2005-04-17 06:20:36 +08:00
|
|
|
#include <linux/pci.h>
|
|
|
|
#include <linux/acpi.h>
|
|
|
|
#include <linux/init.h>
|
2005-10-01 00:34:42 +08:00
|
|
|
#include <linux/irq.h>
|
2007-10-04 06:56:14 +08:00
|
|
|
#include <linux/dmi.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
|
|
|
#include <linux/slab.h>
|
2005-09-13 00:49:24 +08:00
|
|
|
#include <asm/numa.h>
|
2008-12-27 21:02:28 +08:00
|
|
|
#include <asm/pci_x86.h>
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2007-10-04 06:56:51 +08:00
|
|
|
struct pci_root_info {
|
2009-10-07 05:33:49 +08:00
|
|
|
struct acpi_device *bridge;
|
2007-10-04 06:56:51 +08:00
|
|
|
char *name;
|
|
|
|
unsigned int res_num;
|
|
|
|
struct resource *res;
|
|
|
|
struct pci_bus *bus;
|
|
|
|
int busnum;
|
|
|
|
};
|
|
|
|
|
2010-02-24 01:24:41 +08:00
|
|
|
static bool pci_use_crs = true;
|
|
|
|
|
|
|
|
static int __init set_use_crs(const struct dmi_system_id *id)
|
|
|
|
{
|
|
|
|
pci_use_crs = true;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static const struct dmi_system_id pci_use_crs_table[] __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"),
|
|
|
|
},
|
|
|
|
},
|
x86/PCI: use host bridge _CRS info on ASRock ALiveSATA2-GLAN
This DMI quirk turns on "pci=use_crs" for the ALiveSATA2-GLAN because
amd_bus.c doesn't handle this system correctly.
The system has a single HyperTransport I/O chain, but has two PCI host
bridges to buses 00 and 80. amd_bus.c learns the MMIO range associated
with buses 00-ff and that this range is routed to the HT chain hosted at
node 0, link 0:
bus: [00, ff] on node 0 link 0
bus: 00 index 1 [mem 0x80000000-0xfcffffffff]
This includes the address space for both bus 00 and bus 80, and amd_bus.c
assumes it's all routed to bus 00.
We find device 80:01.0, which BIOS left in the middle of that space, but
we don't find a bridge from bus 00 to bus 80, so we conclude that 80:01.0
is unreachable from bus 00, and we move it from the original, working,
address to something outside the bus 00 aperture, which does not work:
pci 0000:80:01.0: reg 10: [mem 0xfebfc000-0xfebfffff 64bit]
pci 0000:80:01.0: BAR 0: assigned [mem 0xfd00000000-0xfd00003fff 64bit]
The BIOS told us everything we need to know to handle this correctly,
so we're better off if we just pay attention, which lets us leave the
80:01.0 device at the original, working, address:
ACPI: PCI Root Bridge [PCI0] (domain 0000 [bus 00-7f])
pci_root PNP0A03:00: host bridge window [mem 0x80000000-0xff37ffff]
ACPI: PCI Root Bridge [PCI1] (domain 0000 [bus 80-ff])
pci_root PNP0A08:00: host bridge window [mem 0xfebfc000-0xfebfffff]
This was a regression between 2.6.33 and 2.6.34. In 2.6.33, amd_bus.c
was used only when we found multiple HT chains. 3e3da00c01d050, which
enabled amd_bus.c even on systems with a single HT chain, caused this
failure.
