linux/arch/x86/pci/common.c

576 lines
13 KiB
C
Raw Normal View History

/*
* Low-Level PCI Support for PC
*
* (c) 1999--2000 Martin Mares <mj@ucw.cz>
*/
#include <linux/sched.h>
#include <linux/pci.h>
#include <linux/ioport.h>
#include <linux/init.h>
[PATCH] PCI: PCI/Cardbus cards hidden, needs pci=assign-busses to fix "In some cases, especially on modern laptops with a lot of PCI and cardbus bridges, we're unable to assign correct secondary/subordinate bus numbers to all cardbus bridges due to BIOS limitations unless we are using "pci=assign-busses" boot option." -- Ivan Kokshaysky (from a patch comment) Without it, Cardbus cards inserted are never seen by PCI because the parent PCI-PCI Bridge of the Cardbus bridge will not pass and translate Type 1 PCI configuration cycles correctly and the system will fail to find and initialise the PCI devices in the system. Reference: PCI-PCI Bridges: PCI Configuration Cycles and PCI Bus Numbering: http://www.science.unitn.it/~fiorella/guidelinux/tlk/node72.html The reason for this is that: ``All PCI busses located behind a PCI-PCI bridge must reside between the secondary bus number and the subordinate bus number (inclusive).'' "pci=assign-busses" makes pcibios_assign_all_busses return 1 and this turns on PCI renumbering during PCI probing. Alan suggested to use DMI automatically set assign-busses on problem systems. The only question for me was where to put it. I put it directly before scanning PCI bus into pcibios_scan_root() because it's called from legacy, acpi and numa and so it can be one place for all systems and configurations which may need it. AMD64 Laptops are also affected and fixed by assign-busses, and the code is also incuded from arch/x86_64/pci/ that place will also work for x86_64 kernels, I only ifdef'-ed the x86-only Laptop in this example. Affected and known or assumed to be fixed with it are (found by googling): * ASUS Z71V and L3s * Samsung X20 * Compaq R3140us and all Compaq R3000 series laptops with TI1620 Controller, also Compaq R4000 series (from a kernel.org bugreport) * HP zv5000z (AMD64 3700+, known that fixup_parent_subordinate_busnr fixes it) * HP zv5200z * IBM ThinkPad 240 * An IBM ThinkPad (1.8 GHz Pentium M) debugged by Pavel Machek gives the correspondig message which detects the possible problem. * MSI S260 / Medion SIM 2100 MD 95600 The patch also expands the "try pci=assign-busses" warning so testers will help us to update the DMI table. Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Cc: Dominik Brodowski <linux@dominikbrodowski.net> Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-02-18 17:36:55 +08:00
#include <linux/dmi.h>
#include <asm/acpi.h>
#include <asm/segment.h>
#include <asm/io.h>
#include <asm/smp.h>
#include "pci.h"
unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2 |
PCI_PROBE_MMCONF;
unsigned int pci_early_dump_regs;
static int pci_bf_sort;
int pci_routeirq;
int pcibios_last_bus = -1;
unsigned long pirq_table_addr;
struct pci_bus *pci_root_bus;
struct pci_raw_ops *raw_pci_ops;
struct pci_raw_ops *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,
};
/*
* legacy, numa, and acpi all want to call pcibios_scan_root
* from their initcalls. This flag prevents that.
*/
int pcibios_scanned;
/*
* This interrupt-safe spinlock protects all accesses to PCI
* configuration space.
*/
DEFINE_SPINLOCK(pci_config_lock);
static int __devinit 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 struct dmi_system_id can_skip_pciprobe_dmi_table[] __devinitdata = {
/*
* 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);
}
PCI: boot parameter to avoid expansion ROM memory allocation Contention for scarce PCI memory resources has been growing due to an increasing number of PCI slots in large multi-node systems. The kernel currently attempts by default to allocate memory for all PCI expansion ROMs so there has also been an increasing number of PCI memory allocation failures seen on these systems. This occurs because the BIOS either (1) provides insufficient PCI memory resource for all the expansion ROMs or (2) provides adequate PCI memory resource for expansion ROMs but provides the space in kernel unexpected BIOS assigned P2P non-prefetch windows. The resulting PCI memory allocation failures may be benign when related to memory requests for expansion ROMs themselves but in some cases they can occur when attempting to allocate space for more critical BARs. This can happen when a successful expansion ROM allocation request consumes memory resource that was intended for a non-ROM BAR. We have seen this happen during PCI hotplug of an adapter that contains a P2P bridge where successful memory allocation for an expansion ROM BAR on device behind the bridge consumed memory that was intended for a non-ROM BAR on the P2P bridge. In all cases the allocation failure messages can be very confusing for users. This patch provides a new 'pci=norom' kernel boot parameter that can be used to disable the default PCI expansion ROM memory resource allocation. This provides a way to avoid the above described issues on systems that do not contain PCI devices for which drivers or user-level applications depend on the default PCI expansion ROM memory resource allocation behavior. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2008-05-13 04:57:46 +08:00
static void __devinit pcibios_fixup_device_resources(struct pci_dev *dev)
{
struct resource *rom_r = &dev->resource[PCI_ROM_RESOURCE];
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 __devinit pcibios_fixup_bus(struct pci_bus *b)
{
PCI: boot parameter to avoid expansion ROM memory allocation Contention for scarce PCI memory resources has been growing due to an increasing number of PCI slots in large multi-node systems. The kernel currently attempts by default to allocate memory for all PCI expansion ROMs so there has also been an increasing number of PCI memory allocation failures seen on these systems. This occurs because the BIOS either (1) provides insufficient PCI memory resource for all the expansion ROMs or (2) provides adequate PCI memory resource for expansion ROMs but provides the space in kernel unexpected BIOS assigned P2P non-prefetch windows. The resulting PCI memory allocation failures may be benign when related to memory requests for expansion ROMs themselves but in some cases they can occur when attempting to allocate space for more critical BARs. This can happen when a successful expansion ROM allocation request consumes memory resource that was intended for a non-ROM BAR. We have seen this happen during PCI hotplug of an adapter that contains a P2P bridge where successful memory allocation for an expansion ROM BAR on device behind the bridge consumed memory that was intended for a non-ROM BAR on the P2P bridge. In all cases the allocation failure messages can be very confusing for users. This patch provides a new 'pci=norom' kernel boot parameter that can be used to disable the default PCI expansion ROM memory resource allocation. This provides a way to avoid the above described issues on systems that do not contain PCI devices for which drivers or user-level applications depend on the default PCI expansion ROM memory resource allocation behavior. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2008-05-13 04:57:46 +08:00
