;
u16 device;
u16 vendor;
bus = hb_probes[i].bus;
slot = hb_probes[i].slot;
id = read_pci_config(bus, slot, 0, PCI_VENDOR_ID);
vendor = id & 0xffff;
device = (id>>16) & 0xffff;
if (vendor != PCI_VENDOR_ID_AMD &&
vendor != PCI_VENDOR_ID_HYGON)
continue;
if (hb_probes[i].device == device) {
found = true;
break;
}
}
if (!found)
return 0;
/*
* We should learn topology and routing information from _PXM and
* _CRS methods in the ACPI namespace. We extract node numbers
* here to work around BIOSes that don't supply _PXM.
*/
for (i = 0; i < AMD_NB_F1_CONFIG_MAP_RANGES; i++) {
int min_bus;
int max_bus;
reg = read_pci_config(bus, slot, 1,
AMD_NB_F1_CONFIG_MAP_REG + (i << 2));
/* Check if that register is enabled for bus range */
if ((reg & 7) != 3)
continue;
min_bus = (reg >> 16) & 0xff;
max_bus = (reg >> 24) & 0xff;
node = (reg >> 4) & 0x07;
link = (reg >> 8) & 0x03;
alloc_pci_root_info(min_bus, max_bus, node, link);
}
/*
* The following code extracts routing information for use on old
* systems where Linux doesn't automatically use host bridge _CRS
* methods (or when the user specifies "pci=nocrs").
*
* We only do this through Fam11h, because _CRS should be enough on
* newer systems.
*/
if (boot_cpu_data.x86 > 0x11)
return 0;
/* get the default node and link for left over res */
reg = read_pci_config(bus, slot, 0, AMD_NB_F0_NODE_ID);
def_node = (reg >> 8) & 0x07;
reg = read_pci_config(bus, slot, 0, AMD_NB_F0_UNIT_ID);
def_link = (reg >> 8) & 0x03;
memset(range, 0, sizeof(range));
add_range(range, RANGE_NUM, 0, 0, 0xffff + 1);
/* io port resource */
for (i = 0; i < 4; i++) {
reg = read_pci_config(bus, slot, 1, 0xc0 + (i << 3));
if (!(reg & 3))
continue;
start = reg & 0xfff000;
reg = read_pci_config(bus, slot, 1, 0xc4 + (i << 3));
node = reg & 0x07;
link = (reg >> 4) & 0x03;
end = (reg & 0xfff000) | 0xfff;
info = find_pci_root_info(node, link);
if (!info)
continue; /* not found */
printk(KERN_DEBUG "node %d link %d: io port [%llx, %llx]\n",
node, link, start, end);
/* kernel only handle 16 bit only */
if (end > 0xffff)
end = 0xffff;
update_res(info, start, end, IORESOURCE_IO, 1);
subtract_range(range, RANGE_NUM, start, end + 1);
}
/* add left over io port range to def node/link, [0, 0xffff] */
/* find the position */
info = find_pci_root_info(def_node, def_link);
if (info) {
for (i = 0; i < RANGE_NUM; i++) {
if (!range[i].end)
continue;
update_res(info, range[i].start, range[i].end - 1,
IORESOURCE_IO, 1);
}
}
memset(range, 0, sizeof(range));
/* 0xfd00000000-0xffffffffff for HT */
end = cap_resource((0xfdULL<<32) - 1);
end++;
add_range(range, RANGE_NUM, 0, 0, end);
/* need to take out [0, TOM) for RAM*/
address = MSR_K8_TOP_MEM1;
rdmsrq(address, val);
end = (val & 0xffffff800000ULL);
printk(KERN_INFO "TOM: %016llx aka %lldM\n", end, end>>20);
if (end < (1ULL<<32))
subtract_range(range, RANGE_NUM, 0, end);
/* get mmconfig */
fam10h_mmconf = amd_get_mmconfig_range(&fam10h_mmconf_res);
/* need to take out mmconf range */
if (fam10h_mmconf) {
printk(KERN_DEBUG "Fam 10h mmconf %pR\n", fam10h_mmconf);
fam10h_mmconf_start = fam10h_mmconf->start;
fam10h_mmconf_end = fam10h_mmconf->end;
subtract_range(range, RANGE_NUM, fam10h_mmconf_start,
fam10h_mmconf_end + 1);
} else {
fam10h_mmconf_start = 0;
fam10h_mmconf_end = 0;
}
/* mmio resource */
for (i = 0; i < 8; i++) {
reg = read_pci_config(bus, slot, 1, 0x80 + (i << 3));
if (!(reg & 3))
continue;
start = reg & 0xffffff00; /* 39:16 on 31:8*/
start <<= 8;
reg = read_pci_config(bus, slot, 1, 0x84 + (i << 3));
node = reg & 0x07;
link = (reg >> 4) & 0x03;
end = (reg & 0xffffff00);
end <<= 8;
end |= 0xffff;
info = find_pci_root_info(node, link);
if (!info)
continue;
printk(KERN_DEBUG "node %d link %d: mmio [%llx, %llx]",
node, link, start, end);
/*
* some sick allocation would have range overlap with fam10h
* mmconf range, so need to update start and end.
*/
if (fam10h_mmconf_end) {
int changed = 0;
u64 endx = 0;
if (start >= fam10h_mmconf_start &&
start <= fam10h_mmconf_end) {
start = fam10h_mmconf_end + 1;
changed = 1;
}
if (end >= fam10h_mmconf_start &&
end <= fam10h_mmconf_end) {
end = fam10h_mmconf_start - 1;
changed = 1;
}
if (start < fam10h_mmconf_start &&
end > fam10h_mmconf_end) {
/* we got a hole */
endx = fam10h_mmconf_start - 1;
update_res(info, start, endx, IORESOURCE_MEM, 0);
subtract_range(range, RANGE_NUM, start,
endx + 1);
printk(KERN_CONT " ==> [%llx, %llx]", start, endx);
start = fam10h_mmconf_end + 1;
changed = 1;
}
if (changed) {
if (start <= end) {
printk(KERN_CONT " %s [%llx, %llx]", endx ? "and" : "==>", start, end);
} else {
printk(KERN_CONT "%s\n", endx?"":" ==> none");
continue;
}
}
}
update_res(info, cap_resource(start), cap_resource(end),
IORESOURCE_MEM, 1);
subtract_range(range, RANGE_NUM, start, end + 1);
printk(KERN_CONT "\n");
}
/* need to take out [4G, TOM2) for RAM*/
/* SYS_CFG */
address = MSR_AMD64_SYSCFG;
rdmsrq(address, val);
/* TOP_MEM2 is enabled? */
if (val & (1<<21)) {
/* TOP_MEM2 */
address = MSR_K8_TOP_MEM2;
rdmsrq(address, val);
end = (val & 0xffffff800000ULL);
printk(KERN_INFO "TOM2: %016llx aka %lldM\n", end, end>>20);
subtract_range(range, RANGE_NUM, 1ULL<<32, end);
}
/*
* add left over mmio range to def node/link ?
* that is tricky, just record range in from start_min to 4G
*/
info = find_pci_root_info(def_node, def_link);
if (info) {
for (i = 0; i < RANGE_NUM; i++) {
if (!range[i].end)
continue;
update_res(info, cap_resource(range[i].start),
cap_resource(range[i].end - 1),
IORESOURCE_MEM, 1);
}
}
list_for_each_entry(info, &pci_root_infos, list) {
int busnum;
struct pci_root_res *root_res;
busnum = info->busn.start;
printk(KERN_DEBUG "bus: %pR on node %x link %x\n",
&info->busn, info->node, info->link);
list_for_each_entry(root_res, &info->resources, list)
printk(KERN_DEBUG "bus: %02x %pR\n",
busnum, &root_res->res);
}
return 0;
}
#define ENABLE_CF8_EXT_CFG (1ULL << 46)
static int amd_bus_cpu_online(unsigned int cpu)
{
u64 reg;
rdmsrq(MSR_AMD64_NB_CFG, reg);
if (!(reg & ENABLE_CF8_EXT_CFG)) {
reg |= ENABLE_CF8_EXT_CFG;
wrmsrq(MSR_AMD64_NB_CFG, reg);
}
return 0;
}
static void __init pci_enable_pci_io_ecs(void)
{
#ifdef CONFIG_AMD_NB
unsigned int i, n;
for (n = i = 0; !n && amd_nb_bus_dev_ranges[i].dev_limit; ++i) {
u8 bus = amd_nb_bus_dev_ranges[i].bus;
u8 slot = amd_nb_bus_dev_ranges[i].dev_base;
u8 limit = amd_nb_bus_dev_ranges[i].dev_limit;
for (; slot < limit; ++slot) {
u32 val = read_pci_config(bus, slot, 3, 0);
if (!early_is_amd_nb(val))
continue;
val = read_pci_config(bus, slot, 3, 0x8c);
if (!(val & (ENABLE_CF8_EXT_CFG >> 32))) {
val |= ENABLE_CF8_EXT_CFG >> 32;
write_pci_config(bus, slot, 3, 0x8c, val);
}
++n;
}
}
#endif
}
static int __init pci_io_ecs_init(void)
{
int ret;
/* assume all cpus from fam10h have IO ECS */
if (boot_cpu_data.x86 < 0x10)
return 0;
/* Try the PCI method first. */
if (early_pci_allowed())
pci_enable_pci_io_ecs();
ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "pci/amd_bus:online",
amd_bus_cpu_online, NULL);
WARN_ON(ret < 0);
pci_probe |= PCI_HAS_IO_ECS;
return 0;
}
static int __init amd_postcore_init(void)
{
if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD &&
boot_cpu_data.x86_vendor != X86_VENDOR_HYGON)
return 0;
early_root_info_init();
pci_io_ecs_init();
return 0;
}
postcore_initcall(amd_postcore_init);
linux-7.2.3/arch/x86/pci/broadcom_bus.c
+112
View File
@@ -0,0 +1,112 @@
// SPDX-License-Identifier: GPL-2.0-or-later
/*
* Read address ranges from a Broadcom CNB20LE Host Bridge
*
* Copyright (c) 2010 Ira W. Snyder <iws@ovro.caltech.edu>
*/
#include <linux/acpi.h>
#include <linux/delay.h>
#include <linux/dmi.h>
#include <linux/pci.h>
#include <linux/init.h>
#include <asm/pci_x86.h>
#include <asm/pci-direct.h>
#include "bus_numa.h"
static void __init cnb20le_res(u8 bus, u8 slot, u8 func)
{
struct pci_root_info *info;
struct pci_root_res *root_res;
struct resource res;
u16 word1, word2;
u8 fbus, lbus;
/* read the PCI bus numbers */
fbus = read_pci_config_byte(bus, slot, func, 0x44);
lbus = read_pci_config_byte(bus, slot, func, 0x45);
info = alloc_pci_root_info(fbus, lbus, 0, 0);
/*
* Add the legacy IDE ports on bus 0
*
* These do not exist anywhere in the bridge registers, AFAICT. I do
* not have the datasheet, so this is the best I can do.
*/
if (fbus == 0) {
update_res(info, 0x01f0, 0x01f7, IORESOURCE_IO, 0);
update_res(info, 0x03f6, 0x03f6, IORESOURCE_IO, 0);
update_res(info, 0x0170, 0x0177, IORESOURCE_IO, 0);
update_res(info, 0x0376, 0x0376, IORESOURCE_IO, 0);
update_res(info, 0xffa0, 0xffaf, IORESOURCE_IO, 0);
}
/* read the non-prefetchable memory window */
word1 = read_pci_config_16(bus, slot, func, 0xc0);
word2 = read_pci_config_16(bus, slot, func, 0xc2);
if (word1 != word2) {
res.start = ((resource_size_t) word1 << 16) | 0x0000;
res.end = ((resource_size_t) word2 << 16) | 0xffff;
res.flags = IORESOURCE_MEM;
update_res(info, res.start, res.end, res.flags, 0);
}
/* read the prefetchable memory window */
word1 = read_pci_config_16(bus, slot, func, 0xc4);
word2 = read_pci_config_16(bus, slot, func, 0xc6);
if (word1 != word2) {
res.start = ((resource_size_t) word1 << 16) | 0x0000;
res.end = ((resource_size_t) word2 << 16) | 0xffff;
res.flags = IORESOURCE_MEM | IORESOURCE_PREFETCH;
update_res(info, res.start, res.end, res.flags, 0);
}
/* read the IO port window */
word1 = read_pci_config_16(bus, slot, func, 0xd0);
word2 = read_pci_config_16(bus, slot, func, 0xd2);
if (word1 != word2) {
res.start = word1;
res.end = word2;
res.flags = IORESOURCE_IO;
update_res(info, res.start, res.end, res.flags, 0);
}
/* print information about this host bridge */
res.start = fbus;
res.end = lbus;
res.flags = IORESOURCE_BUS;
printk(KERN_INFO "CNB20LE PCI Host Bridge (domain 0000 %pR)\n", &res);
list_for_each_entry(root_res, &info->resources, list)
printk(KERN_INFO "host bridge window %pR\n", &root_res->res);
}
static int __init broadcom_postcore_init(void)
{
u8 bus = 0, slot = 0;
u32 id;
u16 vendor, device;
#ifdef CONFIG_ACPI
/*
* We should get host bridge information from ACPI unless the BIOS
* doesn't support it.
*/
if (!acpi_disabled && acpi_os_get_root_pointer())
return 0;
#endif
id = read_pci_config(bus, slot, 0, PCI_VENDOR_ID);
vendor = id & 0xffff;
device = (id >> 16) & 0xffff;
if (vendor == PCI_VENDOR_ID_SERVERWORKS &&
device == PCI_DEVICE_ID_SERVERWORKS_LE) {
cnb20le_res(bus, slot, 0);
cnb20le_res(bus, slot, 1);
}
return 0;
}
postcore_initcall(broadcom_postcore_init);
linux-7.2.3/arch/x86/pci/bus_numa.c
+146
View File
@@ -0,0 +1,146 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/range.h>
#include "bus_numa.h"
LIST_HEAD(pci_root_infos);
static struct pci_root_info *x86_find_pci_root_info(int bus)
{
struct pci_root_info *info;
list_for_each_entry(info, &pci_root_infos, list)
if (info->busn.start == bus)
return info;
return NULL;
}
int x86_pci_root_bus_node(int bus)
{
struct pci_root_info *info = x86_find_pci_root_info(bus);
if (!info)
return NUMA_NO_NODE;
return info->node;
}
void x86_pci_root_bus_resources(int bus, struct list_head *resources)
{
struct pci_root_info *info = x86_find_pci_root_info(bus);
struct pci_root_res *root_res;
struct resource_entry *window;
bool found = false;
if (!info)
goto default_resources;
printk(KERN_DEBUG "PCI: root bus %02x: hardware-probed resources\n",
bus);
/* already added by acpi ? */
resource_list_for_each_entry(window, resources)
if (window->res->flags & IORESOURCE_BUS) {
found = true;
break;
}
if (!found)
pci_add_resource(resources, &info->busn);
list_for_each_entry(root_res, &info->resources, list)
pci_add_resource(resources, &root_res->res);
return;
default_resources:
/*
* We don't have any host bridge aperture information from the
* "native host bridge drivers," e.g., amd_bus or broadcom_bus,
* so fall back to the defaults historically used by pci_create_bus().
*/
printk(KERN_DEBUG "PCI: root bus %02x: using default resources\n", bus);
pci_add_resource(resources, &ioport_resource);
pci_add_resource(resources, &iomem_resource);
}
struct pci_root_info __init *alloc_pci_root_info(int bus_min, int bus_max,
int node, int link)
{
struct pci_root_info *info;
info = kzalloc_obj(*info);
if (!info)
return info;
sprintf(info->name, "PCI Bus #%02x", bus_min);
INIT_LIST_HEAD(&info->resources);
info->busn.name = info->name;
info->busn.start = bus_min;
info->busn.end = bus_max;
info->busn.flags = IORESOURCE_BUS;
info->node = node;
info->link = link;
list_add_tail(&info->list, &pci_root_infos);
return info;
}
void update_res(struct pci_root_info *info, resource_size_t start,
resource_size_t end, unsigned long flags, int merge)
{
struct resource *res;
struct pci_root_res *root_res;
if (start > end)
return;
if (start == RESOURCE_SIZE_MAX)
return;
if (!merge)
goto addit;
/* try to merge it with old one */
list_for_each_entry(root_res, &info->resources, list) {
resource_size_t final_start, final_end;
resource_size_t common_start, common_end;
res = &root_res->res;
if (res->flags != flags)
continue;
common_start = max(res->start, start);
common_end = min(res->end, end);
if (common_start > common_end + 1)
continue;
final_start = min(res->start, start);
final_end = max(res->end, end);
res->start = final_start;
res->end = final_end;
return;
}
addit:
/* need to add that */
root_res = kzalloc_obj(*root_res);
if (!root_res)
return;
res = &root_res->res;
res->name = info->name;
res->flags = flags;
res->start = start;
res->end = end;
list_add_tail(&root_res->list, &info->resources);
}
linux-7.2.3/arch/x86/pci/bus_numa.h
+27
View File
@@ -0,0 +1,27 @@
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef __BUS_NUMA_H
#define __BUS_NUMA_H
/*
* sub bus (transparent) will use entres from 3 to store extra from
* root, so need to make sure we have enough slot there.
*/
struct pci_root_res {
struct list_head list;
struct resource res;
};
struct pci_root_info {
struct list_head list;
char name[12];
struct list_head resources;
struct resource busn;
int node;
int link;
};
extern struct list_head pci_root_infos;
struct pci_root_info *alloc_pci_root_info(int bus_min, int bus_max,
int node, int link);
extern void update_res(struct pci_root_info *info, resource_size_t start,
resource_size_t end, unsigned long flags, int merge);
#endif
linux-7.2.3/arch/x86/pci/ce4100.c
+318
View File
@@ -0,0 +1,318 @@
// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright(c) 2010 Intel Corporation. All rights reserved.
*
* Contact Information:
* Intel Corporation
* 2200 Mission College Blvd.
* Santa Clara, CA 97052
*
* This provides access methods for PCI registers that mis-behave on
* the CE4100. Each register can be assigned a private init, read and
* write routine. The exception to this is the bridge device. The
* bridge device is the only device on bus zero (0) that requires any
* fixup so it is a special case ATM
*/
#include <linux/kernel.h>
#include <linux/pci.h>
#include <linux/init.h>
#include <asm/ce4100.h>
#include <asm/pci_x86.h>
struct sim_reg {
u32 value;
u32 mask;
};
struct sim_dev_reg {
int dev_func;
int reg;
void (*init)(struct sim_dev_reg *reg);
void (*read)(struct sim_dev_reg *reg, u32 *value);
void (*write)(struct sim_dev_reg *reg, u32 value);
struct sim_reg sim_reg;
};
#define MB (1024 * 1024)
#define KB (1024)
#define SIZE_TO_MASK(size) (~(size - 1))
#define DEFINE_REG(device, func, offset, size, init_op, read_op, write_op)\
{ PCI_DEVFN(device, func), offset, init_op, read_op, write_op,\
{0, SIZE_TO_MASK(size)} },
/*
* All read/write functions are called with pci_config_lock held.