This quirk was written by Graham. If we ever enable "pci=use_crs" for
machines from 2006 or earlir, this quirk should be removed.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=16007
Cc: stable@kernel.org
Reported-by: Graham Ramsey <ramsey.graham@ntlworld.com>
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2010-07-24 02:53:27 +08:00
|
|
|
/* 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"),
|
|
|
|
},
|
|
|
|
},
|
2010-02-24 01:24:41 +08:00
|
|
|
{}
|
|
|
|
};
|
|
|
|
|
|
|
|
void __init pci_acpi_crs_quirks(void)
|
|
|
|
{
|
|
|
|
int year;
|
|
|
|
|
|
|
|
if (dmi_get_date(DMI_BIOS_DATE, &year, NULL, NULL) && year < 2008)
|
|
|
|
pci_use_crs = false;
|
|
|
|
|
|
|
|
dmi_check_system(pci_use_crs_table);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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;
|
|
|
|
|
|
|
|
printk(KERN_INFO "PCI: %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");
|
|
|
|
}
|
|
|
|
|
2007-10-04 06:56:51 +08:00
|
|
|
static acpi_status
|
|
|
|
resource_to_addr(struct acpi_resource *resource,
|
|
|
|
struct acpi_resource_address64 *addr)
|
|
|
|
{
|
|
|
|
acpi_status status;
|
2010-04-21 03:52:41 +08:00
|
|
|
struct acpi_resource_memory24 *memory24;
|
|
|
|
struct acpi_resource_memory32 *memory32;
|
|
|
|
struct acpi_resource_fixed_memory32 *fixed_memory32;
|
2007-10-04 06:56:51 +08:00
|
|
|
|
2010-04-21 03:52:41 +08:00
|
|
|
memset(addr, 0, sizeof(*addr));
|
|
|
|
switch (resource->type) {
|
|
|
|
case ACPI_RESOURCE_TYPE_MEMORY24:
|
|
|
|
memory24 = &resource->data.memory24;
|
|
|
|
addr->resource_type = ACPI_MEMORY_RANGE;
|
|
|
|
addr->minimum = memory24->minimum;
|
|
|
|
addr->address_length = memory24->address_length;
|
|
|
|
addr->maximum = addr->minimum + addr->address_length - 1;
|
2007-10-04 06:56:51 +08:00
|
|
|
return AE_OK;
|
2010-04-21 03:52:41 +08:00
|
|
|
case ACPI_RESOURCE_TYPE_MEMORY32:
|
|
|
|
memory32 = &resource->data.memory32;
|
|
|
|
addr->resource_type = ACPI_MEMORY_RANGE;
|
|
|
|
addr->minimum = memory32->minimum;
|
|
|
|
addr->address_length = memory32->address_length;
|
|
|
|
addr->maximum = addr->minimum + addr->address_length - 1;
|
|
|
|
return AE_OK;
|
|
|
|
case ACPI_RESOURCE_TYPE_FIXED_MEMORY32:
|
|
|
|
fixed_memory32 = &resource->data.fixed_memory32;
|
|
|
|
addr->resource_type = ACPI_MEMORY_RANGE;
|
|
|
|
addr->minimum = fixed_memory32->address;
|
|
|
|
addr->address_length = fixed_memory32->address_length;
|
|
|
|
addr->maximum = addr->minimum + addr->address_length - 1;
|
|
|
|
return AE_OK;
|
|
|
|
case ACPI_RESOURCE_TYPE_ADDRESS16:
|
|
|
|
case ACPI_RESOURCE_TYPE_ADDRESS32:
|
|
|
|
case ACPI_RESOURCE_TYPE_ADDRESS64:
|
|
|
|
status = acpi_resource_to_address64(resource, addr);
|
|
|
|
if (ACPI_SUCCESS(status) &&
|
|
|
|
(addr->resource_type == ACPI_MEMORY_RANGE ||
|
|
|
|
addr->resource_type == ACPI_IO_RANGE) &&
|
|
|
|
addr->address_length > 0) {
|
|
|
|
return AE_OK;
|
|
|
|
}
|
|
|
|
break;
|
2007-10-04 06:56:51 +08:00
|
|
|
}
|
|
|
|
return AE_ERROR;
|
|
|
|
}
|
|
|
|
|
|
|
|
static acpi_status
|
|
|
|
count_resource(struct acpi_resource *acpi_res, void *data)
|
|
|
|
{
|
|
|
|
struct pci_root_info *info = data;
|
|
|
|
struct acpi_resource_address64 addr;
|
|
|
|
acpi_status status;
|
|
|
|
|
|
|
|
status = resource_to_addr(acpi_res, &addr);
|
|
|
|
if (ACPI_SUCCESS(status))
|
|
|
|
info->res_num++;
|
|
|
|
return AE_OK;
|
|
|
|
}
|
|
|
|
|
|
|
|
static acpi_status
|
|
|
|
setup_resource(struct acpi_resource *acpi_res, void *data)
|
|
|
|
{
|
|
|
|
struct pci_root_info *info = data;
|
|
|
|
struct resource *res;
|
|
|
|
struct acpi_resource_address64 addr;
|
|
|
|
acpi_status status;
|
|
|
|
unsigned long flags;
|
x86/PCI: compute Address Space length rather than using _LEN
ACPI _CRS Address Space Descriptors have _MIN, _MAX, and _LEN. Linux has
been computing Address Spaces as [_MIN to _MIN + _LEN - 1]. Based on the
tests in the bug reports below, Windows apparently uses [_MIN to _MAX].
Per spec (ACPI 4.0, Table 6-40), for _CRS fixed-size, fixed location
descriptors, "_LEN must be (_MAX - _MIN + 1)", and when that's true, it
doesn't matter which way we compute the end. But of course, there are
BIOSes that don't follow this rule, and we're better off if Linux handles
those exceptions the same way as Windows.
This patch makes Linux use [_MIN to _MAX], as Windows seems to do. This
effectively reverts d558b483d5 and 03db42adfe and replaces them with
simpler code.
https://bugzilla.kernel.org/show_bug.cgi?id=14337 (round)
https://bugzilla.kernel.org/show_bug.cgi?id=15480 (truncate)
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2010-04-28 04:45:43 +08:00
|
|
|
u64 start, end;
|
2009-06-25 10:01:19 +08:00
|
|
|
|
2007-10-04 06:56:51 +08:00
|
|
|
status = resource_to_addr(acpi_res, &addr);
|
|
|
|
if (!ACPI_SUCCESS(status))
|
|
|
|
return AE_OK;
|
|
|
|
|
|
|
|
if (addr.resource_type == ACPI_MEMORY_RANGE) {
|
|
|
|
flags = IORESOURCE_MEM;
|
|
|
|
if (addr.info.mem.caching == ACPI_PREFETCHABLE_MEMORY)
|
|
|
|
flags |= IORESOURCE_PREFETCH;
|
|
|
|
} else if (addr.resource_type == ACPI_IO_RANGE) {
|
|
|
|
flags = IORESOURCE_IO;
|
|
|
|
} else
|
|
|
|
return AE_OK;
|
|
|
|
|
2009-06-25 10:01:19 +08:00
|
|
|
start = addr.minimum + addr.translation_offset;
|
x86/PCI: compute Address Space length rather than using _LEN
ACPI _CRS Address Space Descriptors have _MIN, _MAX, and _LEN. Linux has
been computing Address Spaces as [_MIN to _MIN + _LEN - 1]. Based on the
tests in the bug reports below, Windows apparently uses [_MIN to _MAX].
Per spec (ACPI 4.0, Table 6-40), for _CRS fixed-size, fixed location
descriptors, "_LEN must be (_MAX - _MIN + 1)", and when that's true, it
doesn't matter which way we compute the end. But of course, there are
BIOSes that don't follow this rule, and we're better off if Linux handles
those exceptions the same way as Windows.
This patch makes Linux use [_MIN to _MAX], as Windows seems to do. This
effectively reverts d558b483d5 and 03db42adfe and replaces them with
simpler code.
https://bugzilla.kernel.org/show_bug.cgi?id=14337 (round)
https://bugzilla.kernel.org/show_bug.cgi?id=15480 (truncate)
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2010-04-28 04:45:43 +08:00
|
|
|
end = addr.maximum + addr.translation_offset;
|
2009-05-28 03:41:44 +08:00
|
|
|
|
2009-06-25 10:01:19 +08:00
|
|
|
res = &info->res[info->res_num];
|
|
|
|
res->name = info->name;
|
|
|
|
res->flags = flags;
|
|
|
|
res->start = start;
|
|
|
|
res->end = end;
|
|
|
|
res->child = NULL;
|
|
|
|
|
2010-02-24 01:24:41 +08:00
|
|
|
if (!pci_use_crs) {
|
2009-11-05 01:39:13 +08:00
|
|
|
dev_printk(KERN_DEBUG, &info->bridge->dev,
|
|
|
|
"host bridge window %pR (ignored)\n", res);
|
|
|
|
return AE_OK;
|
|
|
|
}
|
|
|
|
|
2010-09-23 01:09:19 +08:00
|
|
|
info->res_num++;
|
|
|
|
if (addr.translation_offset)
|
|
|
|
dev_info(&info->bridge->dev, "host bridge window %pR "
|
|
|
|
"(PCI address [%#llx-%#llx])\n",
|
|
|
|
res, res->start - addr.translation_offset,
|
|
|
|
res->end - addr.translation_offset);
|
|
|
|
else
|
|
|
|
dev_info(&info->bridge->dev, "host bridge window %pR\n", res);
|
|
|
|
|
|
|
|
return AE_OK;
|
|
|
|
}
|
|
|
|
|
|
|
|
static bool resource_contains(struct resource *res, resource_size_t point)
|
|
|
|
{
|
|
|
|
if (res->start <= point && point <= res->end)
|
|
|
|
return true;
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2011-05-15 01:27:33 +08:00
|
|
|
static void coalesce_windows(struct pci_root_info *info, unsigned long type)
|
2010-09-23 01:09:19 +08:00
|
|
|
{
|
|
|
|
int i, j;
|
|
|
|
struct resource *res1, *res2;
|
|
|
|
|
|
|
|
for (i = 0; i < info->res_num; i++) {
|
|
|
|
res1 = &info->res[i];
|
|
|
|
if (!(res1->flags & type))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
for (j = i + 1; j < info->res_num; j++) {
|
|
|
|
res2 = &info->res[j];
|
|
|
|
if (!(res2->flags & type))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* I don't like throwing away windows because then
|
|
|
|
* our resources no longer match the ACPI _CRS, but
|
|
|
|
* the kernel resource tree doesn't allow overlaps.
|
|
|
|
*/
|
|
|
|
if (resource_contains(res1, res2->start) ||
|
|
|
|
resource_contains(res1, res2->end) ||
|
|
|
|
resource_contains(res2, res1->start) ||
|
|
|
|
resource_contains(res2, res1->end)) {
|
|
|
|
res1->start = min(res1->start, res2->start);
|
|
|
|
res1->end = max(res1->end, res2->end);
|
|
|
|
dev_info(&info->bridge->dev,
|
|
|
|
"host bridge window expanded to %pR; %pR ignored\n",
|
|
|
|
res1, res2);
|
|
|
|
res2->flags = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void add_resources(struct pci_root_info *info)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
struct resource *res, *root, *conflict;
|
|
|
|
|
|
|
|
if (!pci_use_crs)
|
|
|
|
return;
|
|
|
|
|
|
|
|
coalesce_windows(info, IORESOURCE_MEM);
|
|
|
|
coalesce_windows(info, IORESOURCE_IO);
|
|
|
|
|
|
|
|
for (i = 0; i < info->res_num; i++) {
|
|
|
|
res = &info->res[i];
|
|
|
|
|
|
|
|
if (res->flags & IORESOURCE_MEM)
|
|
|
|
root = &iomem_resource;
|
|
|
|
else if (res->flags & IORESOURCE_IO)
|
|
|
|
root = &ioport_resource;
|
2009-10-07 05:33:49 +08:00
|
|
|
else
|
2010-09-23 01:09:19 +08:00
|
|
|
continue;
|
|
|
|
|
|
|
|
conflict = insert_resource_conflict(root, res);
|
|
|
|
if (conflict)
|
2011-07-03 00:47:12 +08:00
|
|
|
dev_info(&info->bridge->dev,
|
|
|
|
"ignoring host bridge window %pR (conflicts with %s %pR)\n",
|
|
|
|
res, conflict->name, conflict);
|
2010-09-23 01:09:19 +08:00
|
|
|
else
|
|
|
|
pci_bus_add_resource(info->bus, res, 0);
|
2007-10-04 06:56:51 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
get_current_resources(struct acpi_device *device, int busnum,
|
2008-02-09 06:00:52 +08:00
|
|
|
int domain, struct pci_bus *bus)
|
2007-10-04 06:56:51 +08:00
|
|
|
{
|
|
|
|
struct pci_root_info info;
|
|
|
|
size_t size;
|
|
|
|
|
2010-02-24 01:24:41 +08:00
|
|
|
if (pci_use_crs)
|
2010-02-24 01:24:36 +08:00
|
|
|
pci_bus_remove_resources(bus);
|
2009-11-05 01:39:13 +08:00
|
|
|
|
2009-10-07 05:33:49 +08:00
|
|
|
info.bridge = device;
|
2007-10-04 06:56:51 +08:00
|
|
|
info.bus = bus;
|
|
|
|
info.res_num = 0;
|
|
|
|
acpi_walk_resources(device->handle, METHOD_NAME__CRS, count_resource,
|
|
|
|
&info);
|
|
|
|
if (!info.res_num)
|
|
|
|
return;
|
|
|
|
|
|
|
|
size = sizeof(*info.res) * info.res_num;
|
|
|
|
info.res = kmalloc(size, GFP_KERNEL);
|
|
|
|
if (!info.res)
|
|
|
|
goto res_alloc_fail;
|
|
|
|
|
2010-05-25 03:13:16 +08:00
|
|
|
info.name = kasprintf(GFP_KERNEL, "PCI Bus %04x:%02x", domain, busnum);
|
2007-10-04 06:56:51 +08:00
|
|
|
if (!info.name)
|
|
|
|
goto name_alloc_fail;
|
|
|
|
|
|
|
|
info.res_num = 0;
|
|
|
|
acpi_walk_resources(device->handle, METHOD_NAME__CRS, setup_resource,
|
|
|
|
&info);
|
|
|
|
|
2010-09-23 01:09:19 +08:00
|
|
|
add_resources(&info);
|
2007-10-04 06:56:51 +08:00
|
|
|
return;
|
|
|
|
|
|
|
|
name_alloc_fail:
|
|
|
|
kfree(info.res);
|
|
|
|
res_alloc_fail:
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2010-03-12 03:20:11 +08:00
|
|
|
struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_pci_root *root)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2010-03-12 03:20:11 +08:00
|
|
|
struct acpi_device *device = root->device;
|
|
|
|
int domain = root->segment;
|
|
|
|
int busnum = root->secondary.start;
|
2005-09-13 00:49:24 +08:00
|
|
|
struct pci_bus *bus;
|
2007-07-22 05:23:39 +08:00
|
|
|
struct pci_sysdata *sd;
|
x86: get mp_bus_to_node early
Currently, on an amd k8 system with multi ht chains, the numa_node of
pci devices under /sys/devices/pci0000:80/* is always 0, even if that
chain is on node 1 or 2 or 3.
Workaround: pcibus_to_node(bus) is used when we want to get the node that
pci_device is on.
In struct device, we already have numa_node member, and we could use
dev_to_node()/set_dev_node() to get and set numa_node in the device.
set_dev_node is called in pci_device_add() with pcibus_to_node(bus),
and pcibus_to_node uses bus->sysdata for nodeid.
The problem is when pci_add_device is called, bus->sysdata is not assigned
correct nodeid yet. The result is that numa_node will always be 0.
pcibios_scan_root and pci_scan_root could take sysdata. So we need to get
mp_bus_to_node mapping before these two are called, and thus
get_mp_bus_to_node could get correct node for sysdata in root bus.
In scanning of the root bus, all child busses will take parent bus sysdata.
So all pci_device->dev.numa_node will be assigned correctly and automatically.
Later we could use dev_to_node(&pci_dev->dev) to get numa_node, and we
could also could make other bus specific device get the correct numa_node
too.
This is an updated version of pci_sysdata and Jeff's pci_domain patch.
[ mingo@elte.hu: build fix ]
Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2008-02-19 19:20:09 +08:00
|
|
|
int node;
|
|
|
|
#ifdef CONFIG_ACPI_NUMA
|
2007-07-22 05:23:39 +08:00
|
|
|
int pxm;
|
x86: get mp_bus_to_node early
Currently, on an amd k8 system with multi ht chains, the numa_node of
pci devices under /sys/devices/pci0000:80/* is always 0, even if that
chain is on node 1 or 2 or 3.
Workaround: pcibus_to_node(bus) is used when we want to get the node that
pci_device is on.
In struct device, we already have numa_node member, and we could use
dev_to_node()/set_dev_node() to get and set numa_node in the device.
set_dev_node is called in pci_device_add() with pcibus_to_node(bus),
and pcibus_to_node uses bus->sysdata for nodeid.
The problem is when pci_add_device is called, bus->sysdata is not assigned
correct nodeid yet. The result is that numa_node will always be 0.
pcibios_scan_root and pci_scan_root could take sysdata. So we need to get
mp_bus_to_node mapping before these two are called, and thus
get_mp_bus_to_node could get correct node for sysdata in root bus.
In scanning of the root bus, all child busses will take parent bus sysdata.
So all pci_device->dev.numa_node will be assigned correctly and automatically.
Later we could use dev_to_node(&pci_dev->dev) to get numa_node, and we
could also could make other bus specific device get the correct numa_node
too.
This is an updated version of pci_sysdata and Jeff's pci_domain patch.
[ mingo@elte.hu: build fix ]
Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2008-02-19 19:20:09 +08:00
|
|
|
#endif
|
2007-07-22 05:23:39 +08:00
|
|
|
|
2007-10-12 04:58:30 +08:00
|
|
|
if (domain && !pci_domains_supported) {
|
2009-11-05 01:32:47 +08:00
|
|
|
printk(KERN_WARNING "pci_bus %04x:%02x: "
|
|
|
|
"ignored (multiple domains not supported)\n",
|
|
|
|
domain, busnum);
|
2007-10-12 04:58:30 +08:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
x86: get mp_bus_to_node early
Currently, on an amd k8 system with multi ht chains, the numa_node of
pci devices under /sys/devices/pci0000:80/* is always 0, even if that
chain is on node 1 or 2 or 3.
Workaround: pcibus_to_node(bus) is used when we want to get the node that
pci_device is on.
In struct device, we already have numa_node member, and we could use
dev_to_node()/set_dev_node() to get and set numa_node in the device.
set_dev_node is called in pci_device_add() with pcibus_to_node(bus),
and pcibus_to_node uses bus->sysdata for nodeid.