struct pci_dev *dev;
pci_read_bridge_bases(b);
PCI: boot parameter to avoid expansion ROM memory allocation Contention for scarce PCI memory resources has been growing due to an increasing number of PCI slots in large multi-node systems. The kernel currently attempts by default to allocate memory for all PCI expansion ROMs so there has also been an increasing number of PCI memory allocation failures seen on these systems. This occurs because the BIOS either (1) provides insufficient PCI memory resource for all the expansion ROMs or (2) provides adequate PCI memory resource for expansion ROMs but provides the space in kernel unexpected BIOS assigned P2P non-prefetch windows. The resulting PCI memory allocation failures may be benign when related to memory requests for expansion ROMs themselves but in some cases they can occur when attempting to allocate space for more critical BARs. This can happen when a successful expansion ROM allocation request consumes memory resource that was intended for a non-ROM BAR. We have seen this happen during PCI hotplug of an adapter that contains a P2P bridge where successful memory allocation for an expansion ROM BAR on device behind the bridge consumed memory that was intended for a non-ROM BAR on the P2P bridge. In all cases the allocation failure messages can be very confusing for users. This patch provides a new 'pci=norom' kernel boot parameter that can be used to disable the default PCI expansion ROM memory resource allocation. This provides a way to avoid the above described issues on systems that do not contain PCI devices for which drivers or user-level applications depend on the default PCI expansion ROM memory resource allocation behavior. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2008-05-13 04:57:46 +08:00
list_for_each_entry(dev, &b->devices, bus_list)
pcibios_fixup_device_resources(dev);
}
PCI: optionally sort device lists breadth-first Problem: New Dell PowerEdge servers have 2 embedded ethernet ports, which are labeled NIC1 and NIC2 on the chassis, in the BIOS setup screens, and in the printed documentation. Assuming no other add-in ethernet ports in the system, Linux 2.4 kernels name these eth0 and eth1 respectively. Many people have come to expect this naming. Linux 2.6 kernels name these eth1 and eth0 respectively (backwards from expectations). I also have reports that various Sun and HP servers have similar behavior. Root cause: Linux 2.4 kernels walk the pci_devices list, which happens to be sorted in breadth-first order (or pcbios_find_device order on i386, which most often is breadth-first also). 2.6 kernels have both the pci_devices list and the pci_bus_type.klist_devices list, the latter is what is walked at driver load time to match the pci_id tables; this klist happens to be in depth-first order. On systems where, for physical routing reasons, NIC1 appears on a lower bus number than NIC2, but NIC2's bridge is discovered first in the depth-first ordering, NIC2 will be discovered before NIC1. If the list were sorted breadth-first, NIC1 would be discovered before NIC2. A PowerEdge 1955 system has the following topology which easily exhibits the difference between depth-first and breadth-first device lists. -[0000:00]-+-00.0 Intel Corporation 5000P Chipset Memory Controller Hub +-02.0-[0000:03-08]--+-00.0-[0000:04-07]--+-00.0-[0000:05-06]----00.0-[0000:06]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC2, 2.4 kernel name eth1, 2.6 kernel name eth0) +-1c.0-[0000:01-02]----00.0-[0000:02]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC1, 2.4 kernel name eth0, 2.6 kernel name eth1) Other factors, such as device driver load order and the presence of PCI slots at various points in the bus hierarchy further complicate this problem; I'm not trying to solve those here, just restore the device order, and thus basic behavior, that 2.4 kernels had. Solution: The solution can come in multiple steps. Suggested fix #1: kernel Patch below optionally sorts the two device lists into breadth-first ordering to maintain compatibility with 2.4 kernels. It adds two new command line options: pci=bfsort pci=nobfsort to force the sort order, or not, as you wish. It also adds DMI checks for the specific Dell systems which exhibit "backwards" ordering, to make them "right". Suggested fix #2: udev rules from userland Many people also have the expectation that embedded NICs are always discovered before add-in NICs (which this patch does not try to do). Using the PCI IRQ Routing Table provided by system BIOS, it's easy to determine which PCI devices are embedded, or if add-in, which PCI slot they're in. I'm working on a tool that would allow udev to name ethernet devices in ascending embedded, slot 1 .. slot N order, subsort by PCI bus/dev/fn breadth-first. It'll be possible to use it independent of udev as well for those distributions that don't use udev in their installers. Suggested fix #3: system board routing rules One can constrain the system board layout to put NIC1 ahead of NIC2 regardless of breadth-first or depth-first discovery order. This adds a significant level of complexity to board routing, and may not be possible in all instances (witness the above systems from several major manufacturers). I don't want to encourage this particular train of thought too far, at the expense of not doing #1 or #2 above. Feedback appreciated. Patch tested on a Dell PowerEdge 1955 blade with 2.6.18. You'll also note I took some liberty and temporarily break the klist abstraction to simplify and speed up the sort algorithm. I think that's both safe and appropriate in this instance. Signed-off-by: Matt Domsch <Matt_Domsch@dell.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-09-30 04:23:23 +08:00
/*
* Only use DMI information to set this if nothing was passed
* on the kernel command line (which was parsed earlier).
*/
static int __devinit set_bf_sort(const struct dmi_system_id *d)
PCI: optionally sort device lists breadth-first Problem: New Dell PowerEdge servers have 2 embedded ethernet ports, which are labeled NIC1 and NIC2 on the chassis, in the BIOS setup screens, and in the printed documentation. Assuming no other add-in ethernet ports in the system, Linux 2.4 kernels name these eth0 and eth1 respectively. Many people have come to expect this naming. Linux 2.6 kernels name these eth1 and eth0 respectively (backwards from expectations). I also have reports that various Sun and HP servers have similar behavior. Root cause: Linux 2.4 kernels walk the pci_devices list, which happens to be sorted in breadth-first order (or pcbios_find_device order on i386, which most often is breadth-first also). 2.6 kernels have both the pci_devices list and the pci_bus_type.klist_devices list, the latter is what is walked at driver load time to match the pci_id tables; this klist happens to be in depth-first order. On systems where, for physical routing reasons, NIC1 appears on a lower bus number than NIC2, but NIC2's bridge is discovered first in the depth-first ordering, NIC2 will be discovered before NIC1. If the list were sorted breadth-first, NIC1 would be discovered before NIC2. A PowerEdge 1955 system has the following topology which easily exhibits the difference between depth-first and breadth-first device lists. -[0000:00]-+-00.0 Intel Corporation 5000P Chipset Memory Controller Hub +-02.0-[0000:03-08]--+-00.0-[0000:04-07]--+-00.0-[0000:05-06]----00.0-[0000:06]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC2, 2.4 kernel name eth1, 2.6 kernel name eth0) +-1c.0-[0000:01-02]----00.0-[0000:02]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC1, 2.4 kernel name eth0, 2.6 kernel name eth1) Other factors, such as device driver load order and the presence of PCI slots at various points in the bus hierarchy further complicate this problem; I'm not trying to solve those here, just restore the device order, and thus basic behavior, that 2.4 kernels had. Solution: The solution can come in multiple steps. Suggested fix #1: kernel Patch below optionally sorts the two device lists into breadth-first ordering to maintain compatibility with 2.4 kernels. It adds two new command line options: pci=bfsort pci=nobfsort to force the sort order, or not, as you wish. It also adds DMI checks for the specific Dell systems which exhibit "backwards" ordering, to make them "right". Suggested fix #2: udev rules from userland Many people also have the expectation that embedded NICs are always discovered before add-in NICs (which this patch does not try to do). Using the PCI IRQ Routing Table provided by system BIOS, it's easy to determine which PCI devices are embedded, or if add-in, which PCI slot they're in. I'm working on a tool that would allow udev to name ethernet devices in ascending embedded, slot 1 .. slot N order, subsort by PCI bus/dev/fn breadth-first. It'll be possible to use it independent of udev as well for those distributions that don't use udev in their installers. Suggested fix #3: system board routing rules One can constrain the system board layout to put NIC1 ahead of NIC2 regardless of breadth-first or depth-first discovery order. This adds a significant level of complexity to board routing, and may not be possible in all instances (witness the above systems from several major manufacturers). I don't want to encourage this particular train of thought too far, at the expense of not doing #1 or #2 above. Feedback appreciated. Patch tested on a Dell PowerEdge 1955 blade with 2.6.18. You'll also note I took some liberty and temporarily break the klist abstraction to simplify and speed up the sort algorithm. I think that's both safe and appropriate in this instance. Signed-off-by: Matt Domsch <Matt_Domsch@dell.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-09-30 04:23:23 +08:00
{
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;
}
[PATCH] PCI: PCI/Cardbus cards hidden, needs pci=assign-busses to fix "In some cases, especially on modern laptops with a lot of PCI and cardbus bridges, we're unable to assign correct secondary/subordinate bus numbers to all cardbus bridges due to BIOS limitations unless we are using "pci=assign-busses" boot option." -- Ivan Kokshaysky (from a patch comment) Without it, Cardbus cards inserted are never seen by PCI because the parent PCI-PCI Bridge of the Cardbus bridge will not pass and translate Type 1 PCI configuration cycles correctly and the system will fail to find and initialise the PCI devices in the system. Reference: PCI-PCI Bridges: PCI Configuration Cycles and PCI Bus Numbering: http://www.science.unitn.it/~fiorella/guidelinux/tlk/node72.html The reason for this is that: ``All PCI busses located behind a PCI-PCI bridge must reside between the secondary bus number and the subordinate bus number (inclusive).'' "pci=assign-busses" makes pcibios_assign_all_busses return 1 and this turns on PCI renumbering during PCI probing. Alan suggested to use DMI automatically set assign-busses on problem systems. The only question for me was where to put it. I put it directly before scanning PCI bus into pcibios_scan_root() because it's called from legacy, acpi and numa and so it can be one place for all systems and configurations which may need it. AMD64 Laptops are also affected and fixed by assign-busses, and the code is also incuded from arch/x86_64/pci/ that place will also work for x86_64 kernels, I only ifdef'-ed the x86-only Laptop in this example. Affected and known or assumed to be fixed with it are (found by googling): * ASUS Z71V and L3s * Samsung X20 * Compaq R3140us and all Compaq R3000 series laptops with TI1620 Controller, also Compaq R4000 series (from a kernel.org bugreport) * HP zv5000z (AMD64 3700+, known that fixup_parent_subordinate_busnr fixes it) * HP zv5200z * IBM ThinkPad 240 * An IBM ThinkPad (1.8 GHz Pentium M) debugged by Pavel Machek gives the correspondig message which detects the possible problem. * MSI S260 / Medion SIM 2100 MD 95600 The patch also expands the "try pci=assign-busses" warning so testers will help us to update the DMI table. Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Cc: Dominik Brodowski <linux@dominikbrodowski.net> Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-02-18 17:36:55 +08:00
/*
* Enable renumbering of PCI bus# ranges to reach all PCI busses (Cardbus)
*/
#ifdef __i386__
static int __devinit assign_all_busses(const struct dmi_system_id *d)
[PATCH] PCI: PCI/Cardbus cards hidden, needs pci=assign-busses to fix "In some cases, especially on modern laptops with a lot of PCI and cardbus bridges, we're unable to assign correct secondary/subordinate bus numbers to all cardbus bridges due to BIOS limitations unless we are using "pci=assign-busses" boot option." -- Ivan Kokshaysky (from a patch comment) Without it, Cardbus cards inserted are never seen by PCI because the parent PCI-PCI Bridge of the Cardbus bridge will not pass and translate Type 1 PCI configuration cycles correctly and the system will fail to find and initialise the PCI devices in the system. Reference: PCI-PCI Bridges: PCI Configuration Cycles and PCI Bus Numbering: http://www.science.unitn.it/~fiorella/guidelinux/tlk/node72.html The reason for this is that: ``All PCI busses located behind a PCI-PCI bridge must reside between the secondary bus number and the subordinate bus number (inclusive).'' "pci=assign-busses" makes pcibios_assign_all_busses return 1 and this turns on PCI renumbering during PCI probing. Alan suggested to use DMI automatically set assign-busses on problem systems. The only question for me was where to put it. I put it directly before scanning PCI bus into pcibios_scan_root() because it's called from legacy, acpi and numa and so it can be one place for all systems and configurations which may need it. AMD64 Laptops are also affected and fixed by assign-busses, and the code is also incuded from arch/x86_64/pci/ that place will also work for x86_64 kernels, I only ifdef'-ed the x86-only Laptop in this example. Affected and known or assumed to be fixed with it are (found by googling): * ASUS Z71V and L3s * Samsung X20 * Compaq R3140us and all Compaq R3000 series laptops with TI1620 Controller, also Compaq R4000 series (from a kernel.org bugreport) * HP zv5000z (AMD64 3700+, known that fixup_parent_subordinate_busnr fixes it) * HP zv5200z * IBM ThinkPad 240 * An IBM ThinkPad (1.8 GHz Pentium M) debugged by Pavel Machek gives the correspondig message which detects the possible problem. * MSI S260 / Medion SIM 2100 MD 95600 The patch also expands the "try pci=assign-busses" warning so testers will help us to update the DMI table. Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Cc: Dominik Brodowski <linux@dominikbrodowski.net> Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-02-18 17:36:55 +08:00
{
pci_probe |= PCI_ASSIGN_ALL_BUSSES;
printk(KERN_INFO "%s detected: enabling PCI bus# renumbering"
" (pci=assign-busses)\n", d->ident);
return 0;
}
#endif
PCI: optionally sort device lists breadth-first Problem: New Dell PowerEdge servers have 2 embedded ethernet ports, which are labeled NIC1 and NIC2 on the chassis, in the BIOS setup screens, and in the printed documentation. Assuming no other add-in ethernet ports in the system, Linux 2.4 kernels name these eth0 and eth1 respectively. Many people have come to expect this naming. Linux 2.6 kernels name these eth1 and eth0 respectively (backwards from expectations). I also have reports that various Sun and HP servers have similar behavior. Root cause: Linux 2.4 kernels walk the pci_devices list, which happens to be sorted in breadth-first order (or pcbios_find_device order on i386, which most often is breadth-first also). 2.6 kernels have both the pci_devices list and the pci_bus_type.klist_devices list, the latter is what is walked at driver load time to match the pci_id tables; this klist happens to be in depth-first order. On systems where, for physical routing reasons, NIC1 appears on a lower bus number than NIC2, but NIC2's bridge is discovered first in the depth-first ordering, NIC2 will be discovered before NIC1. If the list were sorted breadth-first, NIC1 would be discovered before NIC2. A PowerEdge 1955 system has the following topology which easily exhibits the difference between depth-first and breadth-first device lists. -[0000:00]-+-00.0 Intel Corporation 5000P Chipset Memory Controller Hub +-02.0-[0000:03-08]--+-00.0-[0000:04-07]--+-00.0-[0000:05-06]----00.0-[0000:06]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC2, 2.4 kernel name eth1, 2.6 kernel name eth0) +-1c.0-[0000:01-02]----00.0-[0000:02]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC1, 2.4 kernel name eth0, 2.6 kernel name eth1) Other factors, such as device driver load order and the presence of PCI slots at various points in the bus hierarchy further complicate this problem; I'm not trying to solve those here, just restore the device order, and thus basic behavior, that 2.4 kernels had. Solution: The solution can come in multiple steps. Suggested fix #1: kernel Patch below optionally sorts the two device lists into breadth-first ordering to maintain compatibility with 2.4 kernels. It adds two new command line options: pci=bfsort pci=nobfsort to force the sort order, or not, as you wish. It also adds DMI checks for the specific Dell systems which exhibit "backwards" ordering, to make them "right". Suggested fix #2: udev rules from userland Many people also have the expectation that embedded NICs are always discovered before add-in NICs (which this patch does not try to do). Using the PCI IRQ Routing Table provided by system BIOS, it's easy to determine which PCI devices are embedded, or if add-in, which PCI slot they're in. I'm working on a tool that would allow udev to name ethernet devices in ascending embedded, slot 1 .. slot N order, subsort by PCI bus/dev/fn breadth-first. It'll be possible to use it independent of udev as well for those distributions that don't use udev in their installers. Suggested fix #3: system board routing rules One can constrain the system board layout to put NIC1 ahead of NIC2 regardless of breadth-first or depth-first discovery order. This adds a significant level of complexity to board routing, and may not be possible in all instances (witness the above systems from several major manufacturers). I don't want to encourage this particular train of thought too far, at the expense of not doing #1 or #2 above. Feedback appreciated. Patch tested on a Dell PowerEdge 1955 blade with 2.6.18. You'll also note I took some liberty and temporarily break the klist abstraction to simplify and speed up the sort algorithm. I think that's both safe and appropriate in this instance. Signed-off-by: Matt Domsch <Matt_Domsch@dell.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-09-30 04:23:23 +08:00
static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = {
#ifdef __i386__
[PATCH] PCI: PCI/Cardbus cards hidden, needs pci=assign-busses to fix "In some cases, especially on modern laptops with a lot of PCI and cardbus bridges, we're unable to assign correct secondary/subordinate bus numbers to all cardbus bridges due to BIOS limitations unless we are using "pci=assign-busses" boot option." -- Ivan Kokshaysky (from a patch comment) Without it, Cardbus cards inserted are never seen by PCI because the parent PCI-PCI Bridge of the Cardbus bridge will not pass and translate Type 1 PCI configuration cycles correctly and the system will fail to find and initialise the PCI devices in the system. Reference: PCI-PCI Bridges: PCI Configuration Cycles and PCI Bus Numbering: http://www.science.unitn.it/~fiorella/guidelinux/tlk/node72.html The reason for this is that: ``All PCI busses located behind a PCI-PCI bridge must reside between the secondary bus number and the subordinate bus number (inclusive).'' "pci=assign-busses" makes pcibios_assign_all_busses return 1 and this turns on PCI renumbering during PCI probing. Alan suggested to use DMI automatically set assign-busses on problem systems. The only question for me was where to put it. I put it directly before scanning PCI bus into pcibios_scan_root() because it's called from legacy, acpi and numa and so it can be one place for all systems and configurations which may need it. AMD64 Laptops are also affected and fixed by assign-busses, and the code is also incuded from arch/x86_64/pci/ that place will also work for x86_64 kernels, I only ifdef'-ed the x86-only Laptop in this example. Affected and known or assumed to be fixed with it are (found by googling): * ASUS Z71V and L3s * Samsung X20 * Compaq R3140us and all Compaq R3000 series laptops with TI1620 Controller, also Compaq R4000 series (from a kernel.org bugreport) * HP zv5000z (AMD64 3700+, known that fixup_parent_subordinate_busnr fixes it) * HP zv5200z * IBM ThinkPad 240 * An IBM ThinkPad (1.8 GHz Pentium M) debugged by Pavel Machek gives the correspondig message which detects the possible problem. * MSI S260 / Medion SIM 2100 MD 95600 The patch also expands the "try pci=assign-busses" warning so testers will help us to update the DMI table. Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Cc: Dominik Brodowski <linux@dominikbrodowski.net> Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-02-18 17:36:55 +08:00
/*
* 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__ */
PCI: optionally sort device lists breadth-first Problem: New Dell PowerEdge servers have 2 embedded ethernet ports, which are labeled NIC1 and NIC2 on the chassis, in the BIOS setup screens, and in the printed documentation. Assuming no other add-in ethernet ports in the system, Linux 2.4 kernels name these eth0 and eth1 respectively. Many people have come to expect this naming. Linux 2.6 kernels name these eth1 and eth0 respectively (backwards from expectations). I also have reports that various Sun and HP servers have similar behavior. Root cause: Linux 2.4 kernels walk the pci_devices list, which happens to be sorted in breadth-first order (or pcbios_find_device order on i386, which most often is breadth-first also). 2.6 kernels have both the pci_devices list and the pci_bus_type.klist_devices list, the latter is what is walked at driver load time to match the pci_id tables; this klist happens to be in depth-first order. On systems where, for physical routing reasons, NIC1 appears on a lower bus number than NIC2, but NIC2's bridge is discovered first in the depth-first ordering, NIC2 will be discovered before NIC1. If the list were sorted breadth-first, NIC1 would be discovered before NIC2. A PowerEdge 1955 system has the following topology which easily exhibits the difference between depth-first and breadth-first device lists. -[0000:00]-+-00.0 Intel Corporation 5000P Chipset Memory Controller Hub +-02.0-[0000:03-08]--+-00.0-[0000:04-07]--+-00.0-[0000:05-06]----00.0-[0000:06]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC2, 2.4 kernel name eth1, 2.6 kernel name eth0) +-1c.0-[0000:01-02]----00.0-[0000:02]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC1, 2.4 kernel name eth0, 2.6 kernel name eth1) Other factors, such as device driver load order and the presence of PCI slots at various points in the bus hierarchy further complicate this problem; I'm not trying to solve those here, just restore the device order, and thus basic behavior, that 2.4 kernels had. Solution: The solution can come in multiple steps. Suggested fix #1: kernel Patch below optionally sorts the two device lists into breadth-first ordering to maintain compatibility with 2.4 kernels. It adds two new command line options: pci=bfsort pci=nobfsort to force the sort order, or not, as you wish. It also adds DMI checks for the specific Dell systems which exhibit "backwards" ordering, to make them "right". Suggested fix #2: udev rules from userland Many people also have the expectation that embedded NICs are always discovered before add-in NICs (which this patch does not try to do). Using the PCI IRQ Routing Table provided by system BIOS, it's easy to determine which PCI devices are embedded, or if add-in, which PCI slot they're in. I'm working on a tool that would allow udev to name ethernet devices in ascending embedded, slot 1 .. slot N order, subsort by PCI bus/dev/fn breadth-first. It'll be possible to use it independent of udev as well for those distributions that don't use udev in their installers. Suggested fix #3: system board routing rules One can constrain the system board layout to put NIC1 ahead of NIC2 regardless of breadth-first or depth-first discovery order. This adds a significant level of complexity to board routing, and may not be possible in all instances (witness the above systems from several major manufacturers). I don't want to encourage this particular train of thought too far, at the expense of not doing #1 or #2 above. Feedback appreciated. Patch tested on a Dell PowerEdge 1955 blade with 2.6.18. You'll also note I took some liberty and temporarily break the klist abstraction to simplify and speed up the sort algorithm. I think that's both safe and appropriate in this instance. Signed-off-by: Matt Domsch <Matt_Domsch@dell.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-09-30 04:23:23 +08:00
{
.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 = 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"),
},
},
[PATCH] PCI: PCI/Cardbus cards hidden, needs pci=assign-busses to fix "In some cases, especially on modern laptops with a lot of PCI and cardbus bridges, we're unable to assign correct secondary/subordinate bus numbers to all cardbus bridges due to BIOS limitations unless we are using "pci=assign-busses" boot option." -- Ivan Kokshaysky (from a patch comment) Without it, Cardbus cards inserted are never seen by PCI because the parent PCI-PCI Bridge of the Cardbus bridge will not pass and translate Type 1 PCI configuration cycles correctly and the system will fail to find and initialise the PCI devices in the system. Reference: PCI-PCI Bridges: PCI Configuration Cycles and PCI Bus Numbering: http://www.science.unitn.it/~fiorella/guidelinux/tlk/node72.html The reason for this is that: ``All PCI busses located behind a PCI-PCI bridge must reside between the secondary bus number and the subordinate bus number (inclusive).'' "pci=assign-busses" makes pcibios_assign_all_busses return 1 and this turns on PCI renumbering during PCI probing. Alan suggested to use DMI automatically set assign-busses on problem systems. The only question for me was where to put it. I put it directly before scanning PCI bus into pcibios_scan_root() because it's called from legacy, acpi and numa and so it can be one place for all systems and configurations which may need it. AMD64 Laptops are also affected and fixed by assign-busses, and the code is also incuded from arch/x86_64/pci/ that place will also work for x86_64 kernels, I only ifdef'-ed the x86-only Laptop in this example. Affected and known or assumed to be fixed with it are (found by googling): * ASUS Z71V and L3s * Samsung X20 * Compaq R3140us and all Compaq R3000 series laptops with TI1620 Controller, also Compaq R4000 series (from a kernel.org bugreport) * HP zv5000z (AMD64 3700+, known that fixup_parent_subordinate_busnr fixes it) * HP zv5200z * IBM ThinkPad 240 * An IBM ThinkPad (1.8 GHz Pentium M) debugged by Pavel Machek gives the correspondig message which detects the possible problem. * MSI S260 / Medion SIM 2100 MD 95600 The patch also expands the "try pci=assign-busses" warning so testers will help us to update the DMI table. Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Cc: Dominik Brodowski <linux@dominikbrodowski.net> Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-02-18 17:36:55 +08:00
{}
};
void __init dmi_check_pciprobe(void)
{
dmi_check_system(pciprobe_dmi_table);
}
struct pci_bus * __devinit pcibios_scan_root(int busnum)
{
struct pci_bus *bus = NULL;
struct pci_sysdata *sd;
while ((bus = pci_find_next_bus(bus)) != NULL) {
if (bus->number == busnum) {
/* Already scanned */
return bus;
}
}
/* 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) {
printk(KERN_ERR "PCI: OOM, not probing PCI bus %02x\n", busnum);
return NULL;
}
sd->node = get_mp_bus_to_node(busnum);
printk(KERN_DEBUG "PCI: Probing PCI hardware (bus %02x)\n", busnum);
bus = pci_scan_bus_parented(NULL, busnum, &pci_root_ops, sd);
if (!bus)
kfree(sd);
return bus;
}
extern u8 pci_cache_line_size;
int __init pcibios_init(void)
{
struct cpuinfo_x86 *c = &boot_cpu_data;
if (!raw_pci_ops) {
printk(KERN_WARNING "PCI: System does not support PCI\n");
return 0;
}
/*
* Assume PCI cacheline size of 32 bytes for all x86s except K7/K8
* and P4. It's also good for 386/486s (which actually have 16)
* as quite a few PCI devices do not support smaller values.
*/
pci_cache_line_size = 32 >> 2;
if (c->x86 >= 6 && c->x86_vendor == X86_VENDOR_AMD)
pci_cache_line_size = 64 >> 2; /* K7 & K8 */
else if (c->x86 > 6 && c->x86_vendor == X86_VENDOR_INTEL)
pci_cache_line_size = 128 >> 2; /* P4 */
pcibios_resource_survey();
PCI: optionally sort device lists breadth-first Problem: New Dell PowerEdge servers have 2 embedded ethernet ports, which are labeled NIC1 and NIC2 on the chassis, in the BIOS setup screens, and in the printed documentation. Assuming no other add-in ethernet ports in the system, Linux 2.4 kernels name these eth0 and eth1 respectively. Many people have come to expect this naming. Linux 2.6 kernels name these eth1 and eth0 respectively (backwards from expectations). I also have reports that various Sun and HP servers have similar behavior. Root cause: Linux 2.4 kernels walk the pci_devices list, which happens to be sorted in breadth-first order (or pcbios_find_device order on i386, which most often is breadth-first also). 2.6 kernels have both the pci_devices list and the pci_bus_type.klist_devices list, the latter is what is walked at driver load time to match the pci_id tables; this klist happens to be in depth-first order. On systems where, for physical routing reasons, NIC1 appears on a lower bus number than NIC2, but NIC2's bridge is discovered first in the depth-first ordering, NIC2 will be discovered before NIC1. If the list were sorted breadth-first, NIC1 would be discovered before NIC2. A PowerEdge 1955 system has the following topology which easily exhibits the difference between depth-first and breadth-first device lists. -[0000:00]-+-00.0 Intel Corporation 5000P Chipset Memory Controller Hub +-02.0-[0000:03-08]--+-00.0-[0000:04-07]--+-00.0-[0000:05-06]----00.0-[0000:06]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC2, 2.4 kernel name eth1, 2.6 kernel name eth0) +-1c.0-[0000:01-02]----00.0-[0000:02]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC1, 2.4 kernel name eth0, 2.6 kernel name eth1) Other factors, such as device driver load order and the presence of PCI slots at various points in the bus hierarchy further complicate this problem; I'm not trying to solve those here, just restore the device order, and thus basic behavior, that 2.4 kernels had. Solution: The solution can come in multiple steps. Suggested fix #1: kernel Patch below optionally sorts the two device lists into breadth-first ordering to maintain compatibility with 2.4 kernels. It adds two new command line options: pci=bfsort pci=nobfsort to force the sort order, or not, as you wish. It also adds DMI checks for the specific Dell systems which exhibit "backwards" ordering, to make them "right". Suggested fix #2: udev rules from userland Many people also have the expectation that embedded NICs are always discovered before add-in NICs (which this patch does not try to do). Using the PCI IRQ Routing Table provided by system BIOS, it's easy to determine which PCI devices are embedded, or if add-in, which PCI slot they're in. I'm working on a tool that would allow udev to name ethernet devices in ascending embedded, slot 1 .. slot N order, subsort by PCI bus/dev/fn breadth-first. It'll be possible to use it independent of udev as well for those distributions that don't use udev in their installers. Suggested fix #3: system board routing rules One can constrain the system board layout to put NIC1 ahead of NIC2 regardless of breadth-first or depth-first discovery order. This adds a significant level of complexity to board routing, and may not be possible in all instances (witness the above systems from several major manufacturers). I don't want to encourage this particular train of thought too far, at the expense of not doing #1 or #2 above. Feedback appreciated. Patch tested on a Dell PowerEdge 1955 blade with 2.6.18. You'll also note I took some liberty and temporarily break the klist abstraction to simplify and speed up the sort algorithm. I think that's both safe and appropriate in this instance. Signed-off-by: Matt Domsch <Matt_Domsch@dell.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-09-30 04:23:23 +08:00
if (pci_bf_sort >= pci_force_bf)
pci_sort_breadthfirst();
return 0;
}
char * __devinit pcibios_setup(char *str)
{
if (!strcmp(str, "off")) {
pci_probe = 0;
return NULL;
PCI: optionally sort device lists breadth-first Problem: New Dell PowerEdge servers have 2 embedded ethernet ports, which are labeled NIC1 and NIC2 on the chassis, in the BIOS setup screens, and in the printed documentation. Assuming no other add-in ethernet ports in the system, Linux 2.4 kernels name these eth0 and eth1 respectively. Many people have come to expect this naming. Linux 2.6 kernels name these eth1 and eth0 respectively (backwards from expectations). I also have reports that various Sun and HP servers have similar behavior. Root cause: Linux 2.4 kernels walk the pci_devices list, which happens to be sorted in breadth-first order (or pcbios_find_device order on i386, which most often is breadth-first also). 2.6 kernels have both the pci_devices list and the pci_bus_type.klist_devices list, the latter is what is walked at driver load time to match the pci_id tables; this klist happens to be in depth-first order. On systems where, for physical routing reasons, NIC1 appears on a lower bus number than NIC2, but NIC2's bridge is discovered first in the depth-first ordering, NIC2 will be discovered before NIC1. If the list were sorted breadth-first, NIC1 would be discovered before NIC2. A PowerEdge 1955 system has the following topology which easily exhibits the difference between depth-first and breadth-first device lists. -[0000:00]-+-00.0 Intel Corporation 5000P Chipset Memory Controller Hub +-02.0-[0000:03-08]--+-00.0-[0000:04-07]--+-00.0-[0000:05-06]----00.0-[0000:06]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC2, 2.4 kernel name eth1, 2.6 kernel name eth0) +-1c.0-[0000:01-02]----00.0-[0000:02]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC1, 2.4 kernel name eth0, 2.6 kernel name eth1) Other factors, such as device driver load order and the presence of PCI slots at various points in the bus hierarchy further complicate this problem; I'm not trying to solve those here, just restore the device order, and thus basic behavior, that 2.4 kernels had. Solution: The solution can come in multiple steps. Suggested fix #1: kernel Patch below optionally sorts the two device lists into breadth-first ordering to maintain compatibility with 2.4 kernels. It adds two new command line options: pci=bfsort pci=nobfsort to force the sort order, or not, as you wish. It also adds DMI checks for the specific Dell systems which exhibit "backwards" ordering, to make them "right". Suggested fix #2: udev rules from userland Many people also have the expectation that embedded NICs are always discovered before add-in NICs (which this patch does not try to do). Using the PCI IRQ Routing Table provided by system BIOS, it's easy to determine which PCI devices are embedded, or if add-in, which PCI slot they're in. I'm working on a tool that would allow udev to name ethernet devices in ascending embedded, slot 1 .. slot N order, subsort by PCI bus/dev/fn breadth-first. It'll be possible to use it independent of udev as well for those distributions that don't use udev in their installers. Suggested fix #3: system board routing rules One can constrain the system board layout to put NIC1 ahead of NIC2 regardless of breadth-first or depth-first discovery order. This adds a significant level of complexity to board routing, and may not be possible in all instances (witness the above systems from several major manufacturers). I don't want to encourage this particular train of thought too far, at the expense of not doing #1 or #2 above. Feedback appreciated. Patch tested on a Dell PowerEdge 1955 blade with 2.6.18. You'll also note I took some liberty and temporarily break the klist abstraction to simplify and speed up the sort algorithm. I think that's both safe and appropriate in this instance. Signed-off-by: Matt Domsch <Matt_Domsch@dell.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-09-30 04:23:23 +08:00
} 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;
}
#ifndef CONFIG_X86_VISWS
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;
}
#endif
else if (!strcmp(str, "rom")) {
pci_probe |= PCI_ASSIGN_ROMS;
return NULL;
PCI: boot parameter to avoid expansion ROM memory allocation Contention for scarce PCI memory resources has been growing due to an increasing number of PCI slots in large multi-node systems. The kernel currently attempts by default to allocate memory for all PCI expansion ROMs so there has also been an increasing number of PCI memory allocation failures seen on these systems. This occurs because the BIOS either (1) provides insufficient PCI memory resource for all the expansion ROMs or (2) provides adequate PCI memory resource for expansion ROMs but provides the space in kernel unexpected BIOS assigned P2P non-prefetch windows. The resulting PCI memory allocation failures may be benign when related to memory requests for expansion ROMs themselves but in some cases they can occur when attempting to allocate space for more critical BARs. This can happen when a successful expansion ROM allocation request consumes memory resource that was intended for a non-ROM BAR. We have seen this happen during PCI hotplug of an adapter that contains a P2P bridge where successful memory allocation for an expansion ROM BAR on device behind the bridge consumed memory that was intended for a non-ROM BAR on the P2P bridge. In all cases the allocation failure messages can be very confusing for users. This patch provides a new 'pci=norom' kernel boot parameter that can be used to disable the default PCI expansion ROM memory resource allocation. This provides a way to avoid the above described issues on systems that do not contain PCI devices for which drivers or user-level applications depend on the default PCI expansion ROM memory resource allocation behavior. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2008-05-13 04:57:46 +08:00
} else if (!strcmp(str, "norom")) {
pci_probe |= PCI_NOASSIGN_ROMS;
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, "earlydump")) {
pci_early_dump_regs = 1;
return NULL;
} 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;
}
return str;
}
unsigned int pcibios_assign_all_busses(void)
{
return (pci_probe & PCI_ASSIGN_ALL_BUSSES) ? 1 : 0;
}
int pcibios_enable_device(struct pci_dev *dev, int mask)
{
int err;
if ((err = pci_enable_resources(dev, mask)) < 0)
return err;
if (!dev->msi_enabled)
return pcibios_enable_irq(dev);
return 0;
}
void pcibios_disable_device (struct pci_dev *dev)
{
if (!dev->msi_enabled && pcibios_disable_irq)
pcibios_disable_irq(dev);
}
finish i386 and x86-64 sysdata conversion This patch finishes the i386 and x86-64 ->sysdata conversion and hopefully also fixes Riku's and Andy's observed bugs. It is based on Yinghai Lu's and Andy Whitcroft's patches (thanks!) with some changes: - introduce pci_scan_bus_with_sysdata() and use it instead of pci_scan_bus() where appropriate. pci_scan_bus_with_sysdata() will allocate the sysdata structure and then call pci_scan_bus(). - always allocate pci_sysdata dynamically. The whole point of this sysdata work is to make it easy to do root-bus specific things (e.g., support PCI domains and IOMMU's). I dislike using a default struct pci_sysdata in some places and a dynamically allocated pci_sysdata elsewhere - the potential for someone indavertantly changing the default structure is too high. - this patch only makes the minimal changes necessary, i.e., the NUMA node is always initialized to -1. Patches to do the right thing with regards to the NUMA node can build on top of this (either add a 'node' parameter to pci_scan_bus_with_sysdata() or just update the node when it becomes known). The patch was compile tested with various configurations (e.g., NUMAQ, VISWS) and run-time tested on i386 and x86-64. Unfortunately none of my machines exhibited the bugs so caveat emptor. Andy, could you please see if this fixes the NUMA issues you've seen? Riku, does this fix "pci=noacpi" on your laptop? Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com> Cc: Yinghai Lu <yhlu.kernel@gmail.com> Cc: Andi Kleen <ak@suse.de> Cc: Chuck Ebbert <cebbert@redhat.com> Cc: <riku.seppala@kymp.net> Cc: Andy Whitcroft <apw@shadowen.org> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-08-11 04:01:19 +08:00
struct pci_bus * __devinit pci_scan_bus_on_node(int busno, struct pci_ops *ops, int node)
finish i386 and x86-64 sysdata conversion This patch finishes the i386 and x86-64 ->sysdata conversion and hopefully also fixes Riku's and Andy's observed bugs. It is based on Yinghai Lu's and Andy Whitcroft's patches (thanks!) with some changes: - introduce pci_scan_bus_with_sysdata() and use it instead of pci_scan_bus() where appropriate. pci_scan_bus_with_sysdata() will allocate the sysdata structure and then call pci_scan_bus(). - always allocate pci_sysdata dynamically. The whole point of this sysdata work is to make it easy to do root-bus specific things (e.g., support PCI domains and IOMMU's). I dislike using a default struct pci_sysdata in some places and a dynamically allocated pci_sysdata elsewhere - the potential for someone indavertantly changing the default structure is too high. - this patch only makes the minimal changes necessary, i.e., the NUMA node is always initialized to -1. Patches to do the right thing with regards to the NUMA node can build on top of this (either add a 'node' parameter to pci_scan_bus_with_sysdata() or just update the node when it becomes known). The patch was compile tested with various configurations (e.g., NUMAQ, VISWS) and run-time tested on i386 and x86-64. Unfortunately none of my machines exhibited the bugs so caveat emptor. Andy, could you please see if this fixes the NUMA issues you've seen? Riku, does this fix "pci=noacpi" on your laptop? Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com> Cc: Yinghai Lu <yhlu.kernel@gmail.com> Cc: Andi Kleen <ak@suse.de> Cc: Chuck Ebbert <cebbert@redhat.com> Cc: <riku.seppala@kymp.net> Cc: Andy Whitcroft <apw@shadowen.org> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-08-11 04:01:19 +08:00
{
struct pci_bus *bus = NULL;
struct pci_sysdata *sd;
/*
* 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) {
printk(KERN_ERR "PCI: OOM, skipping PCI bus %02x\n", busno);
return NULL;
}
sd->node = node;
bus = pci_scan_bus(busno, ops, sd);
finish i386 and x86-64 sysdata conversion This patch finishes the i386 and x86-64 ->sysdata conversion and hopefully also fixes Riku's and Andy's observed bugs. It is based on Yinghai Lu's and Andy Whitcroft's patches (thanks!) with some changes: - introduce pci_scan_bus_with_sysdata() and use it instead of pci_scan_bus() where appropriate. pci_scan_bus_with_sysdata() will allocate the sysdata structure and then call pci_scan_bus(). - always allocate pci_sysdata dynamically. The whole point of this sysdata work is to make it easy to do root-bus specific things (e.g., support PCI domains and IOMMU's). I dislike using a default struct pci_sysdata in some places and a dynamically allocated pci_sysdata elsewhere - the potential for someone indavertantly changing the default structure is too high. - this patch only makes the minimal changes necessary, i.e., the NUMA node is always initialized to -1. Patches to do the right thing with regards to the NUMA node can build on top of this (either add a 'node' parameter to pci_scan_bus_with_sysdata() or just update the node when it becomes known). The patch was compile tested with various configurations (e.g., NUMAQ, VISWS) and run-time tested on i386 and x86-64. Unfortunately none of my machines exhibited the bugs so caveat emptor. Andy, could you please see if this fixes the NUMA issues you've seen? Riku, does this fix "pci=noacpi" on your laptop? Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com> Cc: Yinghai Lu <yhlu.kernel@gmail.com> Cc: Andi Kleen <ak@suse.de> Cc: Chuck Ebbert <cebbert@redhat.com> Cc: <riku.seppala@kymp.net> Cc: Andy Whitcroft <apw@shadowen.org> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-08-11 04:01:19 +08:00
if (!bus)
kfree(sd);
return bus;
}
struct pci_bus * __devinit pci_scan_bus_with_sysdata(int busno)
{
return pci_scan_bus_on_node(busno, &pci_root_ops, -1);
}