*/
static void reg_init(struct sim_dev_reg *reg)
{
pci_direct_conf1.read(0, 1, reg->dev_func, reg->reg, 4,
&reg->sim_reg.value);
}
static void reg_read(struct sim_dev_reg *reg, u32 *value)
{
*value = reg->sim_reg.value;
}
static void reg_write(struct sim_dev_reg *reg, u32 value)
{
reg->sim_reg.value = (value & reg->sim_reg.mask) |
(reg->sim_reg.value & ~reg->sim_reg.mask);
}
static void sata_reg_init(struct sim_dev_reg *reg)
{
pci_direct_conf1.read(0, 1, PCI_DEVFN(14, 0), 0x10, 4,
&reg->sim_reg.value);
reg->sim_reg.value += 0x400;
}
static void ehci_reg_read(struct sim_dev_reg *reg, u32 *value)
{
reg_read(reg, value);
if (*value != reg->sim_reg.mask)
*value |= 0x100;
}
static void sata_revid_init(struct sim_dev_reg *reg)
{
reg->sim_reg.value = 0x01060100;
reg->sim_reg.mask = 0;
}
static void sata_revid_read(struct sim_dev_reg *reg, u32 *value)
{
reg_read(reg, value);
}
static void reg_noirq_read(struct sim_dev_reg *reg, u32 *value)
{
/* force interrupt pin value to 0 */
*value = reg->sim_reg.value & 0xfff00ff;
}
static struct sim_dev_reg bus1_fixups[] = {
DEFINE_REG(2, 0, 0x10, (16*MB), reg_init, reg_read, reg_write)
DEFINE_REG(2, 0, 0x14, (256), reg_init, reg_read, reg_write)
DEFINE_REG(2, 1, 0x10, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(3, 0, 0x10, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(4, 0, 0x10, (128*KB), reg_init, reg_read, reg_write)
DEFINE_REG(4, 1, 0x10, (128*KB), reg_init, reg_read, reg_write)
DEFINE_REG(6, 0, 0x10, (512*KB), reg_init, reg_read, reg_write)
DEFINE_REG(6, 1, 0x10, (512*KB), reg_init, reg_read, reg_write)
DEFINE_REG(6, 2, 0x10, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(8, 0, 0x10, (1*MB), reg_init, reg_read, reg_write)
DEFINE_REG(8, 1, 0x10, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(8, 2, 0x10, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(9, 0, 0x10 , (1*MB), reg_init, reg_read, reg_write)
DEFINE_REG(9, 0, 0x14, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(10, 0, 0x10, (256), reg_init, reg_read, reg_write)
DEFINE_REG(10, 0, 0x14, (256*MB), reg_init, reg_read, reg_write)
DEFINE_REG(11, 0, 0x10, (256), reg_init, reg_read, reg_write)
DEFINE_REG(11, 0, 0x14, (256), reg_init, reg_read, reg_write)
DEFINE_REG(11, 1, 0x10, (256), reg_init, reg_read, reg_write)
DEFINE_REG(11, 2, 0x10, (256), reg_init, reg_read, reg_write)
DEFINE_REG(11, 2, 0x14, (256), reg_init, reg_read, reg_write)
DEFINE_REG(11, 2, 0x18, (256), reg_init, reg_read, reg_write)
DEFINE_REG(11, 3, 0x10, (256), reg_init, reg_read, reg_write)
DEFINE_REG(11, 3, 0x14, (256), reg_init, reg_read, reg_write)
DEFINE_REG(11, 4, 0x10, (256), reg_init, reg_read, reg_write)
DEFINE_REG(11, 5, 0x10, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(11, 6, 0x10, (256), reg_init, reg_read, reg_write)
DEFINE_REG(11, 7, 0x10, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(11, 7, 0x3c, 256, reg_init, reg_noirq_read, reg_write)
DEFINE_REG(12, 0, 0x10, (128*KB), reg_init, reg_read, reg_write)
DEFINE_REG(12, 0, 0x14, (256), reg_init, reg_read, reg_write)
DEFINE_REG(12, 1, 0x10, (1024), reg_init, reg_read, reg_write)
DEFINE_REG(13, 0, 0x10, (32*KB), reg_init, ehci_reg_read, reg_write)
DEFINE_REG(13, 1, 0x10, (32*KB), reg_init, ehci_reg_read, reg_write)
DEFINE_REG(14, 0, 0x8, 0, sata_revid_init, sata_revid_read, 0)
DEFINE_REG(14, 0, 0x10, 0, reg_init, reg_read, reg_write)
DEFINE_REG(14, 0, 0x14, 0, reg_init, reg_read, reg_write)
DEFINE_REG(14, 0, 0x18, 0, reg_init, reg_read, reg_write)
DEFINE_REG(14, 0, 0x1C, 0, reg_init, reg_read, reg_write)
DEFINE_REG(14, 0, 0x20, 0, reg_init, reg_read, reg_write)
DEFINE_REG(14, 0, 0x24, (0x200), sata_reg_init, reg_read, reg_write)
DEFINE_REG(15, 0, 0x10, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(15, 0, 0x14, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(16, 0, 0x10, (64*KB), reg_init, reg_read, reg_write)
DEFINE_REG(16, 0, 0x14, (64*MB), reg_init, reg_read, reg_write)
DEFINE_REG(16, 0, 0x18, (64*MB), reg_init, reg_read, reg_write)
DEFINE_REG(16, 0, 0x3c, 256, reg_init, reg_noirq_read, reg_write)
DEFINE_REG(17, 0, 0x10, (128*KB), reg_init, reg_read, reg_write)
DEFINE_REG(18, 0, 0x10, (1*KB), reg_init, reg_read, reg_write)
DEFINE_REG(18, 0, 0x3c, 256, reg_init, reg_noirq_read, reg_write)
};
static void __init init_sim_regs(void)
{
int i;
for (i = 0; i < ARRAY_SIZE(bus1_fixups); i++) {
if (bus1_fixups[i].init)
bus1_fixups[i].init(&bus1_fixups[i]);
}
}
static inline void extract_bytes(u32 *value, int reg, int len)
{
uint32_t mask;
*value >>= ((reg & 3) * 8);
mask = 0xFFFFFFFF >> ((4 - len) * 8);
*value &= mask;
}
static int bridge_read(unsigned int devfn, int reg, int len, u32 *value)
{
u32 av_bridge_base, av_bridge_limit;
int retval = 0;
switch (reg) {
/* Make BARs appear to not request any memory. */
case PCI_BASE_ADDRESS_0:
case PCI_BASE_ADDRESS_0 + 1:
case PCI_BASE_ADDRESS_0 + 2:
case PCI_BASE_ADDRESS_0 + 3:
*value = 0;
break;
/* Since subordinate bus number register is hardwired
* to zero and read only, so do the simulation.
*/
case PCI_PRIMARY_BUS:
if (len == 4)
*value = 0x00010100;
break;
case PCI_SUBORDINATE_BUS:
*value = 1;
break;
case PCI_MEMORY_BASE:
case PCI_MEMORY_LIMIT:
/* Get the A/V bridge base address. */
pci_direct_conf1.read(0, 0, devfn,
PCI_BASE_ADDRESS_0, 4, &av_bridge_base);
av_bridge_limit = av_bridge_base + (512*MB - 1);
av_bridge_limit >>= 16;
av_bridge_limit &= 0xFFF0;
av_bridge_base >>= 16;
av_bridge_base &= 0xFFF0;
if (reg == PCI_MEMORY_LIMIT)
*value = av_bridge_limit;
else if (len == 2)
*value = av_bridge_base;
else
*value = (av_bridge_limit << 16) | av_bridge_base;
break;
/* Make prefetchable memory limit smaller than prefetchable
* memory base, so not claim prefetchable memory space.
*/
case PCI_PREF_MEMORY_BASE:
*value = 0xFFF0;
break;
case PCI_PREF_MEMORY_LIMIT:
*value = 0x0;
break;
/* Make IO limit smaller than IO base, so not claim IO space. */
case PCI_IO_BASE:
*value = 0xF0;
break;
case PCI_IO_LIMIT:
*value = 0;
break;
default:
retval = 1;
}
return retval;
}
static int ce4100_bus1_read(unsigned int devfn, int reg, int len, u32 *value)
{
unsigned long flags;
int i;
for (i = 0; i < ARRAY_SIZE(bus1_fixups); i++) {
if (bus1_fixups[i].dev_func == devfn &&
bus1_fixups[i].reg == (reg & ~3) &&
bus1_fixups[i].read) {
raw_spin_lock_irqsave(&pci_config_lock, flags);
bus1_fixups[i].read(&(bus1_fixups[i]), value);
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
extract_bytes(value, reg, len);
return 0;
}
}
return -1;
}
static int ce4100_conf_read(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 *value)
{
WARN_ON(seg);
if (bus == 1 && !ce4100_bus1_read(devfn, reg, len, value))
return 0;
if (bus == 0 && (PCI_DEVFN(1, 0) == devfn) &&
!bridge_read(devfn, reg, len, value))
return 0;
return pci_direct_conf1.read(seg, bus, devfn, reg, len, value);
}
static int ce4100_bus1_write(unsigned int devfn, int reg, int len, u32 value)
{
unsigned long flags;
int i;
for (i = 0; i < ARRAY_SIZE(bus1_fixups); i++) {
if (bus1_fixups[i].dev_func == devfn &&
bus1_fixups[i].reg == (reg & ~3) &&
bus1_fixups[i].write) {
raw_spin_lock_irqsave(&pci_config_lock, flags);
bus1_fixups[i].write(&(bus1_fixups[i]), value);
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
return 0;
}
}
return -1;
}
static int ce4100_conf_write(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 value)
{
WARN_ON(seg);
if (bus == 1 && !ce4100_bus1_write(devfn, reg, len, value))
return 0;
/* Discard writes to A/V bridge BAR. */
if (bus == 0 && PCI_DEVFN(1, 0) == devfn &&
((reg & ~3) == PCI_BASE_ADDRESS_0))
return 0;
return pci_direct_conf1.write(seg, bus, devfn, reg, len, value);
}
static const struct pci_raw_ops ce4100_pci_conf = {
.read = ce4100_conf_read,
.write = ce4100_conf_write,
};
int __init ce4100_pci_init(void)
{
init_sim_regs();
raw_pci_ops = &ce4100_pci_conf;
/* Indicate caller that it should invoke pci_legacy_init() */
return 1;
}
linux-7.2.3/arch/x86/pci/common.c
+734
View File
@@ -0,0 +1,734 @@
// SPDX-License-Identifier: GPL-2.0-only
/*
* Low-Level PCI Support for PC
*
* (c) 1999--2000 Martin Mares <mj@ucw.cz>
*/
#include <linux/sched.h>
#include <linux/pci.h>
#include <linux/pci-acpi.h>
#include <linux/ioport.h>
#include <linux/init.h>
#include <linux/dmi.h>
#include <linux/slab.h>
#include <asm/acpi.h>
#include <asm/segment.h>
#include <asm/io.h>
#include <asm/smp.h>
#include <asm/pci_x86.h>
#include <asm/setup.h>
#include <asm/irqdomain.h>
unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2 |
PCI_PROBE_MMCONF;
static int pci_bf_sort;
int pci_routeirq;
int noioapicquirk;
#ifdef CONFIG_X86_REROUTE_FOR_BROKEN_BOOT_IRQS
int noioapicreroute = 0;
#else
int noioapicreroute = 1;
#endif
int pcibios_last_bus = -1;
unsigned long pirq_table_addr;
const struct pci_raw_ops *__read_mostly raw_pci_ops;
const struct pci_raw_ops *__read_mostly raw_pci_ext_ops;
int raw_pci_read(unsigned int domain, unsigned int bus, unsigned int devfn,
int reg, int len, u32 *val)
{
if (domain == 0 && reg < 256 && raw_pci_ops)
return raw_pci_ops->read(domain, bus, devfn, reg, len, val);
if (raw_pci_ext_ops)
return raw_pci_ext_ops->read(domain, bus, devfn, reg, len, val);
return -EINVAL;
}
int raw_pci_write(unsigned int domain, unsigned int bus, unsigned int devfn,
int reg, int len, u32 val)
{
if (domain == 0 && reg < 256 && raw_pci_ops)
return raw_pci_ops->write(domain, bus, devfn, reg, len, val);
if (raw_pci_ext_ops)
return raw_pci_ext_ops->write(domain, bus, devfn, reg, len, val);
return -EINVAL;
}
static int pci_read(struct pci_bus *bus, unsigned int devfn, int where, int size, u32 *value)
{
return raw_pci_read(pci_domain_nr(bus), bus->number,
devfn, where, size, value);
}
static int pci_write(struct pci_bus *bus, unsigned int devfn, int where, int size, u32 value)
{
return raw_pci_write(pci_domain_nr(bus), bus->number,
devfn, where, size, value);
}
struct pci_ops pci_root_ops = {
.read = pci_read,
.write = pci_write,
};
/*
* This interrupt-safe spinlock protects all accesses to PCI configuration
* space, except for the mmconfig (ECAM) based operations.
*/
DEFINE_RAW_SPINLOCK(pci_config_lock);
static int __init can_skip_ioresource_align(const struct dmi_system_id *d)
{
pci_probe |= PCI_CAN_SKIP_ISA_ALIGN;
printk(KERN_INFO "PCI: %s detected, can skip ISA alignment\n", d->ident);
return 0;
}
static const struct dmi_system_id can_skip_pciprobe_dmi_table[] __initconst = {
/*
* Systems where PCI IO resource ISA alignment can be skipped
* when the ISA enable bit in the bridge control is not set
*/
{
.callback = can_skip_ioresource_align,
.ident = "IBM System x3800",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
DMI_MATCH(DMI_PRODUCT_NAME, "x3800"),
},
},
{
.callback = can_skip_ioresource_align,
.ident = "IBM System x3850",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
DMI_MATCH(DMI_PRODUCT_NAME, "x3850"),
},
},
{
.callback = can_skip_ioresource_align,
.ident = "IBM System x3950",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
DMI_MATCH(DMI_PRODUCT_NAME, "x3950"),
},
},
{}
};
void __init dmi_check_skip_isa_align(void)
{
dmi_check_system(can_skip_pciprobe_dmi_table);
}
static void pcibios_fixup_device_resources(struct pci_dev *dev)
{
struct resource *rom_r = &dev->resource[PCI_ROM_RESOURCE];
struct resource *bar_r;
int bar;
if (pci_probe & PCI_NOASSIGN_BARS) {
/*
* If the BIOS did not assign the BAR, zero out the
* resource so the kernel doesn't attempt to assign
* it later on in pci_assign_unassigned_resources
*/
for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) {
bar_r = &dev->resource[bar];
if (bar_r->start == 0 && bar_r->end != 0) {
bar_r->flags = 0;
bar_r->end = 0;
}
}
}
if (pci_probe & PCI_NOASSIGN_ROMS) {
if (rom_r->parent)
return;
if (rom_r->start) {
/* we deal with BIOS assigned ROM later */
return;
}
rom_r->start = rom_r->end = rom_r->flags = 0;
}
}
/*
* Called after each bus is probed, but before its children
* are examined.
*/
void pcibios_fixup_bus(struct pci_bus *b)
{
struct pci_dev *dev;
pci_read_bridge_bases(b);
list_for_each_entry(dev, &b->devices, bus_list)
pcibios_fixup_device_resources(dev);
}
void pcibios_add_bus(struct pci_bus *bus)
{
acpi_pci_add_bus(bus);
}
void pcibios_remove_bus(struct pci_bus *bus)
{
acpi_pci_remove_bus(bus);
}
/*
* Only use DMI information to set this if nothing was passed
* on the kernel command line (which was parsed earlier).
*/
static int __init set_bf_sort(const struct dmi_system_id *d)
{
if (pci_bf_sort == pci_bf_sort_default) {
pci_bf_sort = pci_dmi_bf;
printk(KERN_INFO "PCI: %s detected, enabling pci=bfsort.\n", d->ident);
}
return 0;
}
static void __init read_dmi_type_b1(const struct dmi_header *dm,
void *private_data)
{
u8 *data = (u8 *)dm + 4;
if (dm->type != 0xB1)
return;
if ((((*(u32 *)data) >> 9) & 0x03) == 0x01)
set_bf_sort((const struct dmi_system_id *)private_data);
}
static int __init find_sort_method(const struct dmi_system_id *d)
{
dmi_walk(read_dmi_type_b1, (void *)d);
return 0;
}
/*
* Enable renumbering of PCI bus# ranges to reach all PCI busses (Cardbus)
*/
#ifdef __i386__
static int __init assign_all_busses(const struct dmi_system_id *d)
{
pci_probe |= PCI_ASSIGN_ALL_BUSSES;
printk(KERN_INFO "%s detected: enabling PCI bus# renumbering"
" (pci=assign-busses)\n", d->ident);
return 0;
}
#endif
static int __init set_scan_all(const struct dmi_system_id *d)
{
printk(KERN_INFO "PCI: %s detected, enabling pci=pcie_scan_all\n",
d->ident);
pci_add_flags(PCI_SCAN_ALL_PCIE_DEVS);
return 0;
}
static const struct dmi_system_id pciprobe_dmi_table[] __initconst = {
#ifdef __i386__
/*
* Laptops which need pci=assign-busses to see Cardbus cards
*/
{
.callback = assign_all_busses,
.ident = "Samsung X20 Laptop",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Samsung Electronics"),
DMI_MATCH(DMI_PRODUCT_NAME, "SX20S"),
},
},
#endif /* __i386__ */
{
.callback = set_bf_sort,
.ident = "Dell PowerEdge 1950",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Dell"),
DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1950"),
},
},
{
.callback = set_bf_sort,
.ident = "Dell PowerEdge 1955",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Dell"),
DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1955"),
},
},
{
.callback = set_bf_sort,
.ident = "Dell PowerEdge 2900",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Dell"),
DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2900"),
},
},
{
.callback = set_bf_sort,
.ident = "Dell PowerEdge 2950",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Dell"),
DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2950"),
},
},
{
.callback = set_bf_sort,
.ident = "Dell PowerEdge R900",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Dell"),
DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R900"),
},
},
{
.callback = find_sort_method,
.ident = "Dell System",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL20p G3",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL20p G3"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL20p G4",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL20p G4"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL30p G1",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL30p G1"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL25p G1",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL25p G1"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL35p G1",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL35p G1"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL45p G1",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL45p G1"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL45p G2",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL45p G2"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL460c G1",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL460c G1"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL465c G1",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL465c G1"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL480c G1",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL480c G1"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant BL685c G1",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant BL685c G1"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant DL360",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL360"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant DL380",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL380"),
},
},
#ifdef __i386__
{
.callback = assign_all_busses,
.ident = "Compaq EVO N800c",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Compaq"),
DMI_MATCH(DMI_PRODUCT_NAME, "EVO N800c"),
},
},
#endif
{
.callback = set_bf_sort,
.ident = "HP ProLiant DL385 G2",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL385 G2"),
},
},
{
.callback = set_bf_sort,
.ident = "HP ProLiant DL585 G2",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "HP"),
DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL585 G2"),
},
},
{
.callback = set_scan_all,
.ident = "Stratus/NEC ftServer",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "Stratus"),
DMI_MATCH(DMI_PRODUCT_NAME, "ftServer"),
},
},
{
.callback = set_scan_all,
.ident = "Stratus/NEC ftServer",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "NEC"),
DMI_MATCH(DMI_PRODUCT_NAME, "Express5800/R32"),
},
},
{
.callback = set_scan_all,
.ident = "Stratus/NEC ftServer",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "NEC"),
DMI_MATCH(DMI_PRODUCT_NAME, "Express5800/R31"),
},
},
{}
};
void __init dmi_check_pciprobe(void)
{
dmi_check_system(pciprobe_dmi_table);
}
void pcibios_scan_root(int busnum)
{
struct pci_bus *bus;
struct pci_sysdata *sd;
LIST_HEAD(resources);
sd = kzalloc_obj(*sd);
if (!sd) {
printk(KERN_ERR "PCI: OOM, skipping PCI bus %02x\n", busnum);
return;
}
sd->node = x86_pci_root_bus_node(busnum);
x86_pci_root_bus_resources(busnum, &resources);
printk(KERN_DEBUG "PCI: Probing PCI hardware (bus %02x)\n", busnum);
bus = pci_scan_root_bus(NULL, busnum, &pci_root_ops, sd, &resources);
if (!bus) {
pci_free_resource_list(&resources);
kfree(sd);
return;
}
pci_bus_add_devices(bus);
}
void __init pcibios_set_cache_line_size(void)
{
struct cpuinfo_x86 *c = &boot_cpu_data;
/*
* Set PCI cacheline size to that of the CPU if the CPU has reported it.
* (For older CPUs that don't support cpuid, we se it to 32 bytes
* It's also good for 386/486s (which actually have 16)
* as quite a few PCI devices do not support smaller values.
*/
if (c->x86_clflush_size > 0) {
pci_dfl_cache_line_size = c->x86_clflush_size >> 2;
printk(KERN_DEBUG "PCI: pci_cache_line_size set to %d bytes\n",
pci_dfl_cache_line_size << 2);
} else {
pci_dfl_cache_line_size = 32 >> 2;
printk(KERN_DEBUG "PCI: Unknown cacheline size. Setting to 32 bytes\n");
}
}
int __init pcibios_init(void)
{
if (!raw_pci_ops && !raw_pci_ext_ops) {
printk(KERN_WARNING "PCI: System does not support PCI\n");
return 0;
}
pcibios_set_cache_line_size();
pcibios_resource_survey();
if (pci_bf_sort >= pci_force_bf)
pci_sort_breadthfirst();
return 0;
}
char *__init pcibios_setup(char *str)
{
if (!strcmp(str, "off")) {
pci_probe = 0;
return NULL;
} else if (!strcmp(str, "bfsort")) {
pci_bf_sort = pci_force_bf;
return NULL;
} else if (!strcmp(str, "nobfsort")) {
pci_bf_sort = pci_force_nobf;
return NULL;
}
#ifdef CONFIG_PCI_BIOS
else if (!strcmp(str, "bios")) {
pci_probe = PCI_PROBE_BIOS;
return NULL;
} else if (!strcmp(str, "nobios")) {
pci_probe &= ~PCI_PROBE_BIOS;
return NULL;
} else if (!strcmp(str, "biosirq")) {
pci_probe |= PCI_BIOS_IRQ_SCAN;
return NULL;
} else if (!strncmp(str, "pirqaddr=", 9)) {
pirq_table_addr = simple_strtoul(str+9, NULL, 0);
return NULL;
}
#endif
#ifdef CONFIG_PCI_DIRECT
else if (!strcmp(str, "conf1")) {
pci_probe = PCI_PROBE_CONF1 | PCI_NO_CHECKS;
return NULL;
}
else if (!strcmp(str, "conf2")) {
pci_probe = PCI_PROBE_CONF2 | PCI_NO_CHECKS;
return NULL;
}
#endif
#ifdef CONFIG_PCI_MMCONFIG
else if (!strcmp(str, "nommconf")) {
pci_probe &= ~PCI_PROBE_MMCONF;
return NULL;
}
else if (!strcmp(str, "check_enable_amd_mmconf")) {
pci_probe |= PCI_CHECK_ENABLE_AMD_MMCONF;
return NULL;
}
#endif
else if (!strcmp(str, "noacpi")) {
acpi_noirq_set();
return NULL;
}
else if (!strcmp(str, "noearly")) {
pci_probe |= PCI_PROBE_NOEARLY;
return NULL;
}
else if (!strcmp(str, "usepirqmask")) {
pci_probe |= PCI_USE_PIRQ_MASK;
return NULL;
} else if (!strncmp(str, "irqmask=", 8)) {
pcibios_irq_mask = simple_strtol(str+8, NULL, 0);
return NULL;
} else if (!strncmp(str, "lastbus=", 8)) {
pcibios_last_bus = simple_strtol(str+8, NULL, 0);
return NULL;
} else if (!strcmp(str, "rom")) {
pci_probe |= PCI_ASSIGN_ROMS;
return NULL;
} else if (!strcmp(str, "norom")) {
pci_probe |= PCI_NOASSIGN_ROMS;
return NULL;
} else if (!strcmp(str, "nobar")) {
pci_probe |= PCI_NOASSIGN_BARS;
return NULL;
} else if (!strcmp(str, "assign-busses")) {
pci_probe |= PCI_ASSIGN_ALL_BUSSES;
return NULL;
} else if (!strcmp(str, "use_crs")) {
pci_probe |= PCI_USE__CRS;
return NULL;
} else if (!strcmp(str, "nocrs")) {
pci_probe |= PCI_ROOT_NO_CRS;
return NULL;
} else if (!strcmp(str, "use_e820")) {
pci_probe |= PCI_USE_E820;
add_taint(TAINT_FIRMWARE_WORKAROUND, LOCKDEP_STILL_OK);
return NULL;
} else if (!strcmp(str, "no_e820")) {
pci_probe |= PCI_NO_E820;
add_taint(TAINT_FIRMWARE_WORKAROUND, LOCKDEP_STILL_OK);
return NULL;
#ifdef CONFIG_PHYS_ADDR_T_64BIT
} else if (!strcmp(str, "big_root_window")) {
pci_probe |= PCI_BIG_ROOT_WINDOW;
return NULL;
#endif
} else if (!strcmp(str, "routeirq")) {
pci_routeirq = 1;
return NULL;
} else if (!strcmp(str, "skip_isa_align")) {
pci_probe |= PCI_CAN_SKIP_ISA_ALIGN;
return NULL;
} else if (!strcmp(str, "noioapicquirk")) {
noioapicquirk = 1;
return NULL;
} else if (!strcmp(str, "ioapicreroute")) {
if (noioapicreroute != -1)
noioapicreroute = 0;
return NULL;
} else if (!strcmp(str, "noioapicreroute")) {
if (noioapicreroute != -1)
noioapicreroute = 1;
return NULL;
}
return str;
}
unsigned int pcibios_assign_all_busses(void)
{
return (pci_probe & PCI_ASSIGN_ALL_BUSSES) ? 1 : 0;
}
static void set_dev_domain_options(struct pci_dev *pdev)
{
if (is_vmd(pdev->bus))
pdev->hotplug_user_indicators = 1;
}
int pcibios_device_add(struct pci_dev *dev)
{
struct pci_setup_rom *rom;
struct irq_domain *msidom;
struct setup_data *data;
u64 pa_data;
pa_data = boot_params.hdr.setup_data;
while (pa_data) {
data = memremap(pa_data, sizeof(*rom), MEMREMAP_WB);
if (!data)
return -ENOMEM;
if (data->type == SETUP_PCI) {
rom = (struct pci_setup_rom *)data;
if ((pci_domain_nr(dev->bus) == rom->segment) &&
(dev->bus->number == rom->bus) &&
(PCI_SLOT(dev->devfn) == rom->device) &&
(PCI_FUNC(dev->devfn) == rom->function) &&
(dev->vendor == rom->vendor) &&
(dev->device == rom->devid)) {
dev->rom = pa_data +
offsetof(struct pci_setup_rom, romdata);
dev->romlen = rom->pcilen;
}
}
pa_data = data->next;
memunmap(data);
}
set_dev_domain_options(dev);
/*
* Setup the initial MSI domain of the device. If the underlying
* bus has a PCI/MSI irqdomain associated use the bus domain,
* otherwise set the default domain. This ensures that special irq
* domains e.g. VMD are preserved. The default ensures initial
* operation if irq remapping is not active. If irq remapping is
* active it will overwrite the domain pointer when the device is
* associated to a remapping domain.
*/
msidom = dev_get_msi_domain(&dev->bus->dev);
if (!msidom)
msidom = x86_pci_msi_default_domain;
dev_set_msi_domain(&dev->dev, msidom);
return 0;
}
int pcibios_enable_device(struct pci_dev *dev, int mask)
{
int err;
if ((err = pci_enable_resources(dev, mask)) < 0)
return err;
if (!pci_dev_msi_enabled(dev))
return pcibios_enable_irq(dev);
return 0;
}
void pcibios_disable_device (struct pci_dev *dev)
{
if (!pci_dev_msi_enabled(dev) && pcibios_disable_irq)
pcibios_disable_irq(dev);
}
#ifdef CONFIG_ACPI_HOTPLUG_IOAPIC
void pcibios_release_device(struct pci_dev *dev)
{
if (atomic_dec_return(&dev->enable_cnt) >= 0)
pcibios_disable_device(dev);
}
#endif
int pci_ext_cfg_avail(void)
{
if (raw_pci_ext_ops)
return 1;
else
return 0;
}
#if IS_ENABLED(CONFIG_VMD)
struct pci_dev *pci_real_dma_dev(struct pci_dev *dev)
{
if (is_vmd(dev->bus))
return to_pci_sysdata(dev->bus)->vmd_dev;
return dev;
}
#endif
linux-7.2.3/arch/x86/pci/direct.c
+315
View File
@@ -0,0 +1,315 @@
// SPDX-License-Identifier: GPL-2.0
/*
* direct.c - Low-level direct PCI config space access
*/
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/dmi.h>
#include <asm/pci_x86.h>
/*
* Functions for accessing PCI base (first 256 bytes) and extended
* (4096 bytes per PCI function) configuration space with type 1
* accesses.
*/
#define PCI_CONF1_ADDRESS(bus, devfn, reg) \
(0x80000000 | ((reg & 0xF00) << 16) | (bus << 16) \
| (devfn << 8) | (reg & 0xFC))
static int pci_conf1_read(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 *value)
{
unsigned long flags;
if (seg || (bus > 255) || (devfn > 255) || (reg > 4095)) {
*value = -1;
return -EINVAL;
}
raw_spin_lock_irqsave(&pci_config_lock, flags);
outl(PCI_CONF1_ADDRESS(bus, devfn, reg), 0xCF8);
switch (len) {
case 1:
*value = inb(0xCFC + (reg & 3));
break;
case 2:
*value = inw(0xCFC + (reg & 2));
break;
case 4:
*value = inl(0xCFC);
break;
}
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
return 0;
}
static int pci_conf1_write(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 value)
{
unsigned long flags;
if (seg || (bus > 255) || (devfn > 255) || (reg > 4095))
return -EINVAL;
raw_spin_lock_irqsave(&pci_config_lock, flags);
outl(PCI_CONF1_ADDRESS(bus, devfn, reg), 0xCF8);
switch (len) {
case 1:
outb((u8)value, 0xCFC + (reg & 3));
break;
case 2:
outw((u16)value, 0xCFC + (reg & 2));
break;
case 4:
outl((u32)value, 0xCFC);
break;
}
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
return 0;
}
#undef PCI_CONF1_ADDRESS
const struct pci_raw_ops pci_direct_conf1 = {
.read = pci_conf1_read,
.write = pci_conf1_write,
};
/*
* Functions for accessing PCI configuration space with type 2 accesses
*/
#define PCI_CONF2_ADDRESS(dev, reg) (u16)(0xC000 | (dev << 8) | reg)
static int pci_conf2_read(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 *value)
{
unsigned long flags;
int dev, fn;
WARN_ON(seg);
if ((bus > 255) || (devfn > 255) || (reg > 255)) {
*value = -1;
return -EINVAL;
}
dev = PCI_SLOT(devfn);
fn = PCI_FUNC(devfn);
if (dev & 0x10)
return PCIBIOS_DEVICE_NOT_FOUND;
raw_spin_lock_irqsave(&pci_config_lock, flags);
outb((u8)(0xF0 | (fn << 1)), 0xCF8);
outb((u8)bus, 0xCFA);
switch (len) {
case 1:
*value = inb(PCI_CONF2_ADDRESS(dev, reg));
break;
case 2:
*value = inw(PCI_CONF2_ADDRESS(dev, reg));
break;
case 4:
*value = inl(PCI_CONF2_ADDRESS(dev, reg));
break;
}
outb(0, 0xCF8);
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
return 0;
}
static int pci_conf2_write(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 value)
{
unsigned long flags;
int dev, fn;
WARN_ON(seg);
if ((bus > 255) || (devfn > 255) || (reg > 255))
return -EINVAL;
dev = PCI_SLOT(devfn);
fn = PCI_FUNC(devfn);
if (dev & 0x10)
return PCIBIOS_DEVICE_NOT_FOUND;
raw_spin_lock_irqsave(&pci_config_lock, flags);
outb((u8)(0xF0 | (fn << 1)), 0xCF8);
outb((u8)bus, 0xCFA);
switch (len) {
case 1:
outb((u8)value, PCI_CONF2_ADDRESS(dev, reg));
break;
case 2:
outw((u16)value, PCI_CONF2_ADDRESS(dev, reg));
break;
case 4:
outl((u32)value, PCI_CONF2_ADDRESS(dev, reg));
break;
}
outb(0, 0xCF8);
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
return 0;
}
#undef PCI_CONF2_ADDRESS
static const struct pci_raw_ops pci_direct_conf2 = {
.read = pci_conf2_read,
.write = pci_conf2_write,
};
/*
* Before we decide to use direct hardware access mechanisms, we try to do some
* trivial checks to ensure it at least _seems_ to be working -- we just test
* whether bus 00 contains a host bridge (this is similar to checking
* techniques used in XFree86, but ours should be more reliable since we
* attempt to make use of direct access hints provided by the PCI BIOS).
*
* This should be close to trivial, but it isn't, because there are buggy
* chipsets (yes, you guessed it, by Intel and Compaq) that have no class ID.
*/
static int __init pci_sanity_check(const struct pci_raw_ops *o)
{
u32 x = 0;
int devfn;
if (pci_probe & PCI_NO_CHECKS)
return 1;
/* Assume Type 1 works for newer systems.
This handles machines that don't have anything on PCI Bus 0. */
if (dmi_get_bios_year() >= 2001)
return 1;
for (devfn = 0; devfn < 0x100; devfn++) {
if (o->read(0, 0, devfn, PCI_CLASS_DEVICE, 2, &x))
continue;
if (x == PCI_CLASS_BRIDGE_HOST || x == PCI_CLASS_DISPLAY_VGA)
return 1;
if (o->read(0, 0, devfn, PCI_VENDOR_ID, 2, &x))
continue;
if (x == PCI_VENDOR_ID_INTEL || x == PCI_VENDOR_ID_COMPAQ)
return 1;
}
DBG(KERN_WARNING "PCI: Sanity check failed\n");
return 0;
}
static int __init pci_check_type1(void)
{
unsigned long flags;
unsigned int tmp;
int works = 0;
local_irq_save(flags);
outb(0x01, 0xCFB);
tmp = inl(0xCF8);
outl(0x80000000, 0xCF8);
if (inl(0xCF8) == 0x80000000 && pci_sanity_check(&pci_direct_conf1)) {
works = 1;
}
outl(tmp, 0xCF8);
local_irq_restore(flags);
return works;
}
static int __init pci_check_type2(void)
{
unsigned long flags;
int works = 0;
local_irq_save(flags);
outb(0x00, 0xCFB);
outb(0x00, 0xCF8);
outb(0x00, 0xCFA);
if (inb(0xCF8) == 0x00 && inb(0xCFA) == 0x00 &&
pci_sanity_check(&pci_direct_conf2)) {
works = 1;
}
local_irq_restore(flags);
return works;
}
void __init pci_direct_init(int type)
{
if (type == 0)
return;
printk(KERN_INFO "PCI: Using configuration type %d for base access\n",
type);
if (type == 1) {
raw_pci_ops = &pci_direct_conf1;
if (raw_pci_ext_ops)
return;
if (!(pci_probe & PCI_HAS_IO_ECS))
return;
printk(KERN_INFO "PCI: Using configuration type 1 "
"for extended access\n");
raw_pci_ext_ops = &pci_direct_conf1;
return;
}
raw_pci_ops = &pci_direct_conf2;
}
int __init pci_direct_probe(void)
{
if ((pci_probe & PCI_PROBE_CONF1) == 0)
goto type2;
if (!request_region(0xCF8, 8, "PCI conf1"))
goto type2;
if (pci_check_type1()) {
raw_pci_ops = &pci_direct_conf1;
port_cf9_safe = true;
return 1;
}
release_region(0xCF8, 8);
type2:
if ((pci_probe & PCI_PROBE_CONF2) == 0)
return 0;
if (!request_region(0xCF8, 4, "PCI conf2"))
return 0;
if (!request_region(0xC000, 0x1000, "PCI conf2"))
goto fail2;
if (pci_check_type2()) {
raw_pci_ops = &pci_direct_conf2;
port_cf9_safe = true;
return 2;
}
release_region(0xC000, 0x1000);
fail2:
release_region(0xCF8, 4);
return 0;
}
linux-7.2.3/arch/x86/pci/early.c
+59
View File
@@ -0,0 +1,59 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/kernel.h>
#include <linux/pci.h>
#include <asm/pci-direct.h>
#include <asm/io.h>
#include <asm/pci_x86.h>
/* Direct PCI access. This is used for PCI accesses in early boot before
the PCI subsystem works. */
u32 read_pci_config(u8 bus, u8 slot, u8 func, u8 offset)
{
u32 v;
outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
v = inl(0xcfc);
return v;
}
u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset)
{
u8 v;
outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
v = inb(0xcfc + (offset&3));
return v;
}
u16 read_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset)
{
u16 v;
outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
v = inw(0xcfc + (offset&2));
return v;
}
void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset,
u32 val)
{
outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
outl(val, 0xcfc);
}
void write_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset, u8 val)
{
outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
outb(val, 0xcfc + (offset&3));
}
void write_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset, u16 val)
{
outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
outw(val, 0xcfc + (offset&2));
}
int early_pci_allowed(void)
{
return (pci_probe & (PCI_PROBE_CONF1|PCI_PROBE_NOEARLY)) ==
PCI_PROBE_CONF1;
}
linux-7.2.3/arch/x86/pci/fixup.c
+1083
View File
File diff suppressed because it is too large Load Diff
linux-7.2.3/arch/x86/pci/i386.c
+412
View File
@@ -0,0 +1,412 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Low-Level PCI Access for i386 machines
*
* Copyright 1993, 1994 Drew Eckhardt
* Visionary Computing
* (Unix and Linux consulting and custom programming)
* Drew@Colorado.EDU
* +1 (303) 786-7975
*
* Drew's work was sponsored by:
* iX Multiuser Multitasking Magazine
* Hannover, Germany
* hm@ix.de
*
* Copyright 1997--2000 Martin Mares <mj@ucw.cz>
*
* For more information, please consult the following manuals (look at
* http://www.pcisig.com/ for how to get them):
*
* PCI BIOS Specification
* PCI Local Bus Specification
* PCI to PCI Bridge Specification
* PCI System Design Guide
*
*/
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/export.h>
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/ioport.h>
#include <linux/errno.h>
#include <linux/memblock.h>
#include <asm/memtype.h>
#include <asm/e820/api.h>
#include <asm/pci_x86.h>
#include <asm/io_apic.h>
/*
* This list of dynamic mappings is for temporarily maintaining
* original BIOS BAR addresses for possible reinstatement.
*/
struct pcibios_fwaddrmap {
struct list_head list;
struct pci_dev *dev;
resource_size_t fw_addr[DEVICE_COUNT_RESOURCE];
};
static LIST_HEAD(pcibios_fwaddrmappings);
static DEFINE_SPINLOCK(pcibios_fwaddrmap_lock);
static bool pcibios_fw_addr_done;
/* Must be called with 'pcibios_fwaddrmap_lock' lock held. */
static struct pcibios_fwaddrmap *pcibios_fwaddrmap_lookup(struct pci_dev *dev)
{
struct pcibios_fwaddrmap *map;
lockdep_assert_held(&pcibios_fwaddrmap_lock);
list_for_each_entry(map, &pcibios_fwaddrmappings, list)
if (map->dev == dev)
return map;
return NULL;
}
static void
pcibios_save_fw_addr(struct pci_dev *dev, int idx, resource_size_t fw_addr)
{
unsigned long flags;
struct pcibios_fwaddrmap *map;
if (pcibios_fw_addr_done)
return;
spin_lock_irqsave(&pcibios_fwaddrmap_lock, flags);
map = pcibios_fwaddrmap_lookup(dev);
if (!map) {
spin_unlock_irqrestore(&pcibios_fwaddrmap_lock, flags);
map = kzalloc_obj(*map);
if (!map)
return;
map->dev = pci_dev_get(dev);
map->fw_addr[idx] = fw_addr;
INIT_LIST_HEAD(&map->list);
spin_lock_irqsave(&pcibios_fwaddrmap_lock, flags);
list_add_tail(&map->list, &pcibios_fwaddrmappings);
} else
map->fw_addr[idx] = fw_addr;
spin_unlock_irqrestore(&pcibios_fwaddrmap_lock, flags);
}
resource_size_t pcibios_retrieve_fw_addr(struct pci_dev *dev, int idx)
{
unsigned long flags;
struct pcibios_fwaddrmap *map;
resource_size_t fw_addr = 0;
if (pcibios_fw_addr_done)
return 0;
spin_lock_irqsave(&pcibios_fwaddrmap_lock, flags);
map = pcibios_fwaddrmap_lookup(dev);
if (map)
fw_addr = map->fw_addr[idx];
spin_unlock_irqrestore(&pcibios_fwaddrmap_lock, flags);
return fw_addr;
}
static void __init pcibios_fw_addr_list_del(void)
{
unsigned long flags;
struct pcibios_fwaddrmap *entry, *next;
spin_lock_irqsave(&pcibios_fwaddrmap_lock, flags);
list_for_each_entry_safe(entry, next, &pcibios_fwaddrmappings, list) {
list_del(&entry->list);
pci_dev_put(entry->dev);
kfree(entry);
}
spin_unlock_irqrestore(&pcibios_fwaddrmap_lock, flags);
pcibios_fw_addr_done = true;
}
static int
skip_isa_ioresource_align(struct pci_dev *dev) {
if ((pci_probe & PCI_CAN_SKIP_ISA_ALIGN) &&
!(dev->bus->bridge_ctl & PCI_BRIDGE_CTL_ISA))
return 1;
return 0;
}
/*
* We need to avoid collisions with `mirrored' VGA ports
* and other strange ISA hardware, so we always want the
* addresses to be allocated in the 0x000-0x0ff region
* modulo 0x400.
*
* Why? Because some silly external IO cards only decode
* the low 10 bits of the IO address. The 0x00-0xff region
* is reserved for motherboard devices that decode all 16
* bits, so it's ok to allocate at, say, 0x2800-0x28ff,
* but we want to try to avoid allocating at 0x2900-0x2bff
* which might have be mirrored at 0x0100-0x03ff..
*/
resource_size_t
pcibios_align_resource(void *data, const struct resource *res,
const struct resource *empty_res,
resource_size_t size, resource_size_t align)
{
struct pci_dev *dev = data;
resource_size_t start = res->start;
if (res->flags & IORESOURCE_IO) {
if (skip_isa_ioresource_align(dev))
return start;
if (start & 0x300)
start = (start + 0x3ff) & ~0x3ff;
} else if (res->flags & IORESOURCE_MEM) {
start = pci_align_resource(dev, res, empty_res, size, align);
/* The low 1MB range is reserved for ISA cards */
if (start < BIOS_END)
start = BIOS_END;
}
return start;
}
EXPORT_SYMBOL(pcibios_align_resource);
/*
* Handle resources of PCI devices. If the world were perfect, we could
* just allocate all the resource regions and do nothing more. It isn't.
* On the other hand, we cannot just re-allocate all devices, as it would
* require us to know lots of host bridge internals. So we attempt to
* keep as much of the original configuration as possible, but tweak it
* when it's found to be wrong.
*
* Known BIOS problems we have to work around:
* - I/O or memory regions not configured
* - regions configured, but not enabled in the command register
* - bogus I/O addresses above 64K used
* - expansion ROMs left enabled (this may sound harmless, but given
* the fact the PCI specs explicitly allow address decoders to be
* shared between expansion ROMs and other resource regions, it's
* at least dangerous)
* - bad resource sizes or overlaps with other regions
*
* Our solution:
* (1) Allocate resources for all buses behind PCI-to-PCI bridges.
* This gives us fixed barriers on where we can allocate.
* (2) Allocate resources for all enabled devices. If there is
* a collision, just mark the resource as unallocated. Also
* disable expansion ROMs during this step.
* (3) Try to allocate resources for disabled devices. If the
* resources were assigned correctly, everything goes well,
* if they weren't, they won't disturb allocation of other
* resources.
* (4) Assign new addresses to resources which were either
* not configured at all or misconfigured. If explicitly
* requested by the user, configure expansion ROM address
* as well.
*/
static void pcibios_allocate_bridge_resources(struct pci_dev *dev)
{
int idx;
struct resource *r;
for (idx = PCI_BRIDGE_RESOURCES; idx < PCI_NUM_RESOURCES; idx++) {
r = &dev->resource[idx];
if (!r->flags)
continue;
if (r->parent) /* Already allocated */
continue;
if (!r->start || pci_claim_bridge_resource(dev, idx) < 0) {
/*
* Something is wrong with the region.
* Invalidate the resource to prevent
* child resource allocations in this
* range.
*/
r->start = r->end = 0;
r->flags = 0;
}
}
}
static void pcibios_allocate_bus_resources(struct pci_bus *bus)
{
struct pci_bus *child;
/* Depth-First Search on bus tree */
if (bus->self)
pcibios_allocate_bridge_resources(bus->self);
list_for_each_entry(child, &bus->children, node)
pcibios_allocate_bus_resources(child);
}
struct pci_check_idx_range {
int start;
int end;
};
static void pcibios_allocate_dev_resources(struct pci_dev *dev, int pass)
{
int idx, disabled, i;
u16 command;
struct resource *r;
struct pci_check_idx_range idx_range[] = {
{ PCI_STD_RESOURCES, PCI_STD_RESOURCE_END },
#ifdef CONFIG_PCI_IOV
{ PCI_IOV_RESOURCES, PCI_IOV_RESOURCE_END },
#endif
};
pci_read_config_word(dev, PCI_COMMAND, &command);
for (i = 0; i < ARRAY_SIZE(idx_range); i++)
for (idx = idx_range[i].start; idx <= idx_range[i].end; idx++) {
r = &dev->resource[idx];
if (r->parent) /* Already allocated */
continue;
if (!r->start) /* Address not assigned at all */
continue;
if (r->flags & IORESOURCE_IO)
disabled = !(command & PCI_COMMAND_IO);
else
disabled = !(command & PCI_COMMAND_MEMORY);
if (pass == disabled) {
dev_dbg(&dev->dev,
"BAR %d: reserving %pr (d=%d, p=%d)\n",
idx, r, disabled, pass);
if (pci_claim_resource(dev, idx) < 0) {
if (r->flags & IORESOURCE_PCI_FIXED) {
dev_info(&dev->dev, "BAR %d %pR is immovable\n",
idx, r);
} else {
/* We'll assign a new address later */
pcibios_save_fw_addr(dev,
idx, r->start);
r->end -= r->start;
r->start = 0;
}
}
}
}
if (!pass) {
r = &dev->resource[PCI_ROM_RESOURCE];
if (r->flags & IORESOURCE_ROM_ENABLE) {
/* Turn the ROM off, leave the resource region,
* but keep it unregistered. */
u32 reg;
dev_dbg(&dev->dev, "disabling ROM %pR\n", r);
r->flags &= ~IORESOURCE_ROM_ENABLE;
pci_read_config_dword(dev, dev->rom_base_reg, &reg);
pci_write_config_dword(dev, dev->rom_base_reg,
reg & ~PCI_ROM_ADDRESS_ENABLE);
}
}
}
static void pcibios_allocate_resources(struct pci_bus *bus, int pass)
{
struct pci_dev *dev;
struct pci_bus *child;
list_for_each_entry(dev, &bus->devices, bus_list) {
pcibios_allocate_dev_resources(dev, pass);
child = dev->subordinate;
if (child)
pcibios_allocate_resources(child, pass);
}
}
static void pcibios_allocate_dev_rom_resource(struct pci_dev *dev)
{
struct resource *r;
/*
* Try to use BIOS settings for ROMs, otherwise let
* pci_assign_unassigned_resources() allocate the new
* addresses.
*/
r = &dev->resource[PCI_ROM_RESOURCE];
if (!r->flags || !r->start)
return;
if (r->parent) /* Already allocated */
return;
if (pci_claim_resource(dev, PCI_ROM_RESOURCE) < 0) {
r->end -= r->start;
r->start = 0;
}
}
static void pcibios_allocate_rom_resources(struct pci_bus *bus)
{
struct pci_dev *dev;
struct pci_bus *child;
list_for_each_entry(dev, &bus->devices, bus_list) {
pcibios_allocate_dev_rom_resource(dev);
child = dev->subordinate;
if (child)
pcibios_allocate_rom_resources(child);
}
}
static int __init pcibios_assign_resources(void)
{
struct pci_bus *bus;
if (!(pci_probe & PCI_ASSIGN_ROMS))
list_for_each_entry(bus, &pci_root_buses, node)
pcibios_allocate_rom_resources(bus);
pci_assign_unassigned_resources();
pcibios_fw_addr_list_del();
return 0;
}
/*
* This is an fs_initcall (one below subsys_initcall) in order to reserve
* resources properly.
*/
fs_initcall(pcibios_assign_resources);
void pcibios_resource_survey_bus(struct pci_bus *bus)
{
dev_printk(KERN_DEBUG, &bus->dev, "Allocating resources\n");
pcibios_allocate_bus_resources(bus);
pcibios_allocate_resources(bus, 0);
pcibios_allocate_resources(bus, 1);
if (!(pci_probe & PCI_ASSIGN_ROMS))
pcibios_allocate_rom_resources(bus);
}
void __init pcibios_resource_survey(void)
{
struct pci_bus *bus;
DBG("PCI: Allocating resources\n");
list_for_each_entry(bus, &pci_root_buses, node)
pcibios_allocate_bus_resources(bus);
list_for_each_entry(bus, &pci_root_buses, node)
pcibios_allocate_resources(bus, 0);
list_for_each_entry(bus, &pci_root_buses, node)
pcibios_allocate_resources(bus, 1);
e820__reserve_resources_late();
/*
* Insert the IO APIC resources after PCI initialization has
* occurred to handle IO APICS that are mapped in on a BAR in
* PCI space, but before trying to assign unassigned pci res.
*/
ioapic_insert_resources();
}
linux-7.2.3/arch/x86/pci/init.c
+51
View File
@@ -0,0 +1,51 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/pci.h>
#include <linux/init.h>
#include <asm/pci_x86.h>
#include <asm/x86_init.h>
#include <asm/irqdomain.h>
/* arch_initcall has too random ordering, so call the initializers
in the right sequence from here. */
static __init int pci_arch_init(void)
{
int type, pcbios = 1;
type = pci_direct_probe();
if (!(pci_probe & PCI_PROBE_NOEARLY))
pci_mmcfg_early_init();
if (x86_init.pci.arch_init)
pcbios = x86_init.pci.arch_init();
/*
* Must happen after x86_init.pci.arch_init(). Xen sets up the
* x86_init.irqs.create_pci_msi_domain there.
*/
x86_create_pci_msi_domain();
if (!pcbios)
return 0;
pci_pcbios_init();
/*
* don't check for raw_pci_ops here because we want pcbios as last
* fallback, yet it's needed to run first to set pcibios_last_bus
* in case legacy PCI probing is used. otherwise detecting peer busses
* fails.
*/
pci_direct_init(type);
if (!raw_pci_ops && !raw_pci_ext_ops)
printk(KERN_ERR
"PCI: Fatal: No config space access function found\n");
dmi_check_pciprobe();
dmi_check_skip_isa_align();
return 0;
}
arch_initcall(pci_arch_init);
linux-7.2.3/arch/x86/pci/intel_mid.c
+406
View File
@@ -0,0 +1,406 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Intel MID PCI support
* Copyright (c) 2008 Intel Corporation
* Jesse Barnes <jesse.barnes@intel.com>
*
* Moorestown has an interesting PCI implementation:
* - configuration space is memory mapped (as defined by MCFG)
* - Lincroft devices also have a real, type 1 configuration space
* - Early Lincroft silicon has a type 1 access bug that will cause
* a hang if non-existent devices are accessed
* - some devices have the "fixed BAR" capability, which means
* they can't be relocated or modified; check for that during
* BAR sizing
*
* So, we use the MCFG space for all reads and writes, but also send
* Lincroft writes to type 1 space. But only read/write if the device
* actually exists, otherwise return all 1s for reads and bit bucket
* the writes.
*/
#include <linux/sched.h>
#include <linux/pci.h>
#include <linux/ioport.h>
#include <linux/init.h>
#include <linux/dmi.h>
#include <linux/acpi.h>
#include <linux/io.h>
#include <linux/smp.h>
#include <asm/cpu_device_id.h>
#include <asm/segment.h>
#include <asm/pci_x86.h>
#include <asm/hw_irq.h>
#include <asm/io_apic.h>
#include <asm/intel-family.h>
#include <asm/intel-mid.h>
#include <asm/acpi.h>
#define PCIE_CAP_OFFSET 0x100
/* Quirks for the listed devices */
#define PCI_DEVICE_ID_INTEL_MRFLD_MMC 0x1190
#define PCI_DEVICE_ID_INTEL_MRFLD_HSU 0x1191
/* Fixed BAR fields */
#define PCIE_VNDR_CAP_ID_FIXED_BAR 0x00 /* Fixed BAR (TBD) */
#define PCI_FIXED_BAR_0_SIZE 0x04
#define PCI_FIXED_BAR_1_SIZE 0x08
#define PCI_FIXED_BAR_2_SIZE 0x0c
#define PCI_FIXED_BAR_3_SIZE 0x10
#define PCI_FIXED_BAR_4_SIZE 0x14
#define PCI_FIXED_BAR_5_SIZE 0x1c
static int pci_soc_mode;
/**
* fixed_bar_cap - return the offset of the fixed BAR cap if found
* @bus: PCI bus
* @devfn: device in question
*
* Look for the fixed BAR cap on @bus and @devfn, returning its offset
* if found or 0 otherwise.
*/
static int fixed_bar_cap(struct pci_bus *bus, unsigned int devfn)
{
int pos;
u32 pcie_cap = 0, cap_data;
pos = PCIE_CAP_OFFSET;
if (!raw_pci_ext_ops)
return 0;
while (pos) {
if (raw_pci_ext_ops->read(pci_domain_nr(bus), bus->number,
devfn, pos, 4, &pcie_cap))
return 0;
if (PCI_EXT_CAP_ID(pcie_cap) == 0x0000 ||
PCI_EXT_CAP_ID(pcie_cap) == 0xffff)
break;
if (PCI_EXT_CAP_ID(pcie_cap) == PCI_EXT_CAP_ID_VNDR) {
raw_pci_ext_ops->read(pci_domain_nr(bus), bus->number,
devfn, pos + 4, 4, &cap_data);
if ((cap_data & 0xffff) == PCIE_VNDR_CAP_ID_FIXED_BAR)
return pos;
}
pos = PCI_EXT_CAP_NEXT(pcie_cap);
}
return 0;
}
static int pci_device_update_fixed(struct pci_bus *bus, unsigned int devfn,
int reg, int len, u32 val, int offset)
{
u32 size;
unsigned int domain, busnum;
int bar = (reg - PCI_BASE_ADDRESS_0) >> 2;
domain = pci_domain_nr(bus);
busnum = bus->number;
if (val == ~0 && len == 4) {
unsigned long decode;
raw_pci_ext_ops->read(domain, busnum, devfn,
offset + 8 + (bar * 4), 4, &size);
/* Turn the size into a decode pattern for the sizing code */
if (size) {
decode = size - 1;
decode |= decode >> 1;
decode |= decode >> 2;
decode |= decode >> 4;
decode |= decode >> 8;
decode |= decode >> 16;
decode++;
decode = ~(decode - 1);
} else {
decode = 0;
}
/*
* If val is all ones, the core code is trying to size the reg,
* so update the mmconfig space with the real size.
*
* Note: this assumes the fixed size we got is a power of two.
*/
return raw_pci_ext_ops->write(domain, busnum, devfn, reg, 4,
decode);
}
/* This is some other kind of BAR write, so just do it. */
return raw_pci_ext_ops->write(domain, busnum, devfn, reg, len, val);
}
/**
* type1_access_ok - check whether to use type 1
* @bus: bus number
* @devfn: device & function in question
* @reg: configuration register offset
*
* If the bus is on a Lincroft chip and it exists, or is not on a Lincroft at
* all, the we can go ahead with any reads & writes. If it's on a Lincroft,
* but doesn't exist, avoid the access altogether to keep the chip from
* hanging.
*/
static bool type1_access_ok(unsigned int bus, unsigned int devfn, int reg)
{
/*
* This is a workaround for A0 LNC bug where PCI status register does
* not have new CAP bit set. can not be written by SW either.
*
* PCI header type in real LNC indicates a single function device, this
* will prevent probing other devices under the same function in PCI
* shim. Therefore, use the header type in shim instead.
*/
if (reg >= 0x100 || reg == PCI_STATUS || reg == PCI_HEADER_TYPE)
return false;
if (bus == 0 && (devfn == PCI_DEVFN(2, 0)
|| devfn == PCI_DEVFN(0, 0)
|| devfn == PCI_DEVFN(3, 0)))
return true;
return false; /* Langwell on others */
}
static int pci_read(struct pci_bus *bus, unsigned int devfn, int where,
int size, u32 *value)
{
if (type1_access_ok(bus->number, devfn, where))
return pci_direct_conf1.read(pci_domain_nr(bus), bus->number,
devfn, where, size, value);
return raw_pci_ext_ops->read(pci_domain_nr(bus), bus->number,
devfn, where, size, value);
}
static int pci_write(struct pci_bus *bus, unsigned int devfn, int where,
int size, u32 value)
{
int offset;
/*
* On MRST, there is no PCI ROM BAR, this will cause a subsequent read
* to ROM BAR return 0 then being ignored.
*/
if (where == PCI_ROM_ADDRESS)
return 0;
/*
* Devices with fixed BARs need special handling:
* - BAR sizing code will save, write ~0, read size, restore
* - so writes to fixed BARs need special handling
* - other writes to fixed BAR devices should go through mmconfig
*/
offset = fixed_bar_cap(bus, devfn);
if (offset &&
(where >= PCI_BASE_ADDRESS_0 && where <= PCI_BASE_ADDRESS_5)) {
return pci_device_update_fixed(bus, devfn, where, size, value,
offset);
}
/*
* On Moorestown update both real & mmconfig space
* Note: early Lincroft silicon can't handle type 1 accesses to
* non-existent devices, so just eat the write in that case.
*/
if (type1_access_ok(bus->number, devfn, where))
return pci_direct_conf1.write(pci_domain_nr(bus), bus->number,
devfn, where, size, value);
return raw_pci_ext_ops->write(pci_domain_nr(bus), bus->number, devfn,
where, size, value);
}
static const struct x86_cpu_id intel_mid_cpu_ids[] = {
X86_MATCH_VFM(INTEL_ATOM_SILVERMONT_MID, NULL),
{}
};
static int intel_mid_pci_irq_enable(struct pci_dev *dev)
{
const struct x86_cpu_id *id;
struct irq_alloc_info info;
bool polarity_low;
u16 model = 0;
int ret;
u8 gsi;
if (dev->irq_managed && dev->irq > 0)
return 0;
ret = pci_read_config_byte(dev, PCI_INTERRUPT_LINE, &gsi);
if (ret) {
dev_warn(&dev->dev, "Failed to read interrupt line: %d\n", ret);
return pcibios_err_to_errno(ret);
}
id = x86_match_cpu(intel_mid_cpu_ids);
if (id)
model = id->model;
switch (model) {
case VFM_MODEL(INTEL_ATOM_SILVERMONT_MID):
polarity_low = false;
/* Special treatment for IRQ0 */
if (gsi == 0) {
/*
* Skip HS UART common registers device since it has
* IRQ0 assigned and not used by the kernel.
*/
if (dev->device == PCI_DEVICE_ID_INTEL_MRFLD_HSU)
return -EBUSY;
/*
* TNG has IRQ0 assigned to eMMC controller. But there
* are also other devices with bogus PCI configuration
* that have IRQ0 assigned. This check ensures that
* eMMC gets it. The rest of devices still could be
* enabled without interrupt line being allocated.
*/
if (dev->device != PCI_DEVICE_ID_INTEL_MRFLD_MMC)
return 0;
}
break;
default:
polarity_low = true;
break;
}
ioapic_set_alloc_attr(&info, dev_to_node(&dev->dev), 1, polarity_low);
/*
* MRST only have IOAPIC, the PCI irq lines are 1:1 mapped to
* IOAPIC RTE entries, so we just enable RTE for the device.
*/
ret = mp_map_gsi_to_irq(gsi, IOAPIC_MAP_ALLOC, &info);
if (ret < 0)
return ret;
dev->irq = ret;
dev->irq_managed = 1;
return 0;
}
static void intel_mid_pci_irq_disable(struct pci_dev *dev)
{
if (!mp_should_keep_irq(&dev->dev) && dev->irq_managed &&
dev->irq > 0) {
mp_unmap_irq(dev->irq);
dev->irq_managed = 0;
}
}
static const struct pci_ops intel_mid_pci_ops __initconst = {
.read = pci_read,
.write = pci_write,
};
/**
* intel_mid_pci_init - installs intel_mid_pci_ops
*
* Moorestown has an interesting PCI implementation (see above).
* Called when the early platform detection installs it.
*/
int __init intel_mid_pci_init(void)
{
pr_info("Intel MID platform detected, using MID PCI ops\n");
pci_mmcfg_late_init();
pcibios_enable_irq = intel_mid_pci_irq_enable;
pcibios_disable_irq = intel_mid_pci_irq_disable;
pci_root_ops = intel_mid_pci_ops;
pci_soc_mode = 1;
/* Continue with standard init */
acpi_noirq_set();
return 1;
}
/*
* Langwell devices are not true PCI devices; they are not subject to 10 ms
* d3 to d0 delay required by PCI spec.
*/
static void pci_d3delay_fixup(struct pci_dev *dev)
{
/*
* PCI fixups are effectively decided compile time. If we have a dual
* SoC/non-SoC kernel we don't want to mangle d3 on non-SoC devices.
*/
if (!pci_soc_mode)
return;
/*
* True PCI devices in Lincroft should allow type 1 access, the rest
* are Langwell fake PCI devices.
*/
if (type1_access_ok(dev->bus->number, dev->devfn, PCI_DEVICE_ID))
return;
dev->d3hot_delay = 0;
}
DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_ANY_ID, pci_d3delay_fixup);
static void mid_power_off_one_device(struct pci_dev *dev)
{
u16 pmcsr;
/*
* Update current state first, otherwise PCI core enforces PCI_D0 in
* pci_set_power_state() for devices which status was PCI_UNKNOWN.
*/
pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
dev->current_state = (pci_power_t __force)(pmcsr & PCI_PM_CTRL_STATE_MASK);
pci_set_power_state(dev, PCI_D3hot);
}
static void mid_power_off_devices(struct pci_dev *dev)
{
int id;
if (!pci_soc_mode)
return;
id = intel_mid_pwr_get_lss_id(dev);
if (id < 0)
return;
/*
* This sets only PMCSR bits. The actual power off will happen in
* arch/x86/platform/intel-mid/pwr.c.
*/
mid_power_off_one_device(dev);
}
DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_ANY_ID, mid_power_off_devices);
/*
* Langwell devices reside at fixed offsets, don't try to move them.
*/
static void pci_fixed_bar_fixup(struct pci_dev *dev)
{
unsigned long offset;
u32 size;
int i;
if (!pci_soc_mode)
return;
/* Must have extended configuration space */
if (dev->cfg_size < PCIE_CAP_OFFSET + 4)
return;
/* Fixup the BAR sizes for fixed BAR devices and make them unmoveable */
offset = fixed_bar_cap(dev->bus, dev->devfn);
if (!offset || PCI_DEVFN(2, 0) == dev->devfn ||
PCI_DEVFN(2, 2) == dev->devfn)
return;
for (i = 0; i < PCI_STD_NUM_BARS; i++) {
pci_read_config_dword(dev, offset + 8 + (i * 4), &size);
dev->resource[i].end = dev->resource[i].start + size - 1;
dev->resource[i].flags |= IORESOURCE_PCI_FIXED;
}
}
DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_ANY_ID, pci_fixed_bar_fixup);
linux-7.2.3/arch/x86/pci/irq.c
+1810
View File
File diff suppressed because it is too large Load Diff
linux-7.2.3/arch/x86/pci/legacy.c
+77
View File
@@ -0,0 +1,77 @@
// SPDX-License-Identifier: GPL-2.0-only
/*
* legacy.c - traditional, old school PCI bus probing
*/
#include <linux/init.h>
#include <linux/export.h>
#include <linux/pci.h>
#include <asm/jailhouse_para.h>
#include <asm/pci_x86.h>
/*
* Discover remaining PCI buses in case there are peer host bridges.