The problem is when pci_add_device is called, bus->sysdata is not assigned
correct nodeid yet. The result is that numa_node will always be 0.
pcibios_scan_root and pci_scan_root could take sysdata. So we need to get
mp_bus_to_node mapping before these two are called, and thus
get_mp_bus_to_node could get correct node for sysdata in root bus.
In scanning of the root bus, all child busses will take parent bus sysdata.
So all pci_device->dev.numa_node will be assigned correctly and automatically.
Later we could use dev_to_node(&pci_dev->dev) to get numa_node, and we
could also could make other bus specific device get the correct numa_node
too.
This is an updated version of pci_sysdata and Jeff's pci_domain patch.
[ mingo@elte.hu: build fix ]
Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2008-02-19 19:20:09 +08:00
|
|
|
node = -1;
|
|
|
|
#ifdef CONFIG_ACPI_NUMA
|
|
|
|
pxm = acpi_get_pxm(device->handle);
|
|
|
|
if (pxm >= 0)
|
|
|
|
node = pxm_to_node(pxm);
|
|
|
|
if (node != -1)
|
|
|
|
set_mp_bus_to_node(busnum, node);
|
|
|
|
else
|
|
|
|
#endif
|
|
|
|
node = get_mp_bus_to_node(busnum);
|
2008-02-21 04:41:52 +08:00
|
|
|
|
|
|
|
if (node != -1 && !node_online(node))
|
|
|
|
node = -1;
|
x86: get mp_bus_to_node early
Currently, on an amd k8 system with multi ht chains, the numa_node of
pci devices under /sys/devices/pci0000:80/* is always 0, even if that
chain is on node 1 or 2 or 3.
Workaround: pcibus_to_node(bus) is used when we want to get the node that
pci_device is on.
In struct device, we already have numa_node member, and we could use
dev_to_node()/set_dev_node() to get and set numa_node in the device.
set_dev_node is called in pci_device_add() with pcibus_to_node(bus),
and pcibus_to_node uses bus->sysdata for nodeid.
The problem is when pci_add_device is called, bus->sysdata is not assigned
correct nodeid yet. The result is that numa_node will always be 0.
pcibios_scan_root and pci_scan_root could take sysdata. So we need to get
mp_bus_to_node mapping before these two are called, and thus
get_mp_bus_to_node could get correct node for sysdata in root bus.
In scanning of the root bus, all child busses will take parent bus sysdata.
So all pci_device->dev.numa_node will be assigned correctly and automatically.
Later we could use dev_to_node(&pci_dev->dev) to get numa_node, and we
could also could make other bus specific device get the correct numa_node
too.
This is an updated version of pci_sysdata and Jeff's pci_domain patch.
[ mingo@elte.hu: build fix ]
Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2008-02-19 19:20:09 +08:00
|
|
|
|
2007-07-22 05:23:39 +08:00
|
|
|
/* Allocate per-root-bus (not per bus) arch-specific data.
|
|
|
|
* TODO: leak; this memory is never freed.
|
|
|
|
* It's arguable whether it's worth the trouble to care.
|
|
|
|
*/
|
|
|
|
sd = kzalloc(sizeof(*sd), GFP_KERNEL);
|
|
|
|
if (!sd) {
|
2009-11-05 01:32:47 +08:00
|
|
|
printk(KERN_WARNING "pci_bus %04x:%02x: "
|
|
|
|
"ignored (out of memory)\n", domain, busnum);
|
2007-07-22 05:23:39 +08:00
|
|
|
return NULL;
|
|
|
|
}
|
2005-09-13 00:49:24 +08:00
|
|
|
|
2007-10-12 04:58:30 +08:00
|
|
|
sd->domain = domain;
|
x86: get mp_bus_to_node early
Currently, on an amd k8 system with multi ht chains, the numa_node of
pci devices under /sys/devices/pci0000:80/* is always 0, even if that
chain is on node 1 or 2 or 3.
Workaround: pcibus_to_node(bus) is used when we want to get the node that
pci_device is on.
In struct device, we already have numa_node member, and we could use
dev_to_node()/set_dev_node() to get and set numa_node in the device.
set_dev_node is called in pci_device_add() with pcibus_to_node(bus),
and pcibus_to_node uses bus->sysdata for nodeid.
The problem is when pci_add_device is called, bus->sysdata is not assigned
correct nodeid yet. The result is that numa_node will always be 0.
pcibios_scan_root and pci_scan_root could take sysdata. So we need to get
mp_bus_to_node mapping before these two are called, and thus
get_mp_bus_to_node could get correct node for sysdata in root bus.
In scanning of the root bus, all child busses will take parent bus sysdata.
So all pci_device->dev.numa_node will be assigned correctly and automatically.
Later we could use dev_to_node(&pci_dev->dev) to get numa_node, and we
could also could make other bus specific device get the correct numa_node
too.
This is an updated version of pci_sysdata and Jeff's pci_domain patch.
[ mingo@elte.hu: build fix ]
Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2008-02-19 19:20:09 +08:00
|
|
|
sd->node = node;
|
2008-04-16 05:34:49 +08:00
|
|
|
/*
|
|
|
|
* Maybe the desired pci bus has been already scanned. In such case
|
|
|
|
* it is unnecessary to scan the pci bus with the given domain,busnum.
|
|
|
|
*/
|
|
|
|
bus = pci_find_bus(domain, busnum);
|
|
|
|
if (bus) {
|
|
|
|
/*
|
|
|
|
* If the desired bus exits, the content of bus->sysdata will
|
|
|
|
* be replaced by sd.
|
|
|
|
*/
|
|
|
|
memcpy(bus->sysdata, sd, sizeof(*sd));
|
|
|
|
kfree(sd);
|
2009-06-25 11:00:12 +08:00
|
|
|
} else {
|
|
|
|
bus = pci_create_bus(NULL, busnum, &pci_root_ops, sd);
|
|
|
|
if (bus) {
|
2009-11-05 01:39:13 +08:00
|
|
|
get_current_resources(device, busnum, domain, bus);
|
2009-06-25 11:00:12 +08:00
|
|
|
bus->subordinate = pci_scan_child_bus(bus);
|
|
|
|
}
|
|
|
|
}
|
2007-07-22 05:23:39 +08:00
|
|
|
|
PCI: Set PCI-E Max Payload Size on fabric
On a given PCI-E fabric, each device, bridge, and root port can have a
different PCI-E maximum payload size. There is a sizable performance
boost for having the largest possible maximum payload size on each PCI-E
device. However, if improperly configured, fatal bus errors can occur.
Thus, it is important to ensure that PCI-E payloads sends by a device
are never larger than the MPS setting of all devices on the way to the
destination.
This can be achieved two ways:
- A conservative approach is to use the smallest common denominator of
the entire tree below a root complex for every device on that fabric.
This means for example that having a 128 bytes MPS USB controller on one
leg of a switch will dramatically reduce performances of a video card or
10GE adapter on another leg of that same switch.
It also means that any hierarchy supporting hotplug slots (including
expresscard or thunderbolt I suppose, dbl check that) will have to be
entirely clamped to 128 bytes since we cannot predict what will be
plugged into those slots, and we cannot change the MPS on a "live"
system.
- A more optimal way is possible, if it falls within a couple of
constraints:
* The top-level host bridge will never generate packets larger than the
smallest TLP (or if it can be controlled independently from its MPS at
least)
* The device will never generate packets larger than MPS (which can be
configured via MRRS)
* No support of direct PCI-E <-> PCI-E transfers between devices without
some additional code to specifically deal with that case
Then we can use an approach that basically ignores downstream requests
and focuses exclusively on upstream requests. In that case, all we need
to care about is that a device MPS is no larger than its parent MPS,
which allows us to keep all switches/bridges to the max MPS supported by
their parent and eventually the PHB.
In this case, your USB controller would no longer "starve" your 10GE
Ethernet and your hotplug slots won't affect your global MPS.
Additionally, the hotplugged devices themselves can be configured to a
larger MPS up to the value configured in the hotplug bridge.
To choose between the two available options, two PCI kernel boot args
have been added to the PCI calls. "pcie_bus_safe" will provide the
former behavior, while "pcie_bus_perf" will perform the latter behavior.
By default, the latter behavior is used.
NOTE: due to the location of the enablement, each arch will need to add
calls to this function. This patch only enables x86.
This patch includes a number of changes recommended by Benjamin
Herrenschmidt.
Tested-by: Jordan_Hargrave@dell.com
Signed-off-by: Jon Mason <mason@myri.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2011-07-21 04:20:54 +08:00
|
|
|
/* 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;
|
2011-09-09 05:41:17 +08:00
|
|
|
list_for_each_entry(child, &bus->children, node) {
|
|
|
|
struct pci_dev *self = child->self;
|
|
|
|
if (!self)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
pcie_bus_configure_settings(child, self->pcie_mpss);
|
|
|
|
}
|
PCI: Set PCI-E Max Payload Size on fabric
On a given PCI-E fabric, each device, bridge, and root port can have a
different PCI-E maximum payload size. There is a sizable performance
boost for having the largest possible maximum payload size on each PCI-E
device. However, if improperly configured, fatal bus errors can occur.
