diff options
Diffstat (limited to 'arch/powerpc')
70 files changed, 1307 insertions, 749 deletions
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 28004f002ec..01feed0e2a1 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -71,15 +71,39 @@ config ARCH_MAY_HAVE_PC_FDC bool default y +config PPC_OF + def_bool y + +config PPC_UDBG_16550 + bool + default n + +config CRASH_DUMP + bool "kernel crash dumps (EXPERIMENTAL)" + depends on PPC_MULTIPLATFORM + depends on EXPERIMENTAL + help + Build a kernel suitable for use as a kdump capture kernel. + The kernel will be linked at a different address than normal, and + so can only be used for Kdump. + + Don't change this unless you know what you are doing. + +config GENERIC_TBSYNC + bool + default y if PPC32 && SMP + default n + menu "Processor support" choice prompt "Processor Type" depends on PPC32 default 6xx -config 6xx +config CLASSIC32 bool "6xx/7xx/74xx" select PPC_FPU + select 6xx help There are four families of PowerPC chips supported. The more common types (601, 603, 604, 740, 750, 7400), the Motorola embedded @@ -93,12 +117,20 @@ config 6xx config PPC_52xx bool "Freescale 52xx" + select 6xx + select PPC_FPU config PPC_82xx bool "Freescale 82xx" + select 6xx + select PPC_FPU config PPC_83xx bool "Freescale 83xx" + select 6xx + select FSL_SOC + select 83xx + select PPC_FPU config 40x bool "AMCC 40x" @@ -134,6 +166,13 @@ config POWER4 depends on PPC64 def_bool y +config 6xx + bool + +# this is temp to handle compat with arch=ppc +config 83xx + bool + config PPC_FPU bool default y if PPC64 @@ -166,7 +205,7 @@ config PHYS_64BIT config ALTIVEC bool "AltiVec Support" - depends on 6xx || POWER4 + depends on CLASSIC32 || POWER4 ---help--- This option enables kernel support for the Altivec extensions to the PowerPC processor. The kernel currently supports saving and restoring @@ -239,7 +278,7 @@ endmenu source "init/Kconfig" menu "Platform support" - depends on PPC64 || 6xx + depends on PPC64 || CLASSIC32 choice prompt "Machine type" @@ -275,6 +314,7 @@ config PPC_PSERIES select PPC_I8259 select PPC_RTAS select RTAS_ERROR_LOGGING + select PPC_UDBG_16550 default y config PPC_CHRP @@ -284,6 +324,7 @@ config PPC_CHRP select PPC_INDIRECT_PCI select PPC_RTAS select PPC_MPC106 + select PPC_UDBG_16550 default y config PPC_PMAC @@ -306,6 +347,7 @@ config PPC_PREP depends on PPC_MULTIPLATFORM && PPC32 && BROKEN select PPC_I8259 select PPC_INDIRECT_PCI + select PPC_UDBG_16550 default y config PPC_MAPLE @@ -314,6 +356,7 @@ config PPC_MAPLE select U3_DART select MPIC_BROKEN_U3 select GENERIC_TBSYNC + select PPC_UDBG_16550 default n help This option enables support for the Maple 970FX Evaluation Board. @@ -324,9 +367,7 @@ config PPC_CELL depends on PPC_MULTIPLATFORM && PPC64 select PPC_RTAS select MMIO_NVRAM - -config PPC_OF - def_bool y + select PPC_UDBG_16550 config XICS depends on PPC_PSERIES @@ -375,17 +416,6 @@ config CELL_IIC bool default y -config CRASH_DUMP - bool "kernel crash dumps (EXPERIMENTAL)" - depends on PPC_MULTIPLATFORM - depends on EXPERIMENTAL - help - Build a kernel suitable for use as a kdump capture kernel. - The kernel will be linked at a different address than normal, and - so can only be used for Kdump. - - Don't change this unless you know what you are doing. - config IBMVIO depends on PPC_PSERIES || PPC_ISERIES bool @@ -401,11 +431,6 @@ config PPC_MPC106 bool default n -config GENERIC_TBSYNC - bool - default y if CONFIG_PPC32 && CONFIG_SMP - default n - source "drivers/cpufreq/Kconfig" config CPU_FREQ_PMAC @@ -486,6 +511,7 @@ endmenu source arch/powerpc/platforms/embedded6xx/Kconfig source arch/powerpc/platforms/4xx/Kconfig +source arch/powerpc/platforms/83xx/Kconfig source arch/powerpc/platforms/85xx/Kconfig source arch/powerpc/platforms/8xx/Kconfig source arch/powerpc/platforms/cell/Kconfig @@ -709,7 +735,7 @@ config PPC_I8259 config PPC_INDIRECT_PCI bool depends on PCI - default y if 40x || 44x || 85xx || 83xx + default y if 40x || 44x || 85xx default n config EISA @@ -718,13 +744,16 @@ config EISA config SBUS bool +config FSL_SOC + bool + # Yes MCA RS/6000s exist but Linux-PPC does not currently support any config MCA bool config PCI - bool "PCI support" if 40x || CPM2 || 83xx || 85xx || PPC_MPC52xx || (EMBEDDED && PPC_ISERIES) - default y if !40x && !CPM2 && !8xx && !APUS && !83xx && !85xx + bool "PCI support" if 40x || CPM2 || PPC_83xx || 85xx || PPC_MPC52xx || (EMBEDDED && PPC_ISERIES) + default y if !40x && !CPM2 && !8xx && !APUS && !PPC_83xx && !85xx default PCI_PERMEDIA if !4xx && !CPM2 && !8xx && APUS default PCI_QSPAN if !4xx && !CPM2 && 8xx help @@ -737,11 +766,6 @@ config PCI_DOMAINS bool default PCI -config MPC83xx_PCI2 - bool " Supprt for 2nd PCI host controller" - depends on PCI && MPC834x - default y if MPC834x_SYS - config PCI_QSPAN bool "QSpan PCI" depends on !4xx && !CPM2 && 8xx diff --git a/arch/powerpc/Kconfig.debug b/arch/powerpc/Kconfig.debug index 30a30bf559e..9254806f703 100644 --- a/arch/powerpc/Kconfig.debug +++ b/arch/powerpc/Kconfig.debug @@ -115,4 +115,46 @@ config PPC_OCP depends on IBM_OCP || XILINX_OCP default y +choice + prompt "Early debugging (dangerous)" + bool + optional + help + Enable early debugging. Careful, if you enable debugging for the + wrong type of machine your kernel _will not boot_. + +config PPC_EARLY_DEBUG_LPAR + bool "LPAR HV Console" + depends on PPC_PSERIES + help + Select this to enable early debugging for a machine with a HVC + console on vterm 0. + +config PPC_EARLY_DEBUG_G5 + bool "Apple G5" + depends on PPC_PMAC64 + help + Select this to enable early debugging for Apple G5 machines. + +config PPC_EARLY_DEBUG_RTAS + bool "RTAS Panel" + depends on PPC_RTAS + help + Select this to enable early debugging via the RTAS panel. + +config PPC_EARLY_DEBUG_MAPLE + bool "Maple real mode" + depends on PPC_MAPLE + help + Select this to enable early debugging for Maple. + +config PPC_EARLY_DEBUG_ISERIES + bool "iSeries HV Console" + depends on PPC_ISERIES + help + Select this to enable early debugging for legacy iSeries. You need + to hit "Ctrl-x Ctrl-x" to see the messages on the console. + +endchoice + endmenu diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile index 5f80e58e5cb..d3654a264ef 100644 --- a/arch/powerpc/Makefile +++ b/arch/powerpc/Makefile @@ -76,8 +76,7 @@ LINUXINCLUDE += $(LINUXINCLUDE-y) CHECKFLAGS += -m$(SZ) -D__powerpc__ -D__powerpc$(SZ)__ ifeq ($(CONFIG_PPC64),y) -GCC_VERSION := $(call cc-version) -GCC_BROKEN_VEC := $(shell if [ $(GCC_VERSION) -lt 0400 ] ; then echo "y"; fi) +GCC_BROKEN_VEC := $(shell if [ $(call cc-version) -lt 0400 ] ; then echo "y"; fi) ifeq ($(CONFIG_POWER4_ONLY),y) ifeq ($(CONFIG_ALTIVEC),y) @@ -189,10 +188,9 @@ TOUT := .tmp_gas_check # Ensure this is binutils 2.12.1 (or 2.12.90.0.7) or later for altivec # instructions. # gcc-3.4 and binutils-2.14 are a fatal combination. -GCC_VERSION := $(call cc-version) checkbin: - @if test "$(GCC_VERSION)" = "0304" ; then \ + @if test "$(call cc-version)" = "0304" ; then \ if ! /bin/echo mftb 5 | $(AS) -v -mppc -many -o $(TOUT) >/dev/null 2>&1 ; then \ echo -n '*** ${VERSION}.${PATCHLEVEL} kernels no longer build '; \ echo 'correctly with gcc-3.4 and your version of binutils.'; \ diff --git a/arch/powerpc/boot/.gitignore b/arch/powerpc/boot/.gitignore new file mode 100644 index 00000000000..45c9ad23526 --- /dev/null +++ b/arch/powerpc/boot/.gitignore @@ -0,0 +1,20 @@ +addnote +infblock.c +infblock.h +infcodes.c +infcodes.h +inffast.c +inffast.h +inflate.c +inftrees.c +inftrees.h +infutil.c +infutil.h +kernel-vmlinux.strip.c +kernel-vmlinux.strip.gz +uImage +zImage +zImage.vmode +zconf.h +zlib.h +zutil.h diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile index 22726aefc8e..b53d677f674 100644 --- a/arch/powerpc/boot/Makefile +++ b/arch/powerpc/boot/Makefile @@ -176,4 +176,4 @@ $(obj)/uImage: $(obj)/vmlinux.gz install: $(CONFIGURE) $(BOOTIMAGE) sh -x $(srctree)/$(src)/install.sh "$(KERNELRELEASE)" vmlinux System.map "$(INSTALL_PATH)" "$(BOOTIMAGE)" -clean-files := $(addprefix $(objtree)/, $(obj-boot) vmlinux.strip) +clean-files += $(addprefix $(objtree)/, $(obj-boot) vmlinux.strip) diff --git a/arch/powerpc/configs/ppc64_defconfig b/arch/powerpc/configs/ppc64_defconfig index 509399eab6f..0b2b55a79c3 100644 --- a/arch/powerpc/configs/ppc64_defconfig +++ b/arch/powerpc/configs/ppc64_defconfig @@ -861,7 +861,7 @@ CONFIG_SERIAL_CORE=y CONFIG_SERIAL_CORE_CONSOLE=y # CONFIG_SERIAL_PMACZILOG is not set CONFIG_SERIAL_ICOM=m -CONFIG_SERIAL_JSM=m +# CONFIG_SERIAL_JSM is not set CONFIG_UNIX98_PTYS=y CONFIG_LEGACY_PTYS=y CONFIG_LEGACY_PTY_COUNT=256 @@ -878,7 +878,7 @@ CONFIG_HVCS=m # # CONFIG_WATCHDOG is not set # CONFIG_RTC is not set -# CONFIG_GEN_RTC is not set +CONFIG_GEN_RTC=y # CONFIG_DTLK is not set # CONFIG_R3964 is not set # CONFIG_APPLICOM is not set diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index 6e03b595b6c..a94699d8dc5 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -4,7 +4,6 @@ ifeq ($(CONFIG_PPC64),y) EXTRA_CFLAGS += -mno-minimal-toc -CFLAGS_ioctl32.o += -Ifs/ endif ifeq ($(CONFIG_PPC32),y) CFLAGS_prom_init.o += -fPIC @@ -12,11 +11,12 @@ CFLAGS_btext.o += -fPIC endif obj-y := semaphore.o cputable.o ptrace.o syscalls.o \ - irq.o align.o signal_32.o pmc.o vdso.o + irq.o align.o signal_32.o pmc.o vdso.o \ + init_task.o process.o obj-y += vdso32/ obj-$(CONFIG_PPC64) += setup_64.o binfmt_elf32.o sys_ppc32.o \ signal_64.o ptrace32.o systbl.o \ - paca.o ioctl32.o cpu_setup_power4.o \ + paca.o cpu_setup_power4.o \ firmware.o sysfs.o idle_64.o obj-$(CONFIG_PPC64) += vdso64/ obj-$(CONFIG_ALTIVEC) += vecemu.o vector.o @@ -45,8 +45,7 @@ extra-$(CONFIG_FSL_BOOKE) := head_fsl_booke.o extra-$(CONFIG_8xx) := head_8xx.o extra-y += vmlinux.lds -obj-y += process.o init_task.o time.o \ - prom.o traps.o setup-common.o udbg.o +obj-y += time.o prom.o traps.o setup-common.o udbg.o obj-$(CONFIG_PPC32) += entry_32.o setup_32.o misc_32.o systbl.o obj-$(CONFIG_PPC64) += misc_64.o dma_64.o iommu.o obj-$(CONFIG_PPC_MULTIPLATFORM) += prom_init.o @@ -55,16 +54,16 @@ obj-$(CONFIG_BOOTX_TEXT) += btext.o obj-$(CONFIG_6xx) += idle_6xx.o obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_KPROBES) += kprobes.o -obj-$(CONFIG_SERIAL_8250) += legacy_serial.o udbg_16550.o +obj-$(CONFIG_PPC_UDBG_16550) += legacy_serial.o udbg_16550.o module-$(CONFIG_PPC64) += module_64.o obj-$(CONFIG_MODULES) += $(module-y) pci64-$(CONFIG_PPC64) += pci_64.o pci_dn.o pci_iommu.o \ pci_direct_iommu.o iomap.o obj-$(CONFIG_PCI) += $(pci64-y) -kexec-$(CONFIG_PPC64) := machine_kexec_64.o +kexec-$(CONFIG_PPC64) := machine_kexec_64.o crash.o kexec-$(CONFIG_PPC32) := machine_kexec_32.o -obj-$(CONFIG_KEXEC) += machine_kexec.o crash.o $(kexec-y) +obj-$(CONFIG_KEXEC) += machine_kexec.o $(kexec-y) ifeq ($(CONFIG_PPC_ISERIES),y) $(obj)/head_64.o: $(obj)/lparmap.s diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c index 4681155121e..5f248e3fdf8 100644 --- a/arch/powerpc/kernel/crash.c +++ b/arch/powerpc/kernel/crash.c @@ -18,7 +18,6 @@ #include <linux/kexec.h> #include <linux/bootmem.h> #include <linux/crash_dump.h> -#include <linux/irq.h> #include <linux/delay.h> #include <linux/elf.h> #include <linux/elfcore.h> @@ -30,6 +29,7 @@ #include <asm/kdump.h> #include <asm/lmb.h> #include <asm/firmware.h> +#include <asm/smp.h> #ifdef DEBUG #include <asm/udbg.h> diff --git a/arch/powerpc/kernel/crash_dump.c b/arch/powerpc/kernel/crash_dump.c index 87effa3f21a..211d72653ea 100644 --- a/arch/powerpc/kernel/crash_dump.c +++ b/arch/powerpc/kernel/crash_dump.c @@ -55,6 +55,7 @@ void __init kdump_setup(void) DBG(" <- kdump_setup()\n"); } +#ifdef CONFIG_PROC_VMCORE static int __init parse_elfcorehdr(char *p) { if (p) @@ -63,6 +64,7 @@ static int __init parse_elfcorehdr(char *p) return 0; } __setup("elfcorehdr=", parse_elfcorehdr); +#endif static int __init parse_savemaxmem(char *p) { diff --git a/arch/powerpc/kernel/ioctl32.c b/arch/powerpc/kernel/ioctl32.c deleted file mode 100644 index 0fa3d27fef0..00000000000 --- a/arch/powerpc/kernel/ioctl32.c +++ /dev/null @@ -1,45 +0,0 @@ -/* - * ioctl32.c: Conversion between 32bit and 64bit native ioctls. - * - * Based on sparc64 ioctl32.c by: - * - * Copyright (C) 1997-2000 Jakub Jelinek (jakub@redhat.com) - * Copyright (C) 1998 Eddie C. Dost (ecd@skynet.be) - * - * ppc64 changes: - * - * Copyright (C) 2000 Ken Aaker (kdaaker@rchland.vnet.ibm.com) - * Copyright (C) 2001 Anton Blanchard (antonb@au.ibm.com) - * - * These routines maintain argument size conversion between 32bit and 64bit - * ioctls. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define INCLUDES -#include "compat_ioctl.c" -#include <linux/syscalls.h> - -#define CODE -#include "compat_ioctl.c" - -#define HANDLE_IOCTL(cmd,handler) { cmd, (ioctl_trans_handler_t)handler, NULL }, -#define COMPATIBLE_IOCTL(cmd) HANDLE_IOCTL(cmd,sys_ioctl) - -#define IOCTL_TABLE_START \ - struct ioctl_trans ioctl_start[] = { -#define IOCTL_TABLE_END \ - }; - -IOCTL_TABLE_START -#include <linux/compat_ioctl.h> -#define DECLARES -#include "compat_ioctl.c" - -IOCTL_TABLE_END - -int ioctl_table_size = ARRAY_SIZE(ioctl_start); diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c index 5368f9c2e6b..cfab48566db 100644 --- a/arch/powerpc/kernel/kprobes.c +++ b/arch/powerpc/kernel/kprobes.c @@ -35,7 +35,6 @@ #include <asm/kdebug.h> #include <asm/sstep.h> -static DECLARE_MUTEX(kprobe_mutex); DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL; DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk); @@ -54,19 +53,17 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p) /* insn must be on a special executable page on ppc64 */ if (!ret) { - down(&kprobe_mutex); p->ainsn.insn = get_insn_slot(); - up(&kprobe_mutex); if (!p->ainsn.insn) ret = -ENOMEM; } - return ret; -} -void __kprobes arch_copy_kprobe(struct kprobe *p) -{ - memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); - p->opcode = *p->addr; + if (!ret) { + memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); + p->opcode = *p->addr; + } + + return ret; } void __kprobes arch_arm_kprobe(struct kprobe *p) @@ -182,6 +179,18 @@ static inline int kprobe_handler(struct pt_regs *regs) kcb->kprobe_status = KPROBE_REENTER; return 1; } else { + if (*addr != BREAKPOINT_INSTRUCTION) { + /* If trap variant, then it belongs not to us */ + kprobe_opcode_t cur_insn = *addr; + if (is_trap(cur_insn)) + goto no_kprobe; + /* The breakpoint instruction was removed by + * another cpu right after we hit, no further + * handling of this interrupt is appropriate + */ + ret = 1; + goto no_kprobe; + } p = __get_cpu_var(current_kprobe); if (p->break_handler && p->break_handler(p, regs)) { goto ss_probe; diff --git a/arch/powerpc/kernel/machine_kexec.c b/arch/powerpc/kernel/machine_kexec.c index a91e40c9ae4..a81ca1b841e 100644 --- a/arch/powerpc/kernel/machine_kexec.c +++ b/arch/powerpc/kernel/machine_kexec.c @@ -14,12 +14,6 @@ #include <linux/threads.h> #include <asm/machdep.h> -/* - * Provide a dummy crash_notes definition until crash dump is implemented. - * This prevents breakage of crash_notes attribute in kernel/ksysfs.c. - */ -note_buf_t crash_notes[NR_CPUS]; - void machine_crash_shutdown(struct pt_regs *regs) { if (ppc_md.machine_crash_shutdown) diff --git a/arch/powerpc/kernel/pci_64.c b/arch/powerpc/kernel/pci_64.c index fc60a773af7..c367520bc1c 100644 --- a/arch/powerpc/kernel/pci_64.c +++ b/arch/powerpc/kernel/pci_64.c @@ -53,6 +53,7 @@ EXPORT_SYMBOL(io_page_mask); #ifdef CONFIG_PPC_MULTIPLATFORM static void fixup_resource(struct resource *res, struct pci_dev *dev); static void do_bus_setup(struct pci_bus *bus); +static void phbs_remap_io(void); #endif /* pci_io_base -- the base address from which io bars are offsets. @@ -251,6 +252,7 @@ void pcibios_free_controller(struct pci_controller *phb) kfree(phb); } +#ifndef CONFIG_PPC_ISERIES void __devinit pcibios_claim_one_bus(struct pci_bus *b) { struct pci_dev *dev; @@ -271,8 +273,10 @@ void __devinit pcibios_claim_one_bus(struct pci_bus *b) list_for_each_entry(child_bus, &b->children, node) pcibios_claim_one_bus(child_bus); } +#ifdef CONFIG_HOTPLUG +EXPORT_SYMBOL_GPL(pcibios_claim_one_bus); +#endif -#ifndef CONFIG_PPC_ISERIES static void __init pcibios_claim_of_setup(void) { struct pci_bus *b; @@ -381,7 +385,7 @@ struct pci_dev *of_create_pci_dev(struct device_node *node, dev->subsystem_vendor = get_int_prop(node, "subsystem-vendor-id", 0); dev->subsystem_device = get_int_prop(node, "subsystem-id", 0); - dev->cfg_size = 256; /*pci_cfg_space_size(dev);*/ + dev->cfg_size = pci_cfg_space_size(dev); sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(bus), dev->bus->number, PCI_SLOT(devfn), PCI_FUNC(devfn)); @@ -1215,7 +1219,7 @@ int remap_bus_range(struct pci_bus *bus) } EXPORT_SYMBOL(remap_bus_range); -void phbs_remap_io(void) +static void phbs_remap_io(void) { struct pci_controller *hose, *tmp; diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c index b2758148a0d..16d9a904f3c 100644 --- a/arch/powerpc/kernel/ppc_ksyms.c +++ b/arch/powerpc/kernel/ppc_ksyms.c @@ -244,7 +244,6 @@ EXPORT_SYMBOL(set_context); extern long mol_trampoline; EXPORT_SYMBOL(mol_trampoline); /* For MOL */ EXPORT_SYMBOL(flush_hash_pages); /* For MOL */ -EXPORT_SYMBOL_GPL(__handle_mm_fault); /* For MOL */ #ifdef CONFIG_SMP extern int mmu_hash_lock; EXPORT_SYMBOL(mmu_hash_lock); /* For MOL */ diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 105d5609ff5..57703994a06 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -201,13 +201,13 @@ int dump_spe(struct pt_regs *regs, elf_vrregset_t *evrregs) } #endif /* CONFIG_SPE */ +#ifndef CONFIG_SMP /* * If we are doing lazy switching of CPU state (FP, altivec or SPE), * and the current task has some state, discard it. */ -static inline void discard_lazy_cpu_state(void) +void discard_lazy_cpu_state(void) { -#ifndef CONFIG_SMP preempt_disable(); if (last_task_used_math == current) last_task_used_math = NULL; @@ -220,9 +220,10 @@ static inline void discard_lazy_cpu_state(void) last_task_used_spe = NULL; #endif preempt_enable(); -#endif /* CONFIG_SMP */ } +#endif /* CONFIG_SMP */ +#ifdef CONFIG_PPC_MERGE /* XXX for now */ int set_dabr(unsigned long dabr) { if (ppc_md.set_dabr) @@ -231,6 +232,7 @@ int set_dabr(unsigned long dabr) mtspr(SPRN_DABR, dabr); return 0; } +#endif #ifdef CONFIG_PPC64 DEFINE_PER_CPU(struct cpu_usage, cpu_usage_array); @@ -424,7 +426,7 @@ void show_regs(struct pt_regs * regs) if (trap == 0x300 || trap == 0x600) printk("DAR: "REG", DSISR: "REG"\n", regs->dar, regs->dsisr); printk("TASK = %p[%d] '%s' THREAD: %p", - current, current->pid, current->comm, current->thread_info); + current, current->pid, current->comm, task_thread_info(current)); #ifdef CONFIG_SMP printk(" CPU: %d", smp_processor_id()); @@ -503,7 +505,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp, { struct pt_regs *childregs, *kregs; extern void ret_from_fork(void); - unsigned long sp = (unsigned long)p->thread_info + THREAD_SIZE; + unsigned long sp = (unsigned long)task_stack_page(p) + THREAD_SIZE; CHECK_FULL_REGS(regs); /* Copy registers */ @@ -516,7 +518,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp, #ifdef CONFIG_PPC32 childregs->gpr[2] = (unsigned long) p; #else - clear_ti_thread_flag(p->thread_info, TIF_32BIT); + clear_tsk_thread_flag(p, TIF_32BIT); #endif p->thread.regs = NULL; /* no user register state */ } else { @@ -588,10 +590,8 @@ void start_thread(struct pt_regs *regs, unsigned long start, unsigned long sp) * set. Do it now. */ if (!current->thread.regs) { - unsigned long childregs = (unsigned long)current->thread_info + - THREAD_SIZE; - childregs -= sizeof(struct pt_regs); - current->thread.regs = (struct pt_regs *)childregs; + struct pt_regs *regs = task_stack_page(current) + THREAD_SIZE; + current->thread.regs = regs - 1; } memset(regs->gpr, 0, sizeof(regs->gpr)); @@ -767,7 +767,7 @@ out: static int validate_sp(unsigned long sp, struct task_struct *p, unsigned long nbytes) { - unsigned long stack_page = (unsigned long)p->thread_info; + unsigned long stack_page = (unsigned long)task_stack_page(p); if (sp >= stack_page + sizeof(struct thread_struct) && sp <= stack_page + THREAD_SIZE - nbytes) diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index 34ab0daec3a..02e2115323e 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -1100,17 +1100,37 @@ static int __init early_init_dt_scan_memory(unsigned long node, static void __init early_reserve_mem(void) { - unsigned long base, size; - unsigned long *reserve_map; + u64 base, size; + u64 *reserve_map; - reserve_map = (unsigned long *)(((unsigned long)initial_boot_params) + + reserve_map = (u64 *)(((unsigned long)initial_boot_params) + initial_boot_params->off_mem_rsvmap); +#ifdef CONFIG_PPC32 + /* + * Handle the case where we might be booting from an old kexec + * image that setup the mem_rsvmap as pairs of 32-bit values + */ + if (*reserve_map > 0xffffffffull) { + u32 base_32, size_32; + u32 *reserve_map_32 = (u32 *)reserve_map; + + while (1) { + base_32 = *(reserve_map_32++); + size_32 = *(reserve_map_32++); + if (size_32 == 0) + break; + DBG("reserving: %lx -> %lx\n", base_32, size_32); + lmb_reserve(base_32, size_32); + } + return; + } +#endif while (1) { base = *(reserve_map++); size = *(reserve_map++); if (size == 0) break; - DBG("reserving: %lx -> %lx\n", base, size); + DBG("reserving: %llx -> %llx\n", base, size); lmb_reserve(base, size); } diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index e381f2fc121..d963a12ec64 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -137,8 +137,8 @@ struct prom_t { }; struct mem_map_entry { - unsigned long base; - unsigned long size; + u64 base; + u64 size; }; typedef u32 cell_t; @@ -897,9 +897,9 @@ static unsigned long __init prom_next_cell(int s, cell_t **cellp) * If problems seem to show up, it would be a good start to track * them down. */ -static void reserve_mem(unsigned long base, unsigned long size) +static void reserve_mem(u64 base, u64 size) { - unsigned long top = base + size; + u64 top = base + size; unsigned long cnt = RELOC(mem_reserve_cnt); if (size == 0) diff --git a/arch/powerpc/kernel/ptrace-common.h b/arch/powerpc/kernel/ptrace-common.h index b1babb72967..5ccbdbe0d5c 100644 --- a/arch/powerpc/kernel/ptrace-common.h +++ b/arch/powerpc/kernel/ptrace-common.h @@ -62,7 +62,7 @@ static inline void set_single_step(struct task_struct *task) struct pt_regs *regs = task->thread.regs; if (regs != NULL) regs->msr |= MSR_SE; - set_ti_thread_flag(task->thread_info, TIF_SINGLESTEP); + set_tsk_thread_flag(task, TIF_SINGLESTEP); } static inline void clear_single_step(struct task_struct *task) @@ -70,7 +70,7 @@ static inline void clear_single_step(struct task_struct *task) struct pt_regs *regs = task->thread.regs; if (regs != NULL) regs->msr &= ~MSR_SE; - clear_ti_thread_flag(task->thread_info, TIF_SINGLESTEP); + clear_tsk_thread_flag(task, TIF_SINGLESTEP); } #ifdef CONFIG_ALTIVEC diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c index ae2e2a31a70..4b9cfe4637b 100644 --- a/arch/powerpc/kernel/rtas.c +++ b/arch/powerpc/kernel/rtas.c @@ -17,6 +17,7 @@ #include <linux/spinlock.h> #include <linux/module.h> #include <linux/init.h> +#include <linux/capability.h> #include <linux/delay.h> #include <asm/prom.h> @@ -29,6 +30,7 @@ #include <asm/delay.h> #include <asm/uaccess.h> #include <asm/lmb.h> +#include <asm/udbg.h> struct rtas_t rtas = { .lock = SPIN_LOCK_UNLOCKED @@ -52,7 +54,7 @@ EXPORT_SYMBOL(rtas_flash_term_hook); * are designed only for very early low-level debugging, which * is why the token is hard-coded to 10. */ -void call_rtas_display_status(unsigned char c) +static void call_rtas_display_status(char c) { struct rtas_args *args = &rtas.args; unsigned long s; @@ -65,14 +67,14 @@ void call_rtas_display_status(unsigned char c) args->nargs = 1; args->nret = 1; args->rets = (rtas_arg_t *)&(args->args[1]); - args->args[0] = (int)c; + args->args[0] = (unsigned char)c; enter_rtas(__pa(args)); spin_unlock_irqrestore(&rtas.lock, s); } -void call_rtas_display_status_delay(unsigned char c) +static void call_rtas_display_status_delay(char c) { static int pending_newline = 0; /* did last write end with unprinted newline? */ static int width = 16; @@ -96,6 +98,11 @@ void call_rtas_display_status_delay(unsigned char c) } } +void __init udbg_init_rtas(void) +{ + udbg_putc = call_rtas_display_status_delay; +} + void rtas_progress(char *s, unsigned short hex) { struct device_node *root; diff --git a/arch/powerpc/kernel/rtas_pci.c b/arch/powerpc/kernel/rtas_pci.c index 45b8109951f..5579f655991 100644 --- a/arch/powerpc/kernel/rtas_pci.c +++ b/arch/powerpc/kernel/rtas_pci.c @@ -72,7 +72,7 @@ static int of_device_available(struct device_node * dn) return 0; } -static int rtas_read_config(struct pci_dn *pdn, int where, int size, u32 *val) +int rtas_read_config(struct pci_dn *pdn, int where, int size, u32 *val) { int returnval = -1; unsigned long buid, addr; diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c index e5d285adb49..db72a92943b 100644 --- a/arch/powerpc/kernel/setup_32.c +++ b/arch/powerpc/kernel/setup_32.c @@ -299,9 +299,7 @@ void __init setup_arch(char **cmdline_p) if (ppc_md.init_early) ppc_md.init_early(); -#ifdef CONFIG_SERIAL_8250 find_legacy_serial_ports(); -#endif finish_device_tree(); smp_setup_cpu_maps(); diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 81567e93126..e29b275e09e 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -33,6 +33,7 @@ #include <linux/unistd.h> #include <linux/serial.h> #include <linux/serial_8250.h> +#include <linux/bootmem.h> #include <asm/io.h> #include <asm/kdump.h> #include <asm/prom.h> @@ -70,33 +71,6 @@ #define DBG(fmt...) #endif -/* - * Here are some early debugging facilities. You can enable one - * but your kernel will not boot on anything else if you do so - */ - -/* This one is for use on LPAR machines that support an HVC console - * on vterm 0 - */ -extern void udbg_init_debug_lpar(void); -/* This one is for use on Apple G5 machines - */ -extern void udbg_init_pmac_realmode(void); -/* That's RTAS panel debug */ -extern void call_rtas_display_status_delay(unsigned char c); -/* Here's maple real mode debug */ -extern void udbg_init_maple_realmode(void); - -#define EARLY_DEBUG_INIT() do {} while(0) - -#if 0 -#define EARLY_DEBUG_INIT() udbg_init_debug_lpar() -#define EARLY_DEBUG_INIT() udbg_init_maple_realmode() -#define EARLY_DEBUG_INIT() udbg_init_pmac_realmode() -#define EARLY_DEBUG_INIT() \ - do { udbg_putc = call_rtas_display_status_delay; } while(0) -#endif - int have_of = 1; int boot_cpuid = 0; int boot_cpuid_phys = 0; @@ -237,11 +211,8 @@ void __init early_setup(unsigned long dt_ptr) struct paca_struct *lpaca = get_paca(); static struct machdep_calls **mach; - /* - * Enable early debugging if any specified (see top of - * this file) - */ - EARLY_DEBUG_INIT(); + /* Enable early debugging if any specified (see udbg.h) */ + udbg_early_init(); DBG(" -> early_setup()\n"); @@ -472,9 +443,7 @@ void __init setup_system(void) * hash table management for us, thus ioremap works. We do that early * so