* We use the number of last PCI bus provided by the PCI BIOS.
*/
static void pcibios_fixup_peer_bridges(void)
{
int n;
if (pcibios_last_bus <= 0 || pcibios_last_bus > 0xff)
return;
DBG("PCI: Peer bridge fixup\n");
for (n=0; n <= pcibios_last_bus; n++)
pcibios_scan_specific_bus(n);
}
int __init pci_legacy_init(void)
{
if (!raw_pci_ops)
return 1;
pr_info("PCI: Probing PCI hardware\n");
pcibios_scan_root(0);
return 0;
}
void pcibios_scan_specific_bus(int busn)
{
int stride = jailhouse_paravirt() ? 1 : 8;
int devfn;
u32 l;
if (pci_find_bus(0, busn))
return;
for (devfn = 0; devfn < 256; devfn += stride) {
if (!raw_pci_read(0, busn, devfn, PCI_VENDOR_ID, 2, &l) &&
l != 0x0000 && l != 0xffff) {
DBG("Found device at %02x:%02x [%04x]\n", busn, devfn, l);
pr_info("PCI: Discovered peer bus %02x\n", busn);
pcibios_scan_root(busn);
return;
}
}
}
EXPORT_SYMBOL_GPL(pcibios_scan_specific_bus);
static int __init pci_subsys_init(void)
{
/*
* The init function returns an non zero value when
* pci_legacy_init should be invoked.
*/
if (x86_init.pci.init()) {
if (pci_legacy_init()) {
pr_info("PCI: System does not support PCI\n");
return -ENODEV;
}
}
pcibios_fixup_peer_bridges();
x86_init.pci.init_irq();
pcibios_init();
return 0;
}
subsys_initcall(pci_subsys_init);
linux-7.2.3/arch/x86/pci/mmconfig-shared.c
+878
View File
@@ -0,0 +1,878 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Low-level direct PCI config space access via ECAM - common code between
* i386 and x86-64.
*
* This code does:
* - known chipset handling
* - ACPI decoding and validation
*
* Per-architecture code takes care of the mappings and accesses
* themselves.
*/
#define pr_fmt(fmt) "PCI: " fmt
#include <linux/acpi.h>
#include <linux/efi.h>
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/bitmap.h>
#include <linux/dmi.h>
#include <linux/slab.h>
#include <linux/mutex.h>
#include <linux/rculist.h>
#include <asm/acpi.h>
#include <asm/e820/api.h>
#include <asm/msr.h>
#include <asm/pci_x86.h>
/* Indicate if the ECAM resources have been placed into the resource table */
static bool pci_mmcfg_running_state;
static bool pci_mmcfg_arch_init_failed;
static DEFINE_MUTEX(pci_mmcfg_lock);
#define pci_mmcfg_lock_held() lock_is_held(&(pci_mmcfg_lock).dep_map)
LIST_HEAD(pci_mmcfg_list);
static void __init pci_mmconfig_remove(struct pci_mmcfg_region *cfg)
{
if (cfg->res.parent)
release_resource(&cfg->res);
list_del(&cfg->list);
kfree(cfg);
}
static void __init free_all_mmcfg(void)
{
struct pci_mmcfg_region *cfg, *tmp;
pci_mmcfg_arch_free();
list_for_each_entry_safe(cfg, tmp, &pci_mmcfg_list, list)
pci_mmconfig_remove(cfg);
}
static void list_add_sorted(struct pci_mmcfg_region *new)
{
struct pci_mmcfg_region *cfg;
/* keep list sorted by segment and starting bus number */
list_for_each_entry_rcu(cfg, &pci_mmcfg_list, list, pci_mmcfg_lock_held()) {
if (cfg->segment > new->segment ||
(cfg->segment == new->segment &&
cfg->start_bus >= new->start_bus)) {
list_add_tail_rcu(&new->list, &cfg->list);
return;
}
}
list_add_tail_rcu(&new->list, &pci_mmcfg_list);
}
static struct pci_mmcfg_region *pci_mmconfig_alloc(int segment, int start,
int end, u64 addr)
{
struct pci_mmcfg_region *new;
struct resource *res;
if (addr == 0)
return NULL;
new = kzalloc_obj(*new);
if (!new)
return NULL;
new->address = addr;
new->segment = segment;
new->start_bus = start;
new->end_bus = end;
res = &new->res;
res->start = addr + PCI_MMCFG_BUS_OFFSET(start);
res->end = addr + PCI_MMCFG_BUS_OFFSET(end + 1) - 1;
res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
snprintf(new->name, PCI_MMCFG_RESOURCE_NAME_LEN,
"PCI ECAM %04x [bus %02x-%02x]", segment, start, end);
res->name = new->name;
return new;
}
struct pci_mmcfg_region *__init pci_mmconfig_add(int segment, int start,
int end, u64 addr)
{
struct pci_mmcfg_region *new;
new = pci_mmconfig_alloc(segment, start, end, addr);
if (!new)
return NULL;
mutex_lock(&pci_mmcfg_lock);
list_add_sorted(new);
mutex_unlock(&pci_mmcfg_lock);
pr_info("ECAM %pR (base %#lx) for domain %04x [bus %02x-%02x]\n",
&new->res, (unsigned long)addr, segment, start, end);
return new;
}
struct pci_mmcfg_region *pci_mmconfig_lookup(int segment, int bus)
{
struct pci_mmcfg_region *cfg;
list_for_each_entry_rcu(cfg, &pci_mmcfg_list, list, pci_mmcfg_lock_held())
if (cfg->segment == segment &&
cfg->start_bus <= bus && bus <= cfg->end_bus)
return cfg;
return NULL;
}
static const char *__init pci_mmcfg_e7520(void)
{
u32 win;
raw_pci_ops->read(0, 0, PCI_DEVFN(0, 0), 0xce, 2, &win);
win = win & 0xf000;
if (win == 0x0000 || win == 0xf000)
return NULL;
if (pci_mmconfig_add(0, 0, 255, win << 16) == NULL)
return NULL;
return "Intel Corporation E7520 Memory Controller Hub";
}
static const char *__init pci_mmcfg_intel_945(void)
{
u32 pciexbar, mask = 0, len = 0;
raw_pci_ops->read(0, 0, PCI_DEVFN(0, 0), 0x48, 4, &pciexbar);
/* Enable bit */
if (!(pciexbar & 1))
return NULL;
/* Size bits */
switch ((pciexbar >> 1) & 3) {
case 0:
mask = 0xf0000000U;
len = 0x10000000U;
break;
case 1:
mask = 0xf8000000U;
len = 0x08000000U;
break;
case 2:
mask = 0xfc000000U;
len = 0x04000000U;
break;
default:
return NULL;
}
/* Errata #2, things break when not aligned on a 256Mb boundary */
/* Can only happen in 64M/128M mode */
if ((pciexbar & mask) & 0x0fffffffU)
return NULL;
/* Don't hit the APIC registers and their friends */
if ((pciexbar & mask) >= 0xf0000000U)
return NULL;
if (pci_mmconfig_add(0, 0, (len >> 20) - 1, pciexbar & mask) == NULL)
return NULL;
return "Intel Corporation 945G/GZ/P/PL Express Memory Controller Hub";
}
static const char *__init pci_mmcfg_amd_fam10h(void)
{
u32 low, high, address;
u64 base, msr;
int i;
unsigned segnbits = 0, busnbits, end_bus;
if (!(pci_probe & PCI_CHECK_ENABLE_AMD_MMCONF))
return NULL;
address = MSR_FAM10H_MMIO_CONF_BASE;
if (rdmsr_safe(address, &low, &high))
return NULL;
msr = high;
msr <<= 32;
msr |= low;
/* ECAM is not enabled */
if (!(msr & FAM10H_MMIO_CONF_ENABLE))
return NULL;
base = msr & (FAM10H_MMIO_CONF_BASE_MASK<<FAM10H_MMIO_CONF_BASE_SHIFT);
busnbits = (msr >> FAM10H_MMIO_CONF_BUSRANGE_SHIFT) &
FAM10H_MMIO_CONF_BUSRANGE_MASK;
/*
* only handle bus 0 ?
* need to skip it
*/
if (!busnbits)
return NULL;
if (busnbits > 8) {
segnbits = busnbits - 8;
busnbits = 8;
}
end_bus = (1 << busnbits) - 1;
for (i = 0; i < (1 << segnbits); i++)
if (pci_mmconfig_add(i, 0, end_bus,
base + (1<<28) * i) == NULL) {
free_all_mmcfg();
return NULL;
}
return "AMD Family 10h NB";
}
static bool __initdata mcp55_checked;
static const char *__init pci_mmcfg_nvidia_mcp55(void)
{
int bus;
int mcp55_mmconf_found = 0;
static const u32 extcfg_regnum __initconst = 0x90;
static const u32 extcfg_regsize __initconst = 4;
static const u32 extcfg_enable_mask __initconst = 1 << 31;
static const u32 extcfg_start_mask __initconst = 0xff << 16;
static const int extcfg_start_shift __initconst = 16;
static const u32 extcfg_size_mask __initconst = 0x3 << 28;
static const int extcfg_size_shift __initconst = 28;
static const int extcfg_sizebus[] __initconst = {
0x100, 0x80, 0x40, 0x20
};
static const u32 extcfg_base_mask[] __initconst = {
0x7ff8, 0x7ffc, 0x7ffe, 0x7fff
};
static const int extcfg_base_lshift __initconst = 25;
/*
* do check if amd fam10h already took over
*/
if (!acpi_disabled || !list_empty(&pci_mmcfg_list) || mcp55_checked)
return NULL;
mcp55_checked = true;
for (bus = 0; bus < 256; bus++) {
u64 base;
u32 l, extcfg;
u16 vendor, device;
int start, size_index, end;
raw_pci_ops->read(0, bus, PCI_DEVFN(0, 0), 0, 4, &l);
vendor = l & 0xffff;
device = (l >> 16) & 0xffff;
if (PCI_VENDOR_ID_NVIDIA != vendor || 0x0369 != device)
continue;
raw_pci_ops->read(0, bus, PCI_DEVFN(0, 0), extcfg_regnum,
extcfg_regsize, &extcfg);
if (!(extcfg & extcfg_enable_mask))
continue;
size_index = (extcfg & extcfg_size_mask) >> extcfg_size_shift;
base = extcfg & extcfg_base_mask[size_index];
/* base could > 4G */
base <<= extcfg_base_lshift;
start = (extcfg & extcfg_start_mask) >> extcfg_start_shift;
end = start + extcfg_sizebus[size_index] - 1;
if (pci_mmconfig_add(0, start, end, base) == NULL)
continue;
mcp55_mmconf_found++;
}
if (!mcp55_mmconf_found)
return NULL;
return "nVidia MCP55";
}
struct pci_mmcfg_hostbridge_probe {
u32 bus;
u32 devfn;
u32 vendor;
u32 device;
const char *(*probe)(void);
};
static const struct pci_mmcfg_hostbridge_probe pci_mmcfg_probes[] __initconst = {
{ 0, PCI_DEVFN(0, 0), PCI_VENDOR_ID_INTEL,
PCI_DEVICE_ID_INTEL_E7520_MCH, pci_mmcfg_e7520 },
{ 0, PCI_DEVFN(0, 0), PCI_VENDOR_ID_INTEL,
PCI_DEVICE_ID_INTEL_82945G_HB, pci_mmcfg_intel_945 },
{ 0, PCI_DEVFN(0x18, 0), PCI_VENDOR_ID_AMD,
0x1200, pci_mmcfg_amd_fam10h },
{ 0xff, PCI_DEVFN(0, 0), PCI_VENDOR_ID_AMD,
0x1200, pci_mmcfg_amd_fam10h },
{ 0, PCI_DEVFN(0, 0), PCI_VENDOR_ID_NVIDIA,
0x0369, pci_mmcfg_nvidia_mcp55 },
};
static void __init pci_mmcfg_check_end_bus_number(void)
{
struct pci_mmcfg_region *cfg, *cfgx;
/* Fixup overlaps */
list_for_each_entry(cfg, &pci_mmcfg_list, list) {
if (cfg->end_bus < cfg->start_bus)
cfg->end_bus = 255;
/* Don't access the list head ! */
if (cfg->list.next == &pci_mmcfg_list)
break;
cfgx = list_entry(cfg->list.next, typeof(*cfg), list);
if (cfg->end_bus >= cfgx->start_bus)
cfg->end_bus = cfgx->start_bus - 1;
}
}
static int __init pci_mmcfg_check_hostbridge(void)
{
u32 l;
u32 bus, devfn;
u16 vendor, device;
int i;
const char *name;
if (!raw_pci_ops)
return 0;
free_all_mmcfg();
for (i = 0; i < ARRAY_SIZE(pci_mmcfg_probes); i++) {
bus = pci_mmcfg_probes[i].bus;
devfn = pci_mmcfg_probes[i].devfn;
raw_pci_ops->read(0, bus, devfn, 0, 4, &l);
vendor = l & 0xffff;
device = (l >> 16) & 0xffff;
name = NULL;
if (pci_mmcfg_probes[i].vendor == vendor &&
pci_mmcfg_probes[i].device == device)
name = pci_mmcfg_probes[i].probe();
if (name)
pr_info("%s with ECAM support\n", name);
}
/* some end_bus_number is crazy, fix it */
pci_mmcfg_check_end_bus_number();
return !list_empty(&pci_mmcfg_list);
}
static acpi_status check_mcfg_resource(struct acpi_resource *res, void *data)
{
struct resource *mcfg_res = data;
struct acpi_resource_address64 address;
acpi_status status;
if (res->type == ACPI_RESOURCE_TYPE_FIXED_MEMORY32) {
struct acpi_resource_fixed_memory32 *fixmem32 =
&res->data.fixed_memory32;
if (!fixmem32)
return AE_OK;
if ((mcfg_res->start >= fixmem32->address) &&
(mcfg_res->end < (fixmem32->address +
fixmem32->address_length))) {
mcfg_res->flags = 1;
return AE_CTRL_TERMINATE;
}
}
if ((res->type != ACPI_RESOURCE_TYPE_ADDRESS32) &&
(res->type != ACPI_RESOURCE_TYPE_ADDRESS64))
return AE_OK;
status = acpi_resource_to_address64(res, &address);
if (ACPI_FAILURE(status) ||
(address.address.address_length <= 0) ||
(address.resource_type != ACPI_MEMORY_RANGE))
return AE_OK;
if ((mcfg_res->start >= address.address.minimum) &&
(mcfg_res->end < (address.address.minimum + address.address.address_length))) {
mcfg_res->flags = 1;
return AE_CTRL_TERMINATE;
}
return AE_OK;
}
static acpi_status find_mboard_resource(acpi_handle handle, u32 lvl,
void *context, void **rv)
{
struct resource *mcfg_res = context;
acpi_walk_resources(handle, METHOD_NAME__CRS,
check_mcfg_resource, context);
if (mcfg_res->flags)
return AE_CTRL_TERMINATE;
return AE_OK;
}
static bool is_acpi_reserved(u64 start, u64 end, enum e820_type not_used)
{
struct resource mcfg_res;
mcfg_res.start = start;
mcfg_res.end = end - 1;
mcfg_res.flags = 0;
acpi_get_devices("PNP0C01", find_mboard_resource, &mcfg_res, NULL);
if (!mcfg_res.flags)
acpi_get_devices("PNP0C02", find_mboard_resource, &mcfg_res,
NULL);
return mcfg_res.flags;
}
static bool is_efi_mmio(struct resource *res)
{
#ifdef CONFIG_EFI
u64 start = res->start;
u64 end = res->start + resource_size(res);
efi_memory_desc_t *md;
u64 size, mmio_start, mmio_end;
for_each_efi_memory_desc(md) {
if (md->type == EFI_MEMORY_MAPPED_IO) {
size = md->num_pages << EFI_PAGE_SHIFT;
mmio_start = md->phys_addr;
mmio_end = mmio_start + size;
if (mmio_start <= start && end <= mmio_end)
return true;
}
}
#endif
return false;
}
typedef bool (*check_reserved_t)(u64 start, u64 end, enum e820_type type);
static bool __ref is_mmconf_reserved(check_reserved_t is_reserved,
struct pci_mmcfg_region *cfg,
struct device *dev, const char *method)
{
u64 addr = cfg->res.start;
u64 size = resource_size(&cfg->res);
u64 old_size = size;
int num_buses;
while (!is_reserved(addr, addr + size, E820_TYPE_RESERVED)) {
size >>= 1;
if (size < (16UL<<20))
break;
}
if (size < (16UL<<20) && size != old_size)
return false;
if (dev)
dev_info(dev, "ECAM %pR reserved as %s\n",
&cfg->res, method);
else
pr_info("ECAM %pR reserved as %s\n", &cfg->res, method);
if (old_size != size) {
/* update end_bus */
cfg->end_bus = cfg->start_bus + ((size>>20) - 1);
num_buses = cfg->end_bus - cfg->start_bus + 1;
cfg->res.end = cfg->res.start +
PCI_MMCFG_BUS_OFFSET(num_buses) - 1;
snprintf(cfg->name, PCI_MMCFG_RESOURCE_NAME_LEN,
"PCI ECAM %04x [bus %02x-%02x]",
cfg->segment, cfg->start_bus, cfg->end_bus);
if (dev)
dev_info(dev, "ECAM %pR (base %#lx) (size reduced!)\n",
&cfg->res, (unsigned long) cfg->address);
else
pr_info("ECAM %pR (base %#lx) for %04x [bus%02x-%02x] (size reduced!)\n",
&cfg->res, (unsigned long) cfg->address,
cfg->segment, cfg->start_bus, cfg->end_bus);
}
return true;
}
static bool __ref pci_mmcfg_reserved(struct device *dev,
struct pci_mmcfg_region *cfg, int early)
{
struct resource *conflict;
if (early) {
/*
* Don't try to do this check unless configuration type 1
* is available. How about type 2?
*/
/*
* 946f2ee5c731 ("Check that MCFG points to an e820
* reserved area") added this E820 check in 2006 to work
* around BIOS defects.
*
* Per PCI Firmware r3.3, sec 4.1.2, ECAM space must be
* reserved by a PNP0C02 resource, but it need not be
* mentioned in E820. Before the ACPI interpreter is
* available, we can't check for PNP0C02 resources, so
* there's no reliable way to verify the region in this
* early check. Keep it only for the old machines that
* motivated 946f2ee5c731.
*/
if (dmi_get_bios_year() < 2016 && raw_pci_ops)
return is_mmconf_reserved(e820__mapped_all, cfg, dev,
"E820 entry");
return true;
}
if (!acpi_disabled) {
if (is_mmconf_reserved(is_acpi_reserved, cfg, dev,
"ACPI motherboard resource"))
return true;
if (dev)
dev_info(dev, FW_INFO "ECAM %pR not reserved in ACPI motherboard resources\n",
&cfg->res);
else
pr_info(FW_INFO "ECAM %pR not reserved in ACPI motherboard resources\n",
&cfg->res);
if (is_efi_mmio(&cfg->res)) {
pr_info("ECAM %pR is EfiMemoryMappedIO; assuming valid\n",
&cfg->res);
conflict = insert_resource_conflict(&iomem_resource,
&cfg->res);
if (conflict)
pr_warn("ECAM %pR conflicts with %s %pR\n",
&cfg->res, conflict->name, conflict);
else
pr_info("ECAM %pR reserved to work around lack of ACPI motherboard _CRS\n",
&cfg->res);
return true;
}
}
/*
* e820__mapped_all() is marked as __init.
* All entries from ACPI MCFG table have been checked at boot time.
* For MCFG information constructed from hotpluggable host bridge's
* _CBA method, just assume it's reserved.
*/
return pci_mmcfg_running_state;
}
static void __init pci_mmcfg_reject_broken(int early)
{
struct pci_mmcfg_region *cfg;
list_for_each_entry(cfg, &pci_mmcfg_list, list) {
if (!pci_mmcfg_reserved(NULL, cfg, early)) {
pr_info("not using ECAM (%pR not reserved)\n",
&cfg->res);
free_all_mmcfg();
return;
}
}
}
static bool __init acpi_mcfg_valid_entry(struct acpi_table_mcfg *mcfg,
struct acpi_mcfg_allocation *cfg)
{
if (cfg->address < 0xFFFFFFFF)
return true;
if (!strncmp(mcfg->header.oem_id, "SGI", 3))
return true;
if ((mcfg->header.revision >= 1) && (dmi_get_bios_year() >= 2010))
return true;
pr_err("ECAM at %#llx for %04x [bus %02x-%02x] is above 4GB, ignored\n",
cfg->address, cfg->pci_segment, cfg->start_bus_number,
cfg->end_bus_number);
return false;
}
static int __init pci_parse_mcfg(struct acpi_table_header *header)
{
struct acpi_table_mcfg *mcfg;
struct acpi_mcfg_allocation *cfg_table, *cfg;
unsigned long i;
int entries;
if (!header)
return -EINVAL;
mcfg = (struct acpi_table_mcfg *)header;
/* how many config structures do we have */
free_all_mmcfg();
entries = 0;
i = header->length - sizeof(struct acpi_table_mcfg);
while (i >= sizeof(struct acpi_mcfg_allocation)) {
entries++;
i -= sizeof(struct acpi_mcfg_allocation);
}
if (entries == 0) {
pr_err("MCFG has no entries\n");
return -ENODEV;
}
cfg_table = (struct acpi_mcfg_allocation *) &mcfg[1];
for (i = 0; i < entries; i++) {
cfg = &cfg_table[i];
if (!acpi_mcfg_valid_entry(mcfg, cfg)) {
free_all_mmcfg();
return -ENODEV;
}
if (pci_mmconfig_add(cfg->pci_segment, cfg->start_bus_number,
cfg->end_bus_number, cfg->address) == NULL) {
pr_warn("no memory for MCFG entries\n");
free_all_mmcfg();
return -ENOMEM;
}
}
return 0;
}
#ifdef CONFIG_ACPI_APEI
extern int (*arch_apei_filter_addr)(int (*func)(__u64 start, __u64 size,
void *data), void *data);
static int pci_mmcfg_for_each_region(int (*func)(__u64 start, __u64 size,
void *data), void *data)
{
struct pci_mmcfg_region *cfg;
int rc;
if (list_empty(&pci_mmcfg_list))
return 0;
list_for_each_entry(cfg, &pci_mmcfg_list, list) {
rc = func(cfg->res.start, resource_size(&cfg->res), data);
if (rc)
return rc;
}
return 0;
}
#define set_apei_filter() (arch_apei_filter_addr = pci_mmcfg_for_each_region)
#else
#define set_apei_filter()
#endif
static void __init __pci_mmcfg_init(int early)
{
pr_debug("%s(%s)\n", __func__, early ? "early" : "late");
pci_mmcfg_reject_broken(early);
if (list_empty(&pci_mmcfg_list))
return;
if (pcibios_last_bus < 0) {
const struct pci_mmcfg_region *cfg;
list_for_each_entry(cfg, &pci_mmcfg_list, list) {
if (cfg->segment)
break;
pcibios_last_bus = cfg->end_bus;
}
}
if (pci_mmcfg_arch_init())
pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF;
else {
free_all_mmcfg();
pci_mmcfg_arch_init_failed = true;
}
}
static int __initdata known_bridge;
void __init pci_mmcfg_early_init(void)
{
pr_debug("%s() pci_probe %#x\n", __func__, pci_probe);
if (pci_probe & PCI_PROBE_MMCONF) {
if (pci_mmcfg_check_hostbridge())
known_bridge = 1;
else
acpi_table_parse(ACPI_SIG_MCFG, pci_parse_mcfg);
__pci_mmcfg_init(1);
set_apei_filter();
}
}
void __init pci_mmcfg_late_init(void)
{
pr_debug("%s() pci_probe %#x\n", __func__, pci_probe);
/* ECAM disabled */
if ((pci_probe & PCI_PROBE_MMCONF) == 0)
return;
if (known_bridge)
return;
/* ECAM hasn't been enabled yet, try again */
if (pci_probe & PCI_PROBE_MASK & ~PCI_PROBE_MMCONF) {
acpi_table_parse(ACPI_SIG_MCFG, pci_parse_mcfg);
__pci_mmcfg_init(0);
}
}
static int __init pci_mmcfg_late_insert_resources(void)
{
struct pci_mmcfg_region *cfg;
pci_mmcfg_running_state = true;
pr_debug("%s() pci_probe %#x\n", __func__, pci_probe);
/* If we are not using ECAM, don't insert the resources. */
if ((pci_probe & PCI_PROBE_MMCONF) == 0)
return 1;
/*
* Attempt to insert the mmcfg resources but not with the busy flag
* marked so it won't cause request errors when __request_region is
* called.
*/
list_for_each_entry(cfg, &pci_mmcfg_list, list) {
if (!cfg->res.parent) {
pr_debug("%s() insert %pR\n", __func__, &cfg->res);
insert_resource(&iomem_resource, &cfg->res);
}
}
return 0;
}
/*
* Perform ECAM resource insertion after PCI initialization to allow for
* misprogrammed MCFG tables that state larger sizes but actually conflict
* with other system resources.
*/
late_initcall(pci_mmcfg_late_insert_resources);
/* Add ECAM information for host bridges */
int pci_mmconfig_insert(struct device *dev, u16 seg, u8 start, u8 end,
phys_addr_t addr)
{
int rc;
struct resource *tmp = NULL;
struct pci_mmcfg_region *cfg;
dev_dbg(dev, "%s(%04x [bus %02x-%02x])\n", __func__, seg, start, end);
if (!(pci_probe & PCI_PROBE_MMCONF) || pci_mmcfg_arch_init_failed)
return -ENODEV;
if (start > end)
return -EINVAL;
mutex_lock(&pci_mmcfg_lock);
cfg = pci_mmconfig_lookup(seg, start);
if (cfg) {
if (cfg->end_bus < end)
dev_info(dev, FW_INFO "ECAM %pR for domain %04x [bus %02x-%02x] only partially covers this bridge\n",
&cfg->res, cfg->segment, cfg->start_bus,
cfg->end_bus);
mutex_unlock(&pci_mmcfg_lock);
return -EEXIST;
}
/*
* Don't move earlier; we must return -EEXIST, not -EINVAL, if
* pci_mmconfig_lookup() finds something
*/
if (!addr) {
mutex_unlock(&pci_mmcfg_lock);
return -EINVAL;
}
rc = -EBUSY;
cfg = pci_mmconfig_alloc(seg, start, end, addr);
if (cfg == NULL) {
dev_warn(dev, "fail to add ECAM (out of memory)\n");
rc = -ENOMEM;
} else if (!pci_mmcfg_reserved(dev, cfg, 0)) {
dev_warn(dev, FW_BUG "ECAM %pR isn't reserved\n",
&cfg->res);
} else {
/* Insert resource if it's not in boot stage */
if (pci_mmcfg_running_state)
tmp = insert_resource_conflict(&iomem_resource,
&cfg->res);
if (tmp) {
dev_warn(dev, "ECAM %pR conflicts with %s %pR\n",
&cfg->res, tmp->name, tmp);
} else if (pci_mmcfg_arch_map(cfg)) {
dev_warn(dev, "fail to map ECAM %pR\n", &cfg->res);
} else {
list_add_sorted(cfg);
dev_info(dev, "ECAM %pR (base %#lx)\n",
&cfg->res, (unsigned long)addr);
cfg = NULL;
rc = 0;
}
}
if (cfg) {
if (cfg->res.parent)
release_resource(&cfg->res);
kfree(cfg);
}
mutex_unlock(&pci_mmcfg_lock);
return rc;
}
/* Delete ECAM information for host bridges */
int pci_mmconfig_delete(u16 seg, u8 start, u8 end)
{
struct pci_mmcfg_region *cfg;
mutex_lock(&pci_mmcfg_lock);
list_for_each_entry_rcu(cfg, &pci_mmcfg_list, list)
if (cfg->segment == seg && cfg->start_bus == start &&
cfg->end_bus == end) {
list_del_rcu(&cfg->list);
synchronize_rcu();
pci_mmcfg_arch_unmap(cfg);
if (cfg->res.parent)
release_resource(&cfg->res);
mutex_unlock(&pci_mmcfg_lock);
kfree(cfg);
return 0;
}
mutex_unlock(&pci_mmcfg_lock);
return -ENOENT;
}
linux-7.2.3/arch/x86/pci/mmconfig_32.c
+156
View File
@@ -0,0 +1,156 @@
// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (C) 2004 Matthew Wilcox <matthew@wil.cx>
* Copyright (C) 2004 Intel Corp.
*/
/*
* mmconfig.c - Low-level direct PCI config space access via MMCONFIG
*/
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/rcupdate.h>
#include <asm/e820/api.h>
#include <asm/pci_x86.h>
/* Assume systems with more busses have correct MCFG */
#define mmcfg_virt_addr ((void __iomem *) fix_to_virt(FIX_PCIE_MCFG))
/* The base address of the last MMCONFIG device accessed */
static u32 mmcfg_last_accessed_device;
static int mmcfg_last_accessed_cpu;
/*
* Functions for accessing PCI configuration space with MMCONFIG accesses
*/
static u32 get_base_addr(unsigned int seg, int bus, unsigned devfn)
{
struct pci_mmcfg_region *cfg = pci_mmconfig_lookup(seg, bus);
if (cfg)
return cfg->address;
return 0;
}
/*
* This is always called under pci_config_lock
*/
static void pci_exp_set_dev_base(unsigned int base, int bus, int devfn)
{
u32 dev_base = base | PCI_MMCFG_BUS_OFFSET(bus) | (devfn << 12);
int cpu = smp_processor_id();
if (dev_base != mmcfg_last_accessed_device ||
cpu != mmcfg_last_accessed_cpu) {
mmcfg_last_accessed_device = dev_base;
mmcfg_last_accessed_cpu = cpu;
set_fixmap_nocache(FIX_PCIE_MCFG, dev_base);
}
}
static int pci_mmcfg_read(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 *value)
{
unsigned long flags;
u32 base;
if ((bus > 255) || (devfn > 255) || (reg > 4095)) {
err: *value = -1;
return -EINVAL;
}
rcu_read_lock();
base = get_base_addr(seg, bus, devfn);
if (!base) {
rcu_read_unlock();
goto err;
}
raw_spin_lock_irqsave(&pci_config_lock, flags);
pci_exp_set_dev_base(base, bus, devfn);
switch (len) {
case 1:
*value = mmio_config_readb(mmcfg_virt_addr + reg);
break;
case 2:
*value = mmio_config_readw(mmcfg_virt_addr + reg);
break;
case 4:
*value = mmio_config_readl(mmcfg_virt_addr + reg);
break;
}
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
rcu_read_unlock();
return 0;
}
static int pci_mmcfg_write(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 value)
{
unsigned long flags;
u32 base;
if ((bus > 255) || (devfn > 255) || (reg > 4095))
return -EINVAL;
rcu_read_lock();
base = get_base_addr(seg, bus, devfn);
if (!base) {
rcu_read_unlock();
return -EINVAL;
}
raw_spin_lock_irqsave(&pci_config_lock, flags);
pci_exp_set_dev_base(base, bus, devfn);
switch (len) {
case 1:
mmio_config_writeb(mmcfg_virt_addr + reg, value);
break;
case 2:
mmio_config_writew(mmcfg_virt_addr + reg, value);
break;
case 4:
mmio_config_writel(mmcfg_virt_addr + reg, value);
break;
}
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
rcu_read_unlock();
return 0;
}
const struct pci_raw_ops pci_mmcfg = {
.read = pci_mmcfg_read,
.write = pci_mmcfg_write,
};
int __init pci_mmcfg_arch_init(void)
{
printk(KERN_INFO "PCI: Using ECAM for extended config space\n");
raw_pci_ext_ops = &pci_mmcfg;
return 1;
}
void __init pci_mmcfg_arch_free(void)
{
}
int pci_mmcfg_arch_map(struct pci_mmcfg_region *cfg)
{
return 0;
}
void pci_mmcfg_arch_unmap(struct pci_mmcfg_region *cfg)
{
unsigned long flags;
/* Invalidate the cached mmcfg map entry. */
raw_spin_lock_irqsave(&pci_config_lock, flags);
mmcfg_last_accessed_device = 0;
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
}
linux-7.2.3/arch/x86/pci/mmconfig_64.c
+154
View File
@@ -0,0 +1,154 @@
// SPDX-License-Identifier: GPL-2.0
/*
* mmconfig.c - Low-level direct PCI config space access via MMCONFIG
*
* This is an 64bit optimized version that always keeps the full mmconfig
* space mapped. This allows lockless config space operation.
*/
#define pr_fmt(fmt) "PCI: " fmt
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/acpi.h>
#include <linux/bitmap.h>
#include <linux/rcupdate.h>
#include <asm/e820/api.h>
#include <asm/pci_x86.h>
static char __iomem *pci_dev_base(unsigned int seg, unsigned int bus, unsigned int devfn)
{
struct pci_mmcfg_region *cfg = pci_mmconfig_lookup(seg, bus);
if (cfg && cfg->virt)
return cfg->virt + (PCI_MMCFG_BUS_OFFSET(bus) | (devfn << 12));
return NULL;
}
static int pci_mmcfg_read(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 *value)
{
char __iomem *addr;
/* Why do we have this when nobody checks it. How about a BUG()!? -AK */
if (unlikely((bus > 255) || (devfn > 255) || (reg > 4095))) {
err: *value = -1;
return -EINVAL;
}
rcu_read_lock();
addr = pci_dev_base(seg, bus, devfn);
if (!addr) {
rcu_read_unlock();
goto err;
}
switch (len) {
case 1:
*value = mmio_config_readb(addr + reg);
break;
case 2:
*value = mmio_config_readw(addr + reg);
break;
case 4:
*value = mmio_config_readl(addr + reg);
break;
}
rcu_read_unlock();
return 0;
}
static int pci_mmcfg_write(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 value)
{
char __iomem *addr;
/* Why do we have this when nobody checks it. How about a BUG()!? -AK */
if (unlikely((bus > 255) || (devfn > 255) || (reg > 4095)))
return -EINVAL;
rcu_read_lock();
addr = pci_dev_base(seg, bus, devfn);
if (!addr) {
rcu_read_unlock();
return -EINVAL;
}
switch (len) {
case 1:
mmio_config_writeb(addr + reg, value);
break;
case 2:
mmio_config_writew(addr + reg, value);
break;
case 4:
mmio_config_writel(addr + reg, value);
break;
}
rcu_read_unlock();
return 0;
}
const struct pci_raw_ops pci_mmcfg = {
.read = pci_mmcfg_read,
.write = pci_mmcfg_write,
};
static void __iomem *mcfg_ioremap(struct pci_mmcfg_region *cfg)
{
void __iomem *addr;
u64 start, size;
int num_buses;
start = cfg->address + PCI_MMCFG_BUS_OFFSET(cfg->start_bus);
num_buses = cfg->end_bus - cfg->start_bus + 1;
size = PCI_MMCFG_BUS_OFFSET(num_buses);
addr = ioremap(start, size);
if (addr)
addr -= PCI_MMCFG_BUS_OFFSET(cfg->start_bus);
return addr;
}
int pci_mmcfg_arch_map(struct pci_mmcfg_region *cfg)
{
cfg->virt = mcfg_ioremap(cfg);
if (!cfg->virt) {
pr_err("can't map ECAM at %pR\n", &cfg->res);
return -ENOMEM;
}
return 0;
}
void pci_mmcfg_arch_unmap(struct pci_mmcfg_region *cfg)
{
if (cfg && cfg->virt) {
iounmap(cfg->virt + PCI_MMCFG_BUS_OFFSET(cfg->start_bus));
cfg->virt = NULL;
}
}
int __init pci_mmcfg_arch_init(void)
{
struct pci_mmcfg_region *cfg;
list_for_each_entry(cfg, &pci_mmcfg_list, list)
if (pci_mmcfg_arch_map(cfg)) {
pci_mmcfg_arch_free();
return 0;
}
raw_pci_ext_ops = &pci_mmcfg;
return 1;
}
void __init pci_mmcfg_arch_free(void)
{
struct pci_mmcfg_region *cfg;
list_for_each_entry(cfg, &pci_mmcfg_list, list)
pci_mmcfg_arch_unmap(cfg);
}
linux-7.2.3/arch/x86/pci/numachip.c
+127
View File
@@ -0,0 +1,127 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Numascale NumaConnect-specific PCI code
*
* Copyright (C) 2012 Numascale AS. All rights reserved.
*
* Send feedback to <support@numascale.com>
*
* PCI accessor functions derived from mmconfig_64.c
*
*/
#include <linux/pci.h>
#include <asm/pci_x86.h>
#include <asm/numachip/numachip.h>
static u8 limit __read_mostly;
static inline char __iomem *pci_dev_base(unsigned int seg, unsigned int bus, unsigned int devfn)
{
struct pci_mmcfg_region *cfg = pci_mmconfig_lookup(seg, bus);
if (cfg && cfg->virt)
return cfg->virt + (PCI_MMCFG_BUS_OFFSET(bus) | (devfn << 12));
return NULL;
}
static int pci_mmcfg_read_numachip(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 *value)
{
char __iomem *addr;
/* Why do we have this when nobody checks it. How about a BUG()!? -AK */
if (unlikely((bus > 255) || (devfn > 255) || (reg > 4095))) {
err: *value = -1;
return -EINVAL;
}
/* Ensure AMD Northbridges don't decode reads to other devices */
if (unlikely(bus == 0 && devfn >= limit)) {
*value = -1;
return 0;
}
rcu_read_lock();
addr = pci_dev_base(seg, bus, devfn);
if (!addr) {
rcu_read_unlock();
goto err;
}
switch (len) {
case 1:
*value = mmio_config_readb(addr + reg);
break;
case 2:
*value = mmio_config_readw(addr + reg);
break;
case 4:
*value = mmio_config_readl(addr + reg);
break;
}
rcu_read_unlock();
return 0;
}
static int pci_mmcfg_write_numachip(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 value)
{
char __iomem *addr;
/* Why do we have this when nobody checks it. How about a BUG()!? -AK */
if (unlikely((bus > 255) || (devfn > 255) || (reg > 4095)))
return -EINVAL;
/* Ensure AMD Northbridges don't decode writes to other devices */
if (unlikely(bus == 0 && devfn >= limit))
return 0;
rcu_read_lock();
addr = pci_dev_base(seg, bus, devfn);
if (!addr) {
rcu_read_unlock();
return -EINVAL;
}
switch (len) {
case 1:
mmio_config_writeb(addr + reg, value);
break;
case 2:
mmio_config_writew(addr + reg, value);
break;
case 4:
mmio_config_writel(addr + reg, value);
break;
}
rcu_read_unlock();
return 0;
}
static const struct pci_raw_ops pci_mmcfg_numachip = {
.read = pci_mmcfg_read_numachip,
.write = pci_mmcfg_write_numachip,
};
int __init pci_numachip_init(void)
{
int ret = 0;
u32 val;
/* For remote I/O, restrict bus 0 access to the actual number of AMD
Northbridges, which starts at device number 0x18 */
ret = raw_pci_read(0, 0, PCI_DEVFN(0x18, 0), 0x60, sizeof(val), &val);
if (ret)
goto out;
/* HyperTransport fabric size in bits 6:4 */
limit = PCI_DEVFN(0x18 + ((val >> 4) & 7) + 1, 0);
/* Use NumaChip PCI accessors for non-extended and extended access */
raw_pci_ops = raw_pci_ext_ops = &pci_mmcfg_numachip;
out:
return ret;
}
linux-7.2.3/arch/x86/pci/olpc.c
+308
View File
@@ -0,0 +1,308 @@
// SPDX-License-Identifier: GPL-2.0-or-later
/*
* Low-level PCI config space access for OLPC systems who lack the VSA
* PCI virtualization software.
*
* Copyright © 2006 Advanced Micro Devices, Inc.
*
* The AMD Geode chipset (ie: GX2 processor, cs5536 I/O companion device)
* has some I/O functions (display, southbridge, sound, USB HCIs, etc)
* that more or less behave like PCI devices, but the hardware doesn't
* directly implement the PCI configuration space headers. AMD provides
* "VSA" (Virtual System Architecture) software that emulates PCI config
* space for these devices, by trapping I/O accesses to PCI config register
* (CF8/CFC) and running some code in System Management Mode interrupt state.
* On the OLPC platform, we don't want to use that VSA code because
* (a) it slows down suspend/resume, and (b) recompiling it requires special
* compilers that are hard to get. So instead of letting the complex VSA
* code simulate the PCI config registers for the on-chip devices, we
* just simulate them the easy way, by inserting the code into the
* pci_write_config and pci_read_config path. Most of the config registers
* are read-only anyway, so the bulk of the simulation is just table lookup.
*/
#include <linux/pci.h>
#include <linux/init.h>
#include <asm/olpc.h>
#include <asm/geode.h>
#include <asm/pci_x86.h>
/*
* In the tables below, the first two line (8 longwords) are the
* size masks that are used when the higher level PCI code determines
* the size of the region by writing ~0 to a base address register
* and reading back the result.
*
* The following lines are the values that are read during normal
* PCI config access cycles, i.e. not after just having written
* ~0 to a base address register.
*/
static const uint32_t lxnb_hdr[] = { /* dev 1 function 0 - devfn = 8 */
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x281022, 0x2200005, 0x6000021, 0x80f808, /* AMD Vendor ID */
0x0, 0x0, 0x0, 0x0, /* No virtual registers, hence no BAR */
0x0, 0x0, 0x0, 0x28100b,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
};
static const uint32_t gxnb_hdr[] = { /* dev 1 function 0 - devfn = 8 */
0xfffffffd, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x28100b, 0x2200005, 0x6000021, 0x80f808, /* NSC Vendor ID */
0xac1d, 0x0, 0x0, 0x0, /* I/O BAR - base of virtual registers */
0x0, 0x0, 0x0, 0x28100b,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
};
static const uint32_t lxfb_hdr[] = { /* dev 1 function 1 - devfn = 9 */
0xff000008, 0xffffc000, 0xffffc000, 0xffffc000,
0xffffc000, 0x0, 0x0, 0x0,
0x20811022, 0x2200003, 0x3000000, 0x0, /* AMD Vendor ID */
0xfd000000, 0xfe000000, 0xfe004000, 0xfe008000, /* FB, GP, VG, DF */
0xfe00c000, 0x0, 0x0, 0x30100b, /* VIP */
0x0, 0x0, 0x0, 0x10e, /* INTA, IRQ14 for graphics accel */
0x0, 0x0, 0x0, 0x0,
0x3d0, 0x3c0, 0xa0000, 0x0, /* VG IO, VG IO, EGA FB, MONO FB */
0x0, 0x0, 0x0, 0x0,
};
static const uint32_t gxfb_hdr[] = { /* dev 1 function 1 - devfn = 9 */
0xff800008, 0xffffc000, 0xffffc000, 0xffffc000,
0x0, 0x0, 0x0, 0x0,
0x30100b, 0x2200003, 0x3000000, 0x0, /* NSC Vendor ID */
0xfd000000, 0xfe000000, 0xfe004000, 0xfe008000, /* FB, GP, VG, DF */
0x0, 0x0, 0x0, 0x30100b,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x3d0, 0x3c0, 0xa0000, 0x0, /* VG IO, VG IO, EGA FB, MONO FB */
0x0, 0x0, 0x0, 0x0,
};
static const uint32_t aes_hdr[] = { /* dev 1 function 2 - devfn = 0xa */
0xffffc000, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x20821022, 0x2a00006, 0x10100000, 0x8, /* NSC Vendor ID */
0xfe010000, 0x0, 0x0, 0x0, /* AES registers */
0x0, 0x0, 0x0, 0x20821022,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
};
static const uint32_t isa_hdr[] = { /* dev f function 0 - devfn = 78 */
0xfffffff9, 0xffffff01, 0xffffffc1, 0xffffffe1,
0xffffff81, 0xffffffc1, 0x0, 0x0,
0x20901022, 0x2a00049, 0x6010003, 0x802000,
0x18b1, 0x1001, 0x1801, 0x1881, /* SMB-8 GPIO-256 MFGPT-64 IRQ-32 */
0x1401, 0x1841, 0x0, 0x20901022, /* PMS-128 ACPI-64 */
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0xaa5b, /* IRQ steering */
0x0, 0x0, 0x0, 0x0,
};
static const uint32_t ac97_hdr[] = { /* dev f function 3 - devfn = 7b */
0xffffff81, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x20931022, 0x2a00041, 0x4010001, 0x0,
0x1481, 0x0, 0x0, 0x0, /* I/O BAR-128 */
0x0, 0x0, 0x0, 0x20931022,
0x0, 0x0, 0x0, 0x205, /* IntB, IRQ5 */
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
};
static const uint32_t ohci_hdr[] = { /* dev f function 4 - devfn = 7c */
0xfffff000, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x20941022, 0x2300006, 0xc031002, 0x0,
0xfe01a000, 0x0, 0x0, 0x0, /* MEMBAR-1000 */
0x0, 0x0, 0x0, 0x20941022,
0x0, 0x40, 0x0, 0x40a, /* CapPtr INT-D, IRQA */
0xc8020001, 0x0, 0x0, 0x0, /* Capabilities - 40 is R/O,
44 is mask 8103 (power control) */
0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
};
static const uint32_t ehci_hdr[] = { /* dev f function 4 - devfn = 7d */
0xfffff000, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0,
0x20951022, 0x2300006, 0xc032002, 0x0,
0xfe01b000, 0x0, 0x0, 0x0, /* MEMBAR-1000 */
0x0, 0x0, 0x0, 0x20951022,
0x0, 0x40, 0x0, 0x40a, /* CapPtr INT-D, IRQA */
0xc8020001, 0x0, 0x0, 0x0, /* Capabilities - 40 is R/O, 44 is
mask 8103 (power control) */
0x01000001, 0x0, 0x0, 0x0, /* EECP - see EHCI spec section 2.1.7 */
0x2020, 0x0, 0x0, 0x0, /* (EHCI page 8) 60 SBRN (R/O),
61 FLADJ (R/W), PORTWAKECAP */
};
static uint32_t ff_loc = ~0;
static uint32_t zero_loc;
static int bar_probing; /* Set after a write of ~0 to a BAR */
static int is_lx;
#define NB_SLOT 0x1 /* Northbridge - GX chip - Device 1 */
#define SB_SLOT 0xf /* Southbridge - CS5536 chip - Device F */
static int is_simulated(unsigned int bus, unsigned int devfn)
{
return (!bus && ((PCI_SLOT(devfn) == NB_SLOT) ||
(PCI_SLOT(devfn) == SB_SLOT)));
}
static uint32_t *hdr_addr(const uint32_t *hdr, int reg)
{
uint32_t addr;
/*
* This is a little bit tricky. The header maps consist of
* 0x20 bytes of size masks, followed by 0x70 bytes of header data.
* In the normal case, when not probing a BAR's size, we want
* to access the header data, so we add 0x20 to the reg offset,
* thus skipping the size mask area.
* In the BAR probing case, we want to access the size mask for
* the BAR, so we subtract 0x10 (the config header offset for
* BAR0), and don't skip the size mask area.
*/
addr = (uint32_t)hdr + reg + (bar_probing ? -0x10 : 0x20);
bar_probing = 0;
return (uint32_t *)addr;
}
static int pci_olpc_read(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, uint32_t *value)
{
uint32_t *addr;
WARN_ON(seg);
/* Use the hardware mechanism for non-simulated devices */
if (!is_simulated(bus, devfn))
return pci_direct_conf1.read(seg, bus, devfn, reg, len, value);
/*
* No device has config registers past 0x70, so we save table space
* by not storing entries for the nonexistent registers
*/
if (reg >= 0x70)
addr = &zero_loc;
else {
switch (devfn) {
case 0x8:
addr = hdr_addr(is_lx ? lxnb_hdr : gxnb_hdr, reg);
break;
case 0x9:
addr = hdr_addr(is_lx ? lxfb_hdr : gxfb_hdr, reg);
break;
case 0xa:
addr = is_lx ? hdr_addr(aes_hdr, reg) : &ff_loc;
break;
case 0x78:
addr = hdr_addr(isa_hdr, reg);
break;
case 0x7b:
addr = hdr_addr(ac97_hdr, reg);
break;
case 0x7c:
addr = hdr_addr(ohci_hdr, reg);
break;
case 0x7d:
addr = hdr_addr(ehci_hdr, reg);
break;
default:
addr = &ff_loc;
break;
}
}
switch (len) {
case 1:
*value = *(uint8_t *)addr;
break;
case 2:
*value = *(uint16_t *)addr;
break;
case 4:
*value = *addr;
break;
default:
BUG();
}
return 0;
}
static int pci_olpc_write(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, uint32_t value)
{
WARN_ON(seg);
/* Use the hardware mechanism for non-simulated devices */
if (!is_simulated(bus, devfn))
return pci_direct_conf1.write(seg, bus, devfn, reg, len, value);
/* XXX we may want to extend this to simulate EHCI power management */
/*
* Mostly we just discard writes, but if the write is a size probe
* (i.e. writing ~0 to a BAR), we remember it and arrange to return
* the appropriate size mask on the next read. This is cheating
* to some extent, because it depends on the fact that the next
* access after such a write will always be a read to the same BAR.
*/
if ((reg >= 0x10) && (reg < 0x2c)) {
/* write is to a BAR */
if (value == ~0)
bar_probing = 1;
} else {
/*
* No warning on writes to ROM BAR, CMD, LATENCY_TIMER,
* CACHE_LINE_SIZE, or PM registers.
*/
if ((reg != PCI_ROM_ADDRESS) && (reg != PCI_COMMAND_MASTER) &&
(reg != PCI_LATENCY_TIMER) &&
(reg != PCI_CACHE_LINE_SIZE) && (reg != 0x44))
printk(KERN_WARNING "OLPC PCI: Config write to devfn"
" %x reg %x value %x\n", devfn, reg, value);
}
return 0;
}
static const struct pci_raw_ops pci_olpc_conf = {
.read = pci_olpc_read,
.write = pci_olpc_write,
};
int __init pci_olpc_init(void)
{
printk(KERN_INFO "PCI: Using configuration type OLPC XO-1\n");
raw_pci_ops = &pci_olpc_conf;
is_lx = is_geode_lx();
return 0;
}
linux-7.2.3/arch/x86/pci/pcbios.c
+443
View File
@@ -0,0 +1,443 @@
// SPDX-License-Identifier: GPL-2.0
/*
* BIOS32 and PCI BIOS handling.
*/
#include <linux/bits.h>
#include <linux/bitfield.h>
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/module.h>
#include <linux/uaccess.h>
#include <asm/pci_x86.h>
#include <asm/e820/types.h>
#include <asm/pci-functions.h>
#include <asm/set_memory.h>
/* BIOS32 signature: "_32_" */
#define BIOS32_SIGNATURE (('_' << 0) + ('3' << 8) + ('2' << 16) + ('_' << 24))
/* PCI signature: "PCI " */
#define PCI_SIGNATURE (('P' << 0) + ('C' << 8) + ('I' << 16) + (' ' << 24))
/* PCI service signature: "$PCI" */
#define PCI_SERVICE (('$' << 0) + ('P' << 8) + ('C' << 16) + ('I' << 24))
/* PCI BIOS hardware mechanism flags */
#define PCIBIOS_HW_TYPE1 0x01
#define PCIBIOS_HW_TYPE2 0x02
#define PCIBIOS_HW_TYPE1_SPEC 0x10
#define PCIBIOS_HW_TYPE2_SPEC 0x20
/*
* Returned in EAX:
* - AH: return code
*/
#define PCIBIOS_RETURN_CODE GENMASK(15, 8)
int pcibios_enabled;
static u8 pcibios_get_return_code(u32 eax)
{
return FIELD_GET(PCIBIOS_RETURN_CODE, eax);
}
/* According to the BIOS specification at:
* http://members.datafast.net.au/dft0802/specs/bios21.pdf, we could
* restrict the x zone to some pages and make it ro. But this may be
* broken on some bios, complex to handle with static_protections.
* We could make the 0xe0000-0x100000 range rox, but this can break
* some ISA mapping.
*
* So we let's an rw and x hole when pcibios is used. This shouldn't
* happen for modern system with mmconfig, and if you don't want it
* you could disable pcibios...
*/
static inline void set_bios_x(void)
{
pcibios_enabled = 1;
set_memory_x(PAGE_OFFSET + BIOS_BEGIN, (BIOS_END - BIOS_BEGIN) >> PAGE_SHIFT);
if (__supported_pte_mask & _PAGE_NX)
printk(KERN_INFO "PCI: PCI BIOS area is rw and x. Use pci=nobios if you want it NX.\n");
}
/*
* This is the standard structure used to identify the entry point
* to the BIOS32 Service Directory, as documented in
* Standard BIOS 32-bit Service Directory Proposal
* Revision 0.4 May 24, 1993
* Phoenix Technologies Ltd.
* Norwood, MA
* and the PCI BIOS specification.
*/
union bios32 {
struct {
unsigned long signature; /* _32_ */
unsigned long entry; /* 32 bit physical address */
unsigned char revision; /* Revision level, 0 */
unsigned char length; /* Length in paragraphs should be 01 */
unsigned char checksum; /* All bytes must add up to zero */
unsigned char reserved[5]; /* Must be zero */
} fields;
char chars[16];
};
/*
* Physical address of the service directory. I don't know if we're
* allowed to have more than one of these or not, so just in case
* we'll make pcibios_present() take a memory start parameter and store
* the array there.
*/
static struct {
unsigned long address;
unsigned short segment;
} bios32_indirect __initdata = { 0, __KERNEL_CS };
/*
* Returns the entry point for the given service, NULL on error
*/
static unsigned long __init bios32_service(unsigned long service)
{
unsigned char return_code; /* %al */
unsigned long address; /* %ebx */
unsigned long length; /* %ecx */
unsigned long entry; /* %edx */
unsigned long flags;
local_irq_save(flags);
__asm__("lcall *(%%edi); cld"
: "=a" (return_code),
"=b" (address),
"=c" (length),
"=d" (entry)
: "0" (service),
"1" (0),
"D" (&bios32_indirect));
local_irq_restore(flags);
switch (return_code) {
case 0:
return address + entry;
case 0x80: /* Not present */
printk(KERN_WARNING "bios32_service(0x%lx): not present\n", service);
return 0;
default: /* Shouldn't happen */
printk(KERN_WARNING "bios32_service(0x%lx): returned 0x%x -- BIOS bug!\n",
service, return_code);
return 0;
}
}
static struct {
unsigned long address;
unsigned short segment;
} pci_indirect __ro_after_init = {
.address = 0,
.segment = __KERNEL_CS,
};
static int pci_bios_present __ro_after_init;
static int __init check_pcibios(void)
{
u32 signature, eax, ebx, ecx;
u8 status, major_ver, minor_ver, hw_mech;
unsigned long flags, pcibios_entry;
if ((pcibios_entry = bios32_service(PCI_SERVICE))) {
pci_indirect.address = pcibios_entry + PAGE_OFFSET;
local_irq_save(flags);
__asm__(
"lcall *(%%edi); cld\n\t"
"jc 1f\n\t"
"xor %%ah, %%ah\n"
"1:"
: "=d" (signature),
"=a" (eax),
"=b" (ebx),
"=c" (ecx)
: "1" (PCIBIOS_PCI_BIOS_PRESENT),
"D" (&pci_indirect)
: "memory");
local_irq_restore(flags);
status = pcibios_get_return_code(eax);
hw_mech = eax & 0xff;
major_ver = (ebx >> 8) & 0xff;
minor_ver = ebx & 0xff;
if (pcibios_last_bus < 0)
pcibios_last_bus = ecx & 0xff;
DBG("PCI: BIOS probe returned s=%02x hw=%02x ver=%02x.%02x l=%02x\n",
status, hw_mech, major_ver, minor_ver, pcibios_last_bus);
if (status || signature != PCI_SIGNATURE) {
printk (KERN_ERR "PCI: BIOS BUG #%x[%08x] found\n",
status, signature);
return 0;
}
printk(KERN_INFO "PCI: PCI BIOS revision %x.%02x entry at 0x%lx, last bus=%d\n",
major_ver, minor_ver, pcibios_entry, pcibios_last_bus);
#ifdef CONFIG_PCI_DIRECT
if (!(hw_mech & PCIBIOS_HW_TYPE1))
pci_probe &= ~PCI_PROBE_CONF1;
if (!(hw_mech & PCIBIOS_HW_TYPE2))
pci_probe &= ~PCI_PROBE_CONF2;
#endif
return 1;
}
return 0;
}
static int pci_bios_read(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 *value)
{
unsigned long result = 0;
unsigned long flags;
unsigned long bx = (bus << 8) | devfn;
u16 number = 0, mask = 0;
WARN_ON(seg);
if (!value || (bus > 255) || (devfn > 255) || (reg > 255))
return -EINVAL;
raw_spin_lock_irqsave(&pci_config_lock, flags);
switch (len) {
case 1:
number = PCIBIOS_READ_CONFIG_BYTE;
mask = 0xff;
break;
case 2:
number = PCIBIOS_READ_CONFIG_WORD;
mask = 0xffff;
break;
case 4:
number = PCIBIOS_READ_CONFIG_DWORD;
break;
}
__asm__("lcall *(%%esi); cld\n\t"
"jc 1f\n\t"
"xor %%ah, %%ah\n"
"1:"
: "=c" (*value),
"=a" (result)
: "1" (number),
"b" (bx),
"D" ((long)reg),
"S" (&pci_indirect));
/*
* Zero-extend the result beyond 8 or 16 bits, do not trust the
* BIOS having done it:
*/
if (mask)
*value &= mask;
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
return pcibios_get_return_code(result);
}
static int pci_bios_write(unsigned int seg, unsigned int bus,
unsigned int devfn, int reg, int len, u32 value)
{
unsigned long result = 0;
unsigned long flags;
unsigned long bx = (bus << 8) | devfn;
u16 number = 0;
WARN_ON(seg);
if ((bus > 255) || (devfn > 255) || (reg > 255))
return -EINVAL;
raw_spin_lock_irqsave(&pci_config_lock, flags);
switch (len) {
case 1:
number = PCIBIOS_WRITE_CONFIG_BYTE;
break;
case 2:
number = PCIBIOS_WRITE_CONFIG_WORD;
break;
case 4:
number = PCIBIOS_WRITE_CONFIG_DWORD;
break;
}
__asm__("lcall *(%%esi); cld\n\t"
"jc 1f\n\t"
"xor %%ah, %%ah\n"
"1:"
: "=a" (result)
: "0" (number),
"c" (value),
"b" (bx),
"D" ((long)reg),
"S" (&pci_indirect));
raw_spin_unlock_irqrestore(&pci_config_lock, flags);
return pcibios_get_return_code(result);
}
/*
* Function table for BIOS32 access
*/
static const struct pci_raw_ops pci_bios_access = {
.read = pci_bios_read,
.write = pci_bios_write
};
/*
* Try to find PCI BIOS.
*/
static const struct pci_raw_ops *__init pci_find_bios(void)
{
union bios32 *check;
unsigned char sum;
int i, length;
/*
* Follow the standard procedure for locating the BIOS32 Service
* directory by scanning the permissible address range from
* 0xe0000 through 0xfffff for a valid BIOS32 structure.
*/
for (check = (union bios32 *) __va(0xe0000);
check <= (union bios32 *) __va(0xffff0);
++check) {
long sig;
if (get_kernel_nofault(sig, &check->fields.signature))
continue;
if (check->fields.signature != BIOS32_SIGNATURE)
continue;
length = check->fields.length * 16;
if (!length)
continue;
sum = 0;
for (i = 0; i < length ; ++i)
sum += check->chars[i];
if (sum != 0)
continue;
if (check->fields.revision != 0) {
printk("PCI: unsupported BIOS32 revision %d at 0x%p\n",
check->fields.revision, check);
continue;
}
DBG("PCI: BIOS32 Service Directory structure at 0x%p\n", check);
if (check->fields.entry >= 0x100000) {
printk("PCI: BIOS32 entry (0x%p) in high memory, "
"cannot use.\n", check);
return NULL;
} else {
unsigned long bios32_entry = check->fields.entry;
DBG("PCI: BIOS32 Service Directory entry at 0x%lx\n",
bios32_entry);
bios32_indirect.address = bios32_entry + PAGE_OFFSET;
set_bios_x();
if (check_pcibios())
return &pci_bios_access;
}
break; /* Hopefully more than one BIOS32 cannot happen... */
}
return NULL;
}
/*
* BIOS Functions for IRQ Routing
*/
struct irq_routing_options {
u16 size;
struct irq_info *table;
u16 segment;
} __attribute__((packed));
struct irq_routing_table * pcibios_get_irq_routing_table(void)
{
struct irq_routing_options opt;
struct irq_routing_table *rt = NULL;
int ret, map;
unsigned long page;
if (!pci_bios_present)
return NULL;
page = __get_free_page(GFP_KERNEL);
if (!page)
return NULL;
opt.table = (struct irq_info *) page;
opt.size = PAGE_SIZE;
opt.segment = __KERNEL_DS;
DBG("PCI: Fetching IRQ routing table... ");
__asm__("push %%es\n\t"
"push %%ds\n\t"
"pop %%es\n\t"
"lcall *(%%esi); cld\n\t"
"pop %%es\n\t"
"jc 1f\n\t"
"xor %%ah, %%ah\n"
"1:"
: "=a" (ret),
"=b" (map),
"=m" (opt)
: "0" (PCIBIOS_GET_ROUTING_OPTIONS),
"1" (0),
"D" ((long) &opt),
"S" (&pci_indirect),
"m" (opt)
: "memory");
DBG("OK ret=%d, size=%d, map=%x\n", ret, opt.size, map);
ret = pcibios_get_return_code(ret);
if (ret) {
printk(KERN_ERR "PCI: Error %02x when fetching IRQ routing table.\n", ret);
} else if (opt.size) {
rt = kmalloc(sizeof(struct irq_routing_table) + opt.size, GFP_KERNEL);
if (rt) {
memset(rt, 0, sizeof(struct irq_routing_table));
rt->size = opt.size + sizeof(struct irq_routing_table);
rt->exclusive_irqs = map;
memcpy(rt->slots, (void *) page, opt.size);
printk(KERN_INFO "PCI: Using BIOS Interrupt Routing Table\n");
}
}
free_page(page);
return rt;
}
EXPORT_SYMBOL(pcibios_get_irq_routing_table);
int pcibios_set_irq_routing(struct pci_dev *dev, int pin, int irq)
{
int ret;
__asm__("lcall *(%%esi); cld\n\t"
"jc 1f\n\t"
"xor %%ah, %%ah\n"
"1:"
: "=a" (ret)
: "0" (PCIBIOS_SET_PCI_HW_INT),
"b" ((dev->bus->number << 8) | dev->devfn),
"c" ((irq << 8) | (pin + 10)),
"S" (&pci_indirect));
return pcibios_get_return_code(ret) == PCIBIOS_SUCCESSFUL;
}
EXPORT_SYMBOL(pcibios_set_irq_routing);
void __init pci_pcbios_init(void)
{
if ((pci_probe & PCI_PROBE_BIOS)
&& ((raw_pci_ops = pci_find_bios()))) {
pci_bios_present = 1;
}
}
linux-7.2.3/arch/x86/pci/xen.c
+587
View File
@@ -0,0 +1,587 @@
// SPDX-License-Identifier: GPL-2.0-only
/*
* Xen PCI - handle PCI (INTx) and MSI infrastructure calls for PV, HVM and
* initial domain support. We also handle the DSDT _PRT callbacks for GSI's
* used in HVM and initial domain mode (PV does not parse ACPI, so it has no
* concept of GSIs). Under PV we hook under the pnbbios API for IRQs and
* 0xcf8 PCI configuration read/write.
