diff options
Diffstat (limited to 'arch')
51 files changed, 3394 insertions, 2297 deletions
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index bf07b6f50fa..07276ac01c2 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -261,36 +261,6 @@ config X86_VOYAGER If you do not specifically know you have a Voyager based machine, say N here, otherwise the kernel you build will not be bootable. -config X86_NUMAQ - bool "NUMAQ (IBM/Sequent)" - depends on SMP && X86_32 - select NUMA - help - This option is used for getting Linux to run on a (IBM/Sequent) NUMA - multiquad box. This changes the way that processors are bootstrapped, - and uses Clustered Logical APIC addressing mode instead of Flat Logical. - You will need a new lynxer.elf file to flash your firmware with - send - email to <Martin.Bligh@us.ibm.com>. - -config X86_SUMMIT - bool "Summit/EXA (IBM x440)" - depends on X86_32 && SMP - help - This option is needed for IBM systems that use the Summit/EXA chipset. - In particular, it is needed for the x440. - - If you don't have one of these computers, you should say N here. - If you want to build a NUMA kernel, you must select ACPI. - -config X86_BIGSMP - bool "Support for other sub-arch SMP systems with more than 8 CPUs" - depends on X86_32 && SMP - help - This option is needed for the systems that have more than 8 CPUs - and if the system is not of any sub-arch type above. - - If you don't have such a system, you should say N here. - config X86_VISWS bool "SGI 320/540 (Visual Workstation)" depends on X86_32 @@ -304,12 +274,33 @@ config X86_VISWS and vice versa. See <file:Documentation/sgi-visws.txt> for details. config X86_GENERICARCH - bool "Generic architecture (Summit, bigsmp, ES7000, default)" + bool "Generic architecture" depends on X86_32 help - This option compiles in the Summit, bigsmp, ES7000, default subarchitectures. - It is intended for a generic binary kernel. - If you want a NUMA kernel, select ACPI. We need SRAT for NUMA. + This option compiles in the NUMAQ, Summit, bigsmp, ES7000, default + subarchitectures. It is intended for a generic binary kernel. + if you select them all, kernel will probe it one by one. and will + fallback to default. + +if X86_GENERICARCH + +config X86_NUMAQ + bool "NUMAQ (IBM/Sequent)" + depends on SMP && X86_32 + select NUMA + help + This option is used for getting Linux to run on a NUMAQ (IBM/Sequent) + NUMA multiquad box. This changes the way that processors are + bootstrapped, and uses Clustered Logical APIC addressing mode instead + of Flat Logical. You will need a new lynxer.elf file to flash your + firmware with - send email to <Martin.Bligh@us.ibm.com>. + +config X86_SUMMIT + bool "Summit/EXA (IBM x440)" + depends on X86_32 && SMP + help + This option is needed for IBM systems that use the Summit/EXA chipset. + In particular, it is needed for the x440. config X86_ES7000 bool "Support for Unisys ES7000 IA32 series" @@ -317,8 +308,15 @@ config X86_ES7000 help Support for Unisys ES7000 systems. Say 'Y' here if this kernel is supposed to run on an IA32-based Unisys ES7000 system. - Only choose this option if you have such a system, otherwise you - should say N here. + +config X86_BIGSMP + bool "Support for big SMP systems with more than 8 CPUs" + depends on X86_32 && SMP + help + This option is needed for the systems that have more than 8 CPUs + and if the system is not of any sub-arch type above. + +endif config X86_RDC321X bool "RDC R-321x SoC" @@ -911,9 +909,9 @@ config X86_PAE config NUMA bool "Numa Memory Allocation and Scheduler Support (EXPERIMENTAL)" depends on SMP - depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || (X86_SUMMIT || X86_GENERICARCH) && ACPI) && EXPERIMENTAL) + depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || X86_GENERICARCH || X86_SUMMIT && ACPI) && EXPERIMENTAL) default n if X86_PC - default y if (X86_NUMAQ || X86_SUMMIT) + default y if (X86_NUMAQ || X86_SUMMIT || X86_GENERICARCH) help Enable NUMA (Non Uniform Memory Access) support. The kernel will try to allocate memory used by a CPU on the @@ -1090,6 +1088,40 @@ config MTRR See <file:Documentation/mtrr.txt> for more information. +config MTRR_SANITIZER + def_bool y + prompt "MTRR cleanup support" + depends on MTRR + help + Convert MTRR layout from continuous to discrete, so some X driver + could add WB entries. + + Say N here if you see bootup problems (boot crash, boot hang, + spontaneous reboots). + + Could be disabled with disable_mtrr_cleanup. Also mtrr_chunk_size + could be used to send largest mtrr entry size for continuous block + to hold holes (aka. UC entries) + + If unsure, say Y. + +config MTRR_SANITIZER_ENABLE_DEFAULT + int "MTRR cleanup enable value (0-1)" + range 0 1 + default "0" + depends on MTRR_SANITIZER + help + Enable mtrr cleanup default value + +config MTRR_SANITIZER_SPARE_REG_NR_DEFAULT + int "MTRR cleanup spare reg num (0-7)" + range 0 7 + default "1" + depends on MTRR_SANITIZER + help + mtrr cleanup spare entries default, it can be changed via + mtrr_spare_reg_nr= + config X86_PAT bool prompt "x86 PAT support" diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 18363374d51..253e7a5706d 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -131,7 +131,7 @@ config 4KSTACKS config X86_FIND_SMP_CONFIG def_bool y - depends on X86_LOCAL_APIC || X86_VOYAGER + depends on X86_MPPARSE || X86_VOYAGER || X86_VISWS depends on X86_32 config X86_MPPARSE diff --git a/arch/x86/Makefile b/arch/x86/Makefile index 3cff3c894cf..d6650131659 100644 --- a/arch/x86/Makefile +++ b/arch/x86/Makefile @@ -117,29 +117,11 @@ mcore-$(CONFIG_X86_VOYAGER) := arch/x86/mach-voyager/ mflags-$(CONFIG_X86_VISWS) := -Iinclude/asm-x86/mach-visws mcore-$(CONFIG_X86_VISWS) := arch/x86/mach-visws/ -# NUMAQ subarch support -mflags-$(CONFIG_X86_NUMAQ) := -Iinclude/asm-x86/mach-numaq -mcore-$(CONFIG_X86_NUMAQ) := arch/x86/mach-default/ - -# BIGSMP subarch support -mflags-$(CONFIG_X86_BIGSMP) := -Iinclude/asm-x86/mach-bigsmp -mcore-$(CONFIG_X86_BIGSMP) := arch/x86/mach-default/ - -#Summit subarch support -mflags-$(CONFIG_X86_SUMMIT) := -Iinclude/asm-x86/mach-summit -mcore-$(CONFIG_X86_SUMMIT) := arch/x86/mach-default/ - # generic subarchitecture mflags-$(CONFIG_X86_GENERICARCH):= -Iinclude/asm-x86/mach-generic fcore-$(CONFIG_X86_GENERICARCH) += arch/x86/mach-generic/ mcore-$(CONFIG_X86_GENERICARCH) := arch/x86/mach-default/ - -# ES7000 subarch support -mflags-$(CONFIG_X86_ES7000) := -Iinclude/asm-x86/mach-es7000 -fcore-$(CONFIG_X86_ES7000) := arch/x86/mach-es7000/ -mcore-$(CONFIG_X86_ES7000) := arch/x86/mach-default/ - # RDC R-321x subarch support mflags-$(CONFIG_X86_RDC321X) := -Iinclude/asm-x86/mach-rdc321x mcore-$(CONFIG_X86_RDC321X) := arch/x86/mach-default/ @@ -160,6 +142,7 @@ KBUILD_AFLAGS += $(mflags-y) head-y := arch/x86/kernel/head_$(BITS).o head-y += arch/x86/kernel/head$(BITS).o +head-y += arch/x86/kernel/head.o head-y += arch/x86/kernel/init_task.o libs-y += arch/x86/lib/ diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c index 90456cee47c..ba0be6a25ff 100644 --- a/arch/x86/boot/compressed/misc.c +++ b/arch/x86/boot/compressed/misc.c @@ -221,10 +221,6 @@ static char *vidmem; static int vidport; static int lines, cols; -#ifdef CONFIG_X86_NUMAQ -void *xquad_portio; -#endif - #include "../../../../lib/inflate.c" static void *malloc(int size) diff --git a/arch/x86/boot/memory.c b/arch/x86/boot/memory.c index acad32eb429..53165c97336 100644 --- a/arch/x86/boot/memory.c +++ b/arch/x86/boot/memory.c @@ -13,6 +13,7 @@ */ #include "boot.h" +#include <linux/kernel.h> #define SMAP 0x534d4150 /* ASCII "SMAP" */ @@ -53,7 +54,7 @@ static int detect_memory_e820(void) count++; desc++; - } while (next && count < E820MAX); + } while (next && count < ARRAY_SIZE(boot_params.e820_map)); return boot_params.e820_entries = count; } diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 77807d4769c..dc3c636d113 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -2,7 +2,7 @@ # Makefile for the linux kernel. # -extra-y := head_$(BITS).o head$(BITS).o init_task.o vmlinux.lds +extra-y := head_$(BITS).o head$(BITS).o head.o init_task.o vmlinux.lds CPPFLAGS_vmlinux.lds += -U$(UTS_MACHINE) @@ -22,7 +22,7 @@ obj-y += setup_$(BITS).o i8259_$(BITS).o setup.o obj-$(CONFIG_X86_32) += sys_i386_32.o i386_ksyms_32.o obj-$(CONFIG_X86_64) += sys_x86_64.o x8664_ksyms_64.o obj-$(CONFIG_X86_64) += syscall_64.o vsyscall_64.o setup64.o -obj-y += bootflag.o e820_$(BITS).o +obj-y += bootflag.o e820_$(BITS).o e820.o obj-y += pci-dma.o quirks.o i8237.o topology.o kdebugfs.o obj-y += alternative.o i8253.o pci-nommu.o obj-$(CONFIG_X86_64) += bugs_64.o diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 33c5216fd3e..caf4ed7ca06 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -338,8 +338,6 @@ acpi_parse_lapic_nmi(struct acpi_subtable_header * header, const unsigned long e #ifdef CONFIG_X86_IO_APIC -struct mp_ioapic_routing mp_ioapic_routing[MAX_IO_APICS]; - static int __init acpi_parse_ioapic(struct acpi_subtable_header * header, const unsigned long end) { @@ -860,6 +858,336 @@ static int __init acpi_parse_madt_lapic_entries(void) #endif /* CONFIG_X86_LOCAL_APIC */ #ifdef CONFIG_X86_IO_APIC +#define MP_ISA_BUS 0 + +#ifdef CONFIG_X86_ES7000 +extern int es7000_plat; +#endif + +static struct { + int apic_id; + int gsi_base; + int gsi_end; + DECLARE_BITMAP(pin_programmed, MP_MAX_IOAPIC_PIN + 1); +} mp_ioapic_routing[MAX_IO_APICS]; + +static int mp_find_ioapic(int gsi) +{ + int i = 0; + + /* Find the IOAPIC that manages this GSI. */ + for (i = 0; i < nr_ioapics; i++) { + if ((gsi >= mp_ioapic_routing[i].gsi_base) + && (gsi <= mp_ioapic_routing[i].gsi_end)) + return i; + } + + printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi); + return -1; +} + +static u8 __init uniq_ioapic_id(u8 id) +{ +#ifdef CONFIG_X86_32 + if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && + !APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) + return io_apic_get_unique_id(nr_ioapics, id); + else + return id; +#else + int i; + DECLARE_BITMAP(used, 256); + bitmap_zero(used, 256); + for (i = 0; i < nr_ioapics; i++) { + struct mp_config_ioapic *ia = &mp_ioapics[i]; + __set_bit(ia->mp_apicid, used); + } + if (!test_bit(id, used)) + return id; + return find_first_zero_bit(used, 256); +#endif +} + +static int bad_ioapic(unsigned long address) +{ + if (nr_ioapics >= MAX_IO_APICS) { + printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded " + "(found %d)\n", MAX_IO_APICS, nr_ioapics); + panic("Recompile kernel with bigger MAX_IO_APICS!\n"); + } + if (!address) { + printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address" + " found in table, skipping!\n"); + return 1; + } + return 0; +} + +void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) +{ + int idx = 0; + + if (bad_ioapic(address)) + return; + + idx = nr_ioapics; + + mp_ioapics[idx].mp_type = MP_IOAPIC; + mp_ioapics[idx].mp_flags = MPC_APIC_USABLE; + mp_ioapics[idx].mp_apicaddr = address; + + set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address); + mp_ioapics[idx].mp_apicid = uniq_ioapic_id(id); +#ifdef CONFIG_X86_32 + mp_ioapics[idx].mp_apicver = io_apic_get_version(idx); +#else + mp_ioapics[idx].mp_apicver = 0; +#endif + /* + * Build basic GSI lookup table to facilitate gsi->io_apic lookups + * and to prevent reprogramming of IOAPIC pins (PCI GSIs). + */ + mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mp_apicid; + mp_ioapic_routing[idx].gsi_base = gsi_base; + mp_ioapic_routing[idx].gsi_end = gsi_base + + io_apic_get_redir_entries(idx); + + printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%lx, " + "GSI %d-%d\n", idx, mp_ioapics[idx].mp_apicid, + mp_ioapics[idx].mp_apicver, mp_ioapics[idx].mp_apicaddr, + mp_ioapic_routing[idx].gsi_base, mp_ioapic_routing[idx].gsi_end); + + nr_ioapics++; +} + +void __init mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) +{ + int ioapic = -1; + int pin = -1; + + /* + * Convert 'gsi' to 'ioapic.pin'. + */ + ioapic = mp_find_ioapic(gsi); + if (ioapic < 0) + return; + pin = gsi - mp_ioapic_routing[ioapic].gsi_base; + + /* + * TBD: This check is for faulty timer entries, where the override + * erroneously sets the trigger to level, resulting in a HUGE + * increase of timer interrupts! + */ + if ((bus_irq == 0) && (trigger == 3)) + trigger = 1; + + mp_irqs[mp_irq_entries].mp_type = MP_INTSRC; + mp_irqs[mp_irq_entries].mp_irqtype = mp_INT; + mp_irqs[mp_irq_entries].mp_irqflag = (trigger << 2) | polarity; + mp_irqs[mp_irq_entries].mp_srcbus = MP_ISA_BUS; + mp_irqs[mp_irq_entries].mp_srcbusirq = bus_irq; /* IRQ */ + mp_irqs[mp_irq_entries].mp_dstapic = + mp_ioapics[ioapic].mp_apicid; /* APIC ID */ + mp_irqs[mp_irq_entries].mp_dstirq = pin; /* INTIN# */ + + if (++mp_irq_entries == MAX_IRQ_SOURCES) + panic("Max # of irq sources exceeded!!\n"); + +} + +void __init mp_config_acpi_legacy_irqs(void) +{ + int i = 0; + int ioapic = -1; + +#if defined (CONFIG_MCA) || defined (CONFIG_EISA) + /* + * Fabricate the legacy ISA bus (bus #31). + */ + mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA; +#endif + set_bit(MP_ISA_BUS, mp_bus_not_pci); + Dprintk("Bus #%d is ISA\n", MP_ISA_BUS); + +#ifdef CONFIG_X86_ES7000 + /* + * Older generations of ES7000 have no legacy identity mappings + */ + if (es7000_plat == 1) + return; +#endif + + /* + * Locate the IOAPIC that manages the ISA IRQs (0-15). + */ + ioapic = mp_find_ioapic(0); + if (ioapic < 0) + return; + + /* + * Use the default configuration for the IRQs 0-15. Unless + * overridden by (MADT) interrupt source override entries. + */ + for (i = 0; i < 16; i++) { + int idx; + + mp_irqs[mp_irq_entries].mp_type = MP_INTSRC; + mp_irqs[mp_irq_entries].mp_irqflag = 0; /* Conforming */ + mp_irqs[mp_irq_entries].mp_srcbus = MP_ISA_BUS; + mp_irqs[mp_irq_entries].mp_dstapic = mp_ioapics[ioapic].mp_apicid; + + for (idx = 0; idx < mp_irq_entries; idx++) { + struct mp_config_intsrc *irq = mp_irqs + idx; + + /* Do we already have a mapping for this ISA IRQ? */ + if (irq->mp_srcbus == MP_ISA_BUS + && irq->mp_srcbusirq == i) + break; + + /* Do we already have a mapping for this IOAPIC pin */ + if ((irq->mp_dstapic == + mp_irqs[mp_irq_entries].mp_dstapic) && + (irq->mp_dstirq == i)) + break; + } + + if (idx != mp_irq_entries) { + printk(KERN_DEBUG "ACPI: IRQ%d used by override.\n", i); + continue; /* IRQ already used */ + } + + mp_irqs[mp_irq_entries].mp_irqtype = mp_INT; + mp_irqs[mp_irq_entries].mp_srcbusirq = i; /* Identity mapped */ + mp_irqs[mp_irq_entries].mp_dstirq = i; + + if (++mp_irq_entries == MAX_IRQ_SOURCES) + panic("Max # of irq sources exceeded!!\n"); + } +} + +int mp_register_gsi(u32 gsi, int triggering, int polarity) +{ + int ioapic; + int ioapic_pin; +#ifdef CONFIG_X86_32 +#define MAX_GSI_NUM 4096 +#define IRQ_COMPRESSION_START 64 + + static int pci_irq = IRQ_COMPRESSION_START; + /* + * Mapping between Global System Interrupts, which + * represent all possible interrupts, and IRQs + * assigned to actual devices. + */ + static int gsi_to_irq[MAX_GSI_NUM]; +#else + + if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC) + return gsi; +#endif + + /* Don't set up the ACPI SCI because it's already set up */ + if (acpi_gbl_FADT.sci_interrupt == gsi) + return gsi; + + ioapic = mp_find_ioapic(gsi); + if (ioapic < 0) { + printk(KERN_WARNING "No IOAPIC for GSI %u\n", gsi); + return gsi; + } + + ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_base; + +#ifdef CONFIG_X86_32 + if (ioapic_renumber_irq) + gsi = ioapic_renumber_irq(ioapic, gsi); +#endif + + /* + * Avoid pin reprogramming. PRTs typically include entries + * with redundant pin->gsi mappings (but unique PCI devices); + * we only program the IOAPIC on the first. + */ + if (ioapic_pin > MP_MAX_IOAPIC_PIN) { + printk(KERN_ERR "Invalid reference to IOAPIC pin " + "%d-%d\n", mp_ioapic_routing[ioapic].apic_id, + ioapic_pin); + return gsi; + } + if (test_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed)) { + Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n", + mp_ioapic_routing[ioapic].apic_id, ioapic_pin); +#ifdef CONFIG_X86_32 + return (gsi < IRQ_COMPRESSION_START ? gsi : gsi_to_irq[gsi]); +#else + return gsi; +#endif + } + + set_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed); +#ifdef CONFIG_X86_32 + /* + * For GSI >= 64, use IRQ compression + */ + if ((gsi >= IRQ_COMPRESSION_START) + && (triggering == ACPI_LEVEL_SENSITIVE)) { + /* + * For PCI devices assign IRQs in order, avoiding gaps + * due to unused I/O APIC pins. + */ + int irq = gsi; + if (gsi < MAX_GSI_NUM) { + /* + * Retain the VIA chipset work-around (gsi > 15), but + * avoid a problem where the 8254 timer (IRQ0) is setup + * via an override (so it's not on pin 0 of the ioapic), + * and at the same time, the pin 0 interrupt is a PCI + * type. The gsi > 15 test could cause these two pins + * to be shared as IRQ0, and they are not shareable. + * So test for this condition, and if necessary, avoid + * the pin collision. + */ + gsi = pci_irq++; + /* + * Don't assign IRQ used by ACPI SCI + */ + if (gsi == acpi_gbl_FADT.sci_interrupt) + gsi = pci_irq++; + gsi_to_irq[irq] = gsi; + } else { + printk(KERN_ERR "GSI %u is too high\n", gsi); + return gsi; + } + } +#endif + io_apic_set_pci_routing(ioapic, ioapic_pin, gsi, + triggering == ACPI_EDGE_SENSITIVE ? 0 : 1, + polarity == ACPI_ACTIVE_HIGH ? 0 : 1); + return gsi; +} + +int mp_config_acpi_gsi(unsigned char number, unsigned int devfn, u8 pin, + u32 gsi, int triggering, int polarity) +{ + struct mpc_config_intsrc intsrc; + int ioapic; + + /* print the entry should happen on mptable identically */ + intsrc.mpc_type = MP_INTSRC; + intsrc.mpc_irqtype = mp_INT; + intsrc.mpc_irqflag = (triggering == ACPI_EDGE_SENSITIVE ? 4 : 0x0c) | + (polarity == ACPI_ACTIVE_HIGH ? 1 : 3); + intsrc.mpc_srcbus = number; + intsrc.mpc_srcbusirq = (((devfn >> 3) & 0x1f) << 2) | ((pin - 1) & 3); + ioapic = mp_find_ioapic(gsi); + intsrc.mpc_dstapic = mp_ioapic_routing[ioapic].apic_id; + intsrc.mpc_dstirq = gsi - mp_ioapic_routing[ioapic].gsi_base; + + MP_intsrc_info(&intsrc); + + return 0; +} + /* * Parse IOAPIC related entries in MADT * returns 0 on success, < 0 on error diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c index 4b99b1bdeb6..954d67931a5 100644 --- a/arch/x86/kernel/apic_32.c +++ b/arch/x86/kernel/apic_32.c @@ -76,6 +76,11 @@ EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok); */ int apic_verbosity; +int pic_mode; + +/* Have we found an MP table */ +int smp_found_config; + static unsigned int calibration_result; static int lapic_next_event(unsigned long delta, @@ -1202,7 +1207,7 @@ void __init init_apic_mappings(void) for (i = 0; i < nr_ioapics; i++) { if (smp_found_config) { - ioapic_phys = mp_ioapics[i].mpc_apicaddr; + ioapic_phys = mp_ioapics[i].mp_apicaddr; if (!ioapic_phys) { printk(KERN_ERR "WARNING: bogus zero IO-APIC " @@ -1513,6 +1518,9 @@ void __cpuinit generic_processor_info(int apicid, int version) */ cpu = 0; + if (apicid > max_physical_apicid) + max_physical_apicid = apicid; + /* * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y * but we need to work other dependencies like SMP_SUSPEND etc @@ -1520,7 +1528,7 @@ void __cpuinit generic_processor_info(int apicid, int version) * if (CPU_HOTPLUG_ENABLED || num_processors > 8) * - Ashok Raj <ashok.raj@intel.com> */ - if (num_processors > 8) { + if (max_physical_apicid >= 8) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_INTEL: if (!APIC_XAPIC(version)) { diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c index 0633cfd0dc2..a4bd8fbb78a 100644 --- a/arch/x86/kernel/apic_64.c +++ b/arch/x86/kernel/apic_64.c @@ -56,6 +56,9 @@ EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok); */ int apic_verbosity; +/* Have we found an MP table */ +int smp_found_config; + static struct resource lapic_resource = { .name = "Local APIC", .flags = IORESOURCE_MEM | IORESOURCE_BUSY, @@ -1090,6 +1093,9 @@ void __cpuinit generic_processor_info(int apicid, int version) */ cpu = 0; } + if (apicid > max_physical_apicid) + max_physical_apicid = apicid; + /* are we being called early in kernel startup? */ if (x86_cpu_to_apicid_early_ptr) { u16 *cpu_to_apicid = x86_cpu_to_apicid_early_ptr; diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 5d241ce94a4..509bd3d9eac 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -37,7 +37,7 @@ static struct fixed_range_block fixed_range_blocks[] = { static unsigned long smp_changes_mask; static struct mtrr_state mtrr_state = {}; static int mtrr_state_set; -static u64 tom2; +u64 mtrr_tom2; #undef MODULE_PARAM_PREFIX #define MODULE_PARAM_PREFIX "mtrr." @@ -139,8 +139,8 @@ u8 mtrr_type_lookup(u64 start, u64 end) } } - if (tom2) { - if (start >= (1ULL<<32) && (end < tom2)) + if (mtrr_tom2) { + if (start >= (1ULL<<32) && (end < mtrr_tom2)) return MTRR_TYPE_WRBACK; } @@ -158,6 +158,20 @@ get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr) rdmsr(MTRRphysMask_MSR(index), vr->mask_lo, vr->mask_hi); } +/* fill the MSR pair relating to a var range */ +void fill_mtrr_var_range(unsigned int index, + u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi) +{ + struct mtrr_var_range *vr; + + vr = mtrr_state.var_ranges; + + vr[index].base_lo = base_lo; + vr[index].base_hi = base_hi; + vr[index].mask_lo = mask_lo; + vr[index].mask_hi = mask_hi; +} + static void get_fixed_ranges(mtrr_type * frs) { @@ -213,13 +227,13 @@ void __init get_mtrr_state(void) mtrr_state.enabled = (lo & 0xc00) >> 10; if (amd_special_default_mtrr()) { - unsigned lo, hi; + unsigned low, high; /* TOP_MEM2 */ - rdmsr(MSR_K8_TOP_MEM2, lo, hi); - tom2 = hi; - tom2 <<= 32; - tom2 |= lo; - tom2 &= 0xffffff8000000ULL; + rdmsr(MSR_K8_TOP_MEM2, low, high); + mtrr_tom2 = high; + mtrr_tom2 <<= 32; + mtrr_tom2 |= low; + mtrr_tom2 &= 0xffffff800000ULL; } if (mtrr_show) { int high_width; @@ -251,9 +265,9 @@ void __init get_mtrr_state(void) else printk(KERN_INFO "MTRR %u disabled\n", i); } - if (tom2) { + if (mtrr_tom2) { printk(KERN_INFO "TOM2: %016llx aka %lldM\n", - tom2, tom2>>20); + mtrr_tom2, mtrr_tom2>>20); } } mtrr_state_set = 1; @@ -328,7 +342,7 @@ static void set_fixed_range(int msr, bool *changed, unsigned int *msrwords) if (lo != msrwords[0] || hi != msrwords[1]) { if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD && - boot_cpu_data.x86 == 15 && + (boot_cpu_data.x86 >= 0x0f && boot_cpu_data.x86 <= 0x11) && ((msrwords[0] | msrwords[1]) & K8_MTRR_RDMEM_WRMEM_MASK)) k8_enable_fixed_iorrs(); mtrr_wrmsr(msr, msrwords[0], msrwords[1]); diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index 6a1e278d932..0642201784e 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -37,6 +37,7 @@ #include <linux/smp.h> #include <linux/cpu.h> #include <linux/mutex.h> +#include <linux/sort.h> #include <asm/e820.h> #include <asm/mtrr.h> @@ -609,6 +610,787 @@ static struct sysdev_driver mtrr_sysdev_driver = { .resume = mtrr_restore, }; +/* should be related to MTRR_VAR_RANGES nums */ +#define RANGE_NUM 256 + +struct res_range { + unsigned long start; + unsigned long end; +}; + +static int __init +add_range(struct res_range *range, int nr_range, unsigned long start, + unsigned long end) +{ + /* out of slots */ + if (nr_range >= RANGE_NUM) + return nr_range; + + range[nr_range].start = start; + range[nr_range].end = end; + + nr_range++; + + return nr_range; +} + +static int __init +add_range_with_merge(struct res_range *range, int nr_range, unsigned long start, + unsigned long end) +{ + int i; + + /* try to merge it with old one */ + for (i = 0; i < nr_range; i++) { + unsigned long final_start, final_end; + unsigned long common_start, common_end; + + if (!range[i].end) + continue; + + common_start = max(range[i].start, start); + common_end = min(range[i].end, end); + if (common_start > common_end + 1) + continue; + + final_start = min(range[i].start, start); + final_end = max(range[i].end, end); + + range[i].start = final_start; + range[i].end = final_end; + return nr_range; + } + + /* need to add that */ + return add_range(range, nr_range, start, end); +} + +static void __init +subtract_range(struct res_range *range, unsigned long start, unsigned long end) +{ + int i, j; + + for (j = 0; j < RANGE_NUM; j++) { + if (!range[j].end) + continue; + + if (start <= range[j].start && end >= range[j].end) { + range[j].start = 0; + range[j].end = 0; + continue; + } + + if (start <= range[j].start && end < range[j].end && + range[j].start < end + 1) { + range[j].start = end + 1; + continue; + } + + + if (start > range[j].start && end >= range[j].end && + range[j].end > start - 1) { + range[j].end = start - 1; + continue; + } + + if (start > range[j].start && end < range[j].end) { + /* find the new spare */ + for (i = 0; i < RANGE_NUM; i++) { + if (range[i].end == 0) + break; + } + if (i < RANGE_NUM) { + range[i].end = range[j].end; + range[i].start = end + 1; + } else { + printk(KERN_ERR "run of slot in ranges\n"); + } + range[j].end = start - 1; + continue; + } + } +} + +static int __init cmp_range(const void *x1, const void *x2) +{ + const struct res_range *r1 = x1; + const struct res_range *r2 = x2; + long start1, start2; + + start1 = r1->start; + start2 = r2->start; + + return start1 - start2; +} + +struct var_mtrr_range_state { + unsigned long base_pfn; + unsigned long size_pfn; + mtrr_type type; +}; + +struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; +static int __initdata debug_print; + +static int __init +x86_get_mtrr_mem_range(struct res_range *range, int nr_range, + unsigned long extra_remove_base, + unsigned long extra_remove_size) +{ + unsigned long i, base, size; + mtrr_type type; + + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type != MTRR_TYPE_WRBACK) + continue; + base = range_state[i].base_pfn; + size = range_state[i].size_pfn; + nr_range = add_range_with_merge(range, nr_range, base, + base + size - 1); + } + if (debug_print) { + printk(KERN_DEBUG "After WB checking\n"); + for (i = 0; i < nr_range; i++) + printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", + range[i].start, range[i].end + 1); + } + + /* take out UC ranges */ + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type != MTRR_TYPE_UNCACHABLE) + continue; + size = range_state[i].size_pfn; + if (!size) + continue; + base = range_state[i].base_pfn; + subtract_range(range, base, base + size - 1); + } + if (extra_remove_size) + subtract_range(range, extra_remove_base, + extra_remove_base + extra_remove_size - 1); + + /* get new range num */ + nr_range = 0; + for (i = 0; i < RANGE_NUM; i++) { + if (!range[i].end) + continue; + nr_range++; + } + if (debug_print) { + printk(KERN_DEBUG "After UC checking\n"); + for (i = 0; i < nr_range; i++) + printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", + range[i].start, range[i].end + 1); + } + + /* sort the ranges */ + sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); + if (debug_print) { + printk(KERN_DEBUG "After sorting\n"); + for (i = 0; i < nr_range; i++) + printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", + range[i].start, range[i].end + 1); + } + + /* clear those is not used */ + for (i = nr_range; i < RANGE_NUM; i++) + memset(&range[i], 0, sizeof(range[i])); + + return nr_range; +} + +static struct res_range __initdata range[RANGE_NUM]; + +#ifdef CONFIG_MTRR_SANITIZER + +static unsigned long __init sum_ranges(struct res_range *range, int nr_range) +{ + unsigned long sum; + int i; + + sum = 0; + for (i = 0; i < nr_range; i++) + sum += range[i].end + 1 - range[i].start; + + return sum; +} + +static int enable_mtrr_cleanup __initdata = + CONFIG_MTRR_SANITIZER_ENABLE_DEFAULT; + +static int __init disable_mtrr_cleanup_setup(char *str) +{ + if (enable_mtrr_cleanup != -1) + enable_mtrr_cleanup = 0; + return 0; +} +early_param("disable_mtrr_cleanup", disable_mtrr_cleanup_setup); + +static int __init enable_mtrr_cleanup_setup(char *str) +{ + if (enable_mtrr_cleanup != -1) + enable_mtrr_cleanup = 1; + return 0; +} +early_param("enble_mtrr_cleanup", enable_mtrr_cleanup_setup); + +struct var_mtrr_state { + unsigned long range_startk; + unsigned long range_sizek; + unsigned long chunk_sizek; + unsigned long gran_sizek; + unsigned int reg; +}; + +static void __init +set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, + unsigned char type, unsigned int address_bits) +{ + u32 base_lo, base_hi, mask_lo, mask_hi; + u64 base, mask; + + if (!sizek) { + fill_mtrr_var_range(reg, 0, 0, 0, 0); + return; + } + + mask = (1ULL << address_bits) - 1; + mask &= ~((((u64)sizek) << 10) - 1); + + base = ((u64)basek) << 10; + + base |= type; + mask |= 0x800; + + base_lo = base & ((1ULL<<32) - 1); + base_hi = base >> 32; + + mask_lo = mask & ((1ULL<<32) - 1); + mask_hi = mask >> 32; + + fill_mtrr_var_range(reg, base_lo, base_hi, mask_lo, mask_hi); +} + +static void __init +save_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, + unsigned char type) +{ + range_state[reg].base_pfn = basek >> (PAGE_SHIFT - 10); + range_state[reg].size_pfn = sizek >> (PAGE_SHIFT - 10); + range_state[reg].type = type; +} + +static void __init +set_var_mtrr_all(unsigned int address_bits) +{ + unsigned long basek, sizek; + unsigned char type; + unsigned int reg; + + for (reg = 0; reg < num_var_ranges; reg++) { + basek = range_state[reg].base_pfn << (PAGE_SHIFT - 10); + sizek = range_state[reg].size_pfn << (PAGE_SHIFT - 10); + type = range_state[reg].type; + + set_var_mtrr(reg, basek, sizek, type, address_bits); + } +} + +static unsigned int __init +range_to_mtrr(unsigned int reg, unsigned long range_startk, + unsigned long range_sizek, unsigned char type) +{ + if (!range_sizek || (reg >= num_var_ranges)) + return reg; + + while (range_sizek) { + unsigned long max_align, align; + unsigned long sizek; + + /* Compute the maximum size I can make a range */ + if (range_startk) + max_align = ffs(range_startk) - 1; + else + max_align = 32; + align = fls(range_sizek) - 1; + if (align > max_align) + align = max_align; + + sizek = 1 << align; + if (debug_print) + printk(KERN_DEBUG "Setting variable MTRR %d, " + "base: %ldMB, range: %ldMB, type %s\n", + reg, range_startk >> 10, sizek >> 10, + (type == MTRR_TYPE_UNCACHABLE)?"UC": + ((type == MTRR_TYPE_WRBACK)?"WB":"Other") + ); + save_var_mtrr(reg++, range_startk, sizek, type); + range_startk += sizek; + range_sizek -= sizek; + if (reg >= num_var_ranges) + break; + } + return reg; +} + +static unsigned __init +range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, + unsigned long sizek) +{ + unsigned long hole_basek, hole_sizek; + unsigned long second_basek, second_sizek; + unsigned long range0_basek, range0_sizek; + unsigned long range_basek, range_sizek; + unsigned long chunk_sizek; + unsigned long gran_sizek; + + hole_basek = 0; + hole_sizek = 0; + second_basek = 0; + second_sizek = 0; + chunk_sizek = state->chunk_sizek; + gran_sizek = state->gran_sizek; + + /* align with gran size, prevent small block used up MTRRs */ + range_basek = ALIGN(state->range_startk, gran_sizek); + if ((range_basek > basek) && basek) + return second_sizek; + state->range_sizek -= (range_basek - state->range_startk); + range_sizek = ALIGN(state->range_sizek, gran_sizek); + + while (range_sizek > state->range_sizek) { + range_sizek -= gran_sizek; + if (!range_sizek) + return 0; + } + state->range_sizek = range_sizek; + + /* try to append some small hole */ + range0_basek = state->range_startk; + range0_sizek = ALIGN(state->range_sizek, chunk_sizek); + if (range0_sizek == state->range_sizek) { + if (debug_print) + printk(KERN_DEBUG "rangeX: %016lx - %016lx\n", + range0_basek<<10, + (range0_basek + state->range_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range0_basek, + state->range_sizek, MTRR_TYPE_WRBACK); + return 0; + } + + range0_sizek -= chunk_sizek; + if (range0_sizek && sizek) { + while (range0_basek + range0_sizek > (basek + sizek)) { + range0_sizek -= chunk_sizek; + if (!range0_sizek) + break; + } + } + + if (range0_sizek) { + if (debug_print) + printk(KERN_DEBUG "range0: %016lx - %016lx\n", + range0_basek<<10, + (range0_basek + range0_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range0_basek, + range0_sizek, MTRR_TYPE_WRBACK); + + } + + range_basek = range0_basek + range0_sizek; + range_sizek = chunk_sizek; + + if (range_basek + range_sizek > basek && + range_basek + range_sizek <= (basek + sizek)) { + /* one hole */ + second_basek = basek; + second_sizek = range_basek + range_sizek - basek; + } + + /* if last piece, only could one hole near end */ + if ((second_basek || !basek) && + range_sizek - (state->range_sizek - range0_sizek) - second_sizek < + (chunk_sizek >> 1)) { + /* + * one hole in middle (second_sizek is 0) or at end + * (second_sizek is 0 ) + */ + hole_sizek = range_sizek - (state->range_sizek - range0_sizek) + - second_sizek; + hole_basek = range_basek + range_sizek - hole_sizek + - second_sizek; + } else { + /* fallback for big hole, or several holes */ + range_sizek = state->range_sizek - range0_sizek; + second_basek = 0; + second_sizek = 0; + } + + if (debug_print) + printk(KERN_DEBUG "range: %016lx - %016lx\n", range_basek<<10, + (range_basek + range_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range_basek, range_sizek, + MTRR_TYPE_WRBACK); + if (hole_sizek) { + if (debug_print) + printk(KERN_DEBUG "hole: %016lx - %016lx\n", + hole_basek<<10, (hole_basek + hole_sizek)<<10); + state->reg = range_to_mtrr(state->reg, hole_basek, hole_sizek, + MTRR_TYPE_UNCACHABLE); + + } + + return second_sizek; +} + +static void __init +set_var_mtrr_range(struct var_mtrr_state *state, unsigned long base_pfn, + unsigned long size_pfn) +{ + unsigned long basek, sizek; + unsigned long second_sizek = 0; + + if (state->reg >= num_var_ranges) + return; + + basek = base_pfn << (PAGE_SHIFT - 10); + sizek = size_pfn << (PAGE_SHIFT - 10); + + /* See if I can merge with the last range */ + if ((basek <= 1024) || + (state->range_startk + state->range_sizek == basek)) { + unsigned long endk = basek + sizek; + state->range_sizek = endk - state->range_startk; + return; + } + /* Write the range mtrrs */ + if (state->range_sizek != 0) + second_sizek = range_to_mtrr_with_hole(state, basek, sizek); + + /* Allocate an msr */ + state->range_startk = basek + second_sizek; + state->range_sizek = sizek - second_sizek; +} + +/* mininum size of mtrr block that can take hole */ +static u64 mtrr_chunk_size __initdata = (256ULL<<20); + +static int __init parse_mtrr_chunk_size_opt(char *p) +{ + if (!p) + return -EINVAL; + mtrr_chunk_size = memparse(p, &p); + return 0; +} +early_param("mtrr_chunk_size", parse_mtrr_chunk_size_opt); + +/* granity of mtrr of block */ +static u64 mtrr_gran_size __initdata; + +static int __init parse_mtrr_gran_size_opt(char *p) +{ + if (!p) + return -EINVAL; + mtrr_gran_size = memparse(p, &p); + return 0; +} +early_param("mtrr_gran_size", parse_mtrr_gran_size_opt); + +static int nr_mtrr_spare_reg __initdata = + CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT; + +static int __init parse_mtrr_spare_reg(char *arg) +{ + if (arg) + nr_mtrr_spare_reg = simple_strtoul(arg, NULL, 0); + return 0; +} + +early_param("mtrr_spare_reg_nr", parse_mtrr_spare_reg); + +static int __init +x86_setup_var_mtrrs(struct res_range *range, int nr_range, + u64 chunk_size, u64 gran_size) +{ + struct var_mtrr_state var_state; + int i; + int num_reg; + + var_state.range_startk = 0; + var_state.range_sizek = 0; + var_state.reg = 0; + var_state.chunk_sizek = chunk_size >> 10; + var_state.gran_sizek = gran_size >> 10; + + memset(range_state, 0, sizeof(range_state)); + + /* Write the range etc */ + for (i = 0; i < nr_range; i++) + set_var_mtrr_range(&var_state, range[i].start, + range[i].end - range[i].start + 1); + + /* Write the last range */ + if (var_state.range_sizek != 0) + range_to_mtrr_with_hole(&var_state, 0, 0); + + num_reg = var_state.reg; + /* Clear out the extra MTRR's */ + while (var_state.reg < num_var_ranges) { + save_var_mtrr(var_state.reg, 0, 0, 0); + var_state.reg++; + } + + return num_reg; +} + +struct mtrr_cleanup_result { + unsigned long gran_sizek; + unsigned long chunk_sizek; + unsigned long lose_cover_sizek; + unsigned int num_reg; + int bad; +}; + +/* + * gran_size: 1M, 2M, ..., 2G + * chunk size: gran_size, ..., 4G + * so we need (2+13)*6 + */ +#define NUM_RESULT 90 +#define PSHIFT (PAGE_SHIFT - 10) + +static struct mtrr_cleanup_result __initdata result[NUM_RESULT]; +static struct res_range __initdata range_new[RANGE_NUM]; +static unsigned long __initdata min_loss_pfn[RANGE_NUM]; + +static int __init mtrr_cleanup(unsigned address_bits) +{ + unsigned long extra_remove_base, extra_remove_size; + unsigned long i, base, size, def, dummy; + mtrr_type type; + int nr_range, nr_range_new; + u64 chunk_size, gran_size; + unsigned long range_sums, range_sums_new; + int index_good; + int num_reg_good; + + /* extra one for all 0 */ + int num[MTRR_NUM_TYPES + 1]; + + if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1) + return 0; + rdmsr(MTRRdefType_MSR, def, dummy); + def &= 0xff; + if (def != MTRR_TYPE_UNCACHABLE) + return 0; + + /* get it and store it aside */ + memset(range_state, 0, sizeof(range_state)); + for (i = 0; i < num_var_ranges; i++) { + mtrr_if->get(i, &base, &size, &type); + range_state[i].base_pfn = base; + range_state[i].size_pfn = size; + range_state[i].type = type; + } + + /* check entries number */ + memset(num, 0, sizeof(num)); + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + size = range_state[i].size_pfn; + if (type >= MTRR_NUM_TYPES) + continue; + if (!size) + type = MTRR_NUM_TYPES; + num[type]++; + } + + /* check if we got UC entries */ + if (!num[MTRR_TYPE_UNCACHABLE]) + return 0; + + /* check if we only had WB and UC */ + if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != + num_var_ranges - num[MTRR_NUM_TYPES]) + return 0; + + memset(range, 0, sizeof(range)); + extra_remove_size = 0; + if (mtrr_tom2) { + extra_remove_base = 1 << (32 - PAGE_SHIFT); + extra_remove_size = + (mtrr_tom2 >> PAGE_SHIFT) - extra_remove_base; + } + nr_range = x86_get_mtrr_mem_range(range, 0, extra_remove_base, + extra_remove_size); + range_sums = sum_ranges(range, nr_range); + printk(KERN_INFO "total RAM coverred: %ldM\n", + range_sums >> (20 - PAGE_SHIFT)); + + if (mtrr_chunk_size && mtrr_gran_size) { + int num_reg; + + debug_print = 1; + /* convert ranges to var ranges state */ + num_reg = x86_setup_var_mtrrs(range, nr_range, mtrr_chunk_size, + mtrr_gran_size); + + /* we got new setting in range_state, check it */ + memset(range_new, 0, sizeof(range_new)); + nr_range_new = x86_get_mtrr_mem_range(range_new, 0, + extra_remove_base, + extra_remove_size); + range_sums_new = sum_ranges(range_new, nr_range_new); + + i = 0; + result[i].chunk_sizek = mtrr_chunk_size >> 10; + result[i].gran_sizek = mtrr_gran_size >> 10; + result[i].num_reg = num_reg; + if (range_sums < range_sums_new) { + result[i].lose_cover_sizek = + (range_sums_new - range_sums) << PSHIFT; + result[i].bad = 1; + } else + result[i].lose_cover_sizek = + (range_sums - range_sums_new) << PSHIFT; + + printk(KERN_INFO "%sgran_size: %ldM \tchunk_size: %ldM \t", + result[i].bad?"*BAD*":" ", result[i].gran_sizek >> 10, + result[i].chunk_sizek >> 10); + printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ldM \n", + result[i].num_reg, result[i].bad?"-":"", + result[i].lose_cover_sizek >> 10); + if (!result[i].bad) { + set_var_mtrr_all(address_bits); + return 1; + } + printk(KERN_INFO "invalid mtrr_gran_size or mtrr_chunk_size, " + "will find optimal one\n"); + debug_print = 0; + memset(result, 0, sizeof(result[0])); + } + + i = 0; + memset(min_loss_pfn, 0xff, sizeof(min_loss_pfn)); + memset(result, 0, sizeof(result)); + for (gran_size = (1ULL<<20); gran_size < (1ULL<<32); gran_size <<= 1) { + for (chunk_size = gran_size; chunk_size < (1ULL<<33); + chunk_size <<= 1) { + int num_reg; + + if (debug_print) + printk(KERN_INFO + "\ngran_size: %lldM chunk_size_size: %lldM\n", + gran_size >> 20, chunk_size >> 20); + if (i >= NUM_RESULT) + continue; + + /* convert ranges to var ranges state */ + num_reg = x86_setup_var_mtrrs(range, nr_range, + chunk_size, gran_size); + + /* we got new setting in range_state, check it */ + memset(range_new, 0, sizeof(range_new)); + nr_range_new = x86_get_mtrr_mem_range(range_new, 0, + extra_remove_base, extra_remove_size); + range_sums_new = sum_ranges(range_new, nr_range_new); + + result[i].chunk_sizek = chunk_size >> 10; + result[i].gran_sizek = gran_size >> 10; + result[i].num_reg = num_reg; + if (range_sums < range_sums_new) { + result[i].lose_cover_sizek = + (range_sums_new - range_sums) << PSHIFT; + result[i].bad = 1; + } else + result[i].lose_cover_sizek = + (range_sums - range_sums_new) << PSHIFT; + + /* double check it */ + if (!result[i].bad && !result[i].lose_cover_sizek) { + if (nr_range_new != nr_range || + memcmp(range, range_new, sizeof(range))) + result[i].bad = 1; + } + + if (!result[i].bad && (range_sums - range_sums_new < + min_loss_pfn[num_reg])) { + min_loss_pfn[num_reg] = + range_sums - range_sums_new; + } + i++; + } + } + + /* print out all */ + for (i = 0; i < NUM_RESULT; i++) { + printk(KERN_INFO "%sgran_size: %ldM \tchunk_size: %ldM \t", + result[i].bad?"*BAD* ":" ", result[i].gran_sizek >> 10, + result[i].chunk_sizek >> 10); + printk(KERN_CONT "num_reg: %d \tlose RAM: %s%ldM\n", + result[i].num_reg, result[i].bad?"-":"", + result[i].lose_cover_sizek >> 10); + } + + /* try to find the optimal index */ + if (nr_mtrr_spare_reg >= num_var_ranges) + nr_mtrr_spare_reg = num_var_ranges - 1; + num_reg_good = -1; + for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) { + if (!min_loss_pfn[i]) { + num_reg_good = i; + break; + } + } + + index_good = -1; + if (num_reg_good != -1) { + for (i = 0; i < NUM_RESULT; i++) { + if (!result[i].bad && + result[i].num_reg == num_reg_good && + !result[i].lose_cover_sizek) { + index_good = i; + break; + } + } + } + + if (index_good != -1) { + printk(KERN_INFO "Found optimal setting for mtrr clean up\n"); + i = index_good; + printk(KERN_INFO "gran_size: %ldM \tchunk_size: %ldM \t", + result[i].gran_sizek >> 10, + result[i].chunk_sizek >> 10); + printk(KERN_CONT "num_reg: %d \tlose RAM: %ldM\n", + result[i].num_reg, + result[i].lose_cover_sizek >> 10); + /* convert ranges to var ranges state */ + chunk_size = result[i].chunk_sizek; + chunk_size <<= 10; + gran_size = result[i].gran_sizek; + gran_size <<= 10; + debug_print = 1; + x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size); + set_var_mtrr_all(address_bits); + return 1; + } + + printk(KERN_INFO "mtrr_cleanup: can not find optimal value\n"); + printk(KERN_INFO "please specify mtrr_gran_size/mtrr_chunk_size\n"); + + return 0; +} +#else +static int __init mtrr_cleanup(unsigned address_bits) +{ + return 0; +} +#endif + +static int __initdata changed_by_mtrr_cleanup; + static int disable_mtrr_trim; static int __init disable_mtrr_trim_setup(char *str) @@ -648,6 +1430,19 @@ int __init amd_special_default_mtrr(void) return 0; } +static u64 __init real_trim_memory(unsigned long start_pfn, + unsigned long limit_pfn) +{ + u64 trim_start, trim_size; + trim_start = start_pfn; + trim_start <<= PAGE_SHIFT; + trim_size = limit_pfn; + trim_size <<= PAGE_SHIFT; + trim_size -= trim_start; + + return update_memory_range(trim_start, trim_size, E820_RAM, + E820_RESERVED); +} /** * mtrr_trim_uncached_memory - trim RAM not covered by MTRRs * @end_pfn: ending page frame number @@ -663,8 +1458,11 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) { unsigned long i, base, size, highest_pfn = 0, def, dummy; mtrr_type type; - u64 trim_start, trim_size; + int nr_range; + u64 total_trim_size; + /* extra one for all 0 */ + int num[MTRR_NUM_TYPES + 1]; /* * Make sure we only trim uncachable memory on machines that * support the Intel MTRR architecture: @@ -676,14 +1474,22 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) if (def != MTRR_TYPE_UNCACHABLE) return 0; - if (amd_special_default_mtrr()) - return 0; + /* get it and store it aside */ + memset(range_state, 0, sizeof(range_state)); + for (i = 0; i < num_var_ranges; i++) { + mtrr_if->get(i, &base, &size, &type); + range_state[i].base_pfn = base; + range_state[i].size_pfn = size; + range_state[i].type = type; + } /* Find highest cached pfn */ for (i = 0; i < num_var_ranges; i++) { - mtrr_if->get(i, &base, &size, &type); + type = range_state[i].type; if (type != MTRR_TYPE_WRBACK) continue; + base = range_state[i].base_pfn; + size = range_state[i].size_pfn; if (highest_pfn < base + size) highest_pfn = base + size; } @@ -698,22 +1504,65 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) return 0; } - if (highest_pfn < end_pfn) { + /* check entries number */ + memset(num, 0, sizeof(num)); + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type >= MTRR_NUM_TYPES) + continue; + size = range_state[i].size_pfn; + if (!size) + type = MTRR_NUM_TYPES; + num[type]++; + } + + /* no entry for WB? */ + if (!num[MTRR_TYPE_WRBACK]) + return 0; + + /* check if we only had WB and UC */ + if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != + num_var_ranges - num[MTRR_NUM_TYPES]) + return 0; + + memset(range, 0, sizeof(range)); + nr_range = 0; + if (mtrr_tom2) { + range[nr_range].start = (1ULL<<(32 - PAGE_SHIFT)); + range[nr_range].end = (mtrr_tom2 >> PAGE_SHIFT) - 1; + if (highest_pfn < range[nr_range].end + 1) + highest_pfn = range[nr_range].end + 1; + nr_range++; + } + nr_range = x86_get_mtrr_mem_range(range, nr_range, 0, 0); + + total_trim_size = 0; + /* check the head */ + if (range[0].start) + total_trim_size += real_trim_memory(0, range[0].start); + /* check the holes */ + for (i = 0; i < nr_range - 1; i++) { + if (range[i].end + 1 < range[i+1].start) + total_trim_size += real_trim_memory(range[i].end + 1, + range[i+1].start); + } + /* check the top */ + i = nr_range - 1; + if (range[i].end + 1 < end_pfn) + total_trim_size += real_trim_memory(range[i].end + 1, + end_pfn); + + if (total_trim_size) { printk(KERN_WARNING "WARNING: BIOS bug: CPU MTRRs don't cover" - " all of memory, losing %luMB of RAM.\n", - (end_pfn - highest_pfn) >> (20 - PAGE_SHIFT)); + " all of memory, losing %lluMB of RAM.\n", + total_trim_size >> 20); - WARN_ON(1); + if (!changed_by_mtrr_cleanup) + WARN_ON(1); printk(KERN_INFO "update e820 for mtrr\n"); - trim_start = highest_pfn; - trim_start <<= PAGE_SHIFT; - trim_size = end_pfn; - trim_size <<= PAGE_SHIFT; - trim_size -= trim_start; - update_memory_range(trim_start, trim_size, E820_RAM, - E820_RESERVED); update_e820(); + return 1; } @@ -729,18 +1578,21 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) */ void __init mtrr_bp_init(void) { + u32 phys_addr; init_ifs(); + phys_addr = 32; + if (cpu_has_mtrr) { mtrr_if = &generic_mtrr_ops; size_or_mask = 0xff000000; /* 36 bits */ size_and_mask = 0x00f00000; + phys_addr = 36; /* This is an AMD specific MSR, but we assume(hope?) that Intel will implement it to when they extend the address bus of the Xeon. */ if (cpuid_eax(0x80000000) >= 0x80000008) { - u32 phys_addr; phys_addr = cpuid_eax(0x80000008) & 0xff; /* CPUID workaround for Intel 0F33/0F34 CPU */ if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && @@ -758,6 +1610,7 @@ void __init mtrr_bp_init(void) don't support PAE */ size_or_mask = 0xfff00000; /* 32 bits */ size_and_mask = 0; + phys_addr = 32; } } else { switch (boot_cpu_data.x86_vendor) { @@ -791,8 +1644,15 @@ void __init mtrr_bp_init(void) if (mtrr_if) { set_num_var_ranges(); init_table(); - if (use_intel()) + if (use_intel()) { get_mtrr_state(); + + if (mtrr_cleanup(phys_addr)) { + changed_by_mtrr_cleanup = 1; + mtrr_if->set_all(); + } + + } } } @@ -829,9 +1689,10 @@ static int __init mtrr_init_finialize(void) { if (!mtrr_if) return 0; - if (use_intel()) - mtrr_state_warn(); - else { + if (use_intel()) { + if (!changed_by_mtrr_cleanup) + mtrr_state_warn(); + } else { /* The CPUs haven't MTRR and seem to not support SMP. They have * specific drivers, we use a tricky method to support * suspend/resume for them. diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h index 2cc77eb6fea..2dc4ec656b2 100644 --- a/arch/x86/kernel/cpu/mtrr/mtrr.h +++ b/arch/x86/kernel/cpu/mtrr/mtrr.h @@ -81,6 +81,8 @@ void set_mtrr_done(struct set_mtrr_context *ctxt); void set_mtrr_cache_disable(struct set_mtrr_context *ctxt); void set_mtrr_prepare_save(struct set_mtrr_context *ctxt); +void fill_mtrr_var_range(unsigned int index, + u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi); void get_mtrr_state(void); extern void set_mtrr_ops(struct mtrr_ops * ops); @@ -92,6 +94,7 @@ extern struct mtrr_ops * mtrr_if; #define use_intel() (mtrr_if && mtrr_if->use_intel_if == 1) extern unsigned int num_var_ranges; +extern u64 mtrr_tom2; void mtrr_state_warn(void); const char *mtrr_attrib_to_str(int x); diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c new file mode 100644 index 00000000000..a706e9057ba --- /dev/null +++ b/arch/x86/kernel/e820.c @@ -0,0 +1,896 @@ +/* + * Handle the memory map. + * The functions here do the job until bootmem takes over. + * + * Getting sanitize_e820_map() in sync with i386 version by applying change: + * - Provisions for empty E820 memory regions (reported by certain BIOSes). + * Alex Achenbach <xela@slit.de>, December 2002. + * Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> + * + */ +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/bootmem.h> +#include <linux/ioport.h> +#include <linux/string.h> +#include <linux/kexec.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/pfn.h> +#include <linux/suspend.h> + +#include <asm/pgtable.h> +#include <asm/page.h> +#include <asm/e820.h> +#include <asm/proto.h> +#include <asm/setup.h> +#include <asm/trampoline.h> + +struct e820map e820; + +/* For PCI or other memory-mapped resources */ +unsigned long pci_mem_start = 0xaeedbabe; +#ifdef CONFIG_PCI +EXPORT_SYMBOL(pci_mem_start); +#endif + +/* + * This function checks if any part of the range <start,end> is mapped + * with type. + */ +int +e820_any_mapped(u64 start, u64 end, unsigned type) +{ + int i; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + + if (type && ei->type != type) + continue; + if (ei->addr >= end || ei->addr + ei->size <= start) + continue; + return 1; + } + return 0; +} +EXPORT_SYMBOL_GPL(e820_any_mapped); + +/* + * This function checks if the entire range <start,end> is mapped with type. + * + * Note: this function only works correct if the e820 table is sorted and + * not-overlapping, which is the case + */ +int __init e820_all_mapped(u64 start, u64 end, unsigned type) +{ + int i; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + + if (type && ei->type != type) + continue; + /* is the region (part) in overlap with the current region ?*/ + if (ei->addr >= end || ei->addr + ei->size <= start) + continue; + + /* if the region is at the beginning of <start,end> we move + * start to the end of the region since it's ok until there + */ + if (ei->addr <= start) + start = ei->addr + ei->size; + /* + * if start is now at or beyond end, we're done, full + * coverage + */ + if (start >= end) + return 1; + } + return 0; +} + +/* + * Add a memory region to the kernel e820 map. + */ +void __init add_memory_region(u64 start, u64 size, int type) +{ + int x = e820.nr_map; + + if (x == ARRAY_SIZE(e820.map)) { + printk(KERN_ERR "Ooops! Too many entries in the memory map!\n"); + return; + } + + e820.map[x].addr = start; + e820.map[x].size = size; + e820.map[x].type = type; + e820.nr_map++; +} + +void __init e820_print_map(char *who) +{ + int i; + + for (i = 0; i < e820.nr_map; i++) { + printk(KERN_INFO " %s: %016Lx - %016Lx ", who, + (unsigned long long) e820.map[i].addr, + (unsigned long long) + (e820.map[i].addr + e820.map[i].size)); + switch (e820.map[i].type) { + case E820_RAM: + printk(KERN_CONT "(usable)\n"); + break; + case E820_RESERVED: + printk(KERN_CONT "(reserved)\n"); + break; + case E820_ACPI: + printk(KERN_CONT "(ACPI data)\n"); + break; + case E820_NVS: + printk(KERN_CONT "(ACPI NVS)\n"); + break; + default: + printk(KERN_CONT "type %u\n", e820.map[i].type); + break; + } + } +} + +/* + * Sanitize the BIOS e820 map. + * + * Some e820 responses include overlapping entries. The following + * replaces the original e820 map with a new one, removing overlaps, + * and resolving conflicting memory types in favor of highest + * numbered type. + * + * The input parameter biosmap points to an array of 'struct + * e820entry' which on entry has elements in the range [0, *pnr_map) + * valid, and which has space for up to max_nr_map entries. + * On return, the resulting sanitized e820 map entries will be in + * overwritten in the same location, starting at biosmap. + * + * The integer pointed to by pnr_map must be valid on entry (the + * current number of valid entries located at biosmap) and will + * be updated on return, with the new number of valid entries + * (something no more than max_nr_map.) + * + * The return value from sanitize_e820_map() is zero if it + * successfully 'sanitized' the map entries passed in, and is -1 + * if it did nothing, which can happen if either of (1) it was + * only passed one map entry, or (2) any of the input map entries + * were invalid (start + size < start, meaning that the size was + * so big the described memory range wrapped around through zero.) + * + * Visually we're performing the following + * (1,2,3,4 = memory types)... + * + * Sample memory map (w/overlaps): + * ____22__________________ + * ______________________4_ + * ____1111________________ + * _44_____________________ + * 11111111________________ + * ____________________33__ + * ___________44___________ + * __________33333_________ + * ______________22________ + * ___________________2222_ + * _________111111111______ + * _____________________11_ + * _________________4______ + * + * Sanitized equivalent (no overlap): + * 1_______________________ + * _44_____________________ + * ___1____________________ + * ____22__________________ + * ______11________________ + * _________1______________ + * __________3_____________ + * ___________44___________ + * _____________33_________ + * _______________2________ + * ________________1_______ + * _________________4______ + * ___________________2____ + * ____________________33__ + * ______________________4_ + */ + +int __init sanitize_e820_map(struct e820entry *biosmap, int max_nr_map, + int *pnr_map) +{ + struct change_member { + struct e820entry *pbios; /* pointer to original bios entry */ + unsigned long long addr; /* address for this change point */ + }; +static struct change_member change_point_list[2*E820_X_MAX] __initdata; +static struct change_member *change_point[2*E820_X_MAX] __initdata; +static struct e820entry *overlap_list[E820_X_MAX] __initdata; +static struct e820entry new_bios[E820_X_MAX] __initdata; + struct change_member *change_tmp; + unsigned long current_type, last_type; + unsigned long long last_addr; + int chgidx, still_changing; + int overlap_entries; + int new_bios_entry; + int old_nr, new_nr, chg_nr; + int i; + + /* if there's only one memory region, don't bother */ + if (*pnr_map < 2) + return -1; + + old_nr = *pnr_map; + BUG_ON(old_nr > max_nr_map); + + /* bail out if we find any unreasonable addresses in bios map */ + for (i = 0; i < old_nr; i++) + if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr) + return -1; + + /* create pointers for initial change-point information (for sorting) */ + for (i = 0; i < 2 * old_nr; i++) + change_point[i] = &change_point_list[i]; + + /* record all known change-points (starting and ending addresses), + omitting those that are for empty memory regions */ + chgidx = 0; + for (i = 0; i < old_nr; i++) { + if (biosmap[i].size != 0) { + change_point[chgidx]->addr = biosmap[i].addr; + change_point[chgidx++]->pbios = &biosmap[i]; + change_point[chgidx]->addr = biosmap[i].addr + + biosmap[i].size; + change_point[chgidx++]->pbios = &biosmap[i]; + } + } + chg_nr = chgidx; + + /* sort change-point list by memory addresses (low -> high) */ + still_changing = 1; + while (still_changing) { + still_changing = 0; + for (i = 1; i < chg_nr; i++) { + unsigned long long curaddr, lastaddr; + unsigned long long curpbaddr, lastpbaddr; + + curaddr = change_point[i]->addr; + lastaddr = change_point[i - 1]->addr; + curpbaddr = change_point[i]->pbios->addr; + lastpbaddr = change_point[i - 1]->pbios->addr; + + /* + * swap entries, when: + * + * curaddr > lastaddr or + * curaddr == lastaddr and curaddr == curpbaddr and + * lastaddr != lastpbaddr + */ + if (curaddr < lastaddr || + (curaddr == lastaddr && curaddr == curpbaddr && + lastaddr != lastpbaddr)) { + change_tmp = change_point[i]; + change_point[i] = change_point[i-1]; + change_point[i-1] = change_tmp; + still_changing = 1; + } + } + } + + /* create a new bios memory map, removing overlaps */ + overlap_entries = 0; /* number of entries in the overlap table */ + new_bios_entry = 0; /* index for creating new bios map entries */ + last_type = 0; /* start with undefined memory type */ + last_addr = 0; /* start with 0 as last starting address */ + + /* loop through change-points, determining affect on the new bios map */ + for (chgidx = 0; chgidx < chg_nr; chgidx++) { + /* keep track of all overlapping bios entries */ + if (change_point[chgidx]->addr == + change_point[chgidx]->pbios->addr) { + /* + * add map entry to overlap list (> 1 entry + * implies an overlap) + */ + overlap_list[overlap_entries++] = + change_point[chgidx]->pbios; + } else { + /* + * remove entry from list (order independent, + * so swap with last) + */ + for (i = 0; i < overlap_entries; i++) { + if (overlap_list[i] == + change_point[chgidx]->pbios) + overlap_list[i] = + overlap_list[overlap_entries-1]; + } + overlap_entries--; + } + /* + * if there are overlapping entries, decide which + * "type" to use (larger value takes precedence -- + * 1=usable, 2,3,4,4+=unusable) + */ + current_type = 0; + for (i = 0; i < overlap_entries; i++) + if (overlap_list[i]->type > current_type) + current_type = overlap_list[i]->type; + /* + * continue building up new bios map based on this + * information + */ + if (current_type != last_type) { + if (last_type != 0) { + new_bios[new_bios_entry].size = + change_point[chgidx]->addr - last_addr; + /* + * move forward only if the new size + * was non-zero + */ + if (new_bios[new_bios_entry].size != 0) + /* + * no more space left for new + * bios entries ? + */ + if (++new_bios_entry >= max_nr_map) + break; + } + if (current_type != 0) { + new_bios[new_bios_entry].addr = + change_point[chgidx]->addr; + new_bios[new_bios_entry].type = current_type; + last_addr = change_point[chgidx]->addr; + } + last_type = current_type; + } + } + /* retain count for new bios entries */ + new_nr = new_bios_entry; + + /* copy new bios mapping into original location */ + memcpy(biosmap, new_bios, new_nr * sizeof(struct e820entry)); + *pnr_map = new_nr; + + return 0; +} + +/* + * Copy the BIOS e820 map into a safe place. + * + * Sanity-check it while we're at it.. + * + * If we're lucky and live on a modern system, the setup code + * will have given us a memory map that we can use to properly + * set up memory. If we aren't, we'll fake a memory map. + */ +int __init copy_e820_map(struct e820entry *biosmap, int nr_map) +{ + /* Only one memory region (or negative)? Ignore it */ + if (nr_map < 2) + return -1; + + do { + u64 start = biosmap->addr; + u64 size = biosmap->size; + u64 end = start + size; + u32 type = biosmap->type; + + /* Overflow in 64 bits? Ignore the memory map. */ + if (start > end) + return -1; + + add_memory_region(start, size, type); + } while (biosmap++, --nr_map); + return 0; +} + +u64 __init update_memory_range(u64 start, u64 size, unsigned old_type, + unsigned new_type) +{ + int i; + u64 real_updated_size = 0; + + BUG_ON(old_type == new_type); + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + u64 final_start, final_end; + if (ei->type != old_type) + continue; + /* totally covered? */ + if (ei->addr >= start && + (ei->addr + ei->size) <= (start + size)) { + ei->type = new_type; + real_updated_size += ei->size; + continue; + } + /* partially covered */ + final_start = max(start, ei->addr); + final_end = min(start + size, ei->addr + ei->size); + if (final_start >= final_end) + continue; + add_memory_region(final_start, final_end - final_start, + new_type); + real_updated_size += final_end - final_start; + } + return real_updated_size; +} + +void __init update_e820(void) +{ + int nr_map; + + nr_map = e820.nr_map; + if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr_map)) + return; + e820.nr_map = nr_map; + printk(KERN_INFO "modified physical RAM map:\n"); + e820_print_map("modified"); +} + +/* + * Search for the biggest gap in the low 32 bits of the e820 + * memory space. We pass this space to PCI to assign MMIO resources + * for hotplug or unconfigured devices in. + * Hopefully the BIOS let enough space left. + */ +__init void e820_setup_gap(void) +{ + unsigned long gapstart, gapsize, round; + unsigned long long last; + int i; + int found = 0; + + last = 0x100000000ull; + gapstart = 0x10000000; + gapsize = 0x400000; + i = e820.nr_map; + while (--i >= 0) { + unsigned long long start = e820.map[i].addr; + unsigned long long end = start + e820.map[i].size; + + /* + * Since "last" is at most 4GB, we know we'll + * fit in 32 bits if this condition is true + */ + if (last > end) { + unsigned long gap = last - end; + + if (gap > gapsize) { + gapsize = gap; + gapstart = end; + found = 1; + } + } + if (start < last) + last = start; + } + +#ifdef CONFIG_X86_64 + if (!found) { + gapstart = (end_pfn << PAGE_SHIFT) + 1024*1024; + printk(KERN_ERR "PCI: Warning: Cannot find a gap in the 32bit " + "address range\n" + KERN_ERR "PCI: Unassigned devices with 32bit resource " + "registers may break!\n"); + } +#endif + + /* + * See how much we want to round up: start off with + * rounding to the next 1MB area. + */ + round = 0x100000; + while ((gapsize >> 4) > round) + round += round; + /* Fun with two's complement */ + pci_mem_start = (gapstart + round) & -round; + + printk(KERN_INFO + "Allocating PCI resources starting at %lx (gap: %lx:%lx)\n", + pci_mem_start, gapstart, gapsize); +} + +#if defined(CONFIG_X86_64) || \ + (defined(CONFIG_X86_32) && defined(CONFIG_HIBERNATION)) +/** + * Find the ranges of physical addresses that do not correspond to + * e820 RAM areas and mark the corresponding pages as nosave for + * hibernation (32 bit) or software suspend and suspend to RAM (64 bit). + * + * This function requires the e820 map to be sorted and without any + * overlapping entries and assumes the first e820 area to be RAM. + */ +void __init e820_mark_nosave_regions(unsigned long limit_pfn) +{ + int i; + unsigned long pfn; + + pfn = PFN_DOWN(e820.map[0].addr + e820.map[0].size); + for (i = 1; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + + if (pfn < PFN_UP(ei->addr)) + register_nosave_region(pfn, PFN_UP(ei->addr)); + + pfn = PFN_DOWN(ei->addr + ei->size); + if (ei->type != E820_RAM) + register_nosave_region(PFN_UP(ei->addr), pfn); + + if (pfn >= limit_pfn) + break; + } +} +#endif + +/* + * Early reserved memory areas. + */ +#define MAX_EARLY_RES 20 + +struct early_res { + u64 start, end; + char name[16]; +}; +static struct early_res early_res[MAX_EARLY_RES] __initdata = { + { 0, PAGE_SIZE, "BIOS data page" }, /* BIOS data page */ +#if defined(CONFIG_X86_64) && defined(CONFIG_X86_TRAMPOLINE) + { TRAMPOLINE_BASE, TRAMPOLINE_BASE + 2 * PAGE_SIZE, "TRAMPOLINE" }, +#endif +#if defined(CONFIG_X86_32) && defined(CONFIG_SMP) + /* + * But first pinch a few for the stack/trampoline stuff + * FIXME: Don't need the extra page at 4K, but need to fix + * trampoline before removing it. (see the GDT stuff) + */ + { PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE" }, + /* + * Has to be in very low memory so we can execute + * real-mode AP code. + */ + { TRAMPOLINE_BASE, TRAMPOLINE_BASE + PAGE_SIZE, "TRAMPOLINE" }, +#endif + {} +}; + +static int __init find_overlapped_early(u64 start, u64 end) +{ + int i; + struct early_res *r; + + for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { + r = &early_res[i]; + if (end > r->start && start < r->end) + break; + } + + return i; +} + +void __init reserve_early(u64 start, u64 end, char *name) +{ + int i; + struct early_res *r; + + i = find_overlapped_early(start, end); + if (i >= MAX_EARLY_RES) + panic("Too many early reservations"); + r = &early_res[i]; + if (r->end) + panic("Overlapping early reservations " + "%llx-%llx %s to %llx-%llx %s\n", + start, end - 1, name?name:"", r->start, + r->end - 1, r->name); + r->start = start; + r->end = end; + if (name) + strncpy(r->name, name, sizeof(r->name) - 1); +} + +void __init free_early(u64 start, u64 end) +{ + struct early_res *r; + int i, j; + + i = find_overlapped_early(start, end); + r = &early_res[i]; + if (i >= MAX_EARLY_RES || r->end != end || r->start != start) + panic("free_early on not reserved area: %llx-%llx!", + start, end - 1); + + for (j = i + 1; j < MAX_EARLY_RES && early_res[j].end; j++) + ; + + memmove(&early_res[i], &early_res[i + 1], + (j - 1 - i) * sizeof(struct early_res)); + + early_res[j - 1].end = 0; +} + +int __init page_is_reserved_early(unsigned long pagenr) +{ + u64 start = (u64)pagenr << PAGE_SHIFT; + int i; + struct early_res *r; + + i = find_overlapped_early(start, start + PAGE_SIZE); + r = &early_res[i]; + return (i < MAX_EARLY_RES && r->end); +} + +void __init early_res_to_bootmem(u64 start, u64 end) +{ + int i; + u64 final_start, final_end; + for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { + struct early_res *r = &early_res[i]; + final_start = max(start, r->start); + final_end = min(end, r->end); + if (final_start >= final_end) + continue; + printk(KERN_INFO " early res: %d [%llx-%llx] %s\n", i, + final_start, final_end - 1, r->name); +#ifdef CONFIG_X86_64 + reserve_bootmem_generic(final_start, final_end - final_start); +#else + reserve_bootmem(final_start, final_end - final_start, + BOOTMEM_DEFAULT); +#endif + } +} + +/* Check for already reserved areas */ +static inline int __init bad_addr(u64 *addrp, u64 size, u64 align) +{ + int i; + u64 addr = *addrp; + int changed = 0; + struct early_res *r; +again: + i = find_overlapped_early(addr, addr + size); + r = &early_res[i]; + if (i < MAX_EARLY_RES && r->end) { + *addrp = addr = round_up(r->end, align); + changed = 1; + goto again; + } + return changed; +} + +/* Check for already reserved areas */ +static inline int __init bad_addr_size(u64 *addrp, u64 *sizep, u64 align) +{ + int i; + u64 addr = *addrp, last; + u64 size = *sizep; + int changed = 0; +again: + last = addr + size; + for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { + struct early_res *r = &early_res[i]; + if (last > r->start && addr < r->start) { + size = r->start - addr; + changed = 1; + goto again; + } + if (last > r->end && addr < r->end) { + addr = round_up(r->end, align); + size = last - addr; + changed = 1; + goto again; + } + if (last <= r->end && addr >= r->start) { + (*sizep)++; + return 0; + } + } + if (changed) { + *addrp = addr; + *sizep = size; + } + return changed; +} + +/* + * Find a free area with specified alignment in a specific range. + */ +u64 __init find_e820_area(u64 start, u64 end, u64 size, u64 align) +{ + int i; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + u64 addr, last; + u64 ei_last; + + if (ei->type != E820_RAM) + continue; + addr = round_up(ei->addr, align); + ei_last = ei->addr + ei->size; + if (addr < start) + addr = round_up(start, align); + if (addr >= ei_last) + continue; + while (bad_addr(&addr, size, align) && addr+size <= ei_last) + ; + last = addr + size; + if (last > ei_last) + continue; + if (last > end) + continue; + return addr; + } + return -1ULL; +} + +/* + * Find next free range after *start + */ +u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align) +{ + int i; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + u64 addr, last; + u64 ei_last; + + if (ei->type != E820_RAM) + continue; + addr = round_up(ei->addr, align); + ei_last = ei->addr + ei->size; + if (addr < start) + addr = round_up(start, align); + if (addr >= ei_last) + continue; + *sizep = ei_last - addr; + while (bad_addr_size(&addr, sizep, align) && + addr + *sizep <= ei_last) + ; + last = addr + *sizep; + if (last > ei_last) + continue; + return addr; + } + return -1UL; + +} + +/* + * pre allocated 4k and reserved it in e820 + */ +u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) +{ + u64 size = 0; + u64 addr; + u64 start; + + start = startt; + while (size < sizet) + start = find_e820_area_size(start, &size, align); + + if (size < sizet) + return 0; + + addr = round_down(start + size - sizet, align); + update_memory_range(addr, sizet, E820_RAM, E820_RESERVED); + printk(KERN_INFO "update e820 for early_reserve_e820\n"); + update_e820(); + + return addr; +} + +#ifdef CONFIG_X86_32 +# ifdef CONFIG_X86_PAE +# define MAX_ARCH_PFN (1ULL<<(36-PAGE_SHIFT)) +# else +# define MAX_ARCH_PFN (1ULL<<(32-PAGE_SHIFT)) +# endif +#else /* CONFIG_X86_32 */ +# define MAX_ARCH_PFN MAXMEM>>PAGE_SHIFT +#endif + +/* + * Last pfn which the user wants to use. + */ +unsigned long __initdata end_user_pfn = MAX_ARCH_PFN; + +/* + * Find the highest page frame number we have available + */ +unsigned long __init e820_end_of_ram(void) +{ + unsigned long last_pfn; + unsigned long max_arch_pfn = MAX_ARCH_PFN; + + last_pfn = find_max_pfn_with_active_regions(); + + if (last_pfn > max_arch_pfn) + last_pfn = max_arch_pfn; + if (last_pfn > end_user_pfn) + last_pfn = end_user_pfn; + + printk(KERN_INFO "last_pfn = %lu max_arch_pfn = %lu\n", + last_pfn, max_arch_pfn); + return last_pfn; +} + +/* + * Finds an active region in the address range from start_pfn to last_pfn and + * returns its range in ei_startpfn and ei_endpfn for the e820 entry. + */ +int __init e820_find_active_region(const struct e820entry *ei, + unsigned long start_pfn, + unsigned long last_pfn, + unsigned long *ei_startpfn, + unsigned long *ei_endpfn) +{ + u64 align = PAGE_SIZE; + + *ei_startpfn = round_up(ei->addr, align) >> PAGE_SHIFT; + *ei_endpfn = round_down(ei->addr + ei->size, align) >> PAGE_SHIFT; + + /* Skip map entries smaller than a page */ + if (*ei_startpfn >= *ei_endpfn) + return 0; + + /* Skip if map is outside the node */ + if (ei->type != E820_RAM || *ei_endpfn <= start_pfn || + *ei_startpfn >= last_pfn) + return 0; + + /* Check for overlaps */ + if (*ei_startpfn < start_pfn) + *ei_startpfn = start_pfn; + if (*ei_endpfn > last_pfn) + *ei_endpfn = last_pfn; + + /* Obey end_user_pfn to save on memmap */ + if (*ei_startpfn >= end_user_pfn) + return 0; + if (*ei_endpfn > end_user_pfn) + *ei_endpfn = end_user_pfn; + + return 1; +} + +/* Walk the e820 map and register active regions within a node */ +void __init e820_register_active_regions(int nid, unsigned long start_pfn, + unsigned long last_pfn) +{ + unsigned long ei_startpfn; + unsigned long ei_endpfn; + int i; + + for (i = 0; i < e820.nr_map; i++) + if (e820_find_active_region(&e820.map[i], + start_pfn, last_pfn, + &ei_startpfn, &ei_endpfn)) + add_active_range(nid, ei_startpfn, ei_endpfn); +} + +/* + * Find the hole size (in bytes) in the memory range. + * @start: starting address of the memory range to scan + * @end: ending address of the memory range to scan + */ +u64 __init e820_hole_size(u64 start, u64 end) +{ + unsigned long start_pfn = start >> PAGE_SHIFT; + unsigned long last_pfn = end >> PAGE_SHIFT; + unsigned long ei_startpfn, ei_endpfn, ram = 0; + int i; + + for (i = 0; i < e820.nr_map; i++) { + if (e820_find_active_region(&e820.map[i], + start_pfn, last_pfn, + &ei_startpfn, &ei_endpfn)) + ram += ei_endpfn - ei_startpfn; + } + return end - start - ((u64)ram << PAGE_SHIFT); +} diff --git a/arch/x86/kernel/e820_32.c b/arch/x86/kernel/e820_32.c index ed733e7cf4e..e8a3b968c9f 100644 --- a/arch/x86/kernel/e820_32.c +++ b/arch/x86/kernel/e820_32.c @@ -9,29 +9,12 @@ #include <linux/mm.h> #include <linux/pfn.h> #include <linux/uaccess.h> -#include <linux/suspend.h> #include <asm/pgtable.h> #include <asm/page.h> #include <asm/e820.h> #include <asm/setup.h> -struct e820map e820; -struct change_member { - struct e820entry *pbios; /* pointer to original bios entry */ - unsigned long long addr; /* address for this change point */ -}; -static struct change_member change_point_list[2*E820MAX] __initdata; -static struct change_member *change_point[2*E820MAX] __initdata; -static struct e820entry *overlap_list[E820MAX] __initdata; -static struct e820entry new_bios[E820MAX] __initdata; -/* For PCI or other memory-mapped resources */ -unsigned long pci_mem_start = 0x10000000; -#ifdef CONFIG_PCI -EXPORT_SYMBOL(pci_mem_start); -#endif -extern int user_defined_memmap; - static struct resource system_rom_resource = { .name = "System ROM", .start = 0xf0000, @@ -224,398 +207,12 @@ void __init init_iomem_resources(struct resource *code_resource, } } -#if defined(CONFIG_PM) && defined(CONFIG_HIBERNATION) -/** - * e820_mark_nosave_regions - Find the ranges of physical addresses that do not - * correspond to e820 RAM areas and mark the corresponding pages as nosave for - * hibernation. - * - * This function requires the e820 map to be sorted and without any - * overlapping entries and assumes the first e820 area to be RAM. - */ -void __init e820_mark_nosave_regions(void) -{ - int i; - unsigned long pfn; - - pfn = PFN_DOWN(e820.map[0].addr + e820.map[0].size); - for (i = 1; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - - if (pfn < PFN_UP(ei->addr)) - register_nosave_region(pfn, PFN_UP(ei->addr)); - - pfn = PFN_DOWN(ei->addr + ei->size); - if (ei->type != E820_RAM) - register_nosave_region(PFN_UP(ei->addr), pfn); - - if (pfn >= max_low_pfn) - break; - } -} -#endif - -void __init add_memory_region(unsigned long long start, - unsigned long long size, int type) -{ - int x; - - x = e820.nr_map; - - if (x == E820MAX) { - printk(KERN_ERR "Ooops! Too many entries in the memory map!\n"); - return; - } - - e820.map[x].addr = start; - e820.map[x].size = size; - e820.map[x].type = type; - e820.nr_map++; -} /* add_memory_region */ - -/* - * Sanitize the BIOS e820 map. - * - * Some e820 responses include overlapping entries. The following - * replaces the original e820 map with a new one, removing overlaps. - * - */ -int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) -{ - struct change_member *change_tmp; - unsigned long current_type, last_type; - unsigned long long last_addr; - int chgidx, still_changing; - int overlap_entries; - int new_bios_entry; - int old_nr, new_nr, chg_nr; - int i; - - /* - Visually we're performing the following (1,2,3,4 = memory types)... - - Sample memory map (w/overlaps): - ____22__________________ - ______________________4_ - ____1111________________ - _44_____________________ - 11111111________________ - ____________________33__ - ___________44___________ - __________33333_________ - ______________22________ - ___________________2222_ - _________111111111______ - _____________________11_ - _________________4______ - - Sanitized equivalent (no overlap): - 1_______________________ - _44_____________________ - ___1____________________ - ____22__________________ - ______11________________ - _________1______________ - __________3_____________ - ___________44___________ - _____________33_________ - _______________2________ - ________________1_______ - _________________4______ - ___________________2____ - ____________________33__ - ______________________4_ - */ - /* if there's only one memory region, don't bother */ - if (*pnr_map < 2) { - return -1; - } - - old_nr = *pnr_map; - - /* bail out if we find any unreasonable addresses in bios map */ - for (i=0; i<old_nr; i++) - if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr) { - return -1; - } - - /* create pointers for initial change-point information (for sorting) */ - for (i=0; i < 2*old_nr; i++) - change_point[i] = &change_point_list[i]; - - /* record all known change-points (starting and ending addresses), - omitting those that are for empty memory regions */ - chgidx = 0; - for (i=0; i < old_nr; i++) { - if (biosmap[i].size != 0) { - change_point[chgidx]->addr = biosmap[i].addr; - change_point[chgidx++]->pbios = &biosmap[i]; - change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size; - change_point[chgidx++]->pbios = &biosmap[i]; - } - } - chg_nr = chgidx; /* true number of change-points */ - - /* sort change-point list by memory addresses (low -> high) */ - still_changing = 1; - while (still_changing) { - still_changing = 0; - for (i=1; i < chg_nr; i++) { - /* if <current_addr> > <last_addr>, swap */ - /* or, if current=<start_addr> & last=<end_addr>, swap */ - if ((change_point[i]->addr < change_point[i-1]->addr) || - ((change_point[i]->addr == change_point[i-1]->addr) && - (change_point[i]->addr == change_point[i]->pbios->addr) && - (change_point[i-1]->addr != change_point[i-1]->pbios->addr)) - ) - { - change_tmp = change_point[i]; - change_point[i] = change_point[i-1]; - change_point[i-1] = change_tmp; - still_changing=1; - } - } - } - - /* create a new bios memory map, removing overlaps */ - overlap_entries=0; /* number of entries in the overlap table */ - new_bios_entry=0; /* index for creating new bios map entries */ - last_type = 0; /* start with undefined memory type */ - last_addr = 0; /* start with 0 as last starting address */ - /* loop through change-points, determining affect on the new bios map */ - for (chgidx=0; chgidx < chg_nr; chgidx++) - { - /* keep track of all overlapping bios entries */ - if (change_point[chgidx]->addr == change_point[chgidx]->pbios->addr) - { - /* add map entry to overlap list (> 1 entry implies an overlap) */ - overlap_list[overlap_entries++]=change_point[chgidx]->pbios; - } - else - { - /* remove entry from list (order independent, so swap with last) */ - for (i=0; i<overlap_entries; i++) - { - if (overlap_list[i] == change_point[chgidx]->pbios) - overlap_list[i] = overlap_list[overlap_entries-1]; - } - overlap_entries--; - } - /* if there are overlapping entries, decide which "type" to use */ - /* (larger value takes precedence -- 1=usable, 2,3,4,4+=unusable) */ - current_type = 0; - for (i=0; i<overlap_entries; i++) - if (overlap_list[i]->type > current_type) - current_type = overlap_list[i]->type; - /* continue building up new bios map based on this information */ - if (current_type != last_type) { - if (last_type != 0) { - new_bios[new_bios_entry].size = - change_point[chgidx]->addr - last_addr; - /* move forward only if the new size was non-zero */ - if (new_bios[new_bios_entry].size != 0) - if (++new_bios_entry >= E820MAX) - break; /* no more space left for new bios entries */ - } - if (current_type != 0) { - new_bios[new_bios_entry].addr = change_point[chgidx]->addr; - new_bios[new_bios_entry].type = current_type; - last_addr=change_point[chgidx]->addr; - } - last_type = current_type; - } - } - new_nr = new_bios_entry; /* retain count for new bios entries */ - - /* copy new bios mapping into original location */ - memcpy(biosmap, new_bios, new_nr*sizeof(struct e820entry)); - *pnr_map = new_nr; - - return 0; -} - -/* - * Copy the BIOS e820 map into a safe place. - * - * Sanity-check it while we're at it.. - * - * If we're lucky and live on a modern system, the setup code - * will have given us a memory map that we can use to properly - * set up memory. If we aren't, we'll fake a memory map. - * - * We check to see that the memory map contains at least 2 elements - * before we'll use it, because the detection code in setup.S may - * not be perfect and most every PC known to man has two memory - * regions: one from 0 to 640k, and one from 1mb up. (The IBM - * thinkpad 560x, for example, does not cooperate with the memory - * detection code.) - */ -int __init copy_e820_map(struct e820entry *biosmap, int nr_map) -{ - /* Only one memory region (or negative)? Ignore it */ - if (nr_map < 2) - return -1; - - do { - u64 start = biosmap->addr; - u64 size = biosmap->size; - u64 end = start + size; - u32 type = biosmap->type; - - /* Overflow in 64 bits? Ignore the memory map. */ - if (start > end) - return -1; - - add_memory_region(start, size, type); - } while (biosmap++, --nr_map); - - return 0; -} - -/* - * Find the highest page frame number we have available - */ -void __init propagate_e820_map(void) -{ - int i; - - max_pfn = 0; - - for (i = 0; i < e820.nr_map; i++) { - unsigned long start, end; - /* RAM? */ - if (e820.map[i].type != E820_RAM) - continue; - start = PFN_UP(e820.map[i].addr); - end = PFN_DOWN(e820.map[i].addr + e820.map[i].size); - if (start >= end) - continue; - if (end > max_pfn) - max_pfn = end; - memory_present(0, start, end); - } -} - -/* - * Register fully available low RAM pages with the bootmem allocator. - */ -void __init register_bootmem_low_pages(unsigned long max_low_pfn) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - unsigned long curr_pfn, last_pfn, size; - /* - * Reserve usable low memory - */ - if (e820.map[i].type != E820_RAM) - continue; - /* - * We are rounding up the start address of usable memory: - */ - curr_pfn = PFN_UP(e820.map[i].addr); - if (curr_pfn >= max_low_pfn) - continue; - /* - * ... and at the end of the usable range downwards: - */ - last_pfn = PFN_DOWN(e820.map[i].addr + e820.map[i].size); - - if (last_pfn > max_low_pfn) - last_pfn = max_low_pfn; - - /* - * .. finally, did all the rounding and playing - * around just make the area go away? - */ - if (last_pfn <= curr_pfn) - continue; - - size = last_pfn - curr_pfn; - free_bootmem(PFN_PHYS(curr_pfn), PFN_PHYS(size)); - } -} - -void __init e820_register_memory(void) -{ - unsigned long gapstart, gapsize, round; - unsigned long long last; - int i; - - /* - * Search for the biggest gap in the low 32 bits of the e820 - * memory space. - */ - last = 0x100000000ull; - gapstart = 0x10000000; - gapsize = 0x400000; - i = e820.nr_map; - while (--i >= 0) { - unsigned long long start = e820.map[i].addr; - unsigned long long end = start + e820.map[i].size; - - /* - * Since "last" is at most 4GB, we know we'll - * fit in 32 bits if this condition is true - */ - if (last > end) { - unsigned long gap = last - end; - - if (gap > gapsize) { - gapsize = gap; - gapstart = end; - } - } - if (start < last) - last = start; - } - - /* - * See how much we want to round up: start off with - * rounding to the next 1MB area. - */ - round = 0x100000; - while ((gapsize >> 4) > round) - round += round; - /* Fun with two's complement */ - pci_mem_start = (gapstart + round) & -round; - - printk("Allocating PCI resources starting at %08lx (gap: %08lx:%08lx)\n", - pci_mem_start, gapstart, gapsize); -} - -void __init print_memory_map(char *who) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - printk(" %s: %016Lx - %016Lx ", who, - e820.map[i].addr, - e820.map[i].addr + e820.map[i].size); - switch (e820.map[i].type) { - case E820_RAM: printk("(usable)\n"); - break; - case E820_RESERVED: - printk("(reserved)\n"); - break; - case E820_ACPI: - printk("(ACPI data)\n"); - break; - case E820_NVS: - printk("(ACPI NVS)\n"); - break; - default: printk("type %u\n", e820.map[i].type); - break; - } - } -} - void __init limit_regions(unsigned long long size) { unsigned long long current_addr; int i; - print_memory_map("limit_regions start"); + e820_print_map("limit_regions start"); for (i = 0; i < e820.nr_map; i++) { current_addr = e820.map[i].addr + e820.map[i].size; if (current_addr < size) @@ -634,63 +231,59 @@ void __init limit_regions(unsigned long long size) e820.nr_map = i + 1; e820.map[i].size -= current_addr - size; } - print_memory_map("limit_regions endfor"); + e820_print_map("limit_regions endfor"); return; } - print_memory_map("limit_regions endfunc"); + e820_print_map("limit_regions endfunc"); } -/* - * This function checks if any part of the range <start,end> is mapped - * with type. - */ -int -e820_any_mapped(u64 start, u64 end, unsigned type) +/* Overridden in paravirt.c if CONFIG_PARAVIRT */ +char * __init __attribute__((weak)) memory_setup(void) { - int i; - for (i = 0; i < e820.nr_map; i++) { - const struct e820entry *ei = &e820.map[i]; - if (type && ei->type != type) - continue; - if (ei->addr >= end || ei->addr + ei->size <= start) - continue; - return 1; - } - return 0; + return machine_specific_memory_setup(); } -EXPORT_SYMBOL_GPL(e820_any_mapped); - - /* - * This function checks if the entire range <start,end> is mapped with type. - * - * Note: this function only works correct if the e820 table is sorted and - * not-overlapping, which is the case - */ -int __init -e820_all_mapped(unsigned long s, unsigned long e, unsigned type) + +void __init setup_memory_map(void) { - u64 start = s; - u64 end = e; - int i; - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - if (type && ei->type != type) - continue; - /* is the region (part) in overlap with the current region ?*/ - if (ei->addr >= end || ei->addr + ei->size <= start) - continue; - /* if the region is at the beginning of <start,end> we move - * start to the end of the region since it's ok until there + printk(KERN_INFO "BIOS-provided physical RAM map:\n"); + e820_print_map(memory_setup()); +} + +static int __initdata user_defined_memmap; + +/* + * "mem=nopentium" disables the 4MB page tables. + * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM + * to <mem>, overriding the bios size. + * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from + * <start> to <start>+<mem>, overriding the bios size. + * + * HPA tells me bootloaders need to parse mem=, so no new + * option should be mem= [also see Documentation/i386/boot.txt] + */ +static int __init parse_mem(char *arg) +{ + if (!arg) + return -EINVAL; + + if (strcmp(arg, "nopentium") == 0) { + setup_clear_cpu_cap(X86_FEATURE_PSE); + } else { + /* If the user specifies memory size, we + * limit the BIOS-provided memory map to + * that size. exactmap can be used to specify + * the exact map. mem=number can be used to + * trim the existing memory map. */ - if (ei->addr <= start) - start = ei->addr + ei->size; - /* if start is now at or beyond end, we're done, full - * coverage */ - if (start >= end) - return 1; /* we're done */ + unsigned long long mem_size; + + mem_size = memparse(arg, &arg); + limit_regions(mem_size); + user_defined_memmap = 1; } return 0; } +early_param("mem", parse_mem); static int __init parse_memmap(char *arg) { @@ -704,8 +297,9 @@ static int __init parse_memmap(char *arg) * size before original memory map is * reset. */ - propagate_e820_map(); - saved_max_pfn = max_pfn; + e820_register_active_regions(0, 0, -1UL); + saved_max_pfn = e820_end_of_ram(); + remove_all_active_ranges(); #endif e820.nr_map = 0; user_defined_memmap = 1; @@ -736,40 +330,12 @@ static int __init parse_memmap(char *arg) return 0; } early_param("memmap", parse_memmap); -void __init update_memory_range(u64 start, u64 size, unsigned old_type, - unsigned new_type) -{ - int i; - - BUG_ON(old_type == new_type); - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - u64 final_start, final_end; - if (ei->type != old_type) - continue; - /* totally covered? */ - if (ei->addr >= start && ei->size <= size) { - ei->type = new_type; - continue; - } - /* partially covered */ - final_start = max(start, ei->addr); - final_end = min(start + size, ei->addr + ei->size); - if (final_start >= final_end) - continue; - add_memory_region(final_start, final_end - final_start, - new_type); +void __init finish_e820_parsing(void) +{ + if (user_defined_memmap) { + printk(KERN_INFO "user-defined physical RAM map:\n"); + e820_print_map("user"); } } -void __init update_e820(void) -{ - u8 nr_map; - nr_map = e820.nr_map; - if (sanitize_e820_map(e820.map, &nr_map)) - return; - e820.nr_map = nr_map; - printk(KERN_INFO "modified physical RAM map:\n"); - print_memory_map("modified"); -} diff --git a/arch/x86/kernel/e820_64.c b/arch/x86/kernel/e820_64.c index 124480c0008..0afee2ca0bf 100644 --- a/arch/x86/kernel/e820_64.c +++ b/arch/x86/kernel/e820_64.c @@ -17,8 +17,8 @@ #include <linux/kexec.h> #include <linux/module.h> #include <linux/mm.h> -#include <linux/suspend.h> #include <linux/pfn.h> +#include <linux/pci.h> #include <asm/pgtable.h> #include <asm/page.h> @@ -29,8 +29,6 @@ #include <asm/kdebug.h> #include <asm/trampoline.h> -struct e820map e820; - /* * PFN of last memory page. */ @@ -44,285 +42,6 @@ unsigned long end_pfn; unsigned long max_pfn_mapped; /* - * Last pfn which the user wants to use. - */ -static unsigned long __initdata end_user_pfn = MAXMEM>>PAGE_SHIFT; - -/* - * Early reserved memory areas. - */ -#define MAX_EARLY_RES 20 - -struct early_res { - unsigned long start, end; - char name[16]; -}; -static struct early_res early_res[MAX_EARLY_RES] __initdata = { - { 0, PAGE_SIZE, "BIOS data page" }, /* BIOS data page */ -#ifdef CONFIG_X86_TRAMPOLINE - { TRAMPOLINE_BASE, TRAMPOLINE_BASE + 2 * PAGE_SIZE, "TRAMPOLINE" }, -#endif - {} -}; - -void __init reserve_early(unsigned long start, unsigned long end, char *name) -{ - int i; - struct early_res *r; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - r = &early_res[i]; - if (end > r->start && start < r->end) - panic("Overlapping early reservations %lx-%lx %s to %lx-%lx %s\n", - start, end - 1, name?name:"", r->start, r->end - 1, r->name); - } - if (i >= MAX_EARLY_RES) - panic("Too many early reservations"); - r = &early_res[i]; - r->start = start; - r->end = end; - if (name) - strncpy(r->name, name, sizeof(r->name) - 1); -} - -void __init free_early(unsigned long start, unsigned long end) -{ - struct early_res *r; - int i, j; - - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - r = &early_res[i]; - if (start == r->start && end == r->end) - break; - } - if (i >= MAX_EARLY_RES || !early_res[i].end) - panic("free_early on not reserved area: %lx-%lx!", start, end); - - for (j = i + 1; j < MAX_EARLY_RES && early_res[j].end; j++) - ; - - memmove(&early_res[i], &early_res[i + 1], - (j - 1 - i) * sizeof(struct early_res)); - - early_res[j - 1].end = 0; -} - -void __init early_res_to_bootmem(unsigned long start, unsigned long end) -{ - int i; - unsigned long final_start, final_end; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - struct early_res *r = &early_res[i]; - final_start = max(start, r->start); - final_end = min(end, r->end); - if (final_start >= final_end) - continue; - printk(KERN_INFO " early res: %d [%lx-%lx] %s\n", i, - final_start, final_end - 1, r->name); - reserve_bootmem_generic(final_start, final_end - final_start); - } -} - -/* Check for already reserved areas */ -static inline int __init -bad_addr(unsigned long *addrp, unsigned long size, unsigned long align) -{ - int i; - unsigned long addr = *addrp, last; - int changed = 0; -again: - last = addr + size; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - struct early_res *r = &early_res[i]; - if (last >= r->start && addr < r->end) { - *addrp = addr = round_up(r->end, align); - changed = 1; - goto again; - } - } - return changed; -} - -/* Check for already reserved areas */ -static inline int __init -bad_addr_size(unsigned long *addrp, unsigned long *sizep, unsigned long align) -{ - int i; - unsigned long addr = *addrp, last; - unsigned long size = *sizep; - int changed = 0; -again: - last = addr + size; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - struct early_res *r = &early_res[i]; - if (last > r->start && addr < r->start) { - size = r->start - addr; - changed = 1; - goto again; - } - if (last > r->end && addr < r->end) { - addr = round_up(r->end, align); - size = last - addr; - changed = 1; - goto again; - } - if (last <= r->end && addr >= r->start) { - (*sizep)++; - return 0; - } - } - if (changed) { - *addrp = addr; - *sizep = size; - } - return changed; -} -/* - * This function checks if any part of the range <start,end> is mapped - * with type. - */ -int -e820_any_mapped(unsigned long start, unsigned long end, unsigned type) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - - if (type && ei->type != type) - continue; - if (ei->addr >= end || ei->addr + ei->size <= start) - continue; - return 1; - } - return 0; -} -EXPORT_SYMBOL_GPL(e820_any_mapped); - -/* - * This function checks if the entire range <start,end> is mapped with type. - * - * Note: this function only works correct if the e820 table is sorted and - * not-overlapping, which is the case - */ -int __init e820_all_mapped(unsigned long start, unsigned long end, - unsigned type) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - - if (type && ei->type != type) - continue; - /* is the region (part) in overlap with the current region ?*/ - if (ei->addr >= end || ei->addr + ei->size <= start) - continue; - - /* if the region is at the beginning of <start,end> we move - * start to the end of the region since it's ok until there - */ - if (ei->addr <= start) - start = ei->addr + ei->size; - /* - * if start is now at or beyond end, we're done, full - * coverage - */ - if (start >= end) - return 1; - } - return 0; -} - -/* - * Find a free area with specified alignment in a specific range. - */ -unsigned long __init find_e820_area(unsigned long start, unsigned long end, - unsigned long size, unsigned long align) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - unsigned long addr, last; - unsigned long ei_last; - - if (ei->type != E820_RAM) - continue; - addr = round_up(ei->addr, align); - ei_last = ei->addr + ei->size; - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - continue; - while (bad_addr(&addr, size, align) && addr+size <= ei_last) - ; - last = addr + size; - if (last > ei_last) - continue; - if (last > end) - continue; - return addr; - } - return -1UL; -} - -/* - * Find next free range after *start - */ -unsigned long __init find_e820_area_size(unsigned long start, - unsigned long *sizep, - unsigned long align) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - unsigned long addr, last; - unsigned long ei_last; - - if (ei->type != E820_RAM) - continue; - addr = round_up(ei->addr, align); - ei_last = ei->addr + ei->size; - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - continue; - *sizep = ei_last - addr; - while (bad_addr_size(&addr, sizep, align) && - addr + *sizep <= ei_last) - ; - last = addr + *sizep; - if (last > ei_last) - continue; - return addr; - } - return -1UL; - -} -/* - * Find the highest page frame number we have available - */ -unsigned long __init e820_end_of_ram(void) -{ - unsigned long end_pfn; - - end_pfn = find_max_pfn_with_active_regions(); - - if (end_pfn > max_pfn_mapped) - max_pfn_mapped = end_pfn; - if (max_pfn_mapped > MAXMEM>>PAGE_SHIFT) - max_pfn_mapped = MAXMEM>>PAGE_SHIFT; - if (end_pfn > end_user_pfn) - end_pfn = end_user_pfn; - if (end_pfn > max_pfn_mapped) - end_pfn = max_pfn_mapped; - - printk(KERN_INFO "max_pfn_mapped = %lu\n", max_pfn_mapped); - return end_pfn; -} - -/* * Mark e820 reserved areas as busy for the resource manager. */ void __init e820_reserve_resources(void) @@ -346,393 +65,6 @@ void __init e820_reserve_resources(void) } } -/* - * Find the ranges of physical addresses that do not correspond to - * e820 RAM areas and mark the corresponding pages as nosave for software - * suspend and suspend to RAM. - * - * This function requires the e820 map to be sorted and without any - * overlapping entries and assumes the first e820 area to be RAM. - */ -void __init e820_mark_nosave_regions(void) -{ - int i; - unsigned long paddr; - - paddr = round_down(e820.map[0].addr + e820.map[0].size, PAGE_SIZE); - for (i = 1; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - - if (paddr < ei->addr) - register_nosave_region(PFN_DOWN(paddr), - PFN_UP(ei->addr)); - - paddr = round_down(ei->addr + ei->size, PAGE_SIZE); - if (ei->type != E820_RAM) - register_nosave_region(PFN_UP(ei->addr), - PFN_DOWN(paddr)); - - if (paddr >= (end_pfn << PAGE_SHIFT)) - break; - } -} - -/* - * Finds an active region in the address range from start_pfn to end_pfn and - * returns its range in ei_startpfn and ei_endpfn for the e820 entry. - */ -static int __init e820_find_active_region(const struct e820entry *ei, - unsigned long start_pfn, - unsigned long end_pfn, - unsigned long *ei_startpfn, - unsigned long *ei_endpfn) -{ - *ei_startpfn = round_up(ei->addr, PAGE_SIZE) >> PAGE_SHIFT; - *ei_endpfn = round_down(ei->addr + ei->size, PAGE_SIZE) >> PAGE_SHIFT; - - /* Skip map entries smaller than a page */ - if (*ei_startpfn >= *ei_endpfn) - return 0; - - /* Check if max_pfn_mapped should be updated */ - if (ei->type != E820_RAM && *ei_endpfn > max_pfn_mapped) - max_pfn_mapped = *ei_endpfn; - - /* Skip if map is outside the node */ - if (ei->type != E820_RAM || *ei_endpfn <= start_pfn || - *ei_startpfn >= end_pfn) - return 0; - - /* Check for overlaps */ - if (*ei_startpfn < start_pfn) - *ei_startpfn = start_pfn; - if (*ei_endpfn > end_pfn) - *ei_endpfn = end_pfn; - - /* Obey end_user_pfn to save on memmap */ - if (*ei_startpfn >= end_user_pfn) - return 0; - if (*ei_endpfn > end_user_pfn) - *ei_endpfn = end_user_pfn; - - return 1; -} - -/* Walk the e820 map and register active regions within a node */ -void __init -e820_register_active_regions(int nid, unsigned long start_pfn, - unsigned long end_pfn) -{ - unsigned long ei_startpfn; - unsigned long ei_endpfn; - int i; - - for (i = 0; i < e820.nr_map; i++) - if (e820_find_active_region(&e820.map[i], - start_pfn, end_pfn, - &ei_startpfn, &ei_endpfn)) - add_active_range(nid, ei_startpfn, ei_endpfn); -} - -/* - * Add a memory region to the kernel e820 map. - */ -void __init add_memory_region(unsigned long start, unsigned long size, int type) -{ - int x = e820.nr_map; - - if (x == E820MAX) { - printk(KERN_ERR "Ooops! Too many entries in the memory map!\n"); - return; - } - - e820.map[x].addr = start; - e820.map[x].size = size; - e820.map[x].type = type; - e820.nr_map++; -} - -/* - * Find the hole size (in bytes) in the memory range. - * @start: starting address of the memory range to scan - * @end: ending address of the memory range to scan - */ -unsigned long __init e820_hole_size(unsigned long start, unsigned long end) -{ - unsigned long start_pfn = start >> PAGE_SHIFT; - unsigned long end_pfn = end >> PAGE_SHIFT; - unsigned long ei_startpfn, ei_endpfn, ram = 0; - int i; - - for (i = 0; i < e820.nr_map; i++) { - if (e820_find_active_region(&e820.map[i], - start_pfn, end_pfn, - &ei_startpfn, &ei_endpfn)) - ram += ei_endpfn - ei_startpfn; - } - return end - start - (ram << PAGE_SHIFT); -} - -static void __init e820_print_map(char *who) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - printk(KERN_INFO " %s: %016Lx - %016Lx ", who, - (unsigned long long) e820.map[i].addr, - (unsigned long long) - (e820.map[i].addr + e820.map[i].size)); - switch (e820.map[i].type) { - case E820_RAM: - printk(KERN_CONT "(usable)\n"); - break; - case E820_RESERVED: - printk(KERN_CONT "(reserved)\n"); - break; - case E820_ACPI: - printk(KERN_CONT "(ACPI data)\n"); - break; - case E820_NVS: - printk(KERN_CONT "(ACPI NVS)\n"); - break; - default: - printk(KERN_CONT "type %u\n", e820.map[i].type); - break; - } - } -} - -/* - * Sanitize the BIOS e820 map. - * - * Some e820 responses include overlapping entries. The following - * replaces the original e820 map with a new one, removing overlaps. - * - */ -static int __init sanitize_e820_map(struct e820entry *biosmap, char *pnr_map) -{ - struct change_member { - struct e820entry *pbios; /* pointer to original bios entry */ - unsigned long long addr; /* address for this change point */ - }; - static struct change_member change_point_list[2*E820MAX] __initdata; - static struct change_member *change_point[2*E820MAX] __initdata; - static struct e820entry *overlap_list[E820MAX] __initdata; - static struct e820entry new_bios[E820MAX] __initdata; - struct change_member *change_tmp; - unsigned long current_type, last_type; - unsigned long long last_addr; - int chgidx, still_changing; - int overlap_entries; - int new_bios_entry; - int old_nr, new_nr, chg_nr; - int i; - - /* - Visually we're performing the following - (1,2,3,4 = memory types)... - - Sample memory map (w/overlaps): - ____22__________________ - ______________________4_ - ____1111________________ - _44_____________________ - 11111111________________ - ____________________33__ - ___________44___________ - __________33333_________ - ______________22________ - ___________________2222_ - _________111111111______ - _____________________11_ - _________________4______ - - Sanitized equivalent (no overlap): - 1_______________________ - _44_____________________ - ___1____________________ - ____22__________________ - ______11________________ - _________1______________ - __________3_____________ - ___________44___________ - _____________33_________ - _______________2________ - ________________1_______ - _________________4______ - ___________________2____ - ____________________33__ - ______________________4_ - */ - - /* if there's only one memory region, don't bother */ - if (*pnr_map < 2) - return -1; - - old_nr = *pnr_map; - - /* bail out if we find any unreasonable addresses in bios map */ - for (i = 0; i < old_nr; i++) - if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr) - return -1; - - /* create pointers for initial change-point information (for sorting) */ - for (i = 0; i < 2 * old_nr; i++) - change_point[i] = &change_point_list[i]; - - /* record all known change-points (starting and ending addresses), - omitting those that are for empty memory regions */ - chgidx = 0; - for (i = 0; i < old_nr; i++) { - if (biosmap[i].size != 0) { - change_point[chgidx]->addr = biosmap[i].addr; - change_point[chgidx++]->pbios = &biosmap[i]; - change_point[chgidx]->addr = biosmap[i].addr + - biosmap[i].size; - change_point[chgidx++]->pbios = &biosmap[i]; - } - } - chg_nr = chgidx; - - /* sort change-point list by memory addresses (low -> high) */ - still_changing = 1; - while (still_changing) { - still_changing = 0; - for (i = 1; i < chg_nr; i++) { - unsigned long long curaddr, lastaddr; - unsigned long long curpbaddr, lastpbaddr; - - curaddr = change_point[i]->addr; - lastaddr = change_point[i - 1]->addr; - curpbaddr = change_point[i]->pbios->addr; - lastpbaddr = change_point[i - 1]->pbios->addr; - - /* - * swap entries, when: - * - * curaddr > lastaddr or - * curaddr == lastaddr and curaddr == curpbaddr and - * lastaddr != lastpbaddr - */ - if (curaddr < lastaddr || - (curaddr == lastaddr && curaddr == curpbaddr && - lastaddr != lastpbaddr)) { - change_tmp = change_point[i]; - change_point[i] = change_point[i-1]; - change_point[i-1] = change_tmp; - still_changing = 1; - } - } - } - - /* create a new bios memory map, removing overlaps */ - overlap_entries = 0; /* number of entries in the overlap table */ - new_bios_entry = 0; /* index for creating new bios map entries */ - last_type = 0; /* start with undefined memory type */ - last_addr = 0; /* start with 0 as last starting address */ - - /* loop through change-points, determining affect on the new bios map */ - for (chgidx = 0; chgidx < chg_nr; chgidx++) { - /* keep track of all overlapping bios entries */ - if (change_point[chgidx]->addr == - change_point[chgidx]->pbios->addr) { - /* - * add map entry to overlap list (> 1 entry - * implies an overlap) - */ - overlap_list[overlap_entries++] = - change_point[chgidx]->pbios; - } else { - /* - * remove entry from list (order independent, - * so swap with last) - */ - for (i = 0; i < overlap_entries; i++) { - if (overlap_list[i] == - change_point[chgidx]->pbios) - overlap_list[i] = - overlap_list[overlap_entries-1]; - } - overlap_entries--; - } - /* - * if there are overlapping entries, decide which - * "type" to use (larger value takes precedence -- - * 1=usable, 2,3,4,4+=unusable) - */ - current_type = 0; - for (i = 0; i < overlap_entries; i++) - if (overlap_list[i]->type > current_type) - current_type = overlap_list[i]->type; - /* - * continue building up new bios map based on this - * information - */ - if (current_type != last_type) { - if (last_type != 0) { - new_bios[new_bios_entry].size = - change_point[chgidx]->addr - last_addr; - /* - * move forward only if the new size - * was non-zero - */ - if (new_bios[new_bios_entry].size != 0) - /* - * no more space left for new - * bios entries ? - */ - if (++new_bios_entry >= E820MAX) - break; - } - if (current_type != 0) { - new_bios[new_bios_entry].addr = - change_point[chgidx]->addr; - new_bios[new_bios_entry].type = current_type; - last_addr = change_point[chgidx]->addr; - } - last_type = current_type; - } - } - /* retain count for new bios entries */ - new_nr = new_bios_entry; - - /* copy new bios mapping into original location */ - memcpy(biosmap, new_bios, new_nr * sizeof(struct e820entry)); - *pnr_map = new_nr; - - return 0; -} - -/* - * Copy the BIOS e820 map into a safe place. - * - * Sanity-check it while we're at it.. - * - * If we're lucky and live on a modern system, the setup code - * will have given us a memory map that we can use to properly - * set up memory. If we aren't, we'll fake a memory map. - */ -static int __init copy_e820_map(struct e820entry *biosmap, int nr_map) -{ - /* Only one memory region (or negative)? Ignore it */ - if (nr_map < 2) - return -1; - - do { - u64 start = biosmap->addr; - u64 size = biosmap->size; - u64 end = start + size; - u32 type = biosmap->type; - - /* Overflow in 64 bits? Ignore the memory map. */ - if (start > end) - return -1; - - add_memory_region(start, size, type); - } while (biosmap++, --nr_map); - return 0; -} - static void early_panic(char *msg) { early_printk(msg); @@ -740,16 +72,21 @@ static void early_panic(char *msg) } /* We're not void only for x86 32-bit compat */ -char * __init machine_specific_memory_setup(void) +char *__init machine_specific_memory_setup(void) { char *who = "BIOS-e820"; + int new_nr; /* * Try to copy the BIOS-supplied E820-map. * * Otherwise fake a memory map; one section from 0k->640k, * the next section from 1mb->appropriate_mem_k */ - sanitize_e820_map(boot_params.e820_map, &boot_params.e820_entries); + new_nr = boot_params.e820_entries; + sanitize_e820_map(boot_params.e820_map, + ARRAY_SIZE(boot_params.e820_map), + &new_nr); + boot_params.e820_entries = new_nr; if (copy_e820_map(boot_params.e820_map, boot_params.e820_entries) < 0) early_panic("Cannot find a valid memory map"); printk(KERN_INFO "BIOS-provided physical RAM map:\n"); @@ -787,7 +124,6 @@ static int __init parse_memmap_opt(char *p) saved_max_pfn = e820_end_of_ram(); remove_all_active_ranges(); #endif - max_pfn_mapped = 0; e820.nr_map = 0; userdef = 1; return 0; @@ -818,9 +154,9 @@ early_param("memmap", parse_memmap_opt); void __init finish_e820_parsing(void) { if (userdef) { - char nr = e820.nr_map; + int nr = e820.nr_map; - if (sanitize_e820_map(e820.map, &nr) < 0) + if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr) < 0) early_panic("Invalid user supplied memory map"); e820.nr_map = nr; @@ -829,109 +165,6 @@ void __init finish_e820_parsing(void) } } -void __init update_memory_range(u64 start, u64 size, unsigned old_type, - unsigned new_type) -{ - int i; - - BUG_ON(old_type == new_type); - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - u64 final_start, final_end; - if (ei->type != old_type) - continue; - /* totally covered? */ - if (ei->addr >= start && ei->size <= size) { - ei->type = new_type; - continue; - } - /* partially covered */ - final_start = max(start, ei->addr); - final_end = min(start + size, ei->addr + ei->size); - if (final_start >= final_end) - continue; - add_memory_region(final_start, final_end - final_start, - new_type); - } -} - -void __init update_e820(void) -{ - u8 nr_map; - - nr_map = e820.nr_map; - if (sanitize_e820_map(e820.map, &nr_map)) - return; - e820.nr_map = nr_map; - printk(KERN_INFO "modified physical RAM map:\n"); - e820_print_map("modified"); -} - -unsigned long pci_mem_start = 0xaeedbabe; -EXPORT_SYMBOL(pci_mem_start); - -/* - * Search for the biggest gap in the low 32 bits of the e820 - * memory space. We pass this space to PCI to assign MMIO resources - * for hotplug or unconfigured devices in. - * Hopefully the BIOS let enough space left. - */ -__init void e820_setup_gap(void) -{ - unsigned long gapstart, gapsize, round; - unsigned long last; - int i; - int found = 0; - - last = 0x100000000ull; - gapstart = 0x10000000; - gapsize = 0x400000; - i = e820.nr_map; - while (--i >= 0) { - unsigned long long start = e820.map[i].addr; - unsigned long long end = start + e820.map[i].size; - - /* - * Since "last" is at most 4GB, we know we'll - * fit in 32 bits if this condition is true - */ - if (last > end) { - unsigned long gap = last - end; - - if (gap > gapsize) { - gapsize = gap; - gapstart = end; - found = 1; - } - } - if (start < last) - last = start; - } - - if (!found) { - gapstart = (end_pfn << PAGE_SHIFT) + 1024*1024; - printk(KERN_ERR "PCI: Warning: Cannot find a gap in the 32bit " - "address range\n" - KERN_ERR "PCI: Unassigned devices with 32bit resource " - "registers may break!\n"); - } - - /* - * See how much we want to round up: start off with - * rounding to the next 1MB area. - */ - round = 0x100000; - while ((gapsize >> 4) > round) - round += round; - /* Fun with two's complement */ - pci_mem_start = (gapstart + round) & -round; - - printk(KERN_INFO - "Allocating PCI resources starting at %lx (gap: %lx:%lx)\n", - pci_mem_start, gapstart, gapsize); -} - int __init arch_get_ram_range(int slot, u64 *addr, u64 *size) { int i; diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index 77d424cf68b..d5c7fcdd186 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -213,6 +213,48 @@ unsigned long efi_get_time(void) eft.minute, eft.second); } +/* + * Tell the kernel about the EFI memory map. This might include + * more than the max 128 entries that can fit in the e820 legacy + * (zeropage) memory map. + */ + +static void __init add_efi_memmap(void) +{ + void *p; + + for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { + efi_memory_desc_t *md = p; + unsigned long long start = md->phys_addr; + unsigned long long size = md->num_pages << EFI_PAGE_SHIFT; + int e820_type; + + if (md->attribute & EFI_MEMORY_WB) + e820_type = E820_RAM; + else + e820_type = E820_RESERVED; + add_memory_region(start, size, e820_type); + } + sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); +} + +void __init efi_reserve_early(void) +{ + unsigned long pmap; + + pmap = boot_params.efi_info.efi_memmap; +#ifdef CONFIG_X86_64 + pmap += (__u64)boot_params.efi_info.efi_memmap_hi << 32; +#endif + memmap.phys_map = (void *)pmap; + memmap.nr_map = boot_params.efi_info.efi_memmap_size / + boot_params.efi_info.efi_memdesc_size; + memmap.desc_version = boot_params.efi_info.efi_memdesc_version; + memmap.desc_size = boot_params.efi_info.efi_memdesc_size; + reserve_early(pmap, pmap + memmap.nr_map * memmap.desc_size, + "EFI memmap"); +} + #if EFI_DEBUG static void __init print_efi_memmap(void) { @@ -242,21 +284,11 @@ void __init efi_init(void) int i = 0; void *tmp; -#ifdef CONFIG_X86_32 efi_phys.systab = (efi_system_table_t *)boot_params.efi_info.efi_systab; - memmap.phys_map = (void *)boot_params.efi_info.efi_memmap; -#else - efi_phys.systab = (efi_system_table_t *) - (boot_params.efi_info.efi_systab | - ((__u64)boot_params.efi_info.efi_systab_hi<<32)); - memmap.phys_map = (void *) - (boot_params.efi_info.efi_memmap | - ((__u64)boot_params.efi_info.efi_memmap_hi<<32)); +#ifdef CONFIG_X86_64 + efi_phys.systab = (void *)efi_phys.systab + + ((__u64)boot_params.efi_info.efi_systab_hi<<32); #endif - memmap.nr_map = boot_params.efi_info.efi_memmap_size / - boot_params.efi_info.efi_memdesc_size; - memmap.desc_version = boot_params.efi_info.efi_memdesc_version; - memmap.desc_size = boot_params.efi_info.efi_memdesc_size; efi.systab = early_ioremap((unsigned long)efi_phys.systab, sizeof(efi_system_table_t)); @@ -370,6 +402,7 @@ void __init efi_init(void) if (memmap.desc_size != sizeof(efi_memory_desc_t)) printk(KERN_WARNING "Kernel-defined memdesc" "doesn't match the one from EFI!\n"); + add_efi_memmap(); /* Setup for EFI runtime service */ reboot_type = BOOT_EFI; diff --git a/arch/x86/kernel/efi_64.c b/arch/x86/kernel/efi_64.c index d0060fdccca..652c5287215 100644 --- a/arch/x86/kernel/efi_64.c +++ b/arch/x86/kernel/efi_64.c @@ -97,13 +97,7 @@ void __init efi_call_phys_epilog(void) early_runtime_code_mapping_set_exec(0); } -void __init efi_reserve_bootmem(void) -{ - reserve_bootmem_generic((unsigned long)memmap.phys_map, - memmap.nr_map * memmap.desc_size); -} - -void __iomem * __init efi_ioremap(unsigned long phys_addr, unsigned long size) +void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size) { static unsigned pages_mapped __initdata; unsigned i, pages; diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c index cbaaf69bedb..1fa8be5bd21 100644 --- a/arch/x86/kernel/genapic_64.c +++ b/arch/x86/kernel/genapic_64.c @@ -51,7 +51,7 @@ void __init setup_apic_routing(void) else #endif - if (num_possible_cpus() <= 8) + if (max_physical_apicid < 8) genapic = &apic_flat; else genapic = &apic_physflat; diff --git a/arch/x86/kernel/head.c b/arch/x86/kernel/head.c new file mode 100644 index 00000000000..a727c0b9819 --- /dev/null +++ b/arch/x86/kernel/head.c @@ -0,0 +1,73 @@ +#include <linux/kernel.h> +#include <linux/init.h> + +#include <asm/setup.h> +#include <asm/bios_ebda.h> + +#define BIOS_LOWMEM_KILOBYTES 0x413 + +/* + * The BIOS places the EBDA/XBDA at the top of conventional + * memory, and usually decreases the reported amount of + * conventional memory (int 0x12) too. This also contains a + * workaround for Dell systems that neglect to reserve EBDA. + * The same workaround also avoids a problem with the AMD768MPX + * chipset: reserve a page before VGA to prevent PCI prefetch + * into it (errata #56). Usually the page is reserved anyways, + * unless you have no PS/2 mouse plugged in. + */ +void __init reserve_ebda_region(void) +{ + unsigned int lowmem, ebda_addr; + + /* To determine the position of the EBDA and the */ + /* end of conventional memory, we need to look at */ + /* the BIOS data area. In a paravirtual environment */ + /* that area is absent. We'll just have to assume */ + /* that the paravirt case can handle memory setup */ + /* correctly, without our help. */ + if (paravirt_enabled()) + return; + + /* end of low (conventional) memory */ + lowmem = *(unsigned short *)__va(BIOS_LOWMEM_KILOBYTES); + lowmem <<= 10; + + /* start of EBDA area */ + ebda_addr = get_bios_ebda(); + + /* Fixup: bios puts an EBDA in the top 64K segment */ + /* of conventional memory, but does not adjust lowmem. */ + if ((lowmem - ebda_addr) <= 0x10000) + lowmem = ebda_addr; + + /* Fixup: bios does not report an EBDA at all. */ + /* Some old Dells seem to need 4k anyhow (bugzilla 2990) */ + if ((ebda_addr == 0) && (lowmem >= 0x9f000)) + lowmem = 0x9f000; + + /* Paranoia: should never happen, but... */ + if ((lowmem == 0) || (lowmem >= 0x100000)) + lowmem = 0x9f000; + + /* reserve all memory between lowmem and the 1MB mark */ + reserve_early(lowmem, 0x100000, "BIOS reserved"); +} + +void __init reserve_setup_data(void) +{ + struct setup_data *data; + u64 pa_data; + char buf[32]; + + if (boot_params.hdr.version < 0x0209) + return; + pa_data = boot_params.hdr.setup_data; + while (pa_data) { + data = early_ioremap(pa_data, sizeof(*data)); + sprintf(buf, "setup data %x", data->type); + reserve_early(pa_data, pa_data+sizeof(*data)+data->len, buf); + pa_data = data->next; + early_iounmap(data, sizeof(*data)); + } +} diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index 3db05905892..fa1d25dd83e 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -8,7 +8,34 @@ #include <linux/init.h> #include <linux/start_kernel.h> +#include <asm/setup.h> +#include <asm/sections.h> +#include <asm/e820.h> +#include <asm/bios_ebda.h> + void __init i386_start_kernel(void) { + reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS"); + +#ifdef CONFIG_BLK_DEV_INITRD + /* Reserve INITRD */ + if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) { + u64 ramdisk_image = boot_params.hdr.ramdisk_image; + u64 ramdisk_size = boot_params.hdr.ramdisk_size; + u64 ramdisk_end = ramdisk_image + ramdisk_size; + reserve_early(ramdisk_image, ramdisk_end, "RAMDISK"); + } +#endif + reserve_early(init_pg_tables_start, init_pg_tables_end, + "INIT_PG_TABLE"); + + reserve_ebda_region(); + + /* + * At this point everything still needed from the boot loader + * or BIOS or kernel text should be early reserved or marked not + * RAM in e820. All other memory is free game. + */ + start_kernel(); } diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index e25c57b8aa8..5fbed459ff3 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -51,74 +51,6 @@ static void __init copy_bootdata(char *real_mode_data) } } -#define BIOS_LOWMEM_KILOBYTES 0x413 - -/* - * The BIOS places the EBDA/XBDA at the top of conventional - * memory, and usually decreases the reported amount of - * conventional memory (int 0x12) too. This also contains a - * workaround for Dell systems that neglect to reserve EBDA. - * The same workaround also avoids a problem with the AMD768MPX - * chipset: reserve a page before VGA to prevent PCI prefetch - * into it (errata #56). Usually the page is reserved anyways, - * unless you have no PS/2 mouse plugged in. - */ -static void __init reserve_ebda_region(void) -{ - unsigned int lowmem, ebda_addr; - - /* To determine the position of the EBDA and the */ - /* end of conventional memory, we need to look at */ - /* the BIOS data area. In a paravirtual environment */ - /* that area is absent. We'll just have to assume */ - /* that the paravirt case can handle memory setup */ - /* correctly, without our help. */ - if (paravirt_enabled()) - return; - - /* end of low (conventional) memory */ - lowmem = *(unsigned short *)__va(BIOS_LOWMEM_KILOBYTES); - lowmem <<= 10; - - /* start of EBDA area */ - ebda_addr = get_bios_ebda(); - - /* Fixup: bios puts an EBDA in the top 64K segment */ - /* of conventional memory, but does not adjust lowmem. */ - if ((lowmem - ebda_addr) <= 0x10000) - lowmem = ebda_addr; - - /* Fixup: bios does not report an EBDA at all. */ - /* Some old Dells seem to need 4k anyhow (bugzilla 2990) */ - if ((ebda_addr == 0) && (lowmem >= 0x9f000)) - lowmem = 0x9f000; - - /* Paranoia: should never happen, but... */ - if ((lowmem == 0) || (lowmem >= 0x100000)) - lowmem = 0x9f000; - - /* reserve all memory between lowmem and the 1MB mark */ - reserve_early(lowmem, 0x100000, "BIOS reserved"); -} - -static void __init reserve_setup_data(void) -{ - struct setup_data *data; - unsigned long pa_data; - char buf[32]; - - if (boot_params.hdr.version < 0x0209) - return; - pa_data = boot_params.hdr.setup_data; - while (pa_data) { - data = early_ioremap(pa_data, sizeof(*data)); - sprintf(buf, "setup data %x", data->type); - reserve_early(pa_data, pa_data+sizeof(*data)+data->len, buf); - pa_data = data->next; - early_iounmap(data, sizeof(*data)); - } -} - void __init x86_64_start_kernel(char * real_mode_data) { int i; diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index f7357cc0162..b98b338aae1 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -194,6 +194,7 @@ default_entry: xorl %ebx,%ebx /* %ebx is kept at zero */ movl $pa(pg0), %edi + movl %edi, pa(init_pg_tables_start) movl $pa(swapper_pg_pmd), %edx movl $PTE_ATTR, %eax 10: @@ -219,6 +220,8 @@ default_entry: jb 10b 1: movl %edi,pa(init_pg_tables_end) + shrl $12, %eax + movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ movl $pa(swapper_pg_fixmap)+PDE_ATTR,%eax @@ -228,6 +231,7 @@ default_entry: page_pde_offset = (__PAGE_OFFSET >> 20); movl $pa(pg0), %edi + movl %edi, pa(init_pg_tables_start) movl $pa(swapper_pg_dir), %edx movl $PTE_ATTR, %eax 10: @@ -249,6 +253,8 @@ page_pde_offset = (__PAGE_OFFSET >> 20); cmpl %ebp,%eax jb 10b movl %edi,pa(init_pg_tables_end) + shrl $12, %eax + movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ movl $pa(swapper_pg_fixmap)+PDE_ATTR,%eax diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c index 4dc8600d9d2..0662817d61b 100644 --- a/arch/x86/kernel/io_apic_32.c +++ b/arch/x86/kernel/io_apic_32.c @@ -72,15 +72,21 @@ int sis_apic_bug = -1; int nr_ioapic_registers[MAX_IO_APICS]; /* I/O APIC entries */ -struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS]; +struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; int nr_ioapics; /* MP IRQ source entries */ -struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; +struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; /* # of MP IRQ source entries */ int mp_irq_entries; +#if defined (CONFIG_MCA) || defined (CONFIG_EISA) +int mp_bus_id_to_type[MAX_MP_BUSSES]; +#endif + +DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); + static int disable_timer_pin_1 __initdata; /* @@ -110,7 +116,7 @@ struct io_apic { static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) { return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) - + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK); + + (mp_ioapics[idx].mp_apicaddr & ~PAGE_MASK); } static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) @@ -801,10 +807,10 @@ static int find_irq_entry(int apic, int pin, int type) int i; for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mpc_irqtype == type && - (mp_irqs[i].mpc_dstapic == mp_ioapics[apic].mpc_apicid || - mp_irqs[i].mpc_dstapic == MP_APIC_ALL) && - mp_irqs[i].mpc_dstirq == pin) + if (mp_irqs[i].mp_irqtype == type && + (mp_irqs[i].mp_dstapic == mp_ioapics[apic].mp_apicid || + mp_irqs[i].mp_dstapic == MP_APIC_ALL) && + mp_irqs[i].mp_dstirq == pin) return i; return -1; @@ -818,13 +824,13 @@ static int __init find_isa_irq_pin(int irq, int type) int i; for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mpc_srcbus; + int lbus = mp_irqs[i].mp_srcbus; if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mpc_irqtype == type) && - (mp_irqs[i].mpc_srcbusirq == irq)) + (mp_irqs[i].mp_irqtype == type) && + (mp_irqs[i].mp_srcbusirq == irq)) - return mp_irqs[i].mpc_dstirq; + return mp_irqs[i].mp_dstirq; } return -1; } @@ -834,17 +840,17 @@ static int __init find_isa_irq_apic(int irq, int type) int i; for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mpc_srcbus; + int lbus = mp_irqs[i].mp_srcbus; if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mpc_irqtype == type) && - (mp_irqs[i].mpc_srcbusirq == irq)) + (mp_irqs[i].mp_irqtype == type) && + (mp_irqs[i].mp_srcbusirq == irq)) break; } if (i < mp_irq_entries) { int apic; for(apic = 0; apic < nr_ioapics; apic++) { - if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic) + if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic) return apic; } } @@ -864,28 +870,28 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin) apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, " "slot:%d, pin:%d.\n", bus, slot, pin); - if (mp_bus_id_to_pci_bus[bus] == -1) { + if (test_bit(bus, mp_bus_not_pci)) { printk(KERN_WARNING "PCI BIOS passed nonexistent PCI bus %d!\n", bus); return -1; } for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mpc_srcbus; + int lbus = mp_irqs[i].mp_srcbus; for (apic = 0; apic < nr_ioapics; apic++) - if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic || - mp_irqs[i].mpc_dstapic == MP_APIC_ALL) + if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic || + mp_irqs[i].mp_dstapic == MP_APIC_ALL) break; if (!test_bit(lbus, mp_bus_not_pci) && - !mp_irqs[i].mpc_irqtype && + !mp_irqs[i].mp_irqtype && (bus == lbus) && - (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) { - int irq = pin_2_irq(i,apic,mp_irqs[i].mpc_dstirq); + (slot == ((mp_irqs[i].mp_srcbusirq >> 2) & 0x1f))) { + int irq = pin_2_irq(i,apic,mp_irqs[i].mp_dstirq); if (!(apic || IO_APIC_IRQ(irq))) continue; - if (pin == (mp_irqs[i].mpc_srcbusirq & 3)) + if (pin == (mp_irqs[i].mp_srcbusirq & 3)) return irq; /* * Use the first all-but-pin matching entry as a @@ -952,7 +958,7 @@ static int EISA_ELCR(unsigned int irq) * EISA conforming in the MP table, that means its trigger type must * be read in from the ELCR */ -#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mpc_srcbusirq)) +#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mp_srcbusirq)) #define default_EISA_polarity(idx) default_ISA_polarity(idx) /* PCI interrupts are always polarity one level triggered, @@ -969,13 +975,13 @@ static int EISA_ELCR(unsigned int irq) static int MPBIOS_polarity(int idx) { - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; int polarity; /* * Determine IRQ line polarity (high active or low active): */ - switch (mp_irqs[idx].mpc_irqflag & 3) + switch (mp_irqs[idx].mp_irqflag & 3) { case 0: /* conforms, ie. bus-type dependent polarity */ { @@ -1012,13 +1018,13 @@ static int MPBIOS_polarity(int idx) static int MPBIOS_trigger(int idx) { - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; int trigger; /* * Determine IRQ trigger mode (edge or level sensitive): */ - switch ((mp_irqs[idx].mpc_irqflag>>2) & 3) + switch ((mp_irqs[idx].mp_irqflag>>2) & 3) { case 0: /* conforms, ie. bus-type dependent */ { @@ -1097,16 +1103,16 @@ static inline int irq_trigger(int idx) static int pin_2_irq(int idx, int apic, int pin) { int irq, i; - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; /* * Debugging check, we are in big trouble if this message pops up! */ - if (mp_irqs[idx].mpc_dstirq != pin) + if (mp_irqs[idx].mp_dstirq != pin) printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n"); if (test_bit(bus, mp_bus_not_pci)) - irq = mp_irqs[idx].mpc_srcbusirq; + irq = mp_irqs[idx].mp_srcbusirq; else { /* * PCI IRQs are mapped in order @@ -1250,12 +1256,12 @@ static void __init setup_IO_APIC_irqs(void) if (first_notcon) { apic_printk(APIC_VERBOSE, KERN_DEBUG " IO-APIC (apicid-pin) %d-%d", - mp_ioapics[apic].mpc_apicid, + mp_ioapics[apic].mp_apicid, pin); first_notcon = 0; } else apic_printk(APIC_VERBOSE, ", %d-%d", - mp_ioapics[apic].mpc_apicid, pin); + mp_ioapics[apic].mp_apicid, pin); continue; } @@ -1357,7 +1363,7 @@ void __init print_IO_APIC(void) printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); for (i = 0; i < nr_ioapics; i++) printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n", - mp_ioapics[i].mpc_apicid, nr_ioapic_registers[i]); + mp_ioapics[i].mp_apicid, nr_ioapic_registers[i]); /* * We are a bit conservative about what we expect. We have to @@ -1376,7 +1382,7 @@ void __init print_IO_APIC(void) reg_03.raw = io_apic_read(apic, 3); spin_unlock_irqrestore(&ioapic_lock, flags); - printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mpc_apicid); + printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mp_apicid); printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw); printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID); printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type); @@ -1716,7 +1722,6 @@ void disable_IO_APIC(void) * by Matt Domsch <Matt_Domsch@dell.com> Tue Dec 21 12:25:05 CST 1999 */ -#ifndef CONFIG_X86_NUMAQ static void __init setup_ioapic_ids_from_mpc(void) { union IO_APIC_reg_00 reg_00; @@ -1726,6 +1731,11 @@ static void __init setup_ioapic_ids_from_mpc(void) unsigned char old_id; unsigned long flags; +#ifdef CONFIG_X86_NUMAQ + if (found_numaq) + return; +#endif + /* * Don't check I/O APIC IDs for xAPIC systems. They have * no meaning without the serial APIC bus. @@ -1749,14 +1759,14 @@ static void __init setup_ioapic_ids_from_mpc(void) reg_00.raw = io_apic_read(apic, 0); spin_unlock_irqrestore(&ioapic_lock, flags); - old_id = mp_ioapics[apic].mpc_apicid; + old_id = mp_ioapics[apic].mp_apicid; - if (mp_ioapics[apic].mpc_apicid >= get_physical_broadcast()) { + if (mp_ioapics[apic].mp_apicid >= get_physical_broadcast()) { printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n", - apic, mp_ioapics[apic].mpc_apicid); + apic, mp_ioapics[apic].mp_apicid); printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", reg_00.bits.ID); - mp_ioapics[apic].mpc_apicid = reg_00.bits.ID; + mp_ioapics[apic].mp_apicid = reg_00.bits.ID; } /* @@ -1765,9 +1775,9 @@ static void __init setup_ioapic_ids_from_mpc(void) * 'stuck on smp_invalidate_needed IPI wait' messages. */ if (check_apicid_used(phys_id_present_map, - mp_ioapics[apic].mpc_apicid)) { + mp_ioapics[apic].mp_apicid)) { printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n", - apic, mp_ioapics[apic].mpc_apicid); + apic, mp_ioapics[apic].mp_apicid); for (i = 0; i < get_physical_broadcast(); i++) if (!physid_isset(i, phys_id_present_map)) break; @@ -1776,13 +1786,13 @@ static void __init setup_ioapic_ids_from_mpc(void) printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", i); physid_set(i, phys_id_present_map); - mp_ioapics[apic].mpc_apicid = i; + mp_ioapics[apic].mp_apicid = i; } else { physid_mask_t tmp; - tmp = apicid_to_cpu_present(mp_ioapics[apic].mpc_apicid); + tmp = apicid_to_cpu_present(mp_ioapics[apic].mp_apicid); apic_printk(APIC_VERBOSE, "Setting %d in the " "phys_id_present_map\n", - mp_ioapics[apic].mpc_apicid); + mp_ioapics[apic].mp_apicid); physids_or(phys_id_present_map, phys_id_present_map, tmp); } @@ -1791,11 +1801,11 @@ static void __init setup_ioapic_ids_from_mpc(void) * We need to adjust the IRQ routing table * if the ID changed. */ - if (old_id != mp_ioapics[apic].mpc_apicid) + if (old_id != mp_ioapics[apic].mp_apicid) for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mpc_dstapic == old_id) - mp_irqs[i].mpc_dstapic - = mp_ioapics[apic].mpc_apicid; + if (mp_irqs[i].mp_dstapic == old_id) + mp_irqs[i].mp_dstapic + = mp_ioapics[apic].mp_apicid; /* * Read the right value from the MPC table and @@ -1803,9 +1813,9 @@ static void __init setup_ioapic_ids_from_mpc(void) */ apic_printk(APIC_VERBOSE, KERN_INFO "...changing IO-APIC physical APIC ID to %d ...", - mp_ioapics[apic].mpc_apicid); + mp_ioapics[apic].mp_apicid); - reg_00.bits.ID = mp_ioapics[apic].mpc_apicid; + reg_00.bits.ID = mp_ioapics[apic].mp_apicid; spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic, 0, reg_00.raw); spin_unlock_irqrestore(&ioapic_lock, flags); @@ -1816,15 +1826,12 @@ static void __init setup_ioapic_ids_from_mpc(void) spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(apic, 0); spin_unlock_irqrestore(&ioapic_lock, flags); - if (reg_00.bits.ID != mp_ioapics[apic].mpc_apicid) + if (reg_00.bits.ID != mp_ioapics[apic].mp_apicid) printk("could not set ID!\n"); else apic_printk(APIC_VERBOSE, " ok.\n"); } } -#else -static void __init setup_ioapic_ids_from_mpc(void) { } -#endif int no_timer_check __initdata; @@ -2347,8 +2354,8 @@ static int ioapic_resume(struct sys_device *dev) spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(dev->id, 0); - if (reg_00.bits.ID != mp_ioapics[dev->id].mpc_apicid) { - reg_00.bits.ID = mp_ioapics[dev->id].mpc_apicid; + if (reg_00.bits.ID != mp_ioapics[dev->id].mp_apicid) { + reg_00.bits.ID = mp_ioapics[dev->id].mp_apicid; io_apic_write(dev->id, 0, reg_00.raw); } spin_unlock_irqrestore(&ioapic_lock, flags); @@ -2781,7 +2788,7 @@ int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int a apic_printk(APIC_DEBUG, KERN_DEBUG "IOAPIC[%d]: Set PCI routing entry " "(%d-%d -> 0x%x -> IRQ %d Mode:%i Active:%i)\n", ioapic, - mp_ioapics[ioapic].mpc_apicid, pin, entry.vector, irq, + mp_ioapics[ioapic].mp_apicid, pin, entry.vector, irq, edge_level, active_high_low); ioapic_register_intr(irq, entry.vector, edge_level); @@ -2802,8 +2809,8 @@ int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity) return -1; for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mpc_irqtype == mp_INT && - mp_irqs[i].mpc_srcbusirq == bus_irq) + if (mp_irqs[i].mp_irqtype == mp_INT && + mp_irqs[i].mp_srcbusirq == bus_irq) break; if (i >= mp_irq_entries) return -1; diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c index ef1a8dfcc52..339cf6f926d 100644 --- a/arch/x86/kernel/io_apic_64.c +++ b/arch/x86/kernel/io_apic_64.c @@ -104,15 +104,17 @@ DEFINE_SPINLOCK(vector_lock); int nr_ioapic_registers[MAX_IO_APICS]; /* I/O APIC entries */ -struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS]; +struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; int nr_ioapics; /* MP IRQ source entries */ -struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; +struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; /* # of MP IRQ source entries */ int mp_irq_entries; +DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); + /* * Rough estimation of how many shared IRQs there are, can * be changed anytime. @@ -140,7 +142,7 @@ struct io_apic { static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) { return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) - + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK); + + (mp_ioapics[idx].mp_apicaddr & ~PAGE_MASK); } static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) @@ -453,10 +455,10 @@ static int find_irq_entry(int apic, int pin, int type) int i; for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mpc_irqtype == type && - (mp_irqs[i].mpc_dstapic == mp_ioapics[apic].mpc_apicid || - mp_irqs[i].mpc_dstapic == MP_APIC_ALL) && - mp_irqs[i].mpc_dstirq == pin) + if (mp_irqs[i].mp_irqtype == type && + (mp_irqs[i].mp_dstapic == mp_ioapics[apic].mp_apicid || + mp_irqs[i].mp_dstapic == MP_APIC_ALL) && + mp_irqs[i].mp_dstirq == pin) return i; return -1; @@ -470,13 +472,13 @@ static int __init find_isa_irq_pin(int irq, int type) int i; for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mpc_srcbus; + int lbus = mp_irqs[i].mp_srcbus; if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mpc_irqtype == type) && - (mp_irqs[i].mpc_srcbusirq == irq)) + (mp_irqs[i].mp_irqtype == type) && + (mp_irqs[i].mp_srcbusirq == irq)) - return mp_irqs[i].mpc_dstirq; + return mp_irqs[i].mp_dstirq; } return -1; } @@ -486,17 +488,17 @@ static int __init find_isa_irq_apic(int irq, int type) int i; for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mpc_srcbus; + int lbus = mp_irqs[i].mp_srcbus; if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mpc_irqtype == type) && - (mp_irqs[i].mpc_srcbusirq == irq)) + (mp_irqs[i].mp_irqtype == type) && + (mp_irqs[i].mp_srcbusirq == irq)) break; } if (i < mp_irq_entries) { int apic; for(apic = 0; apic < nr_ioapics; apic++) { - if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic) + if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic) return apic; } } @@ -516,28 +518,28 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin) apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, slot:%d, pin:%d.\n", bus, slot, pin); - if (mp_bus_id_to_pci_bus[bus] == -1) { + if (test_bit(bus, mp_bus_not_pci)) { apic_printk(APIC_VERBOSE, "PCI BIOS passed nonexistent PCI bus %d!\n", bus); return -1; } for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mpc_srcbus; + int lbus = mp_irqs[i].mp_srcbus; for (apic = 0; apic < nr_ioapics; apic++) - if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic || - mp_irqs[i].mpc_dstapic == MP_APIC_ALL) + if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic || + mp_irqs[i].mp_dstapic == MP_APIC_ALL) break; if (!test_bit(lbus, mp_bus_not_pci) && - !mp_irqs[i].mpc_irqtype && + !mp_irqs[i].mp_irqtype && (bus == lbus) && - (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) { - int irq = pin_2_irq(i,apic,mp_irqs[i].mpc_dstirq); + (slot == ((mp_irqs[i].mp_srcbusirq >> 2) & 0x1f))) { + int irq = pin_2_irq(i,apic,mp_irqs[i].mp_dstirq); if (!(apic || IO_APIC_IRQ(irq))) continue; - if (pin == (mp_irqs[i].mpc_srcbusirq & 3)) + if (pin == (mp_irqs[i].mp_srcbusirq & 3)) return irq; /* * Use the first all-but-pin matching entry as a @@ -565,13 +567,13 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin) static int MPBIOS_polarity(int idx) { - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; int polarity; /* * Determine IRQ line polarity (high active or low active): */ - switch (mp_irqs[idx].mpc_irqflag & 3) + switch (mp_irqs[idx].mp_irqflag & 3) { case 0: /* conforms, ie. bus-type dependent polarity */ if (test_bit(bus, mp_bus_not_pci)) @@ -607,13 +609,13 @@ static int MPBIOS_polarity(int idx) static int MPBIOS_trigger(int idx) { - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; int trigger; /* * Determine IRQ trigger mode (edge or level sensitive): */ - switch ((mp_irqs[idx].mpc_irqflag>>2) & 3) + switch ((mp_irqs[idx].mp_irqflag>>2) & 3) { case 0: /* conforms, ie. bus-type dependent */ if (test_bit(bus, mp_bus_not_pci)) @@ -660,16 +662,16 @@ static inline int irq_trigger(int idx) static int pin_2_irq(int idx, int apic, int pin) { int irq, i; - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; /* * Debugging check, we are in big trouble if this message pops up! */ - if (mp_irqs[idx].mpc_dstirq != pin) + if (mp_irqs[idx].mp_dstirq != pin) printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n"); if (test_bit(bus, mp_bus_not_pci)) { - irq = mp_irqs[idx].mpc_srcbusirq; + irq = mp_irqs[idx].mp_srcbusirq; } else { /* * PCI IRQs are mapped in order @@ -846,7 +848,7 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, apic_printk(APIC_VERBOSE,KERN_DEBUG "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> " "IRQ %d Mode:%i Active:%i)\n", - apic, mp_ioapics[apic].mpc_apicid, pin, cfg->vector, + apic, mp_ioapics[apic].mp_apicid, pin, cfg->vector, irq, trigger, polarity); /* @@ -887,10 +889,10 @@ static void __init setup_IO_APIC_irqs(void) idx = find_irq_entry(apic,pin,mp_INT); if (idx == -1) { if (first_notcon) { - apic_printk(APIC_VERBOSE, KERN_DEBUG " IO-APIC (apicid-pin) %d-%d", mp_ioapics[apic].mpc_apicid, pin); + apic_printk(APIC_VERBOSE, KERN_DEBUG " IO-APIC (apicid-pin) %d-%d", mp_ioapics[apic].mp_apicid, pin); first_notcon = 0; } else - apic_printk(APIC_VERBOSE, ", %d-%d", mp_ioapics[apic].mpc_apicid, pin); + apic_printk(APIC_VERBOSE, ", %d-%d", mp_ioapics[apic].mp_apicid, pin); continue; } if (!first_notcon) { @@ -965,7 +967,7 @@ void __apicdebuginit print_IO_APIC(void) printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); for (i = 0; i < nr_ioapics; i++) printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n", - mp_ioapics[i].mpc_apicid, nr_ioapic_registers[i]); + mp_ioapics[i].mp_apicid, nr_ioapic_registers[i]); /* * We are a bit conservative about what we expect. We have to @@ -983,7 +985,7 @@ void __apicdebuginit print_IO_APIC(void) spin_unlock_irqrestore(&ioapic_lock, flags); printk("\n"); - printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mpc_apicid); + printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mp_apicid); printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw); printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID); @@ -1841,8 +1843,8 @@ static int ioapic_resume(struct sys_device *dev) spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(dev->id, 0); - if (reg_00.bits.ID != mp_ioapics[dev->id].mpc_apicid) { - reg_00.bits.ID = mp_ioapics[dev->id].mpc_apicid; + if (reg_00.bits.ID != mp_ioapics[dev->id].mp_apicid) { + reg_00.bits.ID = mp_ioapics[dev->id].mp_apicid; io_apic_write(dev->id, 0, reg_00.raw); } spin_unlock_irqrestore(&ioapic_lock, flags); @@ -2242,8 +2244,8 @@ int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity) return -1; for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mpc_irqtype == mp_INT && - mp_irqs[i].mpc_srcbusirq == bus_irq) + if (mp_irqs[i].mp_irqtype == mp_INT && + mp_irqs[i].mp_srcbusirq == bus_irq) break; if (i >= mp_irq_entries) return -1; @@ -2336,7 +2338,7 @@ void __init ioapic_init_mappings(void) ioapic_res = ioapic_setup_resources(); for (i = 0; i < nr_ioapics; i++) { if (smp_found_config) { - ioapic_phys = mp_ioapics[i].mpc_apicaddr; + ioapic_phys = mp_ioapics[i].mp_apicaddr; } else { ioapic_phys = (unsigned long) alloc_bootmem_pages(PAGE_SIZE); diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 404683b94e7..1cc7a4b8643 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -25,6 +25,8 @@ #include <asm/proto.h> #include <asm/acpi.h> #include <asm/bios_ebda.h> +#include <asm/e820.h> +#include <asm/trampoline.h> #include <mach_apic.h> #ifdef CONFIG_X86_32 @@ -32,28 +34,6 @@ #include <mach_mpparse.h> #endif -/* Have we found an MP table */ -int smp_found_config; - -/* - * Various Linux-internal data structures created from the - * MP-table. - */ -#if defined (CONFIG_MCA) || defined (CONFIG_EISA) -int mp_bus_id_to_type[MAX_MP_BUSSES]; -#endif - -DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); -int mp_bus_id_to_pci_bus[MAX_MP_BUSSES] = {[0 ... MAX_MP_BUSSES - 1] = -1 }; - -static int mp_current_pci_id; - -int pic_mode; - -/* - * Intel MP BIOS table parsing routines: - */ - /* * Checksum an MP configuration block. */ @@ -69,15 +49,73 @@ static int __init mpf_checksum(unsigned char *mp, int len) } #ifdef CONFIG_X86_NUMAQ +int found_numaq; /* * Have to match translation table entries to main table entries by counter * hence the mpc_record variable .... can't see a less disgusting way of * doing this .... */ +struct mpc_config_translation { + unsigned char mpc_type; + unsigned char trans_len; + unsigned char trans_type; + unsigned char trans_quad; + unsigned char trans_global; + unsigned char trans_local; + unsigned short trans_reserved; +}; + static int mpc_record; static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __cpuinitdata; + +static inline int generate_logical_apicid(int quad, int phys_apicid) +{ + return (quad << 4) + (phys_apicid ? phys_apicid << 1 : 1); +} + + +static inline int mpc_apic_id(struct mpc_config_processor *m, + struct mpc_config_translation *translation_record) +{ + int quad = translation_record->trans_quad; + int logical_apicid = generate_logical_apicid(quad, m->mpc_apicid); + + printk(KERN_DEBUG "Processor #%d %u:%u APIC version %d (quad %d, apic %d)\n", + m->mpc_apicid, + (m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8, + (m->mpc_cpufeature & CPU_MODEL_MASK) >> 4, + m->mpc_apicver, quad, logical_apicid); + return logical_apicid; +} + +int mp_bus_id_to_node[MAX_MP_BUSSES]; + +int mp_bus_id_to_local[MAX_MP_BUSSES]; + +static void mpc_oem_bus_info(struct mpc_config_bus *m, char *name, + struct mpc_config_translation *translation) +{ + int quad = translation->trans_quad; + int local = translation->trans_local; + + mp_bus_id_to_node[m->mpc_busid] = quad; + mp_bus_id_to_local[m->mpc_busid] = local; + printk(KERN_INFO "Bus #%d is %s (node %d)\n", + m->mpc_busid, name, quad); +} + +int quad_local_to_mp_bus_id [NR_CPUS/4][4]; +static void mpc_oem_pci_bus(struct mpc_config_bus *m, + struct mpc_config_translation *translation) +{ + int quad = translation->trans_quad; + int local = translation->trans_local; + + quad_local_to_mp_bus_id[quad][local] = m->mpc_busid; +} + #endif static void __cpuinit MP_processor_info(struct mpc_config_processor *m) @@ -90,7 +128,10 @@ static void __cpuinit MP_processor_info(struct mpc_config_processor *m) return; } #ifdef CONFIG_X86_NUMAQ - apicid = mpc_apic_id(m, translation_table[mpc_record]); + if (found_numaq) + apicid = mpc_apic_id(m, translation_table[mpc_record]); + else + apicid = m->mpc_apicid; #else apicid = m->mpc_apicid; #endif @@ -103,17 +144,18 @@ static void __cpuinit MP_processor_info(struct mpc_config_processor *m) generic_processor_info(apicid, m->mpc_apicver); } +#ifdef CONFIG_X86_IO_APIC static void __init MP_bus_info(struct mpc_config_bus *m) { char str[7]; - memcpy(str, m->mpc_bustype, 6); str[6] = 0; #ifdef CONFIG_X86_NUMAQ - mpc_oem_bus_info(m, str, translation_table[mpc_record]); + if (found_numaq) + mpc_oem_bus_info(m, str, translation_table[mpc_record]); #else - Dprintk("Bus #%d is %s\n", m->mpc_busid, str); + printk(KERN_INFO "Bus #%d is %s\n", m->mpc_busid, str); #endif #if MAX_MP_BUSSES < 256 @@ -132,11 +174,10 @@ static void __init MP_bus_info(struct mpc_config_bus *m) #endif } else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI) - 1) == 0) { #ifdef CONFIG_X86_NUMAQ - mpc_oem_pci_bus(m, translation_table[mpc_record]); + if (found_numaq) + mpc_oem_pci_bus(m, translation_table[mpc_record]); #endif clear_bit(m->mpc_busid, mp_bus_not_pci); - mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id; - mp_current_pci_id++; #if defined(CONFIG_EISA) || defined (CONFIG_MCA) mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI; } else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA) - 1) == 0) { @@ -147,6 +188,7 @@ static void __init MP_bus_info(struct mpc_config_bus *m) } else printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str); } +#endif #ifdef CONFIG_X86_IO_APIC @@ -176,18 +218,89 @@ static void __init MP_ioapic_info(struct mpc_config_ioapic *m) if (bad_ioapic(m->mpc_apicaddr)) return; - mp_ioapics[nr_ioapics] = *m; + mp_ioapics[nr_ioapics].mp_apicaddr = m->mpc_apicaddr; + mp_ioapics[nr_ioapics].mp_apicid = m->mpc_apicid; + mp_ioapics[nr_ioapics].mp_type = m->mpc_type; + mp_ioapics[nr_ioapics].mp_apicver = m->mpc_apicver; + mp_ioapics[nr_ioapics].mp_flags = m->mpc_flags; nr_ioapics++; } -static void __init MP_intsrc_info(struct mpc_config_intsrc *m) +static void print_MP_intsrc_info(struct mpc_config_intsrc *m) { - mp_irqs[mp_irq_entries] = *m; - Dprintk("Int: type %d, pol %d, trig %d, bus %d," + printk(KERN_CONT "Int: type %d, pol %d, trig %d, bus %02x," " IRQ %02x, APIC ID %x, APIC INT %02x\n", m->mpc_irqtype, m->mpc_irqflag & 3, (m->mpc_irqflag >> 2) & 3, m->mpc_srcbus, m->mpc_srcbusirq, m->mpc_dstapic, m->mpc_dstirq); +} + +static void __init print_mp_irq_info(struct mp_config_intsrc *mp_irq) +{ + printk(KERN_CONT "Int: type %d, pol %d, trig %d, bus %02x," + " IRQ %02x, APIC ID %x, APIC INT %02x\n", + mp_irq->mp_irqtype, mp_irq->mp_irqflag & 3, + (mp_irq->mp_irqflag >> 2) & 3, mp_irq->mp_srcbus, + mp_irq->mp_srcbusirq, mp_irq->mp_dstapic, mp_irq->mp_dstirq); +} + +static void assign_to_mp_irq(struct mpc_config_intsrc *m, + struct mp_config_intsrc *mp_irq) +{ + mp_irq->mp_dstapic = m->mpc_dstapic; + mp_irq->mp_type = m->mpc_type; + mp_irq->mp_irqtype = m->mpc_irqtype; + mp_irq->mp_irqflag = m->mpc_irqflag; + mp_irq->mp_srcbus = m->mpc_srcbus; + mp_irq->mp_srcbusirq = m->mpc_srcbusirq; + mp_irq->mp_dstirq = m->mpc_dstirq; +} + +static void __init assign_to_mpc_intsrc(struct mp_config_intsrc *mp_irq, + struct mpc_config_intsrc *m) +{ + m->mpc_dstapic = mp_irq->mp_dstapic; + m->mpc_type = mp_irq->mp_type; + m->mpc_irqtype = mp_irq->mp_irqtype; + m->mpc_irqflag = mp_irq->mp_irqflag; + m->mpc_srcbus = mp_irq->mp_srcbus; + m->mpc_srcbusirq = mp_irq->mp_srcbusirq; + m->mpc_dstirq = mp_irq->mp_dstirq; +} + +static int mp_irq_mpc_intsrc_cmp(struct mp_config_intsrc *mp_irq, + struct mpc_config_intsrc *m) +{ + if (mp_irq->mp_dstapic != m->mpc_dstapic) + return 1; + if (mp_irq->mp_type != m->mpc_type) + return 2; + if (mp_irq->mp_irqtype != m->mpc_irqtype) + return 3; + if (mp_irq->mp_irqflag != m->mpc_irqflag) + return 4; + if (mp_irq->mp_srcbus != m->mpc_srcbus) + return 5; + if (mp_irq->mp_srcbusirq != m->mpc_srcbusirq) + return 6; + if (mp_irq->mp_dstirq != m->mpc_dstirq) + return 7; + + return 0; +} + +void MP_intsrc_info(struct mpc_config_intsrc *m) +{ + int i; + + print_MP_intsrc_info(m); + + for (i = 0; i < mp_irq_entries; i++) { + if (!mp_irq_mpc_intsrc_cmp(&mp_irqs[i], m)) + return; + } + + assign_to_mp_irq(m, &mp_irqs[mp_irq_entries]); if (++mp_irq_entries == MAX_IRQ_SOURCES) panic("Max # of irq sources exceeded!!\n"); } @@ -196,7 +309,7 @@ static void __init MP_intsrc_info(struct mpc_config_intsrc *m) static void __init MP_lintsrc_info(struct mpc_config_lintsrc *m) { - Dprintk("Lint: type %d, pol %d, trig %d, bus %d," + printk(KERN_INFO "Lint: type %d, pol %d, trig %d, bus %02x," " IRQ %02x, APIC ID %x, APIC LINT %02x\n", m->mpc_irqtype, m->mpc_irqflag & 3, (m->mpc_irqflag >> 2) & 3, m->mpc_srcbusid, @@ -266,11 +379,14 @@ static void __init smp_read_mpc_oem(struct mp_config_oemtable *oemtable, } } -static inline void mps_oem_check(struct mp_config_table *mpc, char *oem, +void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem, char *productid) { if (strncmp(oem, "IBM NUMA", 8)) - printk("Warning! May not be a NUMA-Q system!\n"); + printk("Warning! Not a NUMA-Q system!\n"); + else + found_numaq = 1; + if (mpc->mpc_oemptr) smp_read_mpc_oem((struct mp_config_oemtable *)mpc->mpc_oemptr, mpc->mpc_oemsize); @@ -281,12 +397,9 @@ static inline void mps_oem_check(struct mp_config_table *mpc, char *oem, * Read/parse the MPC */ -static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early) +static int __init smp_check_mpc(struct mp_config_table *mpc, char *oem, + char *str) { - char str[16]; - char oem[10]; - int count = sizeof(*mpc); - unsigned char *mpt = ((unsigned char *)mpc) + count; if (memcmp(mpc->mpc_signature, MPC_SIGNATURE, 4)) { printk(KERN_ERR "MPTABLE: bad signature [%c%c%c%c]!\n", @@ -309,19 +422,42 @@ static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early) } memcpy(oem, mpc->mpc_oem, 8); oem[8] = 0; - printk(KERN_INFO "MPTABLE: OEM ID: %s ", oem); + printk(KERN_INFO "MPTABLE: OEM ID: %s\n", oem); memcpy(str, mpc->mpc_productid, 12); str[12] = 0; - printk("Product ID: %s ", str); -#ifdef CONFIG_X86_32 - mps_oem_check(mpc, oem, str); -#endif - printk(KERN_INFO "MPTABLE: Product ID: %s ", str); + printk(KERN_INFO "MPTABLE: Product ID: %s\n", str); printk(KERN_INFO "MPTABLE: APIC at: 0x%X\n", mpc->mpc_lapic); + return 1; +} + +static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early) +{ + char str[16]; + char oem[10]; + + int count = sizeof(*mpc); + unsigned char *mpt = ((unsigned char *)mpc) + count; + + if (!smp_check_mpc(mpc, oem, str)) + return 0; + +#ifdef CONFIG_X86_32 + /* + * need to make sure summit and es7000's mps_oem_check is safe to be + * called early via genericarch 's mps_oem_check + */ + if (early) { +#ifdef CONFIG_X86_NUMAQ + numaq_mps_oem_check(mpc, oem, str); +#endif + } else + mps_oem_check(mpc, oem, str); +#endif + /* save the local APIC address, it might be non-default */ if (!acpi_lapic) mp_lapic_addr = mpc->mpc_lapic; @@ -352,7 +488,9 @@ static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early) { struct mpc_config_bus *m = (struct mpc_config_bus *)mpt; +#ifdef CONFIG_X86_IO_APIC MP_bus_info(m); +#endif mpt += sizeof(*m); count += sizeof(*m); break; @@ -402,6 +540,11 @@ static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early) ++mpc_record; #endif } + +#ifdef CONFIG_X86_GENERICARCH + generic_bigsmp_probe(); +#endif + setup_apic_routing(); if (!num_processors) printk(KERN_ERR "MPTABLE: no processors registered!\n"); @@ -427,7 +570,7 @@ static void __init construct_default_ioirq_mptable(int mpc_default_type) intsrc.mpc_type = MP_INTSRC; intsrc.mpc_irqflag = 0; /* conforming */ intsrc.mpc_srcbus = 0; - intsrc.mpc_dstapic = mp_ioapics[0].mpc_apicid; + intsrc.mpc_dstapic = mp_ioapics[0].mp_apicid; intsrc.mpc_irqtype = mp_INT; @@ -488,40 +631,11 @@ static void __init construct_default_ioirq_mptable(int mpc_default_type) MP_intsrc_info(&intsrc); } -#endif -static inline void __init construct_default_ISA_mptable(int mpc_default_type) +static void construct_ioapic_table(int mpc_default_type) { - struct mpc_config_processor processor; - struct mpc_config_bus bus; -#ifdef CONFIG_X86_IO_APIC struct mpc_config_ioapic ioapic; -#endif - struct mpc_config_lintsrc lintsrc; - int linttypes[2] = { mp_ExtINT, mp_NMI }; - int i; - - /* - * local APIC has default address - */ - mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; - - /* - * 2 CPUs, numbered 0 & 1. - */ - processor.mpc_type = MP_PROCESSOR; - /* Either an integrated APIC or a discrete 82489DX. */ - processor.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01; - processor.mpc_cpuflag = CPU_ENABLED; - processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) | - (boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask; - processor.mpc_featureflag = boot_cpu_data.x86_capability[0]; - processor.mpc_reserved[0] = 0; - processor.mpc_reserved[1] = 0; - for (i = 0; i < 2; i++) { - processor.mpc_apicid = i; - MP_processor_info(&processor); - } + struct mpc_config_bus bus; bus.mpc_type = MP_BUS; bus.mpc_busid = 0; @@ -550,7 +664,6 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type) MP_bus_info(&bus); } -#ifdef CONFIG_X86_IO_APIC ioapic.mpc_type = MP_IOAPIC; ioapic.mpc_apicid = 2; ioapic.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01; @@ -562,7 +675,42 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type) * We set up most of the low 16 IO-APIC pins according to MPS rules. */ construct_default_ioirq_mptable(mpc_default_type); +} +#else +static inline void construct_ioapic_table(int mpc_default_type) { } #endif + +static inline void __init construct_default_ISA_mptable(int mpc_default_type) +{ + struct mpc_config_processor processor; + struct mpc_config_lintsrc lintsrc; + int linttypes[2] = { mp_ExtINT, mp_NMI }; + int i; + + /* + * local APIC has default address + */ + mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; + + /* + * 2 CPUs, numbered 0 & 1. + */ + processor.mpc_type = MP_PROCESSOR; + /* Either an integrated APIC or a discrete 82489DX. */ + processor.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01; + processor.mpc_cpuflag = CPU_ENABLED; + processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) | + (boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask; + processor.mpc_featureflag = boot_cpu_data.x86_capability[0]; + processor.mpc_reserved[0] = 0; + processor.mpc_reserved[1] = 0; + for (i = 0; i < 2; i++) { + processor.mpc_apicid = i; + MP_processor_info(&processor); + } + + construct_ioapic_table(mpc_default_type); + lintsrc.mpc_type = MP_LINTSRC; lintsrc.mpc_irqflag = 0; /* conforming */ lintsrc.mpc_srcbusid = 0; @@ -600,7 +748,7 @@ static void __init __get_smp_config(unsigned early) printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification); -#ifdef CONFIG_X86_32 +#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86_32) if (mpf->mpf_feature2 & (1 << 7)) { printk(KERN_INFO " IMCR and PIC compatibility mode.\n"); pic_mode = 1; @@ -632,7 +780,9 @@ static void __init __get_smp_config(unsigned early) * override the defaults. */ if (!smp_read_mpc(phys_to_virt(mpf->mpf_physptr), early)) { +#ifdef CONFIG_X86_LOCAL_APIC smp_found_config = 0; +#endif printk(KERN_ERR "BIOS bug, MP table errors detected!...\n"); printk(KERN_ERR "... disabling SMP support. " @@ -689,7 +839,7 @@ static int __init smp_scan_config(unsigned long base, unsigned long length, unsigned int *bp = phys_to_virt(base); struct intel_mp_floating *mpf; - Dprintk("Scan SMP from %p for %ld bytes.\n", bp, length); + printk(KERN_DEBUG "Scan SMP from %p for %ld bytes.\n", bp, length); BUILD_BUG_ON(sizeof(*mpf) != 16); while (length > 0) { @@ -699,8 +849,9 @@ static int __init smp_scan_config(unsigned long base, unsigned long length, !mpf_checksum((unsigned char *)bp, 16) && ((mpf->mpf_specification == 1) || (mpf->mpf_specification == 4))) { - +#ifdef CONFIG_X86_LOCAL_APIC smp_found_config = 1; +#endif mpf_found = mpf; #ifdef CONFIG_X86_32 printk(KERN_INFO "found SMP MP-table at [%p] %08lx\n", @@ -790,298 +941,294 @@ void __init find_smp_config(void) __find_smp_config(1); } -/* -------------------------------------------------------------------------- - ACPI-based MP Configuration - -------------------------------------------------------------------------- */ +#ifdef CONFIG_X86_IO_APIC +static u8 __initdata irq_used[MAX_IRQ_SOURCES]; -/* - * Keep this outside and initialized to 0, for !CONFIG_ACPI builds: - */ -int es7000_plat; +static int __init get_MP_intsrc_index(struct mpc_config_intsrc *m) +{ + int i; -#ifdef CONFIG_ACPI + if (m->mpc_irqtype != mp_INT) + return 0; -#ifdef CONFIG_X86_IO_APIC + if (m->mpc_irqflag != 0x0f) + return 0; -#define MP_ISA_BUS 0 + /* not legacy */ -extern struct mp_ioapic_routing mp_ioapic_routing[MAX_IO_APICS]; + for (i = 0; i < mp_irq_entries; i++) { + if (mp_irqs[i].mp_irqtype != mp_INT) + continue; -static int mp_find_ioapic(int gsi) -{ - int i = 0; + if (mp_irqs[i].mp_irqflag != 0x0f) + continue; - /* Find the IOAPIC that manages this GSI. */ - for (i = 0; i < nr_ioapics; i++) { - if ((gsi >= mp_ioapic_routing[i].gsi_base) - && (gsi <= mp_ioapic_routing[i].gsi_end)) - return i; + if (mp_irqs[i].mp_srcbus != m->mpc_srcbus) + continue; + if (mp_irqs[i].mp_srcbusirq != m->mpc_srcbusirq) + continue; + if (irq_used[i]) { + /* already claimed */ + return -2; + } + irq_used[i] = 1; + return i; } - printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi); + /* not found */ return -1; } -static u8 __init uniq_ioapic_id(u8 id) -{ -#ifdef CONFIG_X86_32 - if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && - !APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) - return io_apic_get_unique_id(nr_ioapics, id); - else - return id; -#else - int i; - DECLARE_BITMAP(used, 256); - bitmap_zero(used, 256); - for (i = 0; i < nr_ioapics; i++) { - struct mpc_config_ioapic *ia = &mp_ioapics[i]; - __set_bit(ia->mpc_apicid, used); - } - if (!test_bit(id, used)) - return id; - return find_first_zero_bit(used, 256); +#define SPARE_SLOT_NUM 20 + +static struct mpc_config_intsrc __initdata *m_spare[SPARE_SLOT_NUM]; #endif -} -void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) +static int __init replace_intsrc_all(struct mp_config_table *mpc, + unsigned long mpc_new_phys, + unsigned long mpc_new_length) { - int idx = 0; - - if (bad_ioapic(address)) - return; +#ifdef CONFIG_X86_IO_APIC + int i; + int nr_m_spare = 0; +#endif - idx = nr_ioapics; + int count = sizeof(*mpc); + unsigned char *mpt = ((unsigned char *)mpc) + count; - mp_ioapics[idx].mpc_type = MP_IOAPIC; - mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE; - mp_ioapics[idx].mpc_apicaddr = address; + printk(KERN_INFO "mpc_length %x\n", mpc->mpc_length); + while (count < mpc->mpc_length) { + switch (*mpt) { + case MP_PROCESSOR: + { + struct mpc_config_processor *m = + (struct mpc_config_processor *)mpt; + mpt += sizeof(*m); + count += sizeof(*m); + break; + } + case MP_BUS: + { + struct mpc_config_bus *m = + (struct mpc_config_bus *)mpt; + mpt += sizeof(*m); + count += sizeof(*m); + break; + } + case MP_IOAPIC: + { + mpt += sizeof(struct mpc_config_ioapic); + count += sizeof(struct mpc_config_ioapic); + break; + } + case MP_INTSRC: + { +#ifdef CONFIG_X86_IO_APIC + struct mpc_config_intsrc *m = + (struct mpc_config_intsrc *)mpt; - set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address); - mp_ioapics[idx].mpc_apicid = uniq_ioapic_id(id); -#ifdef CONFIG_X86_32 - mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx); -#else - mp_ioapics[idx].mpc_apicver = 0; + printk(KERN_INFO "OLD "); + print_MP_intsrc_info(m); + i = get_MP_intsrc_index(m); + if (i > 0) { + assign_to_mpc_intsrc(&mp_irqs[i], m); + printk(KERN_INFO "NEW "); + print_mp_irq_info(&mp_irqs[i]); + } else if (!i) { + /* legacy, do nothing */ + } else if (nr_m_spare < SPARE_SLOT_NUM) { + /* + * not found (-1), or duplicated (-2) + * are invalid entries, + * we need to use the slot later + */ + m_spare[nr_m_spare] = m; + nr_m_spare++; + } #endif - /* - * Build basic GSI lookup table to facilitate gsi->io_apic lookups - * and to prevent reprogramming of IOAPIC pins (PCI GSIs). - */ - mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mpc_apicid; - mp_ioapic_routing[idx].gsi_base = gsi_base; - mp_ioapic_routing[idx].gsi_end = gsi_base + - io_apic_get_redir_entries(idx); - - printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%x, " - "GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid, - mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr, - mp_ioapic_routing[idx].gsi_base, mp_ioapic_routing[idx].gsi_end); - - nr_ioapics++; -} + mpt += sizeof(struct mpc_config_intsrc); + count += sizeof(struct mpc_config_intsrc); + break; + } + case MP_LINTSRC: + { + struct mpc_config_lintsrc *m = + (struct mpc_config_lintsrc *)mpt; + mpt += sizeof(*m); + count += sizeof(*m); + break; + } + default: + /* wrong mptable */ + printk(KERN_ERR "Your mptable is wrong, contact your HW vendor!\n"); + printk(KERN_ERR "type %x\n", *mpt); + print_hex_dump(KERN_ERR, " ", DUMP_PREFIX_ADDRESS, 16, + 1, mpc, mpc->mpc_length, 1); + goto out; + } + } -void __init mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) -{ - struct mpc_config_intsrc intsrc; - int ioapic = -1; - int pin = -1; +#ifdef CONFIG_X86_IO_APIC + for (i = 0; i < mp_irq_entries; i++) { + if (irq_used[i]) + continue; - /* - * Convert 'gsi' to 'ioapic.pin'. - */ - ioapic = mp_find_ioapic(gsi); - if (ioapic < 0) - return; - pin = gsi - mp_ioapic_routing[ioapic].gsi_base; + if (mp_irqs[i].mp_irqtype != mp_INT) + continue; - /* - * TBD: This check is for faulty timer entries, where the override - * erroneously sets the trigger to level, resulting in a HUGE - * increase of timer interrupts! - */ - if ((bus_irq == 0) && (trigger == 3)) - trigger = 1; + if (mp_irqs[i].mp_irqflag != 0x0f) + continue; - intsrc.mpc_type = MP_INTSRC; - intsrc.mpc_irqtype = mp_INT; - intsrc.mpc_irqflag = (trigger << 2) | polarity; - intsrc.mpc_srcbus = MP_ISA_BUS; - intsrc.mpc_srcbusirq = bus_irq; /* IRQ */ - intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; /* APIC ID */ - intsrc.mpc_dstirq = pin; /* INTIN# */ + if (nr_m_spare > 0) { + printk(KERN_INFO "*NEW* found "); + nr_m_spare--; + assign_to_mpc_intsrc(&mp_irqs[i], m_spare[nr_m_spare]); + m_spare[nr_m_spare] = NULL; + } else { + struct mpc_config_intsrc *m = + (struct mpc_config_intsrc *)mpt; + count += sizeof(struct mpc_config_intsrc); + if (!mpc_new_phys) { + printk(KERN_INFO "No spare slots, try to append...take your risk, new mpc_length %x\n", count); + } else { + if (count <= mpc_new_length) + printk(KERN_INFO "No spare slots, try to append..., new mpc_length %x\n", count); + else { + printk(KERN_ERR "mpc_new_length %lx is too small\n", mpc_new_length); + goto out; + } + } + assign_to_mpc_intsrc(&mp_irqs[i], m); + mpc->mpc_length = count; + mpt += sizeof(struct mpc_config_intsrc); + } + print_mp_irq_info(&mp_irqs[i]); + } +#endif +out: + /* update checksum */ + mpc->mpc_checksum = 0; + mpc->mpc_checksum -= mpf_checksum((unsigned char *)mpc, + mpc->mpc_length); - MP_intsrc_info(&intsrc); + return 0; } -void __init mp_config_acpi_legacy_irqs(void) -{ - struct mpc_config_intsrc intsrc; - int i = 0; - int ioapic = -1; +int __initdata enable_update_mptable; -#if defined (CONFIG_MCA) || defined (CONFIG_EISA) - /* - * Fabricate the legacy ISA bus (bus #31). - */ - mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA; -#endif - set_bit(MP_ISA_BUS, mp_bus_not_pci); - Dprintk("Bus #%d is ISA\n", MP_ISA_BUS); +static int __init update_mptable_setup(char *str) +{ + enable_update_mptable = 1; + return 0; +} +early_param("update_mptable", update_mptable_setup); - /* - * Older generations of ES7000 have no legacy identity mappings - */ - if (es7000_plat == 1) - return; +static unsigned long __initdata mpc_new_phys; +static unsigned long mpc_new_length __initdata = 4096; - /* - * Locate the IOAPIC that manages the ISA IRQs (0-15). - */ - ioapic = mp_find_ioapic(0); - if (ioapic < 0) - return; +/* alloc_mptable or alloc_mptable=4k */ +static int __initdata alloc_mptable; +static int __init parse_alloc_mptable_opt(char *p) +{ + enable_update_mptable = 1; + alloc_mptable = 1; + if (!p) + return 0; + mpc_new_length = memparse(p, &p); + return 0; +} +early_param("alloc_mptable", parse_alloc_mptable_opt); - intsrc.mpc_type = MP_INTSRC; - intsrc.mpc_irqflag = 0; /* Conforming */ - intsrc.mpc_srcbus = MP_ISA_BUS; -#ifdef CONFIG_X86_IO_APIC - intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; +void __init early_reserve_e820_mpc_new(void) +{ + if (enable_update_mptable && alloc_mptable) { + u64 startt = 0; +#ifdef CONFIG_X86_TRAMPOLINE + startt = TRAMPOLINE_BASE; #endif - /* - * Use the default configuration for the IRQs 0-15. Unless - * overridden by (MADT) interrupt source override entries. - */ - for (i = 0; i < 16; i++) { - int idx; - - for (idx = 0; idx < mp_irq_entries; idx++) { - struct mpc_config_intsrc *irq = mp_irqs + idx; - - /* Do we already have a mapping for this ISA IRQ? */ - if (irq->mpc_srcbus == MP_ISA_BUS - && irq->mpc_srcbusirq == i) - break; - - /* Do we already have a mapping for this IOAPIC pin */ - if ((irq->mpc_dstapic == intsrc.mpc_dstapic) && - (irq->mpc_dstirq == i)) - break; - } - - if (idx != mp_irq_entries) { - printk(KERN_DEBUG "ACPI: IRQ%d used by override.\n", i); - continue; /* IRQ already used */ - } - - intsrc.mpc_irqtype = mp_INT; - intsrc.mpc_srcbusirq = i; /* Identity mapped */ - intsrc.mpc_dstirq = i; - - MP_intsrc_info(&intsrc); + mpc_new_phys = early_reserve_e820(startt, mpc_new_length, 4); } } -int mp_register_gsi(u32 gsi, int triggering, int polarity) +static int __init update_mp_table(void) { - int ioapic; - int ioapic_pin; -#ifdef CONFIG_X86_32 -#define MAX_GSI_NUM 4096 -#define IRQ_COMPRESSION_START 64 + char str[16]; + char oem[10]; + struct intel_mp_floating *mpf; + struct mp_config_table *mpc; + struct mp_config_table *mpc_new; + + if (!enable_update_mptable) + return 0; + + mpf = mpf_found; + if (!mpf) + return 0; - static int pci_irq = IRQ_COMPRESSION_START; /* - * Mapping between Global System Interrupts, which - * represent all possible interrupts, and IRQs - * assigned to actual devices. + * Now see if we need to go further. */ - static int gsi_to_irq[MAX_GSI_NUM]; -#else - - if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC) - return gsi; -#endif + if (mpf->mpf_feature1 != 0) + return 0; - /* Don't set up the ACPI SCI because it's already set up */ - if (acpi_gbl_FADT.sci_interrupt == gsi) - return gsi; + if (!mpf->mpf_physptr) + return 0; - ioapic = mp_find_ioapic(gsi); - if (ioapic < 0) { - printk(KERN_WARNING "No IOAPIC for GSI %u\n", gsi); - return gsi; - } + mpc = phys_to_virt(mpf->mpf_physptr); - ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_base; + if (!smp_check_mpc(mpc, oem, str)) + return 0; -#ifdef CONFIG_X86_32 - if (ioapic_renumber_irq) - gsi = ioapic_renumber_irq(ioapic, gsi); -#endif + printk(KERN_INFO "mpf: %lx\n", virt_to_phys(mpf)); + printk(KERN_INFO "mpf_physptr: %x\n", mpf->mpf_physptr); - /* - * Avoid pin reprogramming. PRTs typically include entries - * with redundant pin->gsi mappings (but unique PCI devices); - * we only program the IOAPIC on the first. - */ - if (ioapic_pin > MP_MAX_IOAPIC_PIN) { - printk(KERN_ERR "Invalid reference to IOAPIC pin " - "%d-%d\n", mp_ioapic_routing[ioapic].apic_id, - ioapic_pin); - return gsi; + if (mpc_new_phys && mpc->mpc_length > mpc_new_length) { + mpc_new_phys = 0; + printk(KERN_INFO "mpc_new_length is %ld, please use alloc_mptable=8k\n", + mpc_new_length); } - if (test_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed)) { - Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n", - mp_ioapic_routing[ioapic].apic_id, ioapic_pin); -#ifdef CONFIG_X86_32 - return (gsi < IRQ_COMPRESSION_START ? gsi : gsi_to_irq[gsi]); -#else - return gsi; -#endif + + if (!mpc_new_phys) { + unsigned char old, new; + /* check if we can change the postion */ + mpc->mpc_checksum = 0; + old = mpf_checksum((unsigned char *)mpc, mpc->mpc_length); + mpc->mpc_checksum = 0xff; + new = mpf_checksum((unsigned char *)mpc, mpc->mpc_length); + if (old == new) { + printk(KERN_INFO "mpc is readonly, please try alloc_mptable instead\n"); + return 0; + } + printk(KERN_INFO "use in-positon replacing\n"); + } else { + mpf->mpf_physptr = mpc_new_phys; + mpc_new = phys_to_virt(mpc_new_phys); + memcpy(mpc_new, mpc, mpc->mpc_length); + mpc = mpc_new; + /* check if we can modify that */ + if (mpc_new_phys - mpf->mpf_physptr) { + struct intel_mp_floating *mpf_new; + /* steal 16 bytes from [0, 1k) */ + printk(KERN_INFO "mpf new: %x\n", 0x400 - 16); + mpf_new = phys_to_virt(0x400 - 16); + memcpy(mpf_new, mpf, 16); + mpf = mpf_new; + mpf->mpf_physptr = mpc_new_phys; + } + mpf->mpf_checksum = 0; + mpf->mpf_checksum -= mpf_checksum((unsigned char *)mpf, 16); + printk(KERN_INFO "mpf_physptr new: %x\n", mpf->mpf_physptr); } - set_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed); -#ifdef CONFIG_X86_32 /* - * For GSI >= 64, use IRQ compression + * only replace the one with mp_INT and + * MP_IRQ_TRIGGER_LEVEL|MP_IRQ_POLARITY_LOW, + * already in mp_irqs , stored by ... and mp_config_acpi_gsi, + * may need pci=routeirq for all coverage */ - if ((gsi >= IRQ_COMPRESSION_START) - && (triggering == ACPI_LEVEL_SENSITIVE)) { - /* - * For PCI devices assign IRQs in order, avoiding gaps - * due to unused I/O APIC pins. - */ - int irq = gsi; - if (gsi < MAX_GSI_NUM) { - /* - * Retain the VIA chipset work-around (gsi > 15), but - * avoid a problem where the 8254 timer (IRQ0) is setup - * via an override (so it's not on pin 0 of the ioapic), - * and at the same time, the pin 0 interrupt is a PCI - * type. The gsi > 15 test could cause these two pins - * to be shared as IRQ0, and they are not shareable. - * So test for this condition, and if necessary, avoid - * the pin collision. - */ - gsi = pci_irq++; - /* - * Don't assign IRQ used by ACPI SCI - */ - if (gsi == acpi_gbl_FADT.sci_interrupt) - gsi = pci_irq++; - gsi_to_irq[irq] = gsi; - } else { - printk(KERN_ERR "GSI %u is too high\n", gsi); - return gsi; - } - } -#endif - io_apic_set_pci_routing(ioapic, ioapic_pin, gsi, - triggering == ACPI_EDGE_SENSITIVE ? 0 : 1, - polarity == ACPI_ACTIVE_HIGH ? 0 : 1); - return gsi; + replace_intsrc_all(mpc, mpc_new_phys, mpc_new_length); + + return 0; } -#endif /* CONFIG_X86_IO_APIC */ -#endif /* CONFIG_ACPI */ +late_initcall(update_mp_table); diff --git a/arch/x86/kernel/numaq_32.c b/arch/x86/kernel/numaq_32.c index e65281b1634..f0f1de1c4a1 100644 --- a/arch/x86/kernel/numaq_32.c +++ b/arch/x86/kernel/numaq_32.c @@ -31,6 +31,8 @@ #include <asm/numaq.h> #include <asm/topology.h> #include <asm/processor.h> +#include <asm/mpspec.h> +#include <asm/e820.h> #define MB_TO_PAGES(addr) ((addr) << (20 - PAGE_SHIFT)) @@ -58,6 +60,8 @@ static void __init smp_dump_qct(void) node_end_pfn[node] = MB_TO_PAGES( eq->hi_shrd_mem_start + eq->hi_shrd_mem_size); + e820_register_active_regions(node, node_start_pfn[node], + node_end_pfn[node]); memory_present(node, node_start_pfn[node], node_end_pfn[node]); node_remap_size[node] = node_memmap_size_bytes(node, @@ -67,13 +71,24 @@ static void __init smp_dump_qct(void) } } -/* - * Unlike Summit, we don't really care to let the NUMA-Q - * fall back to flat mode. Don't compile for NUMA-Q - * unless you really need it! - */ +static __init void early_check_numaq(void) +{ + /* + * Find possible boot-time SMP configuration: + */ + early_find_smp_config(); + /* + * get boot-time SMP configuration: + */ + if (smp_found_config) + early_get_smp_config(); +} + int __init get_memcfg_numaq(void) { + early_check_numaq(); + if (!found_numaq) + return 0; smp_dump_qct(); return 1; } diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 6f80b852a19..45a5e247d45 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -17,6 +17,7 @@ unsigned int num_processors; unsigned disabled_cpus __cpuinitdata; /* Processor that is doing the boot up */ unsigned int boot_cpu_physical_apicid = -1U; +unsigned int max_physical_apicid; EXPORT_SYMBOL(boot_cpu_physical_apicid); DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID; @@ -137,3 +138,25 @@ void __init setup_per_cpu_areas(void) } #endif + +void __init parse_setup_data(void) +{ + struct setup_data *data; + u64 pa_data; + + if (boot_params.hdr.version < 0x0209) + return; + pa_data = boot_params.hdr.setup_data; + while (pa_data) { + data = early_ioremap(pa_data, PAGE_SIZE); + switch (data->type) { + default: + break; + } +#ifndef CONFIG_DEBUG_BOOT_PARAMS + free_early(pa_data, pa_data+sizeof(*data)+data->len); +#endif + pa_data = data->next; + early_iounmap(data, PAGE_SIZE); + } +} diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c index 5a2f8e06388..1d4be07e15e 100644 --- a/arch/x86/kernel/setup_32.c +++ b/arch/x86/kernel/setup_32.c @@ -67,10 +67,12 @@ #include <asm/bios_ebda.h> #include <asm/cacheflush.h> #include <asm/processor.h> +#include <asm/efi.h> /* This value is set up by the early boot code to point to the value immediately after the boot time page tables. It contains a *physical* address, and must not be in the .bss segment! */ +unsigned long init_pg_tables_start __initdata = ~0UL; unsigned long init_pg_tables_end __initdata = ~0UL; /* @@ -237,42 +239,6 @@ static inline void copy_edd(void) } #endif -int __initdata user_defined_memmap; - -/* - * "mem=nopentium" disables the 4MB page tables. - * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM - * to <mem>, overriding the bios size. - * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from - * <start> to <start>+<mem>, overriding the bios size. - * - * HPA tells me bootloaders need to parse mem=, so no new - * option should be mem= [also see Documentation/i386/boot.txt] - */ -static int __init parse_mem(char *arg) -{ - if (!arg) - return -EINVAL; - - if (strcmp(arg, "nopentium") == 0) { - setup_clear_cpu_cap(X86_FEATURE_PSE); - } else { - /* If the user specifies memory size, we - * limit the BIOS-provided memory map to - * that size. exactmap can be used to specify - * the exact map. mem=number can be used to - * trim the existing memory map. - */ - unsigned long long mem_size; - - mem_size = memparse(arg, &arg); - limit_regions(mem_size); - user_defined_memmap = 1; - } - return 0; -} -early_param("mem", parse_mem); - #ifdef CONFIG_PROC_VMCORE /* elfcorehdr= specifies the location of elf core header * stored by the crashed kernel. @@ -395,56 +361,6 @@ unsigned long __init find_max_low_pfn(void) return max_low_pfn; } -#define BIOS_LOWMEM_KILOBYTES 0x413 - -/* - * The BIOS places the EBDA/XBDA at the top of conventional - * memory, and usually decreases the reported amount of - * conventional memory (int 0x12) too. This also contains a - * workaround for Dell systems that neglect to reserve EBDA. - * The same workaround also avoids a problem with the AMD768MPX - * chipset: reserve a page before VGA to prevent PCI prefetch - * into it (errata #56). Usually the page is reserved anyways, - * unless you have no PS/2 mouse plugged in. - */ -static void __init reserve_ebda_region(void) -{ - unsigned int lowmem, ebda_addr; - - /* To determine the position of the EBDA and the */ - /* end of conventional memory, we need to look at */ - /* the BIOS data area. In a paravirtual environment */ - /* that area is absent. We'll just have to assume */ - /* that the paravirt case can handle memory setup */ - /* correctly, without our help. */ - if (paravirt_enabled()) - return; - - /* end of low (conventional) memory */ - lowmem = *(unsigned short *)__va(BIOS_LOWMEM_KILOBYTES); - lowmem <<= 10; - - /* start of EBDA area */ - ebda_addr = get_bios_ebda(); - - /* Fixup: bios puts an EBDA in the top 64K segment */ - /* of conventional memory, but does not adjust lowmem. */ - if ((lowmem - ebda_addr) <= 0x10000) - lowmem = ebda_addr; - - /* Fixup: bios does not report an EBDA at all. */ - /* Some old Dells seem to need 4k anyhow (bugzilla 2990) */ - if ((ebda_addr == 0) && (lowmem >= 0x9f000)) - lowmem = 0x9f000; - - /* Paranoia: should never happen, but... */ - if ((lowmem == 0) || (lowmem >= 0x100000)) - lowmem = 0x9f000; - - /* reserve all memory between lowmem and the 1MB mark */ - reserve_bootmem(lowmem, 0x100000 - lowmem, BOOTMEM_DEFAULT); -} - #ifndef CONFIG_NEED_MULTIPLE_NODES static void __init setup_bootmem_allocator(void); static unsigned long __init setup_memory(void) @@ -462,11 +378,13 @@ static unsigned long __init setup_memory(void) if (max_pfn > max_low_pfn) { highstart_pfn = max_low_pfn; } + memory_present(0, 0, highend_pfn); printk(KERN_NOTICE "%ldMB HIGHMEM available.\n", pages_to_mb(highend_pfn - highstart_pfn)); num_physpages = highend_pfn; high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1; #else + memory_present(0, 0, max_low_pfn); num_physpages = max_low_pfn; high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1; #endif @@ -488,11 +406,12 @@ static void __init zone_sizes_init(void) max_zone_pfns[ZONE_DMA] = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT; max_zone_pfns[ZONE_NORMAL] = max_low_pfn; + remove_all_active_ranges(); #ifdef CONFIG_HIGHMEM max_zone_pfns[ZONE_HIGHMEM] = highend_pfn; - add_active_range(0, 0, highend_pfn); + e820_register_active_regions(0, 0, highend_pfn); #else - add_active_range(0, 0, max_low_pfn); + e820_register_active_regions(0, 0, max_low_pfn); #endif free_area_init_nodes(max_zone_pfns); @@ -558,44 +477,57 @@ static bool do_relocate_initrd = false; static void __init reserve_initrd(void) { - unsigned long ramdisk_image = boot_params.hdr.ramdisk_image; - unsigned long ramdisk_size = boot_params.hdr.ramdisk_size; - unsigned long ramdisk_end = ramdisk_image + ramdisk_size; - unsigned long end_of_lowmem = max_low_pfn << PAGE_SHIFT; - unsigned long ramdisk_here; - - initrd_start = 0; + u64 ramdisk_image = boot_params.hdr.ramdisk_image; + u64 ramdisk_size = boot_params.hdr.ramdisk_size; + u64 ramdisk_end = ramdisk_image + ramdisk_size; + u64 end_of_lowmem = max_low_pfn << PAGE_SHIFT; + u64 ramdisk_here; if (!boot_params.hdr.type_of_loader || !ramdisk_image || !ramdisk_size) return; /* No initrd provided by bootloader */ - if (ramdisk_end < ramdisk_image) { - printk(KERN_ERR "initrd wraps around end of memory, " - "disabling initrd\n"); - return; - } + initrd_start = 0; + if (ramdisk_size >= end_of_lowmem/2) { + free_early(ramdisk_image, ramdisk_end); printk(KERN_ERR "initrd too large to handle, " "disabling initrd\n"); return; } + + printk(KERN_INFO "old RAMDISK: %08llx - %08llx\n", ramdisk_image, + ramdisk_end); + + if (ramdisk_end <= end_of_lowmem) { /* All in lowmem, easy case */ - reserve_bootmem(ramdisk_image, ramdisk_size, BOOTMEM_DEFAULT); + /* + * don't need to reserve again, already reserved early + * in i386_start_kernel + */ initrd_start = ramdisk_image + PAGE_OFFSET; initrd_end = initrd_start+ramdisk_size; return; } /* We need to move the initrd down into lowmem */ - ramdisk_here = (end_of_lowmem - ramdisk_size) & PAGE_MASK; + ramdisk_here = find_e820_area(min_low_pfn<<PAGE_SHIFT, + end_of_lowmem, ramdisk_size, + PAGE_SIZE); + + if (ramdisk_here == -1ULL) + panic("Cannot find place for new RAMDISK of size %lld\n", + ramdisk_size); /* Note: this includes all the lowmem currently occupied by the initrd, we rely on that fact to keep the data intact. */ - reserve_bootmem(ramdisk_here, ramdisk_size, BOOTMEM_DEFAULT); + reserve_early(ramdisk_here, ramdisk_here + ramdisk_size, + "NEW RAMDISK"); initrd_start = ramdisk_here + PAGE_OFFSET; initrd_end = initrd_start + ramdisk_size; + printk(KERN_INFO "Allocated new RAMDISK: %08llx - %08llx\n", + ramdisk_here, ramdisk_here + ramdisk_size); do_relocate_initrd = true; } @@ -604,10 +536,10 @@ static void __init reserve_initrd(void) static void __init relocate_initrd(void) { - unsigned long ramdisk_image = boot_params.hdr.ramdisk_image; - unsigned long ramdisk_size = boot_params.hdr.ramdisk_size; - unsigned long end_of_lowmem = max_low_pfn << PAGE_SHIFT; - unsigned long ramdisk_here; + u64 ramdisk_image = boot_params.hdr.ramdisk_image; + u64 ramdisk_size = boot_params.hdr.ramdisk_size; + u64 end_of_lowmem = max_low_pfn << PAGE_SHIFT; + u64 ramdisk_here; unsigned long slop, clen, mapaddr; char *p, *q; @@ -624,6 +556,10 @@ static void __init relocate_initrd(void) p = (char *)__va(ramdisk_image); memcpy(q, p, clen); q += clen; + /* need to free these low pages...*/ + printk(KERN_INFO "Freeing old partial RAMDISK %08llx-%08llx\n", + ramdisk_image, ramdisk_image + clen - 1); + free_bootmem(ramdisk_image, clen); ramdisk_image += clen; ramdisk_size -= clen; } @@ -642,47 +578,44 @@ static void __init relocate_initrd(void) ramdisk_image += clen; ramdisk_size -= clen; } + /* high pages is not converted by early_res_to_bootmem */ + ramdisk_image = boot_params.hdr.ramdisk_image; + ramdisk_size = boot_params.hdr.ramdisk_size; + printk(KERN_INFO "Copied RAMDISK from %016llx - %016llx to %08llx - %08llx\n", + ramdisk_image, ramdisk_image + ramdisk_size - 1, + ramdisk_here, ramdisk_here + ramdisk_size - 1); } #endif /* CONFIG_BLK_DEV_INITRD */ void __init setup_bootmem_allocator(void) { - unsigned long bootmap_size; + int i; + unsigned long bootmap_size, bootmap; /* * Initialize the boot-time allocator (with low memory only): */ - bootmap_size = init_bootmem(min_low_pfn, max_low_pfn); - - register_bootmem_low_pages(max_low_pfn); - - /* - * Reserve the bootmem bitmap itself as well. We do this in two - * steps (first step was init_bootmem()) because this catches - * the (very unlikely) case of us accidentally initializing the - * bootmem allocator with an invalid RAM area. - */ - reserve_bootmem(__pa_symbol(_text), (PFN_PHYS(min_low_pfn) + - bootmap_size + PAGE_SIZE-1) - __pa_symbol(_text), - BOOTMEM_DEFAULT); - - /* - * reserve physical page 0 - it's a special BIOS page on many boxes, - * enabling clean reboots, SMP operation, laptop functions. - */ - reserve_bootmem(0, PAGE_SIZE, BOOTMEM_DEFAULT); - - /* reserve EBDA region */ - reserve_ebda_region(); - -#ifdef CONFIG_SMP - /* - * But first pinch a few for the stack/trampoline stuff - * FIXME: Don't need the extra page at 4K, but need to fix - * trampoline before removing it. (see the GDT stuff) - */ - reserve_bootmem(PAGE_SIZE, PAGE_SIZE, BOOTMEM_DEFAULT); + bootmap_size = bootmem_bootmap_pages(max_low_pfn)<<PAGE_SHIFT; + bootmap = find_e820_area(min_low_pfn<<PAGE_SHIFT, + max_pfn_mapped<<PAGE_SHIFT, bootmap_size, + PAGE_SIZE); + if (bootmap == -1L) + panic("Cannot find bootmem map of size %ld\n", bootmap_size); + reserve_early(bootmap, bootmap + bootmap_size, "BOOTMAP"); +#ifdef CONFIG_BLK_DEV_INITRD + reserve_initrd(); #endif + bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, max_low_pfn); + printk(KERN_INFO " mapped low ram: 0 - %08lx\n", + max_pfn_mapped<<PAGE_SHIFT); + printk(KERN_INFO " low ram: %08lx - %08lx\n", + min_low_pfn<<PAGE_SHIFT, max_low_pfn<<PAGE_SHIFT); + printk(KERN_INFO " bootmap %08lx - %08lx\n", + bootmap, bootmap + bootmap_size); + for_each_online_node(i) + free_bootmem_with_active_regions(i, max_low_pfn); + early_res_to_bootmem(0, max_low_pfn<<PAGE_SHIFT); + #ifdef CONFIG_ACPI_SLEEP /* * Reserve low memory region for sleep support. @@ -695,10 +628,6 @@ void __init setup_bootmem_allocator(void) */ find_smp_config(); #endif -#ifdef CONFIG_BLK_DEV_INITRD - reserve_initrd(); -#endif - numa_kva_reserve(); reserve_crashkernel(); reserve_ibft_region(); @@ -731,12 +660,6 @@ static void set_mca_bus(int x) static void set_mca_bus(int x) { } #endif -/* Overridden in paravirt.c if CONFIG_PARAVIRT */ -char * __init __attribute__((weak)) memory_setup(void) -{ - return machine_specific_memory_setup(); -} - #ifdef CONFIG_NUMA /* * In the golden day, when everything among i386 and x86_64 will be @@ -764,11 +687,14 @@ void __init setup_arch(char **cmdline_p) pre_setup_arch_hook(); early_cpu_init(); early_ioremap_init(); + reserve_setup_data(); #ifdef CONFIG_EFI if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature, - "EL32", 4)) + "EL32", 4)) { efi_enabled = 1; + efi_reserve_early(); + } #endif ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev); @@ -792,8 +718,7 @@ void __init setup_arch(char **cmdline_p) #endif ARCH_SETUP - printk(KERN_INFO "BIOS-provided physical RAM map:\n"); - print_memory_map(memory_setup()); + setup_memory_map(); copy_edd(); @@ -811,12 +736,11 @@ void __init setup_arch(char **cmdline_p) bss_resource.start = virt_to_phys(&__bss_start); bss_resource.end = virt_to_phys(&__bss_stop)-1; + parse_setup_data(); + parse_early_param(); - if (user_defined_memmap) { - printk(KERN_INFO "user-defined physical RAM map:\n"); - print_memory_map("user"); - } + finish_e820_parsing(); strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE); *cmdline_p = command_line; @@ -824,11 +748,22 @@ void __init setup_arch(char **cmdline_p) if (efi_enabled) efi_init(); + e820_register_active_regions(0, 0, -1UL); + /* + * partially used pages are not usable - thus + * we are rounding upwards: + */ + max_pfn = e820_end_of_ram(); + + /* preallocate 4k for mptable mpc */ + early_reserve_e820_mpc_new(); /* update e820 for memory not covered by WB MTRRs */ - propagate_e820_map(); mtrr_bp_init(); - if (mtrr_trim_uncached_memory(max_pfn)) - propagate_e820_map(); + if (mtrr_trim_uncached_memory(max_pfn)) { + remove_all_active_ranges(); + e820_register_active_regions(0, 0, -1UL); + max_pfn = e820_end_of_ram(); + } max_low_pfn = setup_memory(); @@ -855,9 +790,6 @@ void __init setup_arch(char **cmdline_p) * not to exceed the 8Mb limit. */ -#ifdef CONFIG_SMP - smp_alloc_memory(); /* AP processor realmode stacks in low memory*/ -#endif paging_init(); /* @@ -914,21 +846,20 @@ void __init setup_arch(char **cmdline_p) #ifdef CONFIG_ACPI acpi_boot_init(); - +#endif +#if defined(CONFIG_X86_MPPARSE) || defined(CONFIG_X86_VISWS) + if (smp_found_config) + get_smp_config(); +#endif #if defined(CONFIG_SMP) && defined(CONFIG_X86_PC) if (def_to_bigsmp) printk(KERN_WARNING "More than 8 CPUs detected and " "CONFIG_X86_PC cannot handle it.\nUse " "CONFIG_X86_GENERICARCH or CONFIG_X86_BIGSMP.\n"); #endif -#endif -#ifdef CONFIG_X86_LOCAL_APIC - if (smp_found_config) - get_smp_config(); -#endif - e820_register_memory(); - e820_mark_nosave_regions(); + e820_setup_gap(); + e820_mark_nosave_regions(max_low_pfn); #ifdef CONFIG_VT #if defined(CONFIG_VGA_CONSOLE) diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c index 6dff1286ad8..26d60cc0e37 100644 --- a/arch/x86/kernel/setup_64.c +++ b/arch/x86/kernel/setup_64.c @@ -56,6 +56,7 @@ #include <asm/desc.h> #include <video/edid.h> #include <asm/e820.h> +#include <asm/mpspec.h> #include <asm/dma.h> #include <asm/gart.h> #include <asm/mpspec.h> @@ -271,28 +272,6 @@ void __attribute__((weak)) __init memory_setup(void) machine_specific_memory_setup(); } -static void __init parse_setup_data(void) -{ - struct setup_data *data; - unsigned long pa_data; - - if (boot_params.hdr.version < 0x0209) - return; - pa_data = boot_params.hdr.setup_data; - while (pa_data) { - data = early_ioremap(pa_data, PAGE_SIZE); - switch (data->type) { - default: - break; - } -#ifndef CONFIG_DEBUG_BOOT_PARAMS - free_early(pa_data, pa_data+sizeof(*data)+data->len); -#endif - pa_data = data->next; - early_iounmap(data, PAGE_SIZE); - } -} - #ifdef CONFIG_PCI_MMCONFIG extern void __cpuinit fam10h_check_enable_mmcfg(void); extern void __init check_enable_amd_mmconf_dmi(void); @@ -329,8 +308,10 @@ void __init setup_arch(char **cmdline_p) #endif #ifdef CONFIG_EFI if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature, - "EL64", 4)) + "EL64", 4)) { efi_enabled = 1; + efi_reserve_early(); + } #endif ARCH_SETUP @@ -381,9 +362,13 @@ void __init setup_arch(char **cmdline_p) * we are rounding upwards: */ end_pfn = e820_end_of_ram(); + + /* pre allocte 4k for mptable mpc */ + early_reserve_e820_mpc_new(); /* update e820 for memory not covered by WB MTRRs */ mtrr_bp_init(); if (mtrr_trim_uncached_memory(end_pfn)) { + remove_all_active_ranges(); e820_register_active_regions(0, 0, -1UL); end_pfn = e820_end_of_ram(); } @@ -392,7 +377,7 @@ void __init setup_arch(char **cmdline_p) check_efer(); - max_pfn_mapped = init_memory_mapping(0, (max_pfn_mapped << PAGE_SHIFT)); + max_pfn_mapped = init_memory_mapping(0, (end_pfn << PAGE_SHIFT)); if (efi_enabled) efi_init(); @@ -453,13 +438,12 @@ void __init setup_arch(char **cmdline_p) acpi_reserve_bootmem(); #endif - if (efi_enabled) - efi_reserve_bootmem(); - +#ifdef CONFIG_X86_MPPARSE /* * Find and reserve possible boot-time SMP configuration: */ find_smp_config(); +#endif #ifdef CONFIG_BLK_DEV_INITRD if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) { unsigned long ramdisk_image = boot_params.hdr.ramdisk_image; @@ -502,11 +486,13 @@ void __init setup_arch(char **cmdline_p) init_cpu_to_node(); +#ifdef CONFIG_X86_MPPARSE /* * get boot-time SMP configuration: */ if (smp_found_config) get_smp_config(); +#endif init_apic_mappings(); ioapic_init_mappings(); @@ -516,7 +502,7 @@ void __init setup_arch(char **cmdline_p) * We trust e820 completely. No explicit ROM probing in memory. */ e820_reserve_resources(); - e820_mark_nosave_regions(); + e820_mark_nosave_regions(end_pfn); /* request I/O space for devices used on all i[345]86 PCs */ for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++) diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 3e1cecedde4..83e62137911 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -555,23 +555,6 @@ cpumask_t cpu_coregroup_map(int cpu) return c->llc_shared_map; } -#ifdef CONFIG_X86_32 -/* - * We are called very early to get the low memory for the - * SMP bootup trampoline page. - */ -void __init smp_alloc_memory(void) -{ - trampoline_base = alloc_bootmem_low_pages(PAGE_SIZE); - /* - * Has to be in very low memory so we can execute - * real-mode AP code. - */ - if (__pa(trampoline_base) >= 0x9F000) - BUG(); -} -#endif - static void impress_friends(void) { int cpu; diff --git a/arch/x86/kernel/srat_32.c b/arch/x86/kernel/srat_32.c index 70e4a374b4e..e9d91720a40 100644 --- a/arch/x86/kernel/srat_32.c +++ b/arch/x86/kernel/srat_32.c @@ -31,6 +31,7 @@ #include <asm/srat.h> #include <asm/topology.h> #include <asm/smp.h> +#include <asm/e820.h> /* * proximity macros and definitions @@ -244,12 +245,13 @@ static int __init acpi20_parse_srat(struct acpi_table_srat *sratp) printk("chunk %d nid %d start_pfn %08lx end_pfn %08lx\n", j, chunk->nid, chunk->start_pfn, chunk->end_pfn); node_read_chunk(chunk->nid, chunk); - add_active_range(chunk->nid, chunk->start_pfn, chunk->end_pfn); + e820_register_active_regions(chunk->nid, chunk->start_pfn, + min(chunk->end_pfn, max_pfn)); } for_each_online_node(nid) { unsigned long start = node_start_pfn[nid]; - unsigned long end = node_end_pfn[nid]; + unsigned long end = min(node_end_pfn[nid], max_pfn); memory_present(nid, start, end); node_remap_size[nid] = node_memmap_size_bytes(nid, start, end); @@ -261,7 +263,7 @@ out_fail: struct acpi_static_rsdt { struct acpi_table_rsdt table; - u32 padding[7]; /* Allow for 7 more table entries */ + u32 padding[32]; /* Allow for 32 more table entries */ }; int __init get_memcfg_from_srat(void) @@ -297,7 +299,7 @@ int __init get_memcfg_from_srat(void) } rsdt = (struct acpi_table_rsdt *) - early_ioremap(rsdp->rsdt_physical_address, sizeof(struct acpi_table_rsdt)); + early_ioremap(rsdp->rsdt_physical_address, sizeof(saved_rsdt)); if (!rsdt) { printk(KERN_WARNING @@ -310,6 +312,7 @@ int __init get_memcfg_from_srat(void) if (strncmp(header->signature, ACPI_SIG_RSDT, strlen(ACPI_SIG_RSDT))) { printk(KERN_WARNING "ACPI: RSDT signature incorrect\n"); + early_iounmap(rsdt, sizeof(saved_rsdt)); goto out_err; } @@ -319,37 +322,51 @@ int __init get_memcfg_from_srat(void) * size of RSDT) divided by the size of each entry * (4-byte table pointers). */ - tables = (header->length - sizeof(struct acpi_table_header)) / 4; + tables = (header->length - sizeof(struct acpi_table_header)) / sizeof(u32); if (!tables) goto out_err; memcpy(&saved_rsdt, rsdt, sizeof(saved_rsdt)); - + early_iounmap(rsdt, sizeof(saved_rsdt)); if (saved_rsdt.table.header.length > sizeof(saved_rsdt)) { printk(KERN_WARNING "ACPI: Too big length in RSDT: %d\n", saved_rsdt.table.header.length); goto out_err; } - printk("Begin SRAT table scan....\n"); + printk("Begin SRAT table scan....%d\n", tables); - for (i = 0; i < tables; i++) { + for (i = 0; i < tables; i++){ + int result; + u32 length; /* Map in header, then map in full table length. */ header = (struct acpi_table_header *) early_ioremap(saved_rsdt.table.table_offset_entry[i], sizeof(struct acpi_table_header)); if (!header) break; + + printk(KERN_INFO "ACPI: %4.4s %08lX, %04X\n", + header->signature, + (unsigned long)saved_rsdt.table.table_offset_entry[i], + header->length); + + if (strncmp((char *) &header->signature, ACPI_SIG_SRAT, 4)) { + early_iounmap(header, sizeof(struct acpi_table_header)); + continue; + } + + length = header->length; + early_iounmap(header, sizeof(struct acpi_table_header)); header = (struct acpi_table_header *) - early_ioremap(saved_rsdt.table.table_offset_entry[i], header->length); + early_ioremap(saved_rsdt.table.table_offset_entry[i], length); if (!header) break; - if (strncmp((char *) &header->signature, ACPI_SIG_SRAT, 4)) - continue; - /* we've found the srat table. don't need to look at any more tables */ - return acpi20_parse_srat((struct acpi_table_srat *)header); + result = acpi20_parse_srat((struct acpi_table_srat *)header); + early_iounmap(header, length); + return result; } out_err: remove_all_active_ranges(); diff --git a/arch/x86/kernel/summit_32.c b/arch/x86/kernel/summit_32.c index ae751094eba..d67ce5f044b 100644 --- a/arch/x86/kernel/summit_32.c +++ b/arch/x86/kernel/summit_32.c @@ -36,7 +36,9 @@ static struct rio_table_hdr *rio_table_hdr __initdata; static struct scal_detail *scal_devs[MAX_NUMNODES] __initdata; static struct rio_detail *rio_devs[MAX_NUMNODES*4] __initdata; +#ifndef CONFIG_X86_NUMAQ static int mp_bus_id_to_node[MAX_MP_BUSSES] __initdata; +#endif static int __init setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus) { diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c index abbf199adeb..1106fac6024 100644 --- a/arch/x86/kernel/trampoline.c +++ b/arch/x86/kernel/trampoline.c @@ -2,7 +2,7 @@ #include <asm/trampoline.h> -/* ready for x86_64, no harm for x86, since it will overwrite after alloc */ +/* ready for x86_64 and x86 */ unsigned char *trampoline_base = __va(TRAMPOLINE_BASE); /* diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index 5c7e2fd5207..5e4772907c6 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -1012,6 +1012,7 @@ __init void lguest_init(void) * clobbered. The Launcher places our initial pagetables somewhere at * the top of our physical memory, so we don't need extra space: set * init_pg_tables_end to the end of the kernel. */ + init_pg_tables_start = __pa(pg0); init_pg_tables_end = __pa(pg0); /* Load the %fs segment register (the per-cpu segment register) with @@ -1065,9 +1066,9 @@ __init void lguest_init(void) pm_power_off = lguest_power_off; machine_ops.restart = lguest_restart; - /* Now we're set up, call start_kernel() in init/main.c and we proceed + /* Now we're set up, call i386_start_kernel() in head32.c and we proceed * to boot as normal. It never returns. */ - start_kernel(); + i386_start_kernel(); } /* * This marks the end of stage II of our journey, The Guest. diff --git a/arch/x86/mach-default/setup.c b/arch/x86/mach-default/setup.c index 0c28a071824..56b4c39cb7f 100644 --- a/arch/x86/mach-default/setup.c +++ b/arch/x86/mach-default/setup.c @@ -153,6 +153,7 @@ late_initcall(print_ipi_mode); char * __init machine_specific_memory_setup(void) { char *who; + int new_nr; who = "BIOS-e820"; @@ -163,7 +164,11 @@ char * __init machine_specific_memory_setup(void) * Otherwise fake a memory map; one section from 0k->640k, * the next section from 1mb->appropriate_mem_k */ - sanitize_e820_map(boot_params.e820_map, &boot_params.e820_entries); + new_nr = boot_params.e820_entries; + sanitize_e820_map(boot_params.e820_map, + ARRAY_SIZE(boot_params.e820_map), + &new_nr); + boot_params.e820_entries = new_nr; if (copy_e820_map(boot_params.e820_map, boot_params.e820_entries) < 0) { unsigned long mem_size; diff --git a/arch/x86/mach-es7000/Makefile b/arch/x86/mach-es7000/Makefile index 69dd4da218d..3ef8b43b62f 100644 --- a/arch/x86/mach-es7000/Makefile +++ b/arch/x86/mach-es7000/Makefile @@ -3,4 +3,3 @@ # obj-$(CONFIG_X86_ES7000) := es7000plat.o -obj-$(CONFIG_X86_GENERICARCH) := es7000plat.o diff --git a/arch/x86/mach-es7000/es7000plat.c b/arch/x86/mach-es7000/es7000plat.c index f5d6f7d8b86..4354ce80488 100644 --- a/arch/x86/mach-es7000/es7000plat.c +++ b/arch/x86/mach-es7000/es7000plat.c @@ -52,6 +52,8 @@ static struct mip_reg *host_reg; static int mip_port; static unsigned long mip_addr, host_addr; +int es7000_plat; + /* * GSI override for ES7000 platforms. */ @@ -175,53 +177,6 @@ find_unisys_acpi_oem_table(unsigned long *oem_addr) } #endif -/* - * This file also gets compiled if CONFIG_X86_GENERICARCH is set. Generic - * arch already has got following function definitions (asm-generic/es7000.c) - * hence no need to define these for that case. - */ -#ifndef CONFIG_X86_GENERICARCH -void es7000_sw_apic(void); -void __init enable_apic_mode(void) -{ - es7000_sw_apic(); - return; -} - -__init int mps_oem_check(struct mp_config_table *mpc, char *oem, - char *productid) -{ - if (mpc->mpc_oemptr) { - struct mp_config_oemtable *oem_table = - (struct mp_config_oemtable *)mpc->mpc_oemptr; - if (!strncmp(oem, "UNISYS", 6)) - return parse_unisys_oem((char *)oem_table); - } - return 0; -} -#ifdef CONFIG_ACPI -/* Hook from generic ACPI tables.c */ -int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) -{ - unsigned long oem_addr; - if (!find_unisys_acpi_oem_table(&oem_addr)) { - if (es7000_check_dsdt()) - return parse_unisys_oem((char *)oem_addr); - else { - setup_unisys(); - return 1; - } - } - return 0; -} -#else -int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) -{ - return 0; -} -#endif -#endif /* COFIG_X86_GENERICARCH */ - static void es7000_spin(int n) { diff --git a/arch/x86/mach-generic/Makefile b/arch/x86/mach-generic/Makefile index 19d6d407737..0dbd7803a1d 100644 --- a/arch/x86/mach-generic/Makefile +++ b/arch/x86/mach-generic/Makefile @@ -2,7 +2,11 @@ # Makefile for the generic architecture # -EXTRA_CFLAGS := -Iarch/x86/kernel +EXTRA_CFLAGS := -Iarch/x86/kernel -obj-y := probe.o summit.o bigsmp.o es7000.o default.o -obj-y += ../../x86/mach-es7000/ +obj-y := probe.o default.o +obj-$(CONFIG_X86_NUMAQ) += numaq.o +obj-$(CONFIG_X86_SUMMIT) += summit.o +obj-$(CONFIG_X86_BIGSMP) += bigsmp.o +obj-$(CONFIG_X86_ES7000) += es7000.o +obj-$(CONFIG_X86_ES7000) += ../../x86/mach-es7000/ diff --git a/arch/x86/mach-generic/bigsmp.c b/arch/x86/mach-generic/bigsmp.c index 95fc463056d..59d77171455 100644 --- a/arch/x86/mach-generic/bigsmp.c +++ b/arch/x86/mach-generic/bigsmp.c @@ -23,10 +23,8 @@ static int dmi_bigsmp; /* can be set by dmi scanners */ static int hp_ht_bigsmp(const struct dmi_system_id *d) { -#ifdef CONFIG_X86_GENERICARCH printk(KERN_NOTICE "%s detected: force use of apic=bigsmp\n", d->ident); dmi_bigsmp = 1; -#endif return 0; } @@ -48,7 +46,7 @@ static const struct dmi_system_id bigsmp_dmi_table[] = { static int probe_bigsmp(void) { if (def_to_bigsmp) - dmi_bigsmp = 1; + dmi_bigsmp = 1; else dmi_check_system(bigsmp_dmi_table); return dmi_bigsmp; diff --git a/arch/x86/mach-generic/numaq.c b/arch/x86/mach-generic/numaq.c new file mode 100644 index 00000000000..8091e68764c --- /dev/null +++ b/arch/x86/mach-generic/numaq.c @@ -0,0 +1,41 @@ +/* + * APIC driver for the IBM NUMAQ chipset. + */ +#define APIC_DEFINITION 1 +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/smp.h> +#include <asm/mpspec.h> +#include <asm/genapic.h> +#include <asm/fixmap.h> +#include <asm/apicdef.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/init.h> +#include <asm/mach-numaq/mach_apic.h> +#include <asm/mach-numaq/mach_apicdef.h> +#include <asm/mach-numaq/mach_ipi.h> +#include <asm/mach-numaq/mach_mpparse.h> +#include <asm/mach-numaq/mach_wakecpu.h> +#include <asm/numaq.h> + +static int mps_oem_check(struct mp_config_table *mpc, char *oem, + char *productid) +{ + numaq_mps_oem_check(mpc, oem, productid); + return found_numaq; +} + +static int probe_numaq(void) +{ + /* already know from get_memcfg_numaq() */ + return found_numaq; +} + +/* Hook from generic ACPI tables.c */ +static int acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + return 0; +} + +struct genapic apic_numaq = APIC_INIT("NUMAQ", probe_numaq); diff --git a/arch/x86/mach-generic/probe.c b/arch/x86/mach-generic/probe.c index c5ae751b994..ba18dec4855 100644 --- a/arch/x86/mach-generic/probe.c +++ b/arch/x86/mach-generic/probe.c @@ -16,6 +16,7 @@ #include <asm/apicdef.h> #include <asm/genapic.h> +extern struct genapic apic_numaq; extern struct genapic apic_summit; extern struct genapic apic_bigsmp; extern struct genapic apic_es7000; @@ -24,9 +25,18 @@ extern struct genapic apic_default; struct genapic *genapic = &apic_default; static struct genapic *apic_probe[] __initdata = { +#ifdef CONFIG_X86_NUMAQ + &apic_numaq, +#endif +#ifdef CONFIG_X86_SUMMIT &apic_summit, +#endif +#ifdef CONFIG_X86_BIGSMP &apic_bigsmp, +#endif +#ifdef CONFIG_X86_ES7000 &apic_es7000, +#endif &apic_default, /* must be last */ NULL, }; @@ -54,6 +64,7 @@ early_param("apic", parse_apic); void __init generic_bigsmp_probe(void) { +#if CONFIG_X86_BIGSMP /* * This routine is used to switch to bigsmp mode when * - There is no apic= option specified by the user @@ -67,6 +78,7 @@ void __init generic_bigsmp_probe(void) printk(KERN_INFO "Overriding APIC driver with %s\n", genapic->name); } +#endif } void __init generic_apic_probe(void) @@ -88,7 +100,8 @@ void __init generic_apic_probe(void) /* These functions can switch the APIC even after the initial ->probe() */ -int __init mps_oem_check(struct mp_config_table *mpc, char *oem, char *productid) +int __init mps_oem_check(struct mp_config_table *mpc, char *oem, + char *productid) { int i; for (i = 0; apic_probe[i]; ++i) { diff --git a/arch/x86/mach-visws/mpparse.c b/arch/x86/mach-visws/mpparse.c index 57484e91ab9..a2fb78c0d15 100644 --- a/arch/x86/mach-visws/mpparse.c +++ b/arch/x86/mach-visws/mpparse.c @@ -8,11 +8,6 @@ #include "cobalt.h" #include "mach_apic.h" -/* Have we found an MP table */ -int smp_found_config; - -int pic_mode; - extern unsigned int __cpuinitdata maxcpus; /* @@ -76,7 +71,9 @@ void __init find_smp_config(void) if (ncpus > maxcpus) ncpus = maxcpus; +#ifdef CONFIG_X86_LOCAL_APIC smp_found_config = 1; +#endif while (ncpus--) MP_processor_info(mp++); diff --git a/arch/x86/mach-voyager/setup.c b/arch/x86/mach-voyager/setup.c index 5ae5466b9eb..f4aca9fa954 100644 --- a/arch/x86/mach-voyager/setup.c +++ b/arch/x86/mach-voyager/setup.c @@ -62,6 +62,7 @@ void __init time_init_hook(void) char *__init machine_specific_memory_setup(void) { char *who; + int new_nr; who = "NOT VOYAGER"; @@ -111,7 +112,11 @@ char *__init machine_specific_memory_setup(void) * Otherwise fake a memory map; one section from 0k->640k, * the next section from 1mb->appropriate_mem_k */ - sanitize_e820_map(boot_params.e820_map, &boot_params.e820_entries); + new_nr = boot_params.e820_entries; + sanitize_e820_map(boot_params.e820_map, + ARRAY_SIZE(boot_params.e820_map), + &new_nr); + boot_params.e820_entries = new_nr; if (copy_e820_map(boot_params.e820_map, boot_params.e820_entries) < 0) { unsigned long mem_size; diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c index 8acbf0cdf1a..8dedd01e909 100644 --- a/arch/x86/mach-voyager/voyager_smp.c +++ b/arch/x86/mach-voyager/voyager_smp.c @@ -59,11 +59,6 @@ __u32 voyager_quad_processors = 0; * activity count. Finally exported by i386_ksyms.c */ static int voyager_extended_cpus = 1; -/* Have we found an SMP box - used by time.c to do the profiling - interrupt for timeslicing; do not set to 1 until the per CPU timer - interrupt is active */ -int smp_found_config = 0; - /* Used for the invalidate map that's also checked in the spinlock */ static volatile unsigned long smp_invalidate_needed; @@ -1137,15 +1132,6 @@ void flush_tlb_all(void) on_each_cpu(do_flush_tlb_all, 0, 1, 1); } -/* used to set up the trampoline for other CPUs when the memory manager - * is sorted out */ -void __init smp_alloc_memory(void) -{ - trampoline_base = alloc_bootmem_low_pages(PAGE_SIZE); - if (__pa(trampoline_base) >= 0x93000) - BUG(); -} - /* send a reschedule CPI to one CPU by physical CPU number*/ static void voyager_smp_send_reschedule(int cpu) { diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c index 914ccf98368..accc7c6c57f 100644 --- a/arch/x86/mm/discontig_32.c +++ b/arch/x86/mm/discontig_32.c @@ -38,6 +38,7 @@ #include <asm/setup.h> #include <asm/mmzone.h> #include <asm/bios_ebda.h> +#include <asm/proto.h> struct pglist_data *node_data[MAX_NUMNODES] __read_mostly; EXPORT_SYMBOL(node_data); @@ -59,14 +60,14 @@ unsigned long node_end_pfn[MAX_NUMNODES] __read_mostly; /* * 4) physnode_map - the mapping between a pfn and owning node * physnode_map keeps track of the physical memory layout of a generic - * numa node on a 256Mb break (each element of the array will - * represent 256Mb of memory and will be marked by the node id. so, + * numa node on a 64Mb break (each element of the array will + * represent 64Mb of memory and will be marked by the node id. so, * if the first gig is on node 0, and the second gig is on node 1 * physnode_map will contain: * - * physnode_map[0-3] = 0; - * physnode_map[4-7] = 1; - * physnode_map[8- ] = -1; + * physnode_map[0-15] = 0; + * physnode_map[16-31] = 1; + * physnode_map[32- ] = -1; */ s8 physnode_map[MAX_ELEMENTS] __read_mostly = { [0 ... (MAX_ELEMENTS - 1)] = -1}; EXPORT_SYMBOL(physnode_map); @@ -81,9 +82,9 @@ void memory_present(int nid, unsigned long start, unsigned long end) printk(KERN_DEBUG " "); for (pfn = start; pfn < end; pfn += PAGES_PER_ELEMENT) { physnode_map[pfn / PAGES_PER_ELEMENT] = nid; - printk("%ld ", pfn); + printk(KERN_CONT "%ld ", pfn); } - printk("\n"); + printk(KERN_CONT "\n"); } unsigned long node_memmap_size_bytes(int nid, unsigned long start_pfn, @@ -119,11 +120,11 @@ int __init get_memcfg_numa_flat(void) { printk("NUMA - single node, flat memory mode\n"); - /* Run the memory configuration and find the top of memory. */ - propagate_e820_map(); node_start_pfn[0] = 0; node_end_pfn[0] = max_pfn; + e820_register_active_regions(0, 0, max_pfn); memory_present(0, 0, max_pfn); + node_remap_size[0] = node_memmap_size_bytes(0, 0, max_pfn); /* Indicate there is one node available. */ nodes_clear(node_online_map); @@ -159,9 +160,17 @@ static void __init allocate_pgdat(int nid) if (nid && node_has_online_mem(nid)) NODE_DATA(nid) = (pg_data_t *)node_remap_start_vaddr[nid]; else { - NODE_DATA(nid) = (pg_data_t *)(pfn_to_kaddr(min_low_pfn)); - min_low_pfn += PFN_UP(sizeof(pg_data_t)); + unsigned long pgdat_phys; + pgdat_phys = find_e820_area(min_low_pfn<<PAGE_SHIFT, + (nid ? max_low_pfn:max_pfn_mapped)<<PAGE_SHIFT, + sizeof(pg_data_t), + PAGE_SIZE); + NODE_DATA(nid) = (pg_data_t *)(pfn_to_kaddr(pgdat_phys>>PAGE_SHIFT)); + reserve_early(pgdat_phys, pgdat_phys + sizeof(pg_data_t), + "NODE_DATA"); } + printk(KERN_DEBUG "allocate_pgdat: node %d NODE_DATA %08lx\n", + nid, (unsigned long)NODE_DATA(nid)); } #ifdef CONFIG_DISCONTIGMEM @@ -202,8 +211,12 @@ void __init remap_numa_kva(void) int node; for_each_online_node(node) { + printk(KERN_DEBUG "remap_numa_kva: node %d\n", node); for (pfn=0; pfn < node_remap_size[node]; pfn += PTRS_PER_PTE) { vaddr = node_remap_start_vaddr[node]+(pfn<<PAGE_SHIFT); + printk(KERN_DEBUG "remap_numa_kva: %08lx to pfn %08lx\n", + (unsigned long)vaddr, + node_remap_start_pfn[node] + pfn); set_pmd_pfn((ulong) vaddr, node_remap_start_pfn[node] + pfn, PAGE_KERNEL_LARGE); @@ -215,17 +228,21 @@ static unsigned long calculate_numa_remap_pages(void) { int nid; unsigned long size, reserve_pages = 0; - unsigned long pfn; for_each_online_node(nid) { - unsigned old_end_pfn = node_end_pfn[nid]; + u64 node_end_target; + u64 node_end_final; /* * The acpi/srat node info can show hot-add memroy zones * where memory could be added but not currently present. */ + printk("node %d pfn: [%lx - %lx]\n", + nid, node_start_pfn[nid], node_end_pfn[nid]); if (node_start_pfn[nid] > max_pfn) continue; + if (!node_end_pfn[nid]) + continue; if (node_end_pfn[nid] > max_pfn) node_end_pfn[nid] = max_pfn; @@ -237,39 +254,42 @@ static unsigned long calculate_numa_remap_pages(void) /* now the roundup is correct, convert to PAGE_SIZE pages */ size = size * PTRS_PER_PTE; - /* - * Validate the region we are allocating only contains valid - * pages. - */ - for (pfn = node_end_pfn[nid] - size; - pfn < node_end_pfn[nid]; pfn++) - if (!page_is_ram(pfn)) - break; - - if (pfn != node_end_pfn[nid]) - size = 0; + node_end_target = round_down(node_end_pfn[nid] - size, + PTRS_PER_PTE); + node_end_target <<= PAGE_SHIFT; + do { + node_end_final = find_e820_area(node_end_target, + ((u64)node_end_pfn[nid])<<PAGE_SHIFT, + ((u64)size)<<PAGE_SHIFT, + LARGE_PAGE_BYTES); + node_end_target -= LARGE_PAGE_BYTES; + } while (node_end_final == -1ULL && + (node_end_target>>PAGE_SHIFT) > (node_start_pfn[nid])); + + if (node_end_final == -1ULL) + panic("Can not get kva ram\n"); printk("Reserving %ld pages of KVA for lmem_map of node %d\n", size, nid); node_remap_size[nid] = size; node_remap_offset[nid] = reserve_pages; reserve_pages += size; - printk("Shrinking node %d from %ld pages to %ld pages\n", - nid, node_end_pfn[nid], node_end_pfn[nid] - size); - - if (node_end_pfn[nid] & (PTRS_PER_PTE-1)) { - /* - * Align node_end_pfn[] and node_remap_start_pfn[] to - * pmd boundary. remap_numa_kva will barf otherwise. - */ - printk("Shrinking node %d further by %ld pages for proper alignment\n", - nid, node_end_pfn[nid] & (PTRS_PER_PTE-1)); - size += node_end_pfn[nid] & (PTRS_PER_PTE-1); - } + printk("Shrinking node %d from %ld pages to %lld pages\n", + nid, node_end_pfn[nid], node_end_final>>PAGE_SHIFT); - node_end_pfn[nid] -= size; + /* + * prevent kva address below max_low_pfn want it on system + * with less memory later. + * layout will be: KVA address , KVA RAM + */ + if ((node_end_final>>PAGE_SHIFT) < max_low_pfn) + reserve_early(node_end_final, + node_end_final+(((u64)size)<<PAGE_SHIFT), + "KVA RAM"); + + node_end_pfn[nid] = node_end_final>>PAGE_SHIFT; node_remap_start_pfn[nid] = node_end_pfn[nid]; - shrink_active_range(nid, old_end_pfn, node_end_pfn[nid]); + shrink_active_range(nid, node_end_pfn[nid]); } printk("Reserving total of %ld pages for numa KVA remap\n", reserve_pages); @@ -287,8 +307,7 @@ static void init_remap_allocator(int nid) printk ("node %d will remap to vaddr %08lx - %08lx\n", nid, (ulong) node_remap_start_vaddr[nid], - (ulong) pfn_to_kaddr(highstart_pfn - + node_remap_offset[nid] + node_remap_size[nid])); + (ulong) node_remap_end_vaddr[nid]); } #else void *alloc_remap(int nid, unsigned long size) @@ -315,7 +334,7 @@ unsigned long __init setup_memory(void) { int nid; unsigned long system_start_pfn, system_max_low_pfn; - unsigned long wasted_pages; + long kva_target_pfn; /* * When mapping a NUMA machine we allocate the node_mem_map arrays @@ -324,34 +343,38 @@ unsigned long __init setup_memory(void) * this space and use it to adjust the boundary between ZONE_NORMAL * and ZONE_HIGHMEM. */ + + /* call find_max_low_pfn at first, it could update max_pfn */ + system_max_low_pfn = max_low_pfn = find_max_low_pfn(); + + remove_all_active_ranges(); get_memcfg_numa(); - kva_pages = calculate_numa_remap_pages(); + kva_pages = round_up(calculate_numa_remap_pages(), PTRS_PER_PTE); /* partially used pages are not usable - thus round upwards */ system_start_pfn = min_low_pfn = PFN_UP(init_pg_tables_end); - kva_start_pfn = find_max_low_pfn() - kva_pages; + kva_target_pfn = round_down(max_low_pfn - kva_pages, PTRS_PER_PTE); + do { + kva_start_pfn = find_e820_area(kva_target_pfn<<PAGE_SHIFT, + max_low_pfn<<PAGE_SHIFT, + kva_pages<<PAGE_SHIFT, + PTRS_PER_PTE<<PAGE_SHIFT) >> PAGE_SHIFT; + kva_target_pfn -= PTRS_PER_PTE; + } while (kva_start_pfn == -1UL && kva_target_pfn > min_low_pfn); -#ifdef CONFIG_BLK_DEV_INITRD - /* Numa kva area is below the initrd */ - if (initrd_start) - kva_start_pfn = PFN_DOWN(initrd_start - PAGE_OFFSET) - - kva_pages; -#endif + if (kva_start_pfn == -1UL) + panic("Can not get kva space\n"); - /* - * We waste pages past at the end of the KVA for no good reason other - * than how it is located. This is bad. - */ - wasted_pages = kva_start_pfn & (PTRS_PER_PTE-1); - kva_start_pfn -= wasted_pages; - kva_pages += wasted_pages; - - system_max_low_pfn = max_low_pfn = find_max_low_pfn(); printk("kva_start_pfn ~ %ld find_max_low_pfn() ~ %ld\n", kva_start_pfn, max_low_pfn); printk("max_pfn = %ld\n", max_pfn); + + /* avoid clash with initrd */ + reserve_early(kva_start_pfn<<PAGE_SHIFT, + (kva_start_pfn + kva_pages)<<PAGE_SHIFT, + "KVA PG"); #ifdef CONFIG_HIGHMEM highstart_pfn = highend_pfn = max_pfn; if (max_pfn > system_max_low_pfn) @@ -387,16 +410,8 @@ unsigned long __init setup_memory(void) return max_low_pfn; } -void __init numa_kva_reserve(void) -{ - if (kva_pages) - reserve_bootmem(PFN_PHYS(kva_start_pfn), PFN_PHYS(kva_pages), - BOOTMEM_DEFAULT); -} - void __init zone_sizes_init(void) { - int nid; unsigned long max_zone_pfns[MAX_NR_ZONES]; memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); max_zone_pfns[ZONE_DMA] = @@ -406,15 +421,6 @@ void __init zone_sizes_init(void) max_zone_pfns[ZONE_HIGHMEM] = highend_pfn; #endif - /* If SRAT has not registered memory, register it now */ - if (find_max_pfn_with_active_regions() == 0) { - for_each_online_node(nid) { - if (node_has_online_mem(nid)) - add_active_range(nid, node_start_pfn[nid], - node_end_pfn[nid]); - } - } - free_area_init_nodes(max_zone_pfns); return; } diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index ec30d10154b..0e7bb5e8167 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -289,7 +289,8 @@ static void __init permanent_kmaps_init(pgd_t *pgd_base) void __init add_one_highpage_init(struct page *page, int pfn, int bad_ppro) { - if (page_is_ram(pfn) && !(bad_ppro && page_kills_ppro(pfn))) { + if (page_is_ram(pfn) && !(bad_ppro && page_kills_ppro(pfn)) && + !page_is_reserved_early(pfn)) { ClearPageReserved(page); init_page_count(page); __free_page(page); diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index c5066d519e5..afb07ffb931 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -233,7 +233,7 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, else bootmap_start = round_up(start, PAGE_SIZE); /* - * SMP_CAHCE_BYTES could be enough, but init_bootmem_node like + * SMP_CACHE_BYTES could be enough, but init_bootmem_node like * to use that to align to PAGE_SIZE */ bootmap = early_node_mem(nodeid, bootmap_start, end, diff --git a/arch/x86/pci/Makefile_32 b/arch/x86/pci/Makefile_32 index 89ec35d00ef..962d96c0495 100644 --- a/arch/x86/pci/Makefile_32 +++ b/arch/x86/pci/Makefile_32 @@ -13,10 +13,11 @@ pci-y := fixup.o pci-$(CONFIG_ACPI) += acpi.o pci-y += legacy.o irq.o -# Careful: VISWS and NUMAQ overrule the pci-y above. The colons are +# Careful: VISWS overrule the pci-y above. The colons are # therefor correct. This needs a proper fix by distangling the code. pci-$(CONFIG_X86_VISWS) := visws.o fixup.o -pci-$(CONFIG_X86_NUMAQ) := numa.o irq.o + +pci-$(CONFIG_X86_NUMAQ) += numa.o # Necessary for NUMAQ as well pci-$(CONFIG_NUMA) += mp_bus_to_node.o diff --git a/arch/x86/pci/k8-bus_64.c b/arch/x86/pci/k8-bus_64.c index 5c2799c20e4..bfefdf0f40d 100644 --- a/arch/x86/pci/k8-bus_64.c +++ b/arch/x86/pci/k8-bus_64.c @@ -384,7 +384,7 @@ static int __init early_fill_mp_bus_info(void) /* need to take out [0, TOM) for RAM*/ address = MSR_K8_TOP_MEM1; rdmsrl(address, val); - end = (val & 0xffffff8000000ULL); + end = (val & 0xffffff800000ULL); printk(KERN_INFO "TOM: %016lx aka %ldM\n", end, end>>20); if (end < (1ULL<<32)) update_range(range, 0, end - 1); @@ -478,7 +478,7 @@ static int __init early_fill_mp_bus_info(void) /* TOP_MEM2 */ address = MSR_K8_TOP_MEM2; rdmsrl(address, val); - end = (val & 0xffffff8000000ULL); + end = (val & 0xffffff800000ULL); printk(KERN_INFO "TOM2: %016lx aka %ldM\n", end, end>>20); update_range(range, 1ULL<<32, end - 1); } diff --git a/arch/x86/pci/numa.c b/arch/x86/pci/numa.c index d9afbae5092..99f1ecd485b 100644 --- a/arch/x86/pci/numa.c +++ b/arch/x86/pci/numa.c @@ -6,45 +6,21 @@ #include <linux/init.h> #include <linux/nodemask.h> #include <mach_apic.h> +#include <asm/mpspec.h> #include "pci.h" #define XQUAD_PORTIO_BASE 0xfe400000 #define XQUAD_PORTIO_QUAD 0x40000 /* 256k per quad. */ -int mp_bus_id_to_node[MAX_MP_BUSSES]; #define BUS2QUAD(global) (mp_bus_id_to_node[global]) -int mp_bus_id_to_local[MAX_MP_BUSSES]; #define BUS2LOCAL(global) (mp_bus_id_to_local[global]) -void mpc_oem_bus_info(struct mpc_config_bus *m, char *name, - struct mpc_config_translation *translation) -{ - int quad = translation->trans_quad; - int local = translation->trans_local; - - mp_bus_id_to_node[m->mpc_busid] = quad; - mp_bus_id_to_local[m->mpc_busid] = local; - printk(KERN_INFO "Bus #%d is %s (node %d)\n", - m->mpc_busid, name, quad); -} - -int quad_local_to_mp_bus_id [NR_CPUS/4][4]; #define QUADLOCAL2BUS(quad,local) (quad_local_to_mp_bus_id[quad][local]) -void mpc_oem_pci_bus(struct mpc_config_bus *m, - struct mpc_config_translation *translation) -{ - int quad = translation->trans_quad; - int local = translation->trans_local; - - quad_local_to_mp_bus_id[quad][local] = m->mpc_busid; -} /* Where the IO area was mapped on multiquad, always 0 otherwise */ void *xquad_portio; -#ifdef CONFIG_X86_NUMAQ EXPORT_SYMBOL(xquad_portio); -#endif #define XQUAD_PORT_ADDR(port, quad) (xquad_portio + (XQUAD_PORTIO_QUAD*quad) + port) @@ -179,6 +155,9 @@ static int __init pci_numa_init(void) { int quad; + if (!found_numaq) + return 0; + raw_pci_ops = &pci_direct_conf1_mq; if (pcibios_scanned++) diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index f09c1c69c37..275163f8146 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -1196,6 +1196,7 @@ asmlinkage void __init xen_start_kernel(void) pgd = (pgd_t *)xen_start_info->pt_base; + init_pg_tables_start = __pa(pgd); init_pg_tables_end = __pa(pgd) + xen_start_info->nr_pt_frames*PAGE_SIZE; init_mm.pgd = pgd; /* use the Xen pagetables to start */ @@ -1236,5 +1237,5 @@ asmlinkage void __init xen_start_kernel(void) add_preferred_console("hvc", 0, NULL); /* Start the world */ - start_kernel(); + i386_start_kernel(); } |