that further code can be debugged */ -#ifdef CONFIG_SERIAL_8250 find_legacy_serial_ports(); -#endif /* * "Finish" the device-tree, that is do the actual parsing of @@ -686,3 +655,28 @@ void cpu_die(void) if (ppc_md.cpu_die) ppc_md.cpu_die(); } + +#ifdef CONFIG_SMP +void __init setup_per_cpu_areas(void) +{ + int i; + unsigned long size; + char *ptr; + + /* Copy section for each CPU (we discard the original) */ + size = ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES); +#ifdef CONFIG_MODULES + if (size < PERCPU_ENOUGH_ROOM) + size = PERCPU_ENOUGH_ROOM; +#endif + + for_each_cpu(i) { + ptr = alloc_bootmem_node(NODE_DATA(cpu_to_node(i)), size); + if (!ptr) + panic("Cannot allocate cpu data for CPU %d\n", i); + + paca[i].data_offset = ptr - __per_cpu_start; + memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start); + } +} +#endif diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index d3f0b6d452f..177bba78fb0 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -497,6 +497,15 @@ static long restore_user_regs(struct pt_regs *regs, if (err) return 1; + /* + * Do this before updating the thread state in + * current->thread.fpr/vr/evr. That way, if we get preempted + * and another task grabs the FPU/Altivec/SPE, it won't be + * tempted to save the current CPU state into the thread_struct + * and corrupt what we are writing there. + */ + discard_lazy_cpu_state(); + /* force the process to reload the FP registers from current->thread when it next does FP instructions */ regs->msr &= ~(MSR_FP | MSR_FE0 | MSR_FE1); @@ -538,18 +547,6 @@ static long restore_user_regs(struct pt_regs *regs, return 1; #endif /* CONFIG_SPE */ -#ifndef CONFIG_SMP - preempt_disable(); - if (last_task_used_math == current) - last_task_used_math = NULL; - if (last_task_used_altivec == current) - last_task_used_altivec = NULL; -#ifdef CONFIG_SPE - if (last_task_used_spe == current) - last_task_used_spe = NULL; -#endif - preempt_enable(); -#endif return 0; } diff --git a/arch/powerpc/kernel/signal_64.c b/arch/powerpc/kernel/signal_64.c index 5462bef898f..7b9d999e211 100644 --- a/arch/powerpc/kernel/signal_64.c +++ b/arch/powerpc/kernel/signal_64.c @@ -207,10 +207,20 @@ static long restore_sigcontext(struct pt_regs *regs, sigset_t *set, int sig, if (!sig) regs->gpr[13] = save_r13; - err |= __copy_from_user(¤t->thread.fpr, &sc->fp_regs, FP_REGS_SIZE); if (set != NULL) err |= __get_user(set->sig[0], &sc->oldmask); + /* + * Do this before updating the thread state in + * current->thread.fpr/vr. That way, if we get preempted + * and another task grabs the FPU/Altivec, it won't be + * tempted to save the current CPU state into the thread_struct + * and corrupt what we are writing there. + */ + discard_lazy_cpu_state(); + + err |= __copy_from_user(¤t->thread.fpr, &sc->fp_regs, FP_REGS_SIZE); + #ifdef CONFIG_ALTIVEC err |= __get_user(v_regs, &sc->v_regs); err |= __get_user(msr, &sc->gp_regs[PT_MSR]); @@ -229,14 +239,6 @@ static long restore_sigcontext(struct pt_regs *regs, sigset_t *set, int sig, current->thread.vrsave = 0; #endif /* CONFIG_ALTIVEC */ -#ifndef CONFIG_SMP - preempt_disable(); - if (last_task_used_math == current) - last_task_used_math = NULL; - if (last_task_used_altivec == current) - last_task_used_altivec = NULL; - preempt_enable(); -#endif /* Force reload of FP/VEC */ regs->msr &= ~(MSR_FP | MSR_FE0 | MSR_FE1 | MSR_VEC); diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index d381ec90b75..c8458c531b2 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -338,8 +338,8 @@ static void __init smp_create_idle(unsigned int cpu) #ifdef CONFIG_PPC64 paca[cpu].__current = p; #endif - current_set[cpu] = p->thread_info; - p->thread_info->cpu = cpu; + current_set[cpu] = task_thread_info(p); + task_thread_info(p)->cpu = cpu; } void __init smp_prepare_cpus(unsigned int max_cpus) @@ -375,7 +375,7 @@ void __devinit smp_prepare_boot_cpu(void) #ifdef CONFIG_PPC64 paca[boot_cpuid].__current = current; #endif - current_set[boot_cpuid] = current->thread_info; + current_set[boot_cpuid] = task_thread_info(current); } #ifdef CONFIG_HOTPLUG_CPU diff --git a/arch/powerpc/kernel/sys_ppc32.c b/arch/powerpc/kernel/sys_ppc32.c index 9c921d1c408..475249dc235 100644 --- a/arch/powerpc/kernel/sys_ppc32.c +++ b/arch/powerpc/kernel/sys_ppc32.c @@ -552,30 +552,6 @@ asmlinkage long compat_sys_sched_rr_get_interval(u32 pid, struct compat_timespec return ret; } -asmlinkage int compat_sys_pciconfig_read(u32 bus, u32 dfn, u32 off, u32 len, u32 ubuf) -{ - return sys_pciconfig_read((unsigned long) bus, - (unsigned long) dfn, - (unsigned long) off, - (unsigned long) len, - compat_ptr(ubuf)); -} - -asmlinkage int compat_sys_pciconfig_write(u32 bus, u32 dfn, u32 off, u32 len, u32 ubuf) -{ - return sys_pciconfig_write((unsigned long) bus, - (unsigned long) dfn, - (unsigned long) off, - (unsigned long) len, - compat_ptr(ubuf)); -} - -asmlinkage int compat_sys_pciconfig_iobase(u32 which, u32 in_bus, u32 in_devfn) -{ - return sys_pciconfig_iobase(which, in_bus, in_devfn); -} - - /* Note: it is necessary to treat mode as an unsigned int, * with the corresponding cast to a signed int to insure that the * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) @@ -956,38 +932,6 @@ long ppc32_fadvise64(int fd, u32 unused, u32 offset_high, u32 offset_low, advice); } -long ppc32_timer_create(clockid_t clock, - struct compat_sigevent __user *ev32, - timer_t __user *timer_id) -{ - sigevent_t event; - timer_t t; - long err; - mm_segment_t savefs; - - if (ev32 == NULL) - return sys_timer_create(clock, NULL, timer_id); - - if (get_compat_sigevent(&event, ev32)) - return -EFAULT; - - if (!access_ok(VERIFY_WRITE, timer_id, sizeof(timer_t))) - return -EFAULT; - - savefs = get_fs(); - set_fs(KERNEL_DS); - /* The __user pointer casts are valid due to the set_fs() */ - err = sys_timer_create(clock, - (sigevent_t __user *) &event, - (timer_t __user *) &t); - set_fs(savefs); - - if (err == 0) - err = __put_user(t, timer_id); - - return err; -} - asmlinkage long compat_sys_add_key(const char __user *_type, const char __user *_description, const void __user *_payload, diff --git a/arch/powerpc/kernel/systbl.S b/arch/powerpc/kernel/systbl.S index 65463a1076e..68013179a50 100644 --- a/arch/powerpc/kernel/systbl.S +++ b/arch/powerpc/kernel/systbl.S @@ -239,9 +239,9 @@ SYS32ONLY(ftruncate64) SYSX(sys_ni_syscall,sys_stat64,sys_stat64) SYSX(sys_ni_syscall,sys_lstat64,sys_lstat64) SYSX(sys_ni_syscall,sys_fstat64,sys_fstat64) -COMPAT_SYS(pciconfig_read) -COMPAT_SYS(pciconfig_write) -COMPAT_SYS(pciconfig_iobase) +SYSCALL(pciconfig_read) +SYSCALL(pciconfig_write) +SYSCALL(pciconfig_iobase) SYSCALL(ni_syscall) SYSCALL(getdents64) SYSCALL(pivot_root) @@ -281,7 +281,7 @@ SYSCALL(epoll_create) SYSCALL(epoll_ctl) SYSCALL(epoll_wait) SYSCALL(remap_file_pages) -SYSX(sys_timer_create,ppc32_timer_create,sys_timer_create) +SYSX(sys_timer_create,compat_sys_timer_create,sys_timer_create) COMPAT_SYS(timer_settime) COMPAT_SYS(timer_gettime) SYSCALL(timer_getoverrun) diff --git a/arch/powerpc/kernel/udbg.c b/arch/powerpc/kernel/udbg.c index 558c1ceb2b9..3774e80094f 100644 --- a/arch/powerpc/kernel/udbg.c +++ b/arch/powerpc/kernel/udbg.c @@ -15,11 +15,36 @@ #include <linux/sched.h> #include <linux/console.h> #include <asm/processor.h> +#include <asm/udbg.h> void (*udbg_putc)(char c); int (*udbg_getc)(void); int (*udbg_getc_poll)(void); +/* + * Early debugging facilities. You can enable _one_ of these via .config, + * if you do so your kernel _will not boot_ on anything else. Be careful. + */ +void __init udbg_early_init(void) +{ +#if defined(CONFIG_PPC_EARLY_DEBUG_LPAR) + /* For LPAR machines that have an HVC console on vterm 0 */ + udbg_init_debug_lpar(); +#elif defined(CONFIG_PPC_EARLY_DEBUG_G5) + /* For use on Apple G5 machines */ + udbg_init_pmac_realmode(); +#elif defined(CONFIG_PPC_EARLY_DEBUG_RTAS) + /* RTAS panel debug */ + udbg_init_rtas(); +#elif defined(CONFIG_PPC_EARLY_DEBUG_MAPLE) + /* Maple real mode debug */ + udbg_init_maple_realmode(); +#elif defined(CONFIG_PPC_EARLY_DEBUG_ISERIES) + /* For iSeries - hit Ctrl-x Ctrl-x to see the output */ + udbg_init_iseries(); +#endif +} + /* udbg library, used by xmon et al */ void udbg_puts(const char *s) { diff --git a/arch/powerpc/kernel/udbg_16550.c b/arch/powerpc/kernel/udbg_16550.c index 7541bf44d2d..2da65a9c93f 100644 --- a/arch/powerpc/kernel/udbg_16550.c +++ b/arch/powerpc/kernel/udbg_16550.c @@ -155,7 +155,7 @@ void udbg_maple_real_putc(unsigned char c) } } -void udbg_init_maple_realmode(void) +void __init udbg_init_maple_realmode(void) { udbg_comport = (volatile struct NS16550 __iomem *)0xf40003f8; diff --git a/arch/powerpc/kernel/vdso32/.gitignore b/arch/powerpc/kernel/vdso32/.gitignore new file mode 100644 index 00000000000..e45fba9d0ce --- /dev/null +++ b/arch/powerpc/kernel/vdso32/.gitignore @@ -0,0 +1 @@ +vdso32.lds diff --git a/arch/powerpc/kernel/vdso64/.gitignore b/arch/powerpc/kernel/vdso64/.gitignore new file mode 100644 index 00000000000..3fd18cf9fec --- /dev/null +++ b/arch/powerpc/kernel/vdso64/.gitignore @@ -0,0 +1 @@ +vdso64.lds diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c index 2ffca63602c..7b278d83739 100644 --- a/arch/powerpc/mm/pgtable_64.c +++ b/arch/powerpc/mm/pgtable_64.c @@ -174,7 +174,7 @@ void __iomem * __ioremap(unsigned long addr, unsigned long size, pa = addr & PAGE_MASK; size = PAGE_ALIGN(addr + size) - pa; - if (size == 0) + if ((size == 0) || (pa == 0)) return NULL; if (mem_init_done) { diff --git a/arch/powerpc/platforms/83xx/Kconfig b/arch/powerpc/platforms/83xx/Kconfig new file mode 100644 index 00000000000..b20812d460e --- /dev/null +++ b/arch/powerpc/platforms/83xx/Kconfig @@ -0,0 +1,26 @@ +menu "Platform support" + depends on PPC_83xx + +choice + prompt "Machine Type" + default MPC834x_SYS + +config MPC834x_SYS + bool "Freescale MPC834x SYS" + help + This option enables support for the MPC 834x SYS evaluation board. + + Be aware that PCI buses can only function when SYS board is plugged + into the PIB (Platform IO Board) board from Freescale which provide + 3 PCI slots. The PIBs PCI initialization is the bootloader's + responsiblilty. + +endchoice + +config MPC834x + bool + select PPC_UDBG_16550 + select PPC_INDIRECT_PCI + default y if MPC834x_SYS + +endmenu diff --git a/arch/powerpc/platforms/83xx/Makefile b/arch/powerpc/platforms/83xx/Makefile new file mode 100644 index 00000000000..9d8b28ef334 --- /dev/null +++ b/arch/powerpc/platforms/83xx/Makefile @@ -0,0 +1,4 @@ +# +# Makefile for the PowerPC 83xx linux kernel. +# +obj-$(CONFIG_MPC834x_SYS) += mpc834x_sys.o pci.o diff --git a/arch/powerpc/platforms/Makefile b/arch/powerpc/platforms/Makefile index 8836b3a0066..04073fd987e 100644 --- a/arch/powerpc/platforms/Makefile +++ b/arch/powerpc/platforms/Makefile @@ -7,6 +7,7 @@ endif endif obj-$(CONFIG_PPC_CHRP) += chrp/ obj-$(CONFIG_4xx) += 4xx/ +obj-$(CONFIG_PPC_83xx) += 83xx/ obj-$(CONFIG_85xx) += 85xx/ obj-$(CONFIG_PPC_PSERIES) += pseries/ obj-$(CONFIG_PPC_ISERIES) += iseries/ diff --git a/arch/powerpc/platforms/cell/pervasive.c b/arch/powerpc/platforms/cell/pervasive.c index 85152544c15..e0e051c675d 100644 --- a/arch/powerpc/platforms/cell/pervasive.c +++ b/arch/powerpc/platforms/cell/pervasive.c @@ -142,7 +142,7 @@ static void cbe_idle(void) } } -int cbe_system_reset_exception(struct pt_regs *regs) +static int cbe_system_reset_exception(struct pt_regs *regs) { switch (regs->msr & SRR1_WAKEMASK) { case SRR1_WAKEEE: diff --git a/arch/powerpc/platforms/cell/setup.c b/arch/powerpc/platforms/cell/setup.c index 18e25e65c04..b33a4443f5a 100644 --- a/arch/powerpc/platforms/cell/setup.c +++ b/arch/powerpc/platforms/cell/setup.c @@ -57,7 +57,7 @@ #define DBG(fmt...) #endif -void cell_show_cpuinfo(struct seq_file *m) +static void cell_show_cpuinfo(struct seq_file *m) { struct device_node *root; const char *model = ""; diff --git a/arch/powerpc/platforms/cell/smp.c b/arch/powerpc/platforms/cell/smp.c index de96eadf419..bdf6c5fe58c 100644 --- a/arch/powerpc/platforms/cell/smp.c +++ b/arch/powerpc/platforms/cell/smp.c @@ -86,7 +86,7 @@ static inline int __devinit smp_startup_cpu(unsigned int lcpu) pcpu = get_hard_smp_processor_id(lcpu); /* Fixup atomic count: it exited inside IRQ handler. */ - paca[lcpu].__current->thread_info->preempt_count = 0; + task_thread_info(paca[lcpu].__current)->preempt_count = 0; /* * If the RTAS start-cpu token does not exist then presume the diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index d2ba358c6e3..b3962c3a034 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -138,7 +138,7 @@ static void spufs_prune_dir(struct dentry *dir) { struct dentry *dentry, *tmp; mutex_lock(&dir->d_inode->i_mutex); - list_for_each_entry_safe(dentry, tmp, &dir->d_subdirs, d_child) { + list_for_each_entry_safe(dentry, tmp, &dir->d_subdirs, d_u.d_child) { spin_lock(&dcache_lock); spin_lock(&dentry->d_lock); if (!(d_unhashed(dentry)) && dentry->d_inode) { diff --git a/arch/powerpc/platforms/cell/spufs/switch.c b/arch/powerpc/platforms/cell/spufs/switch.c index 1061c12b2ed..212db28531f 100644 --- a/arch/powerpc/platforms/cell/spufs/switch.c +++ b/arch/powerpc/platforms/cell/spufs/switch.c @@ -169,11 +169,33 @@ static inline void save_mfc_cntl(struct spu_state *csa, struct spu *spu) struct spu_priv2 __iomem *priv2 = spu->priv2; /* Save, Step 8: - * Read and save MFC_CNTL[Ss]. - */ - if (csa) { - csa->priv2.mfc_control_RW = in_be64(&priv2->mfc_control_RW) & - MFC_CNTL_SUSPEND_DMA_STATUS_MASK; + * Suspend DMA and save MFC_CNTL. + */ + switch (in_be64(&priv2->mfc_control_RW) & + MFC_CNTL_SUSPEND_DMA_STATUS_MASK) { + case MFC_CNTL_SUSPEND_IN_PROGRESS: + POLL_WHILE_FALSE((in_be64(&priv2->mfc_control_RW) & + MFC_CNTL_SUSPEND_DMA_STATUS_MASK) == + MFC_CNTL_SUSPEND_COMPLETE); + /* fall through */ + case MFC_CNTL_SUSPEND_COMPLETE: + if (csa) { + csa->priv2.mfc_control_RW = + in_be64(&priv2->mfc_control_RW) | + MFC_CNTL_SUSPEND_DMA_QUEUE; + } + break; + case MFC_CNTL_NORMAL_DMA_QUEUE_OPERATION: + out_be64(&priv2->mfc_control_RW, MFC_CNTL_SUSPEND_DMA_QUEUE); + POLL_WHILE_FALSE((in_be64(&priv2->mfc_control_RW) & + MFC_CNTL_SUSPEND_DMA_STATUS_MASK) == + MFC_CNTL_SUSPEND_COMPLETE); + if (csa) { + csa->priv2.mfc_control_RW = + in_be64(&priv2->mfc_control_RW) & + ~MFC_CNTL_SUSPEND_DMA_QUEUE; + } + break; } } @@ -237,6 +259,8 @@ static inline void save_mfc_decr(struct spu_state *csa, struct spu *spu) eieio(); csa->spu_chnldata_RW[7] = in_be64(&priv2->spu_chnldata_RW); eieio(); + } else { + csa->priv2.mfc_control_RW &= ~MFC_CNTL_DECREMENTER_RUNNING; } } diff --git a/arch/powerpc/platforms/cell/spufs/syscalls.c b/arch/powerpc/platforms/cell/spufs/syscalls.c index d549aa7ebea..e6565a949dd 100644 --- a/arch/powerpc/platforms/cell/spufs/syscalls.c +++ b/arch/powerpc/platforms/cell/spufs/syscalls.c @@ -29,7 +29,9 @@ * value of the spu_status register after the SPU has stopped. * */ -long do_spu_run(struct file *filp, __u32 __user *unpc, __u32 __user *ustatus) +static long do_spu_run(struct file *filp, + __u32 __user *unpc, + __u32 __user *ustatus) { long ret; struct spufs_inode_info *i; diff --git a/arch/powerpc/platforms/embedded6xx/Kconfig b/arch/powerpc/platforms/embedded6xx/Kconfig index 81250090f98..4fdbc9ae876 100644 --- a/arch/powerpc/platforms/embedded6xx/Kconfig +++ b/arch/powerpc/platforms/embedded6xx/Kconfig @@ -144,16 +144,6 @@ config LITE5200 much but it's only been tested on this board version. I think this board is also known as IceCube. -config MPC834x_SYS - bool "Freescale MPC834x SYS" - help - This option enables support for the MPC 834x SYS evaluation board. - - Be aware that PCI buses can only function when SYS board is plugged - into the PIB (Platform IO Board) board from Freescale which provide - 3 PCI slots. The PIBs PCI initialization is the bootloader's - responsiblilty. - config EV64360 bool "Marvell-EV64360BP" help @@ -192,14 +182,6 @@ config 8272 The MPC8272 CPM has a different internal dpram setup than other CPM2 devices -config 83xx - bool - default y if MPC834x_SYS - -config MPC834x - bool - default y if MPC834x_SYS - config CPM2 bool depends on 8260 || MPC8560 || MPC8555 diff --git a/arch/powerpc/platforms/iseries/Makefile b/arch/powerpc/platforms/iseries/Makefile index 127b465308b..ce8c0b943fa 100644 --- a/arch/powerpc/platforms/iseries/Makefile +++ b/arch/powerpc/platforms/iseries/Makefile @@ -1,8 +1,8 @@ EXTRA_CFLAGS += -mno-minimal-toc obj-y += hvlog.o hvlpconfig.o lpardata.o setup.o mf.o lpevents.o \ - hvcall.o proc.o htab.o iommu.o misc.o -obj-$(CONFIG_PCI) += pci.o irq.o vpdinfo.o + hvcall.o proc.o htab.o iommu.o misc.o irq.o +obj-$(CONFIG_PCI) += pci.o vpdinfo.o obj-$(CONFIG_IBMVIO) += vio.o obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_VIOPATH) += viopath.o diff --git a/arch/powerpc/platforms/iseries/iommu.c b/arch/powerpc/platforms/iseries/iommu.c index 2b54eeb2c89..bea0b703f40 100644 --- a/arch/powerpc/platforms/iseries/iommu.c +++ b/arch/powerpc/platforms/iseries/iommu.c @@ -34,6 +34,8 @@ #include <asm/pci-bridge.h> #include <asm/iseries/hv_call_xm.h> +#include "iommu.h" + extern struct list_head iSeries_Global_Device_List; diff --git a/arch/powerpc/platforms/iseries/iommu.h b/arch/powerpc/platforms/iseries/iommu.h new file mode 100644 index 00000000000..cb5658fbe65 --- /dev/null +++ b/arch/powerpc/platforms/iseries/iommu.h @@ -0,0 +1,35 @@ +#ifndef _PLATFORMS_ISERIES_IOMMU_H +#define _PLATFORMS_ISERIES_IOMMU_H + +/* + * Copyright (C) 2005 Stephen Rothwell, IBM Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the: + * Free Software Foundation, Inc., + * 59 Temple Place, Suite 330, + * Boston, MA 02111-1307 USA + */ + +struct device_node; +struct iommu_table; + +/* Creates table for an individual device node */ +extern void iommu_devnode_init_iSeries(struct device_node *dn); + +/* Get table parameters from HV */ +extern void iommu_table_getparms_iSeries(unsigned long busno, + unsigned char slotno, unsigned char virtbus, + struct iommu_table *tbl); + +#endif /* _PLATFORMS_ISERIES_IOMMU_H */ diff --git a/arch/powerpc/platforms/iseries/irq.c b/arch/powerpc/platforms/iseries/irq.c index 42e978e4897..83442ea7747 100644 --- a/arch/powerpc/platforms/iseries/irq.c +++ b/arch/powerpc/platforms/iseries/irq.c @@ -48,6 +48,8 @@ extern void iSeries_smp_message_recv(struct pt_regs *); #endif +#ifdef CONFIG_PCI + enum pci_event_type { pe_bus_created = 0, /* PHB has been created */ pe_bus_error = 1, /* PHB has failed */ @@ -147,20 +149,11 @@ static void int_received(struct pci_event *event, struct pt_regs *regs) static void pci_event_handler(struct HvLpEvent *event, struct pt_regs *regs) { if (event && (event->xType == HvLpEvent_Type_PciIo)) { - switch (event->xFlags.xFunction) { - case HvLpEvent_Function_Int: + if (hvlpevent_is_int(event)) int_received((struct pci_event *)event, regs); - break; - case HvLpEvent_Function_Ack: + else printk(KERN_ERR "pci_event_handler: unexpected ack received\n"); - break; - default: - printk(KERN_ERR - "pci_event_handler: unexpected event function %d\n", - (int)event->xFlags.xFunction); - break; - } } else if (event) printk(KERN_ERR "pci_event_handler: Unrecognized PCI event type 0x%x\n", @@ -334,6 +327,8 @@ int __init iSeries_allocate_IRQ(HvBusNumber bus, return virtirq; } +#endif /* CONFIG_PCI */ + /* * Get the next pending IRQ. */ @@ -353,6 +348,7 @@ int iSeries_get_irq(struct pt_regs *regs) if (hvlpevent_is_pending()) process_hvlpevents(regs); +#ifdef CONFIG_PCI if (num_pending_irqs) { spin_lock(&pending_irqs_lock); for (irq = 0; irq < NR_IRQS; irq++) { @@ -366,6 +362,7 @@ int iSeries_get_irq(struct pt_regs *regs) if (irq >= NR_IRQS) irq = -2; } +#endif return irq; } diff --git a/arch/powerpc/platforms/iseries/lpardata.c b/arch/powerpc/platforms/iseries/lpardata.c index ea72385aaf0..438e2dba63b 100644 --- a/arch/powerpc/platforms/iseries/lpardata.c +++ b/arch/powerpc/platforms/iseries/lpardata.c @@ -93,10 +93,7 @@ struct ItLpNaca itLpNaca = { .xPirEnvironMode = 0, /* Piranha stuff */ .xPirConsoleMode = 0, .xPirDasdMode = 0, - .xLparInstalled = 0, - .xSysPartitioned = 0, - .xHwSyncedTBs = 0, - .xIntProcUtilHmt = 0, + .flags = 0, .xSpVpdFormat = 0, .xIntProcRatio = 0, .xPlicVrmIndex = 0, /* VRM index of PLIC */ diff --git a/arch/powerpc/platforms/iseries/lpevents.c b/arch/powerpc/platforms/iseries/lpevents.c index e9fb98bf895..0b885300d1d 100644 --- a/arch/powerpc/platforms/iseries/lpevents.c +++ b/arch/powerpc/platforms/iseries/lpevents.c @@ -53,7 +53,7 @@ static struct HvLpEvent * get_next_hvlpevent(void) struct HvLpEvent * event; event = (struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr; - if (event->xFlags.xValid) { + if (hvlpevent_is_valid(event)) { /* rmb() needed only for weakly consistent machines (regatta) */ rmb(); /* Set pointer to next potential event */ @@ -84,7 +84,7 @@ int hvlpevent_is_pending(void) next_event = (struct HvLpEvent *)hvlpevent_queue.xSlicCurEventPtr; - return next_event->xFlags.xValid | + return hvlpevent_is_valid(next_event) || hvlpevent_queue.xPlicOverflowIntPending; } @@ -101,18 +101,18 @@ static void hvlpevent_clear_valid(struct HvLpEvent * event) switch (extra) { case 3: tmp = (struct HvLpEvent*)((char*)event + 3 * LpEventAlign); - tmp->xFlags.xValid = 0; + hvlpevent_invalidate(tmp); case 2: tmp = (struct HvLpEvent*)((char*)event + 2 * LpEventAlign); - tmp->xFlags.xValid = 0; + hvlpevent_invalidate(tmp); case 1: tmp = (struct HvLpEvent*)((char*)event + 1 * LpEventAlign); - tmp->xFlags.xValid = 0; + hvlpevent_invalidate(tmp); } mb(); - event->xFlags.xValid = 0; + hvlpevent_invalidate(event); } void process_hvlpevents(struct pt_regs *regs) diff --git a/arch/powerpc/platforms/iseries/mf.c b/arch/powerpc/platforms/iseries/mf.c index 49e7e4b8584..a41d8b78c0c 100644 --- a/arch/powerpc/platforms/iseries/mf.c +++ b/arch/powerpc/platforms/iseries/mf.c @@ -251,10 +251,7 @@ static struct pending_event *new_pending_event(void) } memset(ev, 0, sizeof(struct pending_event)); hev = &ev->event.hp_lp_event; - hev->xFlags.xValid = 1; - hev->xFlags.xAckType = HvLpEvent_AckType_ImmediateAck; - hev->xFlags.xAckInd = HvLpEvent_AckInd_DoAck; - hev->xFlags.xFunction = HvLpEvent_Function_Int; + hev->flags = HV_LP_EVENT_VALID | HV_LP_EVENT_DO_ACK | HV_LP_EVENT_INT; hev->xType = HvLpEvent_Type_MachineFac; hev->xSourceLp = HvLpConfig_getLpIndex(); hev->xTargetLp = primary_lp; @@ -518,17 +515,10 @@ static void handle_ack(struct io_mf_lp_event *event) static void hv_handler(struct HvLpEvent *event, struct pt_regs *regs) { if ((event != NULL) && (event->xType == HvLpEvent_Type_MachineFac)) { - switch(event->xFlags.xFunction) { - case HvLpEvent_Function_Ack: + if (hvlpevent_is_ack(event)) handle_ack((struct io_mf_lp_event *)event); - break; - case HvLpEvent_Function_Int: + else handle_int((struct io_mf_lp_event *)event); - break; - default: - printk(KERN_ERR "mf.c: non ack/int event received\n"); - break; - } } else printk(KERN_ERR "mf.c: alien event received\n"); } diff --git a/arch/powerpc/platforms/iseries/pci.c b/arch/powerpc/platforms/iseries/pci.c index dafc518fbb8..a19833b880e 100644 --- a/arch/powerpc/platforms/iseries/pci.c +++ b/arch/powerpc/platforms/iseries/pci.c @@ -43,6 +43,7 @@ #include "irq.h" #include "pci.h" #include "call_pci.h" +#include "iommu.h" extern unsigned long io_page_mask; diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c index ad5ef80500c..c6bbe5c2510 100644 --- a/arch/powerpc/platforms/iseries/setup.c +++ b/arch/powerpc/platforms/iseries/setup.c @@ -52,6 +52,7 @@ #include <asm/iseries/mf.h> #include <asm/iseries/hv_lp_event.h> #include <asm/iseries/lpar_map.h> +#include <asm/udbg.h> #include "naca.h" #include "setup.h" @@ -62,10 +63,8 @@ #include "call_sm.h" #include "call_hpt.h" -extern void hvlog(char *fmt, ...); - #ifdef DEBUG -#define DBG(fmt...) hvlog(fmt) +#define DBG(fmt...) udbg_printf(fmt) #else #define DBG(fmt...) #endif @@ -474,14 +473,6 @@ static unsigned long __init build_iSeries_Memory_Map(void) printk("HPT absolute addr = %016lx, size = %dK\n", chunk_to_addr(hptFirstChunk), hptSizeChunks * 256); - ppc64_pft_size = __ilog2(hptSizePages * HW_PAGE_SIZE); - - /* - * The actual hashed page table is in the hypervisor, - * we have no direct access - */ - htab_address = NULL; - /* * Determine if absolute memory has any * holes so that we can interpret the @@ -861,6 +852,11 @@ void dt_prop_u64_list(struct iseries_flat_dt *dt, char *name, u64 *data, int n) dt_prop(dt, name, (char *)data, sizeof(u64) * n); } +void dt_prop_u32_list(struct iseries_flat_dt *dt, char *name, u32 *data, int n) +{ + dt_prop(dt, name, (char *)data, sizeof(u32) * n); +} + void dt_prop_empty(struct iseries_flat_dt *dt, char *name) { dt_prop(dt, name, NULL, 0); @@ -872,6 +868,7 @@ void dt_cpus(struct iseries_flat_dt *dt) unsigned char *p; unsigned int i, index; struct IoHriProcessorVpd *d; + u32 pft_size[2]; /* yuck */ snprintf(buf, 32, "PowerPC,%s", cur_cpu_spec->cpu_name); @@ -882,6 +879,9 @@ void dt_cpus(struct iseries_flat_dt *dt) dt_prop_u32(dt, "#address-cells", 1); dt_prop_u32(dt, "#size-cells", 0); + pft_size[0] = 0; /* NUMA CEC cookie, 0 for non NUMA */ + pft_size[1] = __ilog2(HvCallHpt_getHptPages() * HW_PAGE_SIZE); + for (i = 0; i < NR_CPUS; i++) { if (paca[i].lppaca.dyn_proc_status >= 2) continue; @@ -908,6 +908,8 @@ void dt_cpus(struct iseries_flat_dt *dt) dt_prop_u32(dt, "reg", i); + dt_prop_u32_list(dt, "ibm,pft-size", pft_size, 2); + dt_end_node(dt); } @@ -984,3 +986,16 @@ static int __init early_parsemem(char *p) return 0; } early_param("mem", early_parsemem); + +static void hvputc(char c) +{ + if (c == '\n') + hvputc('\r'); + + HvCall_writeLogBuffer(&c, 1); +} + +void __init udbg_init_iseries(void) +{ + udbg_putc = hvputc; +} diff --git a/arch/powerpc/platforms/iseries/vio.c b/arch/powerpc/platforms/iseries/vio.c index 384360ee06e..ad36ab0639f 100644 --- a/arch/powerpc/platforms/iseries/vio.c +++ b/arch/powerpc/platforms/iseries/vio.c @@ -22,6 +22,8 @@ #include <asm/iseries/hv_lp_config.h> #include <asm/iseries/hv_call_xm.h> +#include "iommu.h" + struct device *iSeries_vio_dev = &vio_bus_device.dev; EXPORT_SYMBOL(iSeries_vio_dev); diff --git a/arch/powerpc/platforms/iseries/viopath.c b/arch/powerpc/platforms/iseries/viopath.c index 84267269559..622a30149b4 100644 --- a/arch/powerpc/platforms/iseries/viopath.c +++ b/arch/powerpc/platforms/iseries/viopath.c @@ -270,7 +270,7 @@ static void handleMonitorEvent(struct HvLpEvent *event) * First see if this is just a normal monitor message from the * other partition */ - if (event->xFlags.xFunction == HvLpEvent_Function_Int) { + if (hvlpevent_is_int(event)) { remoteLp = event->xSourceLp; if (!viopathStatus[remoteLp].isActive) sendMonMsg(remoteLp); @@ -331,13 +331,12 @@ static void handleConfig(struct HvLpEvent *event) { if (!event) return; - if (event->xFlags.xFunction == HvLpEvent_Function_Int) { + if (hvlpevent_is_int(event)) { printk(VIOPATH_KERN_WARN "unexpected config request from partition %d", event->xSourceLp); - if ((event->xFlags.xFunction == HvLpEvent_Function_Int) && - (event->xFlags.xAckInd == HvLpEvent_AckInd_DoAck)) { + if (hvlpevent_need_ack(event)) { event->xRc = HvLpEvent_Rc_InvalidSubtype; HvCallEvent_ackLpEvent(event); } @@ -377,7 +376,7 @@ static void vio_handleEvent(struct HvLpEvent *event, struct pt_regs *regs) int subtype = (event->xSubtype & VIOMAJOR_SUBTYPE_MASK) >> VIOMAJOR_SUBTYPE_SHIFT; - if (event->xFlags.xFunction == HvLpEvent_Function_Int) { + if (hvlpevent_is_int(event)) { remoteLp = event->xSourceLp; /* * The isActive is checked because if the hosting partition @@ -436,8 +435,7 @@ static void vio_handleEvent(struct HvLpEvent *event, struct pt_regs *regs) "unexpected virtual io event subtype %d from partition %d\n", event->xSubtype, remoteLp); /* No handler. Ack if necessary */ - if ((event->xFlags.xFunction == HvLpEvent_Function_Int) && - (event->xFlags.xAckInd == HvLpEvent_AckInd_DoAck)) { + if (hvlpevent_is_int(event) && hvlpevent_need_ack(event)) { event->xRc = HvLpEvent_Rc_InvalidSubtype; HvCallEvent_ackLpEvent(event); } diff --git a/arch/powerpc/platforms/maple/setup.c b/arch/powerpc/platforms/maple/setup.c index dd73e38bfb7..a1cb4d23672 100644 --- a/arch/powerpc/platforms/maple/setup.c +++ b/arch/powerpc/platforms/maple/setup.c @@ -71,9 +71,6 @@ #define DBG(fmt...) #endif -extern void generic_find_legacy_serial_ports(u64 *physport, - unsigned int *default_speed); - static void maple_restart(char *cmd) { unsigned int maple_nvram_base; diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c index 3b1a9d4fcbc..89c4c363616 100644 --- a/arch/powerpc/platforms/powermac/setup.c +++ b/arch/powerpc/platforms/powermac/setup.c @@ -278,7 +278,7 @@ static void __init l2cr_init(void) } #endif -void __init pmac_setup_arch(void) +static void __init pmac_setup_arch(void) { struct device_node *cpu, *ic; int *fp; diff --git a/arch/powerpc/platforms/powermac/udbg_scc.c b/arch/powerpc/platforms/powermac/udbg_scc.c index e87d53acfb6..c4352a8db64 100644 --- a/arch/powerpc/platforms/powermac/udbg_scc.c +++ b/arch/powerpc/platforms/powermac/udbg_scc.c @@ -153,7 +153,7 @@ static void udbg_real_scc_putc(char c) udbg_real_scc_putc('\r'); } -void udbg_init_pmac_realmode(void) +void __init udbg_init_pmac_realmode(void) { sccc = (volatile u8 __iomem *)0x80013020ul; sccd = (volatile u8 __iomem *)0x80013030ul; diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile index 6accdd15550..61616d14407 100644 --- a/arch/powerpc/platforms/pseries/Makefile +++ b/arch/powerpc/platforms/pseries/Makefile @@ -4,7 +4,7 @@ obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_IBMVIO) += vio.o obj-$(CONFIG_XICS) += xics.o obj-$(CONFIG_SCANLOG) += scanlog.o -obj-$(CONFIG_EEH) += eeh.o eeh_event.o +obj-$(CONFIG_EEH) += eeh.o eeh_cache.o eeh_driver.o eeh_event.o obj-$(CONFIG_HVC_CONSOLE) += hvconsole.o obj-$(CONFIG_HVCS) += hvcserver.o diff --git a/arch/powerpc/platforms/pseries/eeh.c b/arch/powerpc/platforms/pseries/eeh.c index 7fbfd16d72b..83578313ee7 100644 --- a/arch/powerpc/platforms/pseries/eeh.c +++ b/arch/powerpc/platforms/pseries/eeh.c @@ -76,15 +76,14 @@ */ #define EEH_MAX_FAILS 100000 -/* Misc forward declaraions */ -static void eeh_save_bars(struct pci_dev * pdev, struct pci_dn *pdn); - /* RTAS tokens */ static int ibm_set_eeh_option; static int ibm_set_slot_reset; static int ibm_read_slot_reset_state; static int ibm_read_slot_reset_state2; static int ibm_slot_error_detail; +static int ibm_get_config_addr_info; +static int ibm_configure_bridge; int eeh_subsystem_enabled; EXPORT_SYMBOL(eeh_subsystem_enabled); @@ -98,308 +97,23 @@ static DEFINE_SPINLOCK(slot_errbuf_lock); static int eeh_error_buf_size; /* System monitoring statistics */ -static DEFINE_PER_CPU(unsigned long, no_device); -static DEFINE_PER_CPU(unsigned long, no_dn); -static DEFINE_PER_CPU(unsigned long, no_cfg_addr); -static DEFINE_PER_CPU(unsigned long, ignored_check); -static DEFINE_PER_CPU(unsigned long, total_mmio_ffs); -static DEFINE_PER_CPU(unsigned long, false_positives); -static DEFINE_PER_CPU(unsigned long, ignored_failures); -static DEFINE_PER_CPU(unsigned long, slot_resets); - -/** - * The pci address cache subsystem. This subsystem places - * PCI device address resources into a red-black tree, sorted - * according to the address range, so that given only an i/o - * address, the corresponding PCI device can be **quickly** - * found. It is safe to perform an address lookup in an interrupt - * context; this ability is an important feature. - * - * Currently, the only customer of this code is the EEH subsystem; - * thus, this code has been somewhat tailored to suit EEH better. - * In particular, the cache does *not* hold the addresses of devices - * for which EEH is not enabled. - * - * (Implementation Note: The RB tree seems to be better/faster - * than any hash algo I could think of for this problem, even - * with the penalty of slow pointer chases for d-cache misses). - */ -struct pci_io_addr_range -{ - struct rb_node rb_node; - unsigned long addr_lo; - unsigned long addr_hi; - struct pci_dev *pcidev; - unsigned int flags; -}; - -static struct pci_io_addr_cache -{ - struct rb_root rb_root; - spinlock_t piar_lock; -} pci_io_addr_cache_root; - -static inline struct pci_dev *__pci_get_device_by_addr(unsigned long addr) -{ - struct rb_node *n = pci_io_addr_cache_root.rb_root.rb_node; - - while (n) { - struct pci_io_addr_range *piar; - piar = rb_entry(n, struct pci_io_addr_range, rb_node); - - if (addr < piar->addr_lo) { - n = n->rb_left; - } else { - if (addr > piar->addr_hi) { - n = n->rb_right; - } else { - pci_dev_get(piar->pcidev); - return piar->pcidev; - } - } - } - - return NULL; -} - -/** - * pci_get_device_by_addr - Get device, given only address - * @addr: mmio (PIO) phys address or i/o port number - * - * Given an mmio phys address, or a port number, find a pci device - * that implements this address. Be sure to pci_dev_put the device - * when finished. I/O port numbers are assumed to be offset - * from zero (that is, they do *not* have pci_io_addr added in). - * It is safe to call this function within an interrupt. - */ -static struct pci_dev *pci_get_device_by_addr(unsigned long addr) -{ - struct pci_dev *dev; - unsigned long flags; - - spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags); - dev = __pci_get_device_by_addr(addr); - spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags); - return dev; -} - -#ifdef DEBUG -/* - * Handy-dandy debug print routine, does nothing more - * than print out the contents of our addr cache. - */ -static void pci_addr_cache_print(struct pci_io_addr_cache *cache) -{ - struct rb_node *n; - int cnt = 0; - - n = rb_first(&cache->rb_root); - while (n) { - struct pci_io_addr_range *piar; - piar = rb_entry(n, struct pci_io_addr_range, rb_node); - printk(KERN_DEBUG "PCI: %s addr range %d [%lx-%lx]: %s\n", - (piar->flags & IORESOURCE_IO) ? "i/o" : "mem", cnt, - piar->addr_lo, piar->addr_hi, pci_name(piar->pcidev)); - cnt++; - n = rb_next(n); - } -} -#endif - -/* Insert address range into the rb tree. */ -static struct pci_io_addr_range * -pci_addr_cache_insert(struct pci_dev *dev, unsigned long alo, - unsigned long ahi, unsigned int flags) -{ - struct rb_node **p = &pci_io_addr_cache_root.rb_root.rb_node; - struct rb_node *parent = NULL; - struct pci_io_addr_range *piar; - - /* Walk tree, find a place to insert into tree */ - while (*p) { - parent = *p; - piar = rb_entry(parent, struct pci_io_addr_range, rb_node); - if (ahi < piar->addr_lo) { - p = &parent->rb_left; - } else if (alo > piar->addr_hi) { - p = &parent->rb_right; - } else { - if (dev != piar->pcidev || - alo != piar->addr_lo || ahi != piar->addr_hi) { - printk(KERN_WARNING "PIAR: overlapping address range\n"); - } - return piar; - } - } - piar = (struct pci_io_addr_range *)kmalloc(sizeof(struct pci_io_addr_range), GFP_ATOMIC); - if (!piar) - return NULL; - - piar->addr_lo = alo; - piar->addr_hi = ahi; - piar->pcidev = dev; - piar->flags = flags; - -#ifdef DEBUG - printk(KERN_DEBUG "PIAR: insert range=[%lx:%lx] dev=%s\n", - alo, ahi, pci_name (dev)); -#endif - - rb_link_node(&piar->rb_node, parent, p); - rb_insert_color(&piar->rb_node, &pci_io_addr_cache_root.rb_root); - - return piar; -} - -static void __pci_addr_cache_insert_device(struct pci_dev *dev) -{ - struct device_node *dn; - struct pci_dn *pdn; - int i; - int inserted = 0; - - dn = pci_device_to_OF_node(dev); - if (!dn) { - printk(KERN_WARNING "PCI: no pci dn found for dev=%s\n", pci_name(dev)); - return; - } - - /* Skip any devices for which EEH is not enabled. */ - pdn = PCI_DN(dn); - if (!(pdn->eeh_mode & EEH_MODE_SUPPORTED) || - pdn->eeh_mode & EEH_MODE_NOCHECK) { -#ifdef DEBUG - printk(KERN_INFO "PCI: skip building address cache for=%s - %s\n", - pci_name(dev), pdn->node->full_name); -#endif - return; - } - - /* The cache holds a reference to the device... */ - pci_dev_get(dev); - - /* Walk resources on this device, poke them into the tree */ - for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) { - unsigned long start = pci_resource_start(dev,i); - unsigned long end = pci_resource_end(dev,i); - unsigned int flags = pci_resource_flags(dev,i); - - /* We are interested only bus addresses, not dma or other stuff */ - if (0 == (flags & (IORESOURCE_IO | IORESOURCE_MEM))) - continue; - if (start == 0 || ~start == 0 || end == 0 || ~end == 0) - continue; - pci_addr_cache_insert(dev, start, end, flags); - inserted = 1; - } - - /* If there was nothing to add, the cache has no reference... */ - if (!inserted) - pci_dev_put(dev); -} - -/** - * pci_addr_cache_insert_device - Add a device to the address cache - * @dev: PCI device whose I/O addresses we are interested in. - * - * In order to support the fast lookup of devices based on addresses, - * we maintain a cache of devices that can be quickly searched. - * This routine adds a device to that cache. - */ -static void pci_addr_cache_insert_device(struct pci_dev *dev) -{ - unsigned long flags; - - spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags); - __pci_addr_cache_insert_device(dev); - spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags); -} - -static inline void __pci_addr_cache_remove_device(struct pci_dev *dev) -{ - struct rb_node *n; - int removed = 0; - -restart: - n = rb_first(&pci_io_addr_cache_root.rb_root); - while (n) { - struct pci_io_addr_range *piar; - piar = rb_entry(n, struct pci_io_addr_range, rb_node); - - if (piar->pcidev == dev) { - rb_erase(n, &pci_io_addr_cache_root.rb_root); - removed = 1; - kfree(piar); - goto restart; - } - n = rb_next(n); - } - - /* The cache no longer holds its reference to this device... */ - if (removed) - pci_dev_put(dev); -} - -/** - * pci_addr_cache_remove_device - remove pci device from addr cache - * @dev: device to remove - * - * Remove a device from the addr-cache tree. - * This is potentially expensive, since it will walk - * the tree multiple times (once per resource). - * But so what; device removal doesn't need to be that fast. - */ -static void pci_addr_cache_remove_device(struct pci_dev *dev) -{ - unsigned long flags; - - spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags); - __pci_addr_cache_remove_device(dev); - spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags); -} - -/** - * pci_addr_cache_build - Build a cache of I/O addresses - * - * Build a cache of pci i/o addresses. This cache will be used to - * find the pci device that corresponds to a given address. - * This routine scans all pci busses to build the cache. - * Must be run late in boot process, after the pci controllers - * have been scaned for devices (after all device resources are known). - */ -void __init pci_addr_cache_build(void) -{ - struct device_node *dn; - struct pci_dev *dev = NULL; - - if (!eeh_subsystem_enabled) - return; - - spin_lock_init(&pci_io_addr_cache_root.piar_lock); - - while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { - /* Ignore PCI bridges ( XXX why ??) */ - if ((dev->class >> 16) == PCI_BASE_CLASS_BRIDGE) { - continue; - } - pci_addr_cache_insert_device(dev); - - /* Save the BAR's; firmware doesn't restore these after EEH reset */ - dn = pci_device_to_OF_node(dev); - eeh_save_bars(dev, PCI_DN(dn)); - } +static unsigned long no_device; +static unsigned long no_dn; +static unsigned long no_cfg_addr; +static unsigned long ignored_check; +static unsigned long total_mmio_ffs; +static unsigned long false_positives; +static unsigned long ignored_failures; +static unsigned long slot_resets; -#ifdef DEBUG - /* Verify tree built up above, echo back the list of addrs. */ - pci_addr_cache_print(&pci_io_addr_cache_root); -#endif -} +#define IS_BRIDGE(class_code) (((class_code)<<16) == PCI_BASE_CLASS_BRIDGE) /* --------------------------------------------------------------- */ -/* Above lies the PCI Address Cache. Below lies the EEH event infrastructure */ +/* Below lies the EEH event infrastructure */ void eeh_slot_error_detail (struct pci_dn *pdn, int severity) { + int config_addr; unsigned long flags; int rc; @@ -407,8 +121,13 @@ void eeh_slot_error_detail (struct pci_dn *pdn, int severity) spin_lock_irqsave(&slot_errbuf_lock, flags); memset(slot_errbuf, 0, eeh_error_buf_size); + /* Use PE configuration address, if present */ + config_addr = pdn->eeh_config_addr; + if (pdn->eeh_pe_config_addr) + config_addr = pdn->eeh_pe_config_addr; + rc = rtas_call(ibm_slot_error_detail, - 8, 1, NULL, pdn->eeh_config_addr, + 8, 1, NULL, config_addr, BUID_HI(pdn->phb->buid), BUID_LO(pdn->phb->buid), NULL, 0, virt_to_phys(slot_errbuf), @@ -428,6 +147,7 @@ void eeh_slot_error_detail (struct pci_dn *pdn, int severity) static int read_slot_reset_state(struct pci_dn *pdn, int rets[]) { int token, outputs; + int config_addr; if (ibm_read_slot_reset_state2 != RTAS_UNKNOWN_SERVICE) { token = ibm_read_slot_reset_state2; @@ -438,7 +158,12 @@ static int read_slot_reset_state(struct pci_dn *pdn, int rets[]) outputs = 3; } - return rtas_call(token, 3, outputs, rets, pdn->eeh_config_addr, + /* Use PE configuration address, if present */ + config_addr = pdn->eeh_config_addr; + if (pdn->eeh_pe_config_addr) + config_addr = pdn->eeh_pe_config_addr; + + return rtas_call(token, 3, outputs, rets, config_addr, BUID_HI(pdn->phb->buid), BUID_LO(pdn->phb->buid)); } @@ -462,7 +187,7 @@ static inline unsigned long eeh_token_to_phys(unsigned long token) /** * Return the "partitionable endpoint" (pe) under which this device lies */ -static struct device_node * find_device_pe(struct device_node *dn) +struct device_node * find_device_pe(struct device_node *dn) { while ((dn->parent) && PCI_DN(dn->parent) && (PCI_DN(dn->parent)->eeh_mode & EEH_MODE_SUPPORTED)) { @@ -483,8 +208,14 @@ static void __eeh_mark_slot (struct device_node *dn, int mode_flag) { while (dn) { if (PCI_DN(dn)) { + /* Mark the pci device driver too */ + struct pci_dev *dev = PCI_DN(dn)->pcidev; + PCI_DN(dn)->eeh_mode |= mode_flag; + if (dev && dev->driver) + dev->error_state = pci_channel_io_frozen; + if (dn->child) __eeh_mark_slot (dn->child, mode_flag); } @@ -495,6 +226,11 @@ static void __eeh_mark_slot (struct device_node *dn, int mode_flag) void eeh_mark_slot (struct device_node *dn, int mode_flag) { dn = find_device_pe (dn); + + /* Back up one, since config addrs might be shared */ + if (PCI_DN(dn) && PCI_DN(dn)->eeh_pe_config_addr) + dn = dn->parent; + PCI_DN(dn)->eeh_mode |= mode_flag; __eeh_mark_slot (dn->child, mode_flag); } @@ -516,7 +252,13 @@ void eeh_clear_slot (struct device_node *dn, int mode_flag) { unsigned long flags; spin_lock_irqsave(&confirm_error_lock, flags); + dn = find_device_pe (dn); + + /* Back up one, since config addrs might be shared */ + if (PCI_DN(dn) && PCI_DN(dn)->eeh_pe_config_addr) + dn = dn->parent; + PCI_DN(dn)->eeh_mode &= ~mode_flag; PCI_DN(dn)->eeh_check_count = 0; __eeh_clear_slot (dn->child, mode_flag); @@ -544,15 +286,16 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) int rets[3]; unsigned long flags; struct pci_dn *pdn; + enum pci_channel_state state; int rc = 0; - __get_cpu_var(total_mmio_ffs)++; + total_mmio_ffs++; if (!eeh_subsystem_enabled) return 0; if (!dn) { - __get_cpu_var(no_dn)++; + no_dn++; return 0; } pdn = PCI_DN(dn); @@ -560,7 +303,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) /* Access to IO BARs might get this far and still not want checking. */ if (!(pdn->eeh_mode & EEH_MODE_SUPPORTED) || pdn->eeh_mode & EEH_MODE_NOCHECK) { - __get_cpu_var(ignored_check)++; + ignored_check++; #ifdef DEBUG printk ("EEH:ignored check (%x) for %s %s\n", pdn->eeh_mode, pci_name (dev), dn->full_name); @@ -568,8 +311,8 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) return 0; } - if (!pdn->eeh_config_addr) { - __get_cpu_var(no_cfg_addr)++; + if (!pdn->eeh_config_addr && !pdn->eeh_pe_config_addr) { + no_cfg_addr++; return 0; } @@ -611,7 +354,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) if (ret != 0) { printk(KERN_WARNING "EEH: read_slot_reset_state() failed; rc=%d dn=%s\n", ret, dn->full_name); - __get_cpu_var(false_positives)++; + false_positives++; rc = 0; goto dn_unlock; } @@ -620,14 +363,14 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) if (rets[1] != 1) { printk(KERN_WARNING "EEH: event on unsupported device, rc=%d dn=%s\n", ret, dn->full_name); - __get_cpu_var(false_positives)++; + false_positives++; rc = 0; goto dn_unlock; } /* If not the kind of error we know about, punt. */ if (rets[0] != 2 && rets[0] != 4 && rets[0] != 5) { - __get_cpu_var(false_positives)++; + false_positives++; rc = 0; goto dn_unlock; } @@ -635,12 +378,12 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) /* Note that config-io to empty slots may fail; * we recognize empty because they don't have children. */ if ((rets[0] == 5) && (dn->child == NULL)) { - __get_cpu_var(false_positives)++; + false_positives++; rc = 0; goto dn_unlock; } - __get_cpu_var(slot_resets)++; + slot_resets++; /* Avoid repeated reports of this failure, including problems * with other functions on this device, and functions under @@ -648,8 +391,13 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) eeh_mark_slot (dn, EEH_MODE_ISOLATED); spin_unlock_irqrestore(&confirm_error_lock, flags); - eeh_send_failure_event (dn, dev, rets[0], rets[2]); - + state = pci_channel_io_normal; + if ((rets[0] == 2) || (rets[0] == 4)) + state = pci_channel_io_frozen; + if (rets[0] == 5) + state = pci_channel_io_perm_failure; + eeh_send_failure_event (dn, dev, state, rets[2]); + /* Most EEH events are due to device driver bugs. Having * a stack trace will help the device-driver authors figure * out what happened. So print that out. */ @@ -685,7 +433,7 @@ unsigned long eeh_check_failure(const volatile void __iomem *token, unsigned lon addr = eeh_token_to_phys((unsigned long __force) token); dev = pci_get_device_by_addr(addr); if (!dev) { - __get_cpu_var(no_device)++; + no_device++; return val; } @@ -716,11 +464,16 @@ eeh_slot_availability(struct pci_dn *pdn) if (rc) return rc; if (rets[1] == 0) return -1; /* EEH is not supported */ - if (rets[0] == 0) return 0; /* Oll Korrect */ + if (rets[0] == 0) return 0; /* Oll Korrect */ if (rets[0] == 5) { if (rets[2] == 0) return -1; /* permanently unavailable */ return rets[2]; /* number of millisecs to wait */ } + if (rets[0] == 1) + return 250; + + printk (KERN_ERR "EEH: Slot unavailable: rc=%d, rets=%d %d %d\n", + rc, rets[0], rets[1], rets[2]); return -1; } @@ -737,6 +490,7 @@ eeh_slot_availability(struct pci_dn *pdn) static void rtas_pci_slot_reset(struct pci_dn *pdn, int state) { + int config_addr; int rc; BUG_ON (pdn==NULL); @@ -747,8 +501,13 @@ rtas_pci_slot_reset(struct pci_dn *pdn, int state) return; } + /* Use PE configuration address, if present */ + config_addr = pdn->eeh_config_addr; + if (pdn->eeh_pe_config_addr) + config_addr = pdn->eeh_pe_config_addr; + rc = rtas_call(ibm_set_slot_reset,4,1, NULL, - pdn->eeh_config_addr, + config_addr, BUID_HI(pdn->phb->buid), BUID_LO(pdn->phb->buid), state); @@ -761,9 +520,11 @@ rtas_pci_slot_reset(struct pci_dn *pdn, int state) /** rtas_set_slot_reset -- assert the pci #RST line for 1/4 second * dn -- device node to be reset. + * + * Return 0 if success, else a non-zero value. */ -void +int rtas_set_slot_reset(struct pci_dn *pdn) { int i, rc; @@ -793,10 +554,21 @@ rtas_set_slot_reset(struct pci_dn *pdn) * ready to be used; if not, wait for recovery. */ for (i=0; i<10; i++) { rc = eeh_slot_availability (pdn); - if (rc <= 0) break; + if (rc < 0) + printk (KERN_ERR "EEH: failed (%d) to reset slot %s\n", rc, pdn->node->full_name); + if (rc == 0) + return 0; + if (rc < 0) + return -1; msleep (rc+100); } + + rc = eeh_slot_availability (pdn); + if (rc) + printk (KERN_ERR "EEH: timeout resetting slot %s\n", pdn->node->full_name); + + return rc; } /* ------------------------------------------------------- */ @@ -851,7 +623,7 @@ void eeh_restore_bars(struct pci_dn *pdn) if (!pdn) return; - if (! pdn->eeh_is_bridge) + if ((pdn->eeh_mode & EEH_MODE_SUPPORTED) && !IS_BRIDGE(pdn->class_code)) __restore_bars (pdn); dn = pdn->node->child; @@ -869,30 +641,30 @@ void eeh_restore_bars(struct pci_dn *pdn) * PCI devices are added individuallly; but, for the restore, * an entire slot is reset at a time. */ -static void eeh_save_bars(struct pci_dev * pdev, struct pci_dn *pdn) +static void eeh_save_bars(struct pci_dn *pdn) { int i; - if (!pdev || !pdn ) + if (!pdn ) return; for (i = 0; i < 16; i++) - pci_read_config_dword(pdev, i * 4, &pdn->config_space[i]); - - if (pdev->hdr_type == PCI_HEADER_TYPE_BRIDGE) - pdn->eeh_is_bridge = 1; + rtas_read_config(pdn, i * 4, 4, &pdn->config_space[i]); } void rtas_configure_bridge(struct pci_dn *pdn) { - int token = rtas_token ("ibm,configure-bridge"); + int config_addr; int rc; - if (token == RTAS_UNKNOWN_SERVICE) - return; - rc = rtas_call(token,3,1, NULL, - pdn->eeh_config_addr, + /* Use PE configuration address, if present */ + config_addr = pdn->eeh_config_addr; + if (pdn->eeh_pe_config_addr) + config_addr = pdn->eeh_pe_config_addr; + + rc = rtas_call(ibm_configure_bridge,3,1, NULL, + config_addr, BUID_HI(pdn->phb->buid), BUID_LO(pdn->phb->buid)); if (rc) { @@ -927,6 +699,7 @@ static void *early_enable_eeh(struct device_node *dn, void *data) int enable; struct pci_dn *pdn = PCI_DN(dn); + pdn->class_code = 0; pdn->eeh_mode = 0; pdn->eeh_check_count = 0; pdn->eeh_freeze_count = 0; @@ -943,6 +716,7 @@ static void *early_enable_eeh(struct device_node *dn, void *data) pdn->eeh_mode |= EEH_MODE_NOCHECK; return NULL; } + pdn->class_code = *class_code; /* * Now decide if we are going to "Disable" EEH checking @@ -953,8 +727,10 @@ static void *early_enable_eeh(struct device_node *dn, void *data) * But there are a few cases like display devices that make sense. */ enable = 1; /* i.e. we will do checking */ +#if 0 if ((*class_code >> 16) == PCI_BASE_CLASS_DISPLAY) enable = 0; +#endif if (!enable) pdn->eeh_mode |= EEH_MODE_NOCHECK; @@ -973,8 +749,22 @@ static void *early_enable_eeh(struct device_node *dn, void *data) eeh_subsystem_enabled = 1; pdn->eeh_mode |= EEH_MODE_SUPPORTED; pdn->eeh_config_addr = regs[0]; + + /* If the newer, better, ibm,get-config-addr-info is supported, + * then use that instead. */ + pdn->eeh_pe_config_addr = 0; + if (ibm_get_config_addr_info != RTAS_UNKNOWN_SERVICE) { + unsigned int rets[2]; + ret = rtas_call (ibm_get_config_addr_info, 4, 2, rets, + pdn->eeh_config_addr, + info->buid_hi, info->buid_lo, + 0); + if (ret == 0) + pdn->eeh_pe_config_addr = rets[0]; + } #ifdef DEBUG - printk(KERN_DEBUG "EEH: %s: eeh enabled\n", dn->full_name); + printk(KERN_DEBUG "EEH: %s: eeh enabled, config=%x pe_config=%x\n", + dn->full_name, pdn->eeh_config_addr, pdn->eeh_pe_config_addr); #endif } else { @@ -993,6 +783,7 @@ static void *early_enable_eeh(struct device_node *dn, void *data) dn->full_name); } + eeh_save_bars(pdn); return NULL; } @@ -1026,6 +817,8 @@ void __init eeh_init(void) ibm_read_slot_reset_state2 = rtas_token("ibm,read-slot-reset-state2"); ibm_read_slot_reset_state = rtas_token("ibm,read-slot-reset-state"); ibm_slot_error_detail = rtas_token("ibm,slot-error-detail"); + ibm_get_config_addr_info = rtas_token("ibm,get-config-addr-info"); + ibm_configure_bridge = rtas_token ("ibm,configure-bridge"); if (ibm_set_eeh_option == RTAS_UNKNOWN_SERVICE) return; @@ -1080,12 +873,10 @@ void eeh_add_device_early(struct device_node *dn) if (!dn || !PCI_DN(dn)) return; phb = PCI_DN(dn)->phb; - if (NULL == phb || 0 == phb->buid) { - printk(KERN_WARNING "EEH: Expected buid but found none for %s\n", - dn->full_name); - dump_stack(); + + /* USB Bus children of PCI devices will not have BUID's */ + if (NULL == phb || 0 == phb->buid) return; - } info.buid_hi = BUID_HI(phb->buid); info.buid_lo = BUID_LO(phb->buid); @@ -1127,7 +918,6 @@ void eeh_add_device_late(struct pci_dev *dev) pdn->pcidev = dev; pci_addr_cache_insert_device (dev); - eeh_save_bars(dev, pdn); } EXPORT_SYMBOL_GPL(eeh_add_device_late); @@ -1175,25 +965,9 @@ EXPORT_SYMBOL_GPL(eeh_remove_bus_device); static int proc_eeh_show(struct seq_file *m, void *v) { - unsigned int cpu; - unsigned long ffs = 0, positives = 0, failures = 0; - unsigned long resets = 0; - unsigned long no_dev = 0, no_dn = 0, no_cfg = 0, no_check = 0; - - for_each_cpu(cpu) { - ffs += per_cpu(total_mmio_ffs, cpu); - positives += per_cpu(false_positives, cpu); - failures += per_cpu(ignored_failures, cpu); - resets += per_cpu(slot_resets, cpu); - no_dev += per_cpu(no_device, cpu); - no_dn += per_cpu(no_dn, cpu); - no_cfg += per_cpu(no_cfg_addr, cpu); - no_check += per_cpu(ignored_check, cpu); - } - if (0 == eeh_subsystem_enabled) { seq_printf(m, "EEH Subsystem is globally disabled\n"); - seq_printf(m, "eeh_total_mmio_ffs=%ld\n", ffs); + seq_printf(m, "eeh_total_mmio_ffs=%ld\n", total_mmio_ffs); } else { seq_printf(m, "EEH Subsystem is enabled\n"); seq_printf(m, @@ -1205,8 +979,10 @@ static int proc_eeh_show(struct seq_file *m, void *v) "eeh_false_positives=%ld\n" "eeh_ignored_failures=%ld\n" "eeh_slot_resets=%ld\n", - no_dev, no_dn, no_cfg, no_check, - ffs, positives, failures, resets); + no_device, no_dn, no_cfg_addr, + ignored_check, total_mmio_ffs, + false_positives, ignored_failures, + slot_resets); } return 0; diff --git a/arch/powerpc/platforms/pseries/eeh_cache.c b/arch/powerpc/platforms/pseries/eeh_cache.c new file mode 100644 index 00000000000..d4a402c5866 --- /dev/null +++ b/arch/powerpc/platforms/pseries/eeh_cache.c @@ -0,0 +1,316 @@ +/* + * eeh_cache.c + * PCI address cache; allows the lookup of PCI devices based on I/O address + * + * Copyright (C) 2004 Linas Vepstas <linas@austin.ibm.com> IBM Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include <linux/list.h> +#include <linux/pci.h> +#include <linux/rbtree.h> +#include <linux/spinlock.h> +#include <asm/atomic.h> +#include <asm/pci-bridge.h> +#include <asm/ppc-pci.h> + +#undef DEBUG + +/** + * The pci address cache subsystem. This subsystem places + * PCI device address resources into a red-black tree, sorted + * according to the address range, so that given only an i/o + * address, the corresponding PCI device can be **quickly** + * found. It is safe to perform an address lookup in an interrupt + * context; this ability is an important feature. + * + * Currently, the only customer of this code is the EEH subsystem; + * thus, this code has been somewhat tailored to suit EEH better. + * In particular, the cache does *not* hold the addresses of devices + * for which EEH is not enabled. + * + * (Implementation Note: The RB tree seems to be better/faster + * than any hash algo I could think of for this problem, even + * with the penalty of slow pointer chases for d-cache misses). + */ +struct pci_io_addr_range +{ + struct rb_node rb_node; + unsigned long addr_lo; + unsigned long addr_hi; + struct pci_dev *pcidev; + unsigned int flags; +}; + +static struct pci_io_addr_cache +{ + struct rb_root rb_root; + spinlock_t piar_lock; +} pci_io_addr_cache_root; + +static inline struct pci_dev *__pci_get_device_by_addr(unsigned long addr) +{ + struct rb_node *n = pci_io_addr_cache_root.rb_root.rb_node; + + while (n) { + struct pci_io_addr_range *piar; + piar = rb_entry(n, struct pci_io_addr_range, rb_node); + + if (addr < piar->addr_lo) { + n = n->rb_left; + } else { + if (addr > piar->addr_hi) { + n = n->rb_right; + } else { + pci_dev_get(piar->pcidev); + return piar->pcidev; + } + } + } + + return NULL; +} + +/** + * pci_get_device_by_addr - Get device, given only address + * @addr: mmio (PIO) phys address or i/o port number + * + * Given an mmio phys address, or a port number, find a pci device + * that implements this address. Be sure to pci_dev_put the device + * when finished. I/O port numbers are assumed to be offset + * from zero (that is, they do *not* have pci_io_addr added in). + * It is safe to call this function within an interrupt. + */ +struct pci_dev *pci_get_device_by_addr(unsigned long addr) +{ + struct pci_dev *dev; + unsigned long flags; + + spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags); + dev = __pci_get_device_by_addr(addr); + spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags); + return dev; +} + +#ifdef DEBUG +/* + * Handy-dandy debug print routine, does nothing more + * than print out the contents of our addr cache. + */ +static void pci_addr_cache_print(struct pci_io_addr_cache *cache) +{ + struct rb_node *n; + int cnt = 0; + + n = rb_first(&cache->rb_root); + while (n) { + struct pci_io_addr_range *piar; + piar = rb_entry(n, struct pci_io_addr_range, rb_node); + printk(KERN_DEBUG "PCI: %s addr range %d [%lx-%lx]: %s\n", + (piar->flags & IORESOURCE_IO) ? "i/o" : "mem", cnt, + piar->addr_lo, piar->addr_hi, pci_name(piar->pcidev)); + cnt++; + n = rb_next(n); + } +} +#endif + +/* Insert address range into the rb tree. */ +static struct pci_io_addr_range * +pci_addr_cache_insert(struct pci_dev *dev, unsigned long alo, + unsigned long ahi, unsigned int flags) +{ + struct rb_node **p = &pci_io_addr_cache_root.rb_root.rb_node; + struct rb_node *parent = NULL; + struct pci_io_addr_range *piar; + + /* Walk tree, find a place to insert into tree */ + while (*p) { + parent = *p; + piar = rb_entry(parent, struct pci_io_addr_range, rb_node); + if (ahi < piar->addr_lo) { + p = &parent->rb_left; + } else if (alo > piar->addr_hi) { + p = &parent->rb_right; + } else { + if (dev != piar->pcidev || + alo != piar->addr_lo || ahi != piar->addr_hi) { + printk(KERN_WARNING "PIAR: overlapping address range\n"); + } + return piar; + } + } + piar = (struct pci_io_addr_range *)kmalloc(sizeof(struct pci_io_addr_range), GFP_ATOMIC); + if (!piar) + return NULL; + + piar->addr_lo = alo; + piar->addr_hi = ahi; + piar->pcidev = dev; + piar->flags = flags; + +#ifdef DEBUG + printk(KERN_DEBUG "PIAR: insert range=[%lx:%lx] dev=%s\n", + alo, ahi, pci_name (dev)); +#endif + + rb_link_node(&piar->rb_node, parent, p); + rb_insert_color(&piar->rb_node, &pci_io_addr_cache_root.rb_root); + + return piar; +} + +static void __pci_addr_cache_insert_device(struct pci_dev *dev) +{ + struct device_node *dn; + struct pci_dn *pdn; + int i; + int inserted = 0; + + dn = pci_device_to_OF_node(dev); + if (!dn) { + printk(KERN_WARNING "PCI: no pci dn found for dev=%s\n", pci_name(dev)); + return; + } + + /* Skip any devices for which EEH is not enabled. */ + pdn = PCI_DN(dn); + if (!(pdn->eeh_mode & EEH_MODE_SUPPORTED) || + pdn->eeh_mode & EEH_MODE_NOCHECK) { +#ifdef DEBUG + printk(KERN_INFO "PCI: skip building address cache for=%s - %s\n", + pci_name(dev), pdn->node->full_name); +#endif + return; + } + + /* The cache holds a reference to the device... */ + pci_dev_get(dev); + + /* Walk resources on this device, poke them into the tree */ + for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) { + unsigned long start = pci_resource_start(dev,i); + unsigned long end = pci_resource_end(dev,i); + unsigned int flags = pci_resource_flags(dev,i); + + /* We are interested only bus addresses, not dma or other stuff */ + if (0 == (flags & (IORESOURCE_IO | IORESOURCE_MEM))) + continue; + if (start == 0 || ~start == 0 || end == 0 || ~end == 0) + continue; + pci_addr_cache_insert(dev, start, end, flags); + inserted = 1; + } + + /* If there was nothing to add, the cache has no reference... */ + if (!inserted) + pci_dev_put(dev); +} + +/** + * pci_addr_cache_insert_device - Add a device to the address cache + * @dev: PCI device whose I/O addresses we are interested in. + * + * In order to support the fast lookup of devices based on addresses, + * we maintain a cache of devices that can be quickly searched. + * This routine adds a device to that cache. + */ +void pci_addr_cache_insert_device(struct pci_dev *dev) +{ + unsigned long flags; + + spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags); + __pci_addr_cache_insert_device(dev); + spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags); +} + +static inline void __pci_addr_cache_remove_device(struct pci_dev *dev) +{ + struct rb_node *n; + int removed = 0; + +restart: + n = rb_first(&pci_io_addr_cache_root.rb_root); + while (n) { + struct pci_io_addr_range *piar; + piar = rb_entry(n, struct pci_io_addr_range, rb_node); + + if (piar->pcidev == dev) { + rb_erase(n, &pci_io_addr_cache_root.rb_root); + removed = 1; + kfree(piar); + goto restart; + } + n = rb_next(n); + } + + /* The cache no longer holds its reference to this device... */ + if (removed) + pci_dev_put(dev); +} + +/** + * pci_addr_cache_remove_device - remove pci device from addr cache + * @dev: device to remove + * + * Remove a device from the addr-cache tree. + * This is potentially expensive, since it will walk + * the tree multiple times (once per resource). + * But so what; device removal doesn't need to be that fast. + */ +void pci_addr_cache_remove_device(struct pci_dev *dev) +{ + unsigned long flags; + + spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags); + __pci_addr_cache_remove_device(dev); + spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags); +} + +/** + * pci_addr_cache_build - Build a cache of I/O addresses + * + * Build a cache of pci i/o addresses. This cache will be used to + * find the pci device that corresponds to a given address. + * This routine scans all pci busses to build the cache. + * Must be run late in boot process, after the pci controllers + * have been scaned for devices (after all device resources are known). + */ +void __init pci_addr_cache_build(void) +{ + struct device_node *dn; + struct pci_dev *dev = NULL; + + spin_lock_init(&pci_io_addr_cache_root.piar_lock); + + while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { + /* Ignore PCI bridges */ + if ((dev->class >> 16) == PCI_BASE_CLASS_BRIDGE) + continue; + + pci_addr_cache_insert_device(dev); + + dn = pci_device_to_OF_node(dev); + pci_dev_get (dev); /* matching put is in eeh_remove_device() */ + PCI_DN(dn)->pcidev = dev; + } + +#ifdef DEBUG + /* Verify tree built up above, echo back the list of addrs. */ + pci_addr_cache_print(&pci_io_addr_cache_root); +#endif +} + diff --git a/arch/powerpc/platforms/pseries/eeh_driver.c b/arch/powerpc/platforms/pseries/eeh_driver.c new file mode 100644 index 00000000000..6373372932b --- /dev/null +++ b/arch/powerpc/platforms/pseries/eeh_driver.c @@ -0,0 +1,376 @@ +/* + * PCI Error Recovery Driver for RPA-compliant PPC64 platform. + * Copyright (C) 2004, 2005 Linas Vepstas <linas@linas.org> + * + * All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for more + * details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + * + * Send feedback to <linas@us.ibm.com> + * + */ +#include <linux/delay.h> +#include <linux/irq.h> +#include <linux/interrupt.h> +#include <linux/notifier.h> +#include <linux/pci.h> +#include <asm/eeh.h> +#include <asm/eeh_event.h> +#include <asm/ppc-pci.h> +#include <asm/pci-bridge.h> +#include <asm/prom.h> +#include <asm/rtas.h> + + +static inline const char * pcid_name (struct pci_dev *pdev) +{ + if (pdev->dev.driver) + return pdev->dev.driver->name; + return ""; +} + +#ifdef DEBUG +static void print_device_node_tree (struct pci_dn *pdn, int dent) +{ + int i; + if (!pdn) return; + for (i=0;i<dent; i++) + printk(" "); + printk("dn=%s mode=%x \tcfg_addr=%x pe_addr=%x \tfull=%s\n", + pdn->node->name, pdn->eeh_mode, pdn->eeh_config_addr, + pdn->eeh_pe_config_addr, pdn->node->full_name); + dent += 3; + struct device_node *pc = pdn->node->child; + while (pc) { + print_device_node_tree(PCI_DN(pc), dent); + pc = pc->sibling; + } +} +#endif + +/** + * irq_in_use - return true if this irq is being used + */ +static int irq_in_use(unsigned int irq) +{ + int rc = 0; + unsigned long flags; + struct irq_desc *desc = irq_desc + irq; + + spin_lock_irqsave(&desc->lock, flags); + if (desc->action) + rc = 1; + spin_unlock_irqrestore(&desc->lock, flags); + return rc; +} + +/* ------------------------------------------------------- */ +/** eeh_report_error - report an EEH error to each device, + * collect up and merge the device responses. + */ + +static void eeh_report_error(struct pci_dev *dev, void *userdata) +{ + enum pci_ers_result rc, *res = userdata; + struct pci_driver *driver = dev->driver; + + dev->error_state = pci_channel_io_frozen; + + if (!driver) + return; + + if (irq_in_use (dev->irq)) { + struct device_node *dn = pci_device_to_OF_node(dev); + PCI_DN(dn)->eeh_mode |= EEH_MODE_IRQ_DISABLED; + disable_irq_nosync(dev->irq); + } + if (!driver->err_handler) + return; + if (!driver->err_handler->error_detected) + return; + + rc = driver->err_handler->error_detected (dev, pci_channel_io_frozen); + if (*res == PCI_ERS_RESULT_NONE) *res = rc; + if (*res == PCI_ERS_RESULT_NEED_RESET) return; + if (*res == PCI_ERS_RESULT_DISCONNECT && + rc == PCI_ERS_RESULT_NEED_RESET) *res = rc; +} + +/** eeh_report_reset -- tell this device that the pci slot + * has been reset. + */ + +static void eeh_report_reset(struct pci_dev *dev, void *userdata) +{ + struct pci_driver *driver = dev->driver; + struct device_node *dn = pci_device_to_OF_node(dev); + + if (!driver) + return; + + if ((PCI_DN(dn)->eeh_mode) & EEH_MODE_IRQ_DISABLED) { + PCI_DN(dn)->eeh_mode &= ~EEH_MODE_IRQ_DISABLED; + enable_irq(dev->irq); + } + if (!driver->err_handler) + return; + if (!driver->err_handler->slot_reset) + return; + + driver->err_handler->slot_reset(dev); +} + +static void eeh_report_resume(struct pci_dev *dev, void *userdata) +{ + struct pci_driver *driver = dev->driver; + + dev->error_state = pci_channel_io_normal; + + if (!driver) + return; + if (!driver->err_handler) + return; + if (!driver->err_handler->resume) + return; + + driver->err_handler->resume(dev); +} + +static void eeh_report_failure(struct pci_dev *dev, void *userdata) +{ + struct pci_driver *driver = dev->driver; + + dev->error_state = pci_channel_io_perm_failure; + + if (!driver) + return; + + if (irq_in_use (dev->irq)) { + struct device_node *dn = pci_device_to_OF_node(dev); + PCI_DN(dn)->eeh_mode |= EEH_MODE_IRQ_DISABLED; + disable_irq_nosync(dev->irq); + } + if (!driver->err_handler) + return; + if (!driver->err_handler->error_detected) + return; + driver->err_handler->error_detected(dev, pci_channel_io_perm_failure); +} + +/* ------------------------------------------------------- */ +/** + * handle_eeh_events -- reset a PCI device after hard lockup. + * + * pSeries systems will isolate a PCI slot if the PCI-Host + * bridge detects address or data parity errors, DMA's + * occuring to wild addresses (which usually happen due to + * bugs in device drivers or in PCI adapter firmware). + * Slot isolations also occur if #SERR, #PERR or other misc + * PCI-related errors are detected. + * + * Recovery process consists of unplugging the device driver + * (which generated hotplug events to userspace), then issuing + * a PCI #RST to the device, then reconfiguring the PCI config + * space for all bridges & devices under this slot, and then + * finally restarting the device drivers (which cause a second + * set of hotplug events to go out to userspace). + */ + +/** + * eeh_reset_device() -- perform actual reset of a pci slot + * Args: bus: pointer to the pci bus structure corresponding + * to the isolated slot. A non-null value will + * cause all devices under the bus to be removed + * and then re-added. + * pe_dn: pointer to a "Partionable Endpoint" device node. + * This is the top-level structure on which pci + * bus resets can be performed. + */ + +static int eeh_reset_device (struct pci_dn *pe_dn, struct pci_bus *bus) +{ + int rc; + if (bus) + pcibios_remove_pci_devices(bus); + + /* Reset the pci controller. (Asserts RST#; resets config space). + * Reconfigure bridges and devices. Don't try to bring the system + * up if the reset failed for some reason. */ + rc = rtas_set_slot_reset(pe_dn); + if (rc) + return rc; + + /* New-style config addrs might be shared across multiple devices, + * Walk over all functions on this device */ + if (pe_dn->eeh_pe_config_addr) { + struct device_node *pe = pe_dn->node; + pe = pe->parent->child; + while (pe) { + struct pci_dn *ppe = PCI_DN(pe); + if (pe_dn->eeh_pe_config_addr == ppe->eeh_pe_config_addr) { + rtas_configure_bridge(ppe); + eeh_restore_bars(ppe); + } + pe = pe->sibling; + } + } else { + rtas_configure_bridge(pe_dn); + eeh_restore_bars(pe_dn); + } + + /* Give the system 5 seconds to finish running the user-space + * hotplug shutdown scripts, e.g. ifdown for ethernet. Yes, + * this is a hack, but if we don't do this, and try to bring + * the device up before the scripts have taken it down, + * potentially weird things happen. + */ + if (bus) { + ssleep (5); + pcibios_add_pci_devices(bus); + } + + return 0; +} + +/* The longest amount of time to wait for a pci device + * to come back on line, in seconds. + */ +#define MAX_WAIT_FOR_RECOVERY 15 + +void handle_eeh_events (struct eeh_event *event) +{ + struct device_node *frozen_dn; + struct pci_dn *frozen_pdn; + struct pci_bus *frozen_bus; + int rc = 0; + enum pci_ers_result result = PCI_ERS_RESULT_NONE; + + frozen_dn = find_device_pe(event->dn); + frozen_bus = pcibios_find_pci_bus(frozen_dn); + + if (!frozen_dn) { + printk(KERN_ERR "EEH: Error: Cannot find partition endpoint for %s\n", + pci_name(event->dev)); + return; + } + + /* There are two different styles for coming up with the PE. + * In the old style, it was the highest EEH-capable device + * which was always an EADS pci bridge. In the new style, + * there might not be any EADS bridges, and even when there are, + * the firmware marks them as "EEH incapable". So another + * two-step is needed to find the pci bus.. */ + if (!frozen_bus) + frozen_bus = pcibios_find_pci_bus (frozen_dn->parent); + + if (!frozen_bus) { + printk(KERN_ERR "EEH: Cannot find PCI bus for %s\n", + frozen_dn->full_name); + return; + } + +#if 0 + /* We may get "permanent failure" messages on empty slots. + * These are false alarms. Empty slots have no child dn. */ + if ((event->state == pci_channel_io_perm_failure) && (frozen_device == NULL)) + return; +#endif + + frozen_pdn = PCI_DN(frozen_dn); + frozen_pdn->eeh_freeze_count++; + + if (frozen_pdn->eeh_freeze_count > EEH_MAX_ALLOWED_FREEZES) + goto hard_fail; + + /* If the reset state is a '5' and the time to reset is 0 (infinity) + * or is more then 15 seconds, then mark this as a permanent failure. + */ + if ((event->state == pci_channel_io_perm_failure) && + ((event->time_unavail <= 0) || + (event->time_unavail > MAX_WAIT_FOR_RECOVERY*1000))) + goto hard_fail; + + eeh_slot_error_detail(frozen_pdn, 1 /* Temporary Error */); + printk(KERN_WARNING + "EEH: This PCI device has failed %d times since last reboot: %s - %s\n", + frozen_pdn->eeh_freeze_count, + pci_name (frozen_pdn->pcidev), + pcid_name(frozen_pdn->pcidev)); + + /* Walk the various device drivers attached to this slot through + * a reset sequence, giving each an opportunity to do what it needs + * to accomplish the reset. Each child gets a report of the + * status ... if any child can't handle the reset, then the entire + * slot is dlpar removed and added. + */ + pci_walk_bus(frozen_bus, eeh_report_error, &result); + + /* If all device drivers were EEH-unaware, then shut + * down all of the device drivers, and hope they + * go down willingly, without panicing the system. + */ + if (result == PCI_ERS_RESULT_NONE) { + rc = eeh_reset_device(frozen_pdn, frozen_bus); + if (rc) + goto hard_fail; + } + + /* If any device called out for a reset, then reset the slot */ + if (result == PCI_ERS_RESULT_NEED_RESET) { + rc = eeh_reset_device(frozen_pdn, NULL); + if (rc) + goto hard_fail; + pci_walk_bus(frozen_bus, eeh_report_reset, 0); + } + + /* If all devices reported they can proceed, the re-enable PIO */ + if (result == PCI_ERS_RESULT_CAN_RECOVER) { + /* XXX Not supported; we brute-force reset the device */ + rc = eeh_reset_device(frozen_pdn, NULL); + if (rc) + goto hard_fail; + pci_walk_bus(frozen_bus, eeh_report_reset, 0); + } + + /* Tell all device drivers that they can resume operations */ + pci_walk_bus(frozen_bus, eeh_report_resume, 0); + + return; + +hard_fail: + /* + * About 90% of all real-life EEH failures in the field + * are due to poorly seated PCI cards. Only 10% or so are + * due to actual, failed cards. + */ + printk(KERN_ERR + "EEH: PCI device %s - %s has failed %d times \n" + "and has been permanently disabled. Please try reseating\n" + "this device or replacing it.\n", + pci_name (frozen_pdn->pcidev), + pcid_name(frozen_pdn->pcidev), + frozen_pdn->eeh_freeze_count); + + eeh_slot_error_detail(frozen_pdn, 2 /* Permanent Error */); + + /* Notify all devices that they're about to go down. */ + pci_walk_bus(frozen_bus, eeh_report_failure, 0); + + /* Shut down the device drivers for good. */ + pcibios_remove_pci_devices(frozen_bus); +} + +/* ---------- end of file ---------- */ diff --git a/arch/powerpc/platforms/pseries/eeh_event.c b/arch/powerpc/platforms/pseries/eeh_event.c index 92497333c2b..9a9961f2748 100644 --- a/arch/powerpc/platforms/pseries/eeh_event.c +++ b/arch/powerpc/platforms/pseries/eeh_event.c @@ -21,6 +21,7 @@ #include <linux/list.h> #include <linux/pci.h> #include <asm/eeh_event.h> +#include <asm/ppc-pci.h> /** Overview: * EEH error states may be detected within exception handlers; @@ -37,31 +38,6 @@ static void eeh_thread_launcher(void *); DECLARE_WORK(eeh_event_wq, eeh_thread_launcher, NULL); /** - * eeh_panic - call panic() for an eeh event that cannot be handled. - * The philosophy of this routine is that it is better to panic and - * halt the OS than it is to risk possible data corruption by - * oblivious device drivers that don't know better. - * - * @dev pci device that had an eeh event - * @reset_state current reset state of the device slot - */ -static void eeh_panic(struct pci_dev *dev, int reset_state) -{ - /* - * Since the panic_on_oops sysctl is used to halt the system - * in light of potential corruption, we can use it here. - */ - if (panic_on_oops) { - panic("EEH: MMIO failure (%d) on device:%s\n", reset_state, - pci_name(dev)); - } - else { - printk(KERN_INFO "EEH: Ignored MMIO failure (%d) on device:%s\n", - reset_state, pci_name(dev)); - } -} - -/** * eeh_event_handler - dispatch EEH events. The detection of a frozen * slot can occur inside an interrupt, where it can be hard to do * anything about it. The goal of this routine is to pull these @@ -82,10 +58,16 @@ static int eeh_event_handler(void * dummy) spin_lock_irqsave(&eeh_eventlist_lock, flags); event = NULL; + + /* Unqueue the event, get ready to process. */ if (!list_empty(&eeh_eventlist)) { event = list_entry(eeh_eventlist.next, struct eeh_event, list); list_del(&event->list); } + + if (event) + eeh_mark_slot(event->dn, EEH_MODE_RECOVERING); + spin_unlock_irqrestore(&eeh_eventlist_lock, flags); if (event == NULL) break; @@ -93,8 +75,11 @@ static int eeh_event_handler(void * dummy) printk(KERN_INFO "EEH: Detected PCI bus error on device %s\n", pci_name(event->dev)); - eeh_panic (event->dev, event->state); + handle_eeh_events(event); + + eeh_clear_slot(event->dn, EEH_MODE_RECOVERING); + pci_dev_put(event->dev); kfree(event); } @@ -122,7 +107,7 @@ static void eeh_thread_launcher(void *dummy) */ int eeh_send_failure_event (struct device_node *dn, struct pci_dev *dev, - int state, + enum pci_channel_state state, int time_unavail) { unsigned long flags; diff --git a/arch/powerpc/platforms/pseries/hvcserver.c b/arch/powerpc/platforms/pseries/hvcserver.c index 4d584172055..22bfb5c89db 100644 --- a/arch/powerpc/platforms/pseries/hvcserver.c +++ b/arch/powerpc/platforms/pseries/hvcserver.c @@ -40,7 +40,7 @@ MODULE_VERSION(HVCS_ARCH_VERSION); * functions aren't performance sensitive, so this conversion isn't an * issue. */ -int hvcs_convert(long to_convert) +static int hvcs_convert(long to_convert) { switch (to_convert) { case H_Success: @@ -91,7 +91,7 @@ int hvcs_free_partner_info(struct list_head *head) EXPORT_SYMBOL(hvcs_free_partner_info); /* Helper function for hvcs_get_partner_info */ -int hvcs_next_partner(uint32_t unit_address, +static int hvcs_next_partner(uint32_t unit_address, unsigned long last_p_partition_ID, unsigned long last_p_unit_address, unsigned long *pi_buff) diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index 169f9148789..48cfbfc43f9 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -51,8 +51,6 @@ #define DBG(fmt...) -extern int is_python(struct device_node *); - static void tce_build_pSeries(struct iommu_table *tbl, long index, long npages, unsigned long uaddr, enum dma_data_direction direction) diff --git a/arch/powerpc/platforms/pseries/scanlog.c b/arch/powerpc/platforms/pseries/scanlog.c index 2edc947f7c4..50643496eb6 100644 --- a/arch/powerpc/platforms/pseries/scanlog.c +++ b/arch/powerpc/platforms/pseries/scanlog.c @@ -192,7 +192,7 @@ struct file_operations scanlog_fops = { .release = scanlog_release, }; -int __init scanlog_init(void) +static int __init scanlog_init(void) { struct proc_dir_entry *ent; @@ -222,7 +222,7 @@ int __init scanlog_init(void) return 0; } -void __exit scanlog_cleanup(void) +static void __exit scanlog_cleanup(void) { if (proc_ppc64_scan_log_dump) { kfree(proc_ppc64_scan_log_dump->data); diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c index 8903cf63236..68b7f086d63 100644 --- a/arch/powerpc/platforms/pseries/setup.c +++ b/arch/powerpc/platforms/pseries/setup.c @@ -86,7 +86,7 @@ static void pseries_dedicated_idle(void); struct mpic *pSeries_mpic; -void pSeries_show_cpuinfo(struct seq_file *m) +static void pSeries_show_cpuinfo(struct seq_file *m) { struct device_node *root; const char *model = ""; diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c index 25181c594d7..8e6b1ed1396 100644 --- a/arch/powerpc/platforms/pseries/smp.c +++ b/arch/powerpc/platforms/pseries/smp.c @@ -93,7 +93,7 @@ static int query_cpu_stopped(unsigned int pcpu) return cpu_status; } -int pSeries_cpu_disable(void) +static int pSeries_cpu_disable(void) { int cpu = smp_processor_id(); @@ -109,7 +109,7 @@ int pSeries_cpu_disable(void) return 0; } -void pSeries_cpu_die(unsigned int cpu) +static void pSeries_cpu_die(unsigned int cpu) { int tries; int cpu_status; @@ -282,7 +282,7 @@ static inline int __devinit smp_startup_cpu(unsigned int lcpu) pcpu = get_hard_smp_processor_id(lcpu); /* Fixup atomic count: it exited inside IRQ handler. */ - paca[lcpu].__current->thread_info->preempt_count = 0; + task_thread_info(paca[lcpu].__current)->preempt_count = 0; /* * If the RTAS start-cpu token does not exist then presume the diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c index 0c0cfa32eb5..fd823c7c9ac 100644 --- a/arch/powerpc/platforms/pseries/xics.c +++ b/arch/powerpc/platforms/pseries/xics.c @@ -381,7 +381,7 @@ int xics_get_irq(struct pt_regs *regs) #ifdef CONFIG_SMP -irqreturn_t xics_ipi_action(int irq, void *dev_id, struct pt_regs *regs) +static irqreturn_t xics_ipi_action(int irq, void *dev_id, struct pt_regs *regs) { int cpu = smp_processor_id(); diff --git a/arch/powerpc/sysdev/Makefile b/arch/powerpc/sysdev/Makefile index 14b9abde2d2..0ae841347a0 100644 --- a/arch/powerpc/sysdev/Makefile +++ b/arch/powerpc/sysdev/Makefile @@ -6,4 +6,4 @@ obj-$(CONFIG_BOOKE) += dcr.o obj-$(CONFIG_40x) += dcr.o obj-$(CONFIG_U3_DART) += dart_iommu.o obj-$(CONFIG_MMIO_NVRAM) += mmio_nvram.o -obj-$(CONFIG_83xx) += ipic.o +obj-$(CONFIG_PPC_83xx) += ipic.o diff --git a/arch/powerpc/sysdev/dart_iommu.c b/arch/powerpc/sysdev/dart_iommu.c index e00b46b9514..977de9db875 100644 --- a/arch/powerpc/sysdev/dart_iommu.c +++ b/arch/powerpc/sysdev/dart_iommu.c @@ -139,7 +139,6 @@ static void dart_build(struct iommu_table *tbl, long index, *(dp++) = DARTMAP_VALID | (rpn & DARTMAP_RPNMASK); - rpn++; uaddr += DART_PAGE_SIZE; } diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c index 22612ed5379..7d02fa2a899 100644 --- a/arch/powerpc/xmon/xmon.c +++ b/arch/powerpc/xmon/xmon.c @@ -311,7 +311,7 @@ static void release_output_lock(void) } #endif -int xmon_core(struct pt_regs *regs, int fromipi) +static int xmon_core(struct pt_regs *regs, int fromipi) { int cmd = 0; unsigned long msr; @@ -528,7 +528,7 @@ xmon_irq(int irq, void *d, struct pt_regs *regs) return IRQ_HANDLED; } -int xmon_bpt(struct pt_regs *regs) +static int xmon_bpt(struct pt_regs *regs) { struct bpt *bp; unsigned long offset; @@ -554,7 +554,7 @@ int xmon_bpt(struct pt_regs *regs) return 1; } -int xmon_sstep(struct pt_regs *regs) +static int xmon_sstep(struct pt_regs *regs) { if (user_mode(regs)) return 0; @@ -562,7 +562,7 @@ int xmon_sstep(struct pt_regs *regs) return 1; } -int xmon_dabr_match(struct pt_regs *regs) +static int xmon_dabr_match(struct pt_regs *regs) { if ((regs->msr & (MSR_IR|MSR_PR|MSR_SF)) != (MSR_IR|MSR_SF)) return 0; @@ -572,7 +572,7 @@ int xmon_dabr_match(struct pt_regs *regs) return 1; } -int xmon_iabr_match(struct pt_regs *regs) +static int xmon_iabr_match(struct pt_regs *regs) { if ((regs->msr & (MSR_IR|MSR_PR|MSR_SF)) != (MSR_IR|MSR_SF)) return 0; @@ -582,7 +582,7 @@ int xmon_iabr_match(struct pt_regs *regs) return 1; } -int xmon_ipi(struct pt_regs *regs) +static int xmon_ipi(struct pt_regs *regs) { #ifdef CONFIG_SMP if (in_xmon && !cpu_isset(smp_processor_id(), cpus_in_xmon)) @@ -591,7 +591,7 @@ int xmon_ipi(struct pt_regs *regs) return 0; } -int xmon_fault_handler(struct pt_regs *regs) +static int xmon_fault_handler(struct pt_regs *regs) { struct bpt *bp; unsigned long offset; |