Thus, it is important to ensure that PCI-E payloads sends by a device
are never larger than the MPS setting of all devices on the way to the
destination.
This can be achieved two ways:
- A conservative approach is to use the smallest common denominator of
the entire tree below a root complex for every device on that fabric.
This means for example that having a 128 bytes MPS USB controller on one
leg of a switch will dramatically reduce performances of a video card or
10GE adapter on another leg of that same switch.
It also means that any hierarchy supporting hotplug slots (including
expresscard or thunderbolt I suppose, dbl check that) will have to be
entirely clamped to 128 bytes since we cannot predict what will be
plugged into those slots, and we cannot change the MPS on a "live"
system.
- A more optimal way is possible, if it falls within a couple of
constraints:
* The top-level host bridge will never generate packets larger than the
smallest TLP (or if it can be controlled independently from its MPS at
least)
* The device will never generate packets larger than MPS (which can be
configured via MRRS)
* No support of direct PCI-E <-> PCI-E transfers between devices without
some additional code to specifically deal with that case
Then we can use an approach that basically ignores downstream requests
and focuses exclusively on upstream requests. In that case, all we need
to care about is that a device MPS is no larger than its parent MPS,
which allows us to keep all switches/bridges to the max MPS supported by
their parent and eventually the PHB.
In this case, your USB controller would no longer "starve" your 10GE
Ethernet and your hotplug slots won't affect your global MPS.
Additionally, the hotplugged devices themselves can be configured to a
larger MPS up to the value configured in the hotplug bridge.
To choose between the two available options, two PCI kernel boot args
have been added to the PCI calls. "pcie_bus_safe" will provide the
former behavior, while "pcie_bus_perf" will perform the latter behavior.
By default, the latter behavior is used.
NOTE: due to the location of the enablement, each arch will need to add
calls to this function. This patch only enables x86.
This patch includes a number of changes recommended by Benjamin
Herrenschmidt.
Tested-by: Jordan_Hargrave@dell.com
Signed-off-by: Jon Mason <mason@myri.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2011-07-21 04:20:54 +08:00
|
|
|
}
|
|
|
|
|
2007-07-22 05:23:39 +08:00
|
|
|
if (!bus)
|
|
|
|
kfree(sd);
|
|
|
|
|
2008-04-19 16:30:16 +08:00
|
|
|
if (bus && node != -1) {
|
2005-09-13 00:49:24 +08:00
|
|
|
#ifdef CONFIG_ACPI_NUMA
|
2008-04-19 16:30:16 +08:00
|
|
|
if (pxm >= 0)
|
2008-12-19 07:34:51 +08:00
|
|
|
dev_printk(KERN_DEBUG, &bus->dev,
|
|
|
|
"on NUMA node %d (pxm %d)\n", node, pxm);
|
2008-04-19 16:30:16 +08:00
|
|
|
#else
|
2008-12-19 07:34:51 +08:00
|
|
|
dev_printk(KERN_DEBUG, &bus->dev, "on NUMA node %d\n", node);
|
2005-09-13 00:49:24 +08:00
|
|
|
#endif
|
2008-04-19 16:30:16 +08:00
|
|
|
}
|
2007-10-04 06:56:51 +08:00
|
|
|
|
2005-09-13 00:49:24 +08:00
|
|
|
return bus;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2008-07-03 04:50:29 +08:00
|
|
|
int __init pci_acpi_init(void)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
struct pci_dev *dev = NULL;
|
|
|
|
|
|
|
|
if (acpi_noirq)
|
2009-08-29 22:24:51 +08:00
|
|
|
return -ENODEV;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
printk(KERN_INFO "PCI: Using ACPI for IRQ routing\n");
|
|
|
|
acpi_irq_penalty_init();
|
|
|
|
pcibios_enable_irq = acpi_pci_irq_enable;
|
2005-07-28 11:02:00 +08:00
|
|
|
pcibios_disable_irq = acpi_pci_irq_disable;
|
2009-08-29 23:47:33 +08:00
|
|
|
x86_init.pci.init_irq = x86_init_noop;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
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().
|
|
|
|
*/
|
|
|
|
printk(KERN_INFO "PCI: Routing PCI interrupts for all devices because \"pci=routeirq\" specified\n");
|
2005-11-07 15:39:36 +08:00
|
|
|
for_each_pci_dev(dev)
|
2005-04-17 06:20:36 +08:00
|
|
|
acpi_pci_irq_enable(dev);
|
2008-02-19 00:44:13 +08:00
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|