diff options
Diffstat (limited to 'arch/powerpc')
97 files changed, 4604 insertions, 978 deletions
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 4bb2e9310a56..4e40c122bf26 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -626,20 +626,6 @@ config ADVANCED_OPTIONS comment "Default settings for advanced configuration options are used" depends on !ADVANCED_OPTIONS -config HIGHMEM_START_BOOL - bool "Set high memory pool address" - depends on ADVANCED_OPTIONS && HIGHMEM - help - This option allows you to set the base address of the kernel virtual - area used to map high memory pages. This can be useful in - optimizing the layout of kernel virtual memory. - - Say N here unless you know what you are doing. - -config HIGHMEM_START - hex "Virtual start address of high memory pool" if HIGHMEM_START_BOOL - default "0xfe000000" - config LOWMEM_SIZE_BOOL bool "Set maximum low memory" depends on ADVANCED_OPTIONS @@ -656,21 +642,76 @@ config LOWMEM_SIZE hex "Maximum low memory size (in bytes)" if LOWMEM_SIZE_BOOL default "0x30000000" +config RELOCATABLE + bool "Build a relocatable kernel (EXPERIMENTAL)" + depends on EXPERIMENTAL && ADVANCED_OPTIONS && FLATMEM && FSL_BOOKE + help + This builds a kernel image that is capable of running at the + location the kernel is loaded at (some alignment restrictions may + exist). + + One use is for the kexec on panic case where the recovery kernel + must live at a different physical address than the primary + kernel. + + Note: If CONFIG_RELOCATABLE=y, then the kernel runs from the address + it has been loaded at and the compile time physical addresses + CONFIG_PHYSICAL_START is ignored. However CONFIG_PHYSICAL_START + setting can still be useful to bootwrappers that need to know the + load location of the kernel (eg. u-boot/mkimage). + +config PAGE_OFFSET_BOOL + bool "Set custom page offset address" + depends on ADVANCED_OPTIONS + help + This option allows you to set the kernel virtual address at which + the kernel will map low memory. This can be useful in optimizing + the virtual memory layout of the system. + + Say N here unless you know what you are doing. + +config PAGE_OFFSET + hex "Virtual address of memory base" if PAGE_OFFSET_BOOL + default "0xc0000000" + config KERNEL_START_BOOL bool "Set custom kernel base address" depends on ADVANCED_OPTIONS help This option allows you to set the kernel virtual address at which - the kernel will map low memory (the kernel image will be linked at - this address). This can be useful in optimizing the virtual memory - layout of the system. + the kernel will be loaded. Normally this should match PAGE_OFFSET + however there are times (like kdump) that one might not want them + to be the same. Say N here unless you know what you are doing. config KERNEL_START hex "Virtual address of kernel base" if KERNEL_START_BOOL + default PAGE_OFFSET if PAGE_OFFSET_BOOL + default "0xc2000000" if CRASH_DUMP default "0xc0000000" +config PHYSICAL_START_BOOL + bool "Set physical address where the kernel is loaded" + depends on ADVANCED_OPTIONS && FLATMEM && FSL_BOOKE + help + This gives the physical address where the kernel is loaded. + + Say N here unless you know what you are doing. + +config PHYSICAL_START + hex "Physical address where the kernel is loaded" if PHYSICAL_START_BOOL + default "0x02000000" if PPC_STD_MMU && CRASH_DUMP + default "0x00000000" + +config PHYSICAL_ALIGN + hex + default "0x10000000" if FSL_BOOKE + help + This value puts the alignment restrictions on physical address + where kernel is loaded and run from. Kernel is compiled for an + address which meets above alignment restriction. + config TASK_SIZE_BOOL bool "Set custom user task size" depends on ADVANCED_OPTIONS @@ -717,9 +758,17 @@ config PIN_TLB endmenu if PPC64 +config PAGE_OFFSET + hex + default "0xc000000000000000" config KERNEL_START hex + default "0xc000000002000000" if