*
* Author: Ryan Wilson <hap9@epoch.ncsc.mil>
* Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
* Stefano Stabellini <stefano.stabellini@eu.citrix.com>
*/
#include <linux/export.h>
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/acpi.h>
#include <linux/io.h>
#include <asm/io_apic.h>
#include <asm/pci_x86.h>
#include <asm/cpuid/api.h>
#include <asm/xen/hypervisor.h>
#include <xen/features.h>
#include <xen/events.h>
#include <xen/pci.h>
#include <asm/xen/pci.h>
#include <asm/xen/cpuid.h>
#include <asm/apic.h>
#include <asm/acpi.h>
#include <asm/i8259.h>
static int xen_pcifront_enable_irq(struct pci_dev *dev)
{
int rc;
int share = 1;
int pirq;
u8 gsi;
rc = pci_read_config_byte(dev, PCI_INTERRUPT_LINE, &gsi);
if (rc) {
dev_warn(&dev->dev, "Xen PCI: failed to read interrupt line: %d\n",
rc);
return pcibios_err_to_errno(rc);
}
/* In PV DomU the Xen PCI backend puts the PIRQ in the interrupt line.*/
pirq = gsi;
if (gsi < nr_legacy_irqs())
share = 0;
rc = xen_bind_pirq_gsi_to_irq(gsi, pirq, share, "pcifront");
if (rc < 0) {
dev_warn(&dev->dev, "Xen PCI: failed to bind GSI%d (PIRQ%d) to IRQ: %d\n",
gsi, pirq, rc);
return rc;
}
dev->irq = rc;
dev_info(&dev->dev, "Xen PCI mapped GSI%d to IRQ%d\n", gsi, dev->irq);
return 0;
}
#ifdef CONFIG_ACPI
static int xen_register_pirq(u32 gsi, int triggering, bool set_pirq)
{
int rc, pirq = -1, irq;
struct physdev_map_pirq map_irq;
int shareable = 0;
char *name;
irq = xen_irq_from_gsi(gsi);
if (irq > 0)
return irq;
if (set_pirq)
pirq = gsi;
map_irq.domid = DOMID_SELF;
map_irq.type = MAP_PIRQ_TYPE_GSI;
map_irq.index = gsi;
map_irq.pirq = pirq;
rc = HYPERVISOR_physdev_op(PHYSDEVOP_map_pirq, &map_irq);
if (rc) {
printk(KERN_WARNING "xen map irq failed %d\n", rc);
return -1;
}
if (triggering == ACPI_EDGE_SENSITIVE) {
shareable = 0;
name = "ioapic-edge";
} else {
shareable = 1;
name = "ioapic-level";
}
irq = xen_bind_pirq_gsi_to_irq(gsi, map_irq.pirq, shareable, name);
if (irq < 0)
goto out;
printk(KERN_DEBUG "xen: --> pirq=%d -> irq=%d (gsi=%d)\n", map_irq.pirq, irq, gsi);
out:
return irq;
}
static int acpi_register_gsi_xen_hvm(struct device *dev, u32 gsi,
int trigger, int polarity)
{
if (!xen_hvm_domain())
return -1;
return xen_register_pirq(gsi, trigger,
false /* no mapping of GSI to PIRQ */);
}
#ifdef CONFIG_XEN_PV_DOM0
static int xen_register_gsi(u32 gsi, int triggering, int polarity)
{
int rc, irq;
struct physdev_setup_gsi setup_gsi;
if (!xen_pv_domain())
return -1;
printk(KERN_DEBUG "xen: registering gsi %u triggering %d polarity %d\n",
gsi, triggering, polarity);
irq = xen_register_pirq(gsi, triggering, true);
setup_gsi.gsi = gsi;
setup_gsi.triggering = (triggering == ACPI_EDGE_SENSITIVE ? 0 : 1);
setup_gsi.polarity = (polarity == ACPI_ACTIVE_HIGH ? 0 : 1);
rc = HYPERVISOR_physdev_op(PHYSDEVOP_setup_gsi, &setup_gsi);
if (rc == -EEXIST)
printk(KERN_INFO "Already setup the GSI :%d\n", gsi);
else if (rc) {
printk(KERN_ERR "Failed to setup GSI :%d, err_code:%d\n",
gsi, rc);
}
return irq;
}
static int acpi_register_gsi_xen(struct device *dev, u32 gsi,
int trigger, int polarity)
{
return xen_register_gsi(gsi, trigger, polarity);
}
#endif
#endif
#if defined(CONFIG_PCI_MSI)
#include <linux/msi.h>
struct xen_pci_frontend_ops *xen_pci_frontend;
EXPORT_SYMBOL_GPL(xen_pci_frontend);
struct xen_msi_ops {
int (*setup_msi_irqs)(struct pci_dev *dev, int nvec, int type);
void (*teardown_msi_irqs)(struct pci_dev *dev);
};
static struct xen_msi_ops xen_msi_ops __ro_after_init;
static int xen_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
{
int irq, ret, i;
struct msi_desc *msidesc;
int *v;
if (type == PCI_CAP_ID_MSI && nvec > 1)
return 1;
v = kzalloc_objs(int, max(1, nvec));
if (!v)
return -ENOMEM;
if (type == PCI_CAP_ID_MSIX)
ret = xen_pci_frontend_enable_msix(dev, v, nvec);
else
ret = xen_pci_frontend_enable_msi(dev, v);
if (ret)
goto error;
i = 0;
msi_for_each_desc(msidesc, &dev->dev, MSI_DESC_NOTASSOCIATED) {
irq = xen_bind_pirq_msi_to_irq(dev, msidesc, v[i],
(type == PCI_CAP_ID_MSI) ? nvec : 1,
(type == PCI_CAP_ID_MSIX) ?
"pcifront-msi-x" :
"pcifront-msi",
DOMID_SELF);
if (irq < 0) {
ret = irq;
goto free;
}
i++;
}
kfree(v);
return msi_device_populate_sysfs(&dev->dev);
error:
if (ret == -ENOSYS)
dev_err(&dev->dev, "Xen PCI frontend has not registered MSI/MSI-X support!\n");
else if (ret)
dev_err(&dev->dev, "Xen PCI frontend error: %d!\n", ret);
free:
kfree(v);
return ret;
}
static void xen_msi_compose_msg(struct pci_dev *pdev, unsigned int pirq,
struct msi_msg *msg)
{
/*
* We set vector == 0 to tell the hypervisor we don't care about
* it, but we want a pirq setup instead. We use the dest_id fields
* to pass the pirq that we want.
*/
memset(msg, 0, sizeof(*msg));
msg->address_hi = X86_MSI_BASE_ADDRESS_HIGH;
msg->arch_addr_hi.destid_8_31 = pirq >> 8;
msg->arch_addr_lo.destid_0_7 = pirq & 0xFF;
msg->arch_addr_lo.base_address = X86_MSI_BASE_ADDRESS_LOW;
msg->arch_data.delivery_mode = APIC_DELIVERY_MODE_EXTINT;
}
static int xen_hvm_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
{
int irq, pirq;
struct msi_desc *msidesc;
struct msi_msg msg;
if (type == PCI_CAP_ID_MSI && nvec > 1)
return 1;
msi_for_each_desc(msidesc, &dev->dev, MSI_DESC_NOTASSOCIATED) {
pirq = xen_allocate_pirq_msi(dev, msidesc);
if (pirq < 0) {
irq = -ENODEV;
goto error;
}
xen_msi_compose_msg(dev, pirq, &msg);
__pci_write_msi_msg(msidesc, &msg);
dev_dbg(&dev->dev, "xen: msi bound to pirq=%d\n", pirq);
irq = xen_bind_pirq_msi_to_irq(dev, msidesc, pirq,
(type == PCI_CAP_ID_MSI) ? nvec : 1,
(type == PCI_CAP_ID_MSIX) ?
"msi-x" : "msi",
DOMID_SELF);
if (irq < 0)
goto error;
dev_dbg(&dev->dev,
"xen: msi --> pirq=%d --> irq=%d\n", pirq, irq);
}
return msi_device_populate_sysfs(&dev->dev);
error:
dev_err(&dev->dev, "Failed to create MSI%s! ret=%d!\n",
type == PCI_CAP_ID_MSI ? "" : "-X", irq);
return irq;
}
#ifdef CONFIG_XEN_PV_DOM0
static bool __read_mostly pci_seg_supported = true;
static int xen_initdom_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
{
int ret = 0;
struct msi_desc *msidesc;
msi_for_each_desc(msidesc, &dev->dev, MSI_DESC_NOTASSOCIATED) {
struct physdev_map_pirq map_irq;
domid_t domid;
domid = ret = xen_find_device_domain_owner(dev);
/* N.B. Casting int's -ENODEV to uint16_t results in 0xFFED,
* hence check ret value for < 0. */
if (ret < 0)
domid = DOMID_SELF;
memset(&map_irq, 0, sizeof(map_irq));
map_irq.domid = domid;
map_irq.type = MAP_PIRQ_TYPE_MSI_SEG;
map_irq.index = -1;
map_irq.pirq = -1;
map_irq.bus = dev->bus->number |
(pci_domain_nr(dev->bus) << 16);
map_irq.devfn = dev->devfn;
if (type == PCI_CAP_ID_MSI && nvec > 1) {
map_irq.type = MAP_PIRQ_TYPE_MULTI_MSI;
map_irq.entry_nr = nvec;
} else if (type == PCI_CAP_ID_MSIX) {
int pos;
unsigned long flags;
u32 table_offset, bir;
pos = dev->msix_cap;
pci_read_config_dword(dev, pos + PCI_MSIX_TABLE,
&table_offset);
bir = (u8)(table_offset & PCI_MSIX_TABLE_BIR);
flags = pci_resource_flags(dev, bir);
if (!flags || (flags & IORESOURCE_UNSET))
return -EINVAL;
map_irq.table_base = pci_resource_start(dev, bir);
map_irq.entry_nr = msidesc->msi_index;
}
ret = -EINVAL;
if (pci_seg_supported)
ret = HYPERVISOR_physdev_op(PHYSDEVOP_map_pirq,
&map_irq);
if (type == PCI_CAP_ID_MSI && nvec > 1 && ret) {
/*
* If MAP_PIRQ_TYPE_MULTI_MSI is not available
* there's nothing else we can do in this case.
* Just set ret > 0 so driver can retry with
* single MSI.
*/
ret = 1;
goto out;
}
if (ret == -EINVAL && !pci_domain_nr(dev->bus)) {
map_irq.type = MAP_PIRQ_TYPE_MSI;
map_irq.index = -1;
map_irq.pirq = -1;
map_irq.bus = dev->bus->number;
ret = HYPERVISOR_physdev_op(PHYSDEVOP_map_pirq,
&map_irq);
if (ret != -EINVAL)
pci_seg_supported = false;
}
if (ret) {
dev_warn(&dev->dev, "xen map irq failed %d for %d domain\n",
ret, domid);
goto out;
}
ret = xen_bind_pirq_msi_to_irq(dev, msidesc, map_irq.pirq,
(type == PCI_CAP_ID_MSI) ? nvec : 1,
(type == PCI_CAP_ID_MSIX) ? "msi-x" : "msi",
domid);
if (ret < 0)
goto out;
}
ret = msi_device_populate_sysfs(&dev->dev);
out:
return ret;
}
bool xen_initdom_restore_msi(struct pci_dev *dev)
{
int ret = 0;
if (!xen_initial_domain())
return true;
if (pci_seg_supported) {
struct physdev_pci_device restore_ext;
restore_ext.seg = pci_domain_nr(dev->bus);
restore_ext.bus = dev->bus->number;
restore_ext.devfn = dev->devfn;
ret = HYPERVISOR_physdev_op(PHYSDEVOP_restore_msi_ext,
&restore_ext);
if (ret == -ENOSYS)
pci_seg_supported = false;
WARN(ret && ret != -ENOSYS, "restore_msi_ext -> %d\n", ret);
}
if (!pci_seg_supported) {
struct physdev_restore_msi restore;
restore.bus = dev->bus->number;
restore.devfn = dev->devfn;
ret = HYPERVISOR_physdev_op(PHYSDEVOP_restore_msi, &restore);
WARN(ret && ret != -ENOSYS, "restore_msi -> %d\n", ret);
}
return false;
}
#else /* CONFIG_XEN_PV_DOM0 */
#define xen_initdom_setup_msi_irqs NULL
#endif /* !CONFIG_XEN_PV_DOM0 */
static void xen_teardown_msi_irqs(struct pci_dev *dev)
{
struct msi_desc *msidesc;
int i;
msi_for_each_desc(msidesc, &dev->dev, MSI_DESC_ASSOCIATED) {
for (i = 0; i < msidesc->nvec_used; i++)
xen_destroy_irq(msidesc->irq + i);
msidesc->irq = 0;
}
msi_device_destroy_sysfs(&dev->dev);
}
static void xen_pv_teardown_msi_irqs(struct pci_dev *dev)
{
if (dev->msix_enabled)
xen_pci_frontend_disable_msix(dev);
else
xen_pci_frontend_disable_msi(dev);
xen_teardown_msi_irqs(dev);
}
static int xen_msi_domain_alloc_irqs(struct irq_domain *domain,
struct device *dev, int nvec)
{
int type;
if (WARN_ON_ONCE(!dev_is_pci(dev)))
return -EINVAL;
type = to_pci_dev(dev)->msix_enabled ? PCI_CAP_ID_MSIX : PCI_CAP_ID_MSI;
return xen_msi_ops.setup_msi_irqs(to_pci_dev(dev), nvec, type);
}
static void xen_msi_domain_free_irqs(struct irq_domain *domain,
struct device *dev)
{
if (WARN_ON_ONCE(!dev_is_pci(dev)))
return;
xen_msi_ops.teardown_msi_irqs(to_pci_dev(dev));
}
static struct msi_domain_ops xen_pci_msi_domain_ops = {
.domain_alloc_irqs = xen_msi_domain_alloc_irqs,
.domain_free_irqs = xen_msi_domain_free_irqs,
};
static struct msi_domain_info xen_pci_msi_domain_info = {
.flags = MSI_FLAG_PCI_MSIX | MSI_FLAG_FREE_MSI_DESCS |
MSI_FLAG_DEV_SYSFS | MSI_FLAG_NO_MASK,
.ops = &xen_pci_msi_domain_ops,
};
/*
* This irq domain is a blatant violation of the irq domain design, but
* distangling XEN into real irq domains is not a job for mere mortals with
* limited XENology. But it's the least dangerous way for a mere mortal to
* get rid of the arch_*_msi_irqs() hackery in order to store the irq
* domain pointer in struct device. This irq domain wrappery allows to do
* that without breaking XEN terminally.
*/
static __init struct irq_domain *xen_create_pci_msi_domain(void)
{
struct irq_domain *d = NULL;
struct fwnode_handle *fn;
fn = irq_domain_alloc_named_fwnode("XEN-MSI");
if (fn)
d = msi_create_irq_domain(fn, &xen_pci_msi_domain_info, NULL);
/* FIXME: No idea how to survive if this fails */
BUG_ON(!d);
return d;
}
static __init void xen_setup_pci_msi(void)
{
if (xen_pv_domain()) {
if (xen_initial_domain())
xen_msi_ops.setup_msi_irqs = xen_initdom_setup_msi_irqs;
else
xen_msi_ops.setup_msi_irqs = xen_setup_msi_irqs;
xen_msi_ops.teardown_msi_irqs = xen_pv_teardown_msi_irqs;
} else if (xen_hvm_domain()) {
xen_msi_ops.setup_msi_irqs = xen_hvm_setup_msi_irqs;
xen_msi_ops.teardown_msi_irqs = xen_teardown_msi_irqs;
} else {
WARN_ON_ONCE(1);
return;
}
/*
* Override the PCI/MSI irq domain init function. No point
* in allocating the native domain and never use it.
*/
x86_init.irqs.create_pci_msi_domain = xen_create_pci_msi_domain;
}
#else /* CONFIG_PCI_MSI */
static inline void xen_setup_pci_msi(void) { }
#endif /* CONFIG_PCI_MSI */
int __init pci_xen_init(void)
{
if (!xen_pv_domain() || xen_initial_domain())
return -ENODEV;
printk(KERN_INFO "PCI: setting up Xen PCI frontend stub\n");
pcibios_set_cache_line_size();
pcibios_enable_irq = xen_pcifront_enable_irq;
pcibios_disable_irq = NULL;
/* Keep ACPI out of the picture */
acpi_noirq_set();
xen_setup_pci_msi();
return 0;
}
#ifdef CONFIG_PCI_MSI
static void __init xen_hvm_msi_init(void)
{
if (!apic_is_disabled) {
/*
* If hardware supports (x2)APIC virtualization (as indicated
* by hypervisor's leaf 4) then we don't need to use pirqs/
* event channels for MSI handling and instead use regular
* APIC processing
*/
uint32_t eax = cpuid_eax(xen_cpuid_base() + 4);
if (((eax & XEN_HVM_CPUID_X2APIC_VIRT) && x2apic_mode) ||
((eax & XEN_HVM_CPUID_APIC_ACCESS_VIRT) && boot_cpu_has(X86_FEATURE_APIC)))
return;
}
xen_setup_pci_msi();
}
#endif
int __init pci_xen_hvm_init(void)
{
if (!xen_have_vector_callback || !xen_feature(XENFEAT_hvm_pirqs))
return 0;
#ifdef CONFIG_ACPI
/*
* We don't want to change the actual ACPI delivery model,
* just how GSIs get registered.
*/
__acpi_register_gsi = acpi_register_gsi_xen_hvm;
__acpi_unregister_gsi = NULL;
#endif
#ifdef CONFIG_PCI_MSI
/*
* We need to wait until after x2apic is initialized
* before we can set MSI IRQ ops.
*/
x86_platform.apic_post_init = xen_hvm_msi_init;
#endif
return 0;
}
#ifdef CONFIG_XEN_PV_DOM0
int __init pci_xen_initial_domain(void)
{
int irq;
xen_setup_pci_msi();
__acpi_register_gsi = acpi_register_gsi_xen;
__acpi_unregister_gsi = NULL;
/*
* Pre-allocate the legacy IRQs. Use NR_LEGACY_IRQS here
* because we don't have a PIC and thus nr_legacy_irqs() is zero.
*/
for (irq = 0; irq < NR_IRQS_LEGACY; irq++) {
int trigger, polarity;
if (acpi_get_override_irq(irq, &trigger, &polarity) == -1)
continue;
xen_register_pirq(irq,
trigger ? ACPI_LEVEL_SENSITIVE : ACPI_EDGE_SENSITIVE,
true /* Map GSI to PIRQ */);
}
if (0 == nr_ioapics) {
for (irq = 0; irq < nr_legacy_irqs(); irq++)
xen_bind_pirq_gsi_to_irq(irq, irq, 0, "xt-pic");
}
return 0;
}
#endif