CRASH_DUMP default "0xc000000000000000" +config PHYSICAL_START + hex + default "0x02000000" if CRASH_DUMP + default "0x00000000" endif source "net/Kconfig" @@ -754,3 +803,4 @@ config PPC_CLOCK config PPC_LIB_RHEAP bool +source "arch/powerpc/kvm/Kconfig" diff --git a/arch/powerpc/Kconfig.debug b/arch/powerpc/Kconfig.debug index a86d8d853214..807a2dce6263 100644 --- a/arch/powerpc/Kconfig.debug +++ b/arch/powerpc/Kconfig.debug @@ -151,6 +151,9 @@ config BOOTX_TEXT config PPC_EARLY_DEBUG bool "Early debugging (dangerous)" + # PPC_EARLY_DEBUG on 440 leaves AS=1 mappings above the TLB high water + # mark, which doesn't work with current 440 KVM. + depends on !KVM help Say Y to enable some early debugging facilities that may be available for your processor/board combination. Those facilities are hacks diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile index e2ec4a91ccef..9dcdc036cdf7 100644 --- a/arch/powerpc/Makefile +++ b/arch/powerpc/Makefile @@ -145,6 +145,7 @@ core-y += arch/powerpc/kernel/ \ arch/powerpc/platforms/ core-$(CONFIG_MATH_EMULATION) += arch/powerpc/math-emu/ core-$(CONFIG_XMON) += arch/powerpc/xmon/ +core-$(CONFIG_KVM) += arch/powerpc/kvm/ drivers-$(CONFIG_OPROFILE) += arch/powerpc/oprofile/ diff --git a/arch/powerpc/boot/.gitignore b/arch/powerpc/boot/.gitignore index 5ef2bdf8d189..2347294ff35b 100644 --- a/arch/powerpc/boot/.gitignore +++ b/arch/powerpc/boot/.gitignore @@ -27,6 +27,7 @@ zImage.chrp zImage.coff zImage.coff.lds zImage.ep* +zImage.iseries zImage.*lds zImage.miboot zImage.pmac diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile index 5ba50c673390..7822d25c9d31 100644 --- a/arch/powerpc/boot/Makefile +++ b/arch/powerpc/boot/Makefile @@ -40,7 +40,7 @@ $(obj)/ebony.o: BOOTCFLAGS += -mcpu=405 $(obj)/cuboot-taishan.o: BOOTCFLAGS += -mcpu=405 $(obj)/cuboot-katmai.o: BOOTCFLAGS += -mcpu=405 $(obj)/treeboot-walnut.o: BOOTCFLAGS += -mcpu=405 -$(obj)/virtex405-head.o: BOOTCFLAGS += -mcpu=405 +$(obj)/virtex405-head.o: BOOTAFLAGS += -mcpu=405 zlib := inffast.c inflate.c inftrees.c diff --git a/arch/powerpc/boot/dts/canyonlands.dts b/arch/powerpc/boot/dts/canyonlands.dts index 6f3d38a1554f..39634124929b 100644 --- a/arch/powerpc/boot/dts/canyonlands.dts +++ b/arch/powerpc/boot/dts/canyonlands.dts @@ -142,8 +142,45 @@ #address-cells = <2>; #size-cells = <1>; clock-frequency = <0>; /* Filled in by U-Boot */ + /* ranges property is supplied by U-Boot */ interrupts = <6 4>; interrupt-parent = <&UIC1>; + + nor_flash@0,0 { + compatible = "amd,s29gl512n", "cfi-flash"; + bank-width = <2>; + reg = <0 000000 4000000>; + #address-cells = <1>; + #size-cells = <1>; + partition@0 { + label = "kernel"; + reg = <0 1e0000>; + }; + partition@1e0000 { + label = "dtb"; + reg = <1e0000 20000>; + }; + partition@200000 { + label = "ramdisk"; + reg = <200000 1400000>; + }; + partition@1600000 { + label = "jffs2"; + reg = <1600000 400000>; + }; + partition@1a00000 { + label = "user"; + reg = <1a00000 2560000>; + }; + partition@3f60000 { + label = "env"; + reg = <3f60000 40000>; + }; + partition@3fa0000 { + label = "u-boot"; + reg = <3fa0000 60000>; + }; + }; }; UART0: serial@ef600300 { diff --git a/arch/powerpc/boot/dts/glacier.dts b/arch/powerpc/boot/dts/glacier.dts index 958a5ca53d35..0f2fc077d8db 100644 --- a/arch/powerpc/boot/dts/glacier.dts +++ b/arch/powerpc/boot/dts/glacier.dts @@ -145,8 +145,45 @@ #address-cells = <2>; #size-cells = <1>; clock-frequency = <0>; /* Filled in by U-Boot */ + /* ranges property is supplied by U-Boot */ interrupts = <6 4>; interrupt-parent = <&UIC1>; + + nor_flash@0,0 { + compatible = "amd,s29gl512n", "cfi-flash"; + bank-width = <2>; + reg = <0 000000 4000000>; + #address-cells = <1>; + #size-cells = <1>; + partition@0 { + label = "kernel"; + reg = <0 1e0000>; + }; + partition@1e0000 { + label = "dtb"; + reg = <1e0000 20000>; + }; + partition@200000 { + label = "ramdisk"; + reg = <200000 1400000>; + }; + partition@1600000 { + label = "jffs2"; + reg = <1600000 400000>; + }; + partition@1a00000 { + label = "user"; + reg = <1a00000 2560000>; + }; + partition@3f60000 { + label = "env"; + reg = <3f60000 40000>; + }; + partition@3fa0000 { + label = "u-boot"; + reg = <3fa0000 60000>; + }; + }; }; UART0: serial@ef600300 { diff --git a/arch/powerpc/boot/dts/mpc8610_hpcd.dts b/arch/powerpc/boot/dts/mpc8610_hpcd.dts index 16c947b8a721..1f2f1e0a5571 100644 --- a/arch/powerpc/boot/dts/mpc8610_hpcd.dts +++ b/arch/powerpc/boot/dts/mpc8610_hpcd.dts @@ -45,6 +45,11 @@ reg = <0x00000000 0x20000000>; // 512M at 0x0 }; + board-control@e8000000 { + compatible = "fsl,fpga-pixis"; + reg = <0xe8000000 32>; // pixis at 0xe8000000 + }; + soc@e0000000 { #address-cells = <1>; #size-cells = <1>; @@ -104,6 +109,13 @@ interrupt-parent = <&mpic>; }; + display@2c000 { + compatible = "fsl,diu"; + reg = <0x2c000 100>; + interrupts = <72 2>; + interrupt-parent = <&mpic>; + }; + mpic: interrupt-controller@40000 { clock-frequency = <0>; interrupt-controller; diff --git a/arch/powerpc/boot/ns16550.c b/arch/powerpc/boot/ns16550.c index aef3bdc89160..8c9ead94be06 100644 --- a/arch/powerpc/boot/ns16550.c +++ b/arch/powerpc/boot/ns16550.c @@ -55,10 +55,15 @@ static u8 ns16550_tstc(void) int ns16550_console_init(void *devp, struct serial_console_data *scdp) { int n; + u32 reg_offset; if (dt_get_virtual_reg(devp, (void **)®_base, 1) < 1) return -1; + n = getprop(devp, "reg-offset", ®_offset, sizeof(reg_offset)); + if (n == sizeof(reg_offset)) + reg_base += reg_offset; + n = getprop(devp, "reg-shift", ®_shift, sizeof(reg_shift)); if (n != sizeof(reg_shift)) reg_shift = 0; diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index ce1e8d24e747..9177b21b1a95 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -106,4 +106,13 @@ PHONY += systbl_chk systbl_chk: $(src)/systbl_chk.sh $(obj)/systbl_chk.i $(call cmd,systbl_chk) +$(obj)/built-in.o: prom_init_check + +quiet_cmd_prom_init_check = CALL $< + cmd_prom_init_check = $(CONFIG_SHELL) $< "$(NM)" "$(obj)/prom_init.o" + +PHONY += prom_init_check +prom_init_check: $(src)/prom_init_check.sh $(obj)/prom_init.o + $(call cmd,prom_init_check) + clean-files := vmlinux.lds diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c index 292c6d8db0e1..62134845af08 100644 --- a/arch/powerpc/kernel/asm-offsets.c +++ b/arch/powerpc/kernel/asm-offsets.c @@ -23,6 +23,9 @@ #include <linux/mm.h> #include <linux/suspend.h> #include <linux/hrtimer.h> +#ifdef CONFIG_KVM +#include <linux/kvm_host.h> +#endif #ifdef CONFIG_PPC64 #include <linux/time.h> #include <linux/hardirq.h> @@ -93,10 +96,7 @@ int main(void) DEFINE(TI_LOCAL_FLAGS, offsetof(struct thread_info, local_flags)); DEFINE(TI_PREEMPT, offsetof(struct thread_info, preempt_count)); DEFINE(TI_TASK, offsetof(struct thread_info, task)); -#ifdef CONFIG_PPC32 - DEFINE(TI_EXECDOMAIN, offsetof(struct thread_info, exec_domain)); DEFINE(TI_CPU, offsetof(struct thread_info, cpu)); -#endif /* CONFIG_PPC32 */ #ifdef CONFIG_PPC64 DEFINE(DCACHEL1LINESIZE, offsetof(struct ppc64_caches, dline_size)); @@ -165,13 +165,9 @@ int main(void) /* Interrupt register frame */ DEFINE(STACK_FRAME_OVERHEAD, STACK_FRAME_OVERHEAD); -#ifndef CONFIG_PPC64 - DEFINE(INT_FRAME_SIZE, STACK_FRAME_OVERHEAD + sizeof(struct pt_regs)); -#else /* CONFIG_PPC64 */ + DEFINE(INT_FRAME_SIZE, STACK_INT_FRAME_SIZE); +#ifdef CONFIG_PPC64 DEFINE(SWITCH_FRAME_SIZE, STACK_FRAME_OVERHEAD + sizeof(struct pt_regs)); - /* 288 = # of volatile regs, int & fp, for leaf routines */ - /* which do not stack a frame. See the PPC64 ABI. */ - DEFINE(INT_FRAME_SIZE, STACK_FRAME_OVERHEAD + sizeof(struct pt_regs) + 288); /* Create extra stack space for SRR0 and SRR1 when calling prom/rtas. */ DEFINE(PROM_FRAME_SIZE, STACK_FRAME_OVERHEAD + sizeof(struct pt_regs) + 16); DEFINE(RTAS_FRAME_SIZE, STACK_FRAME_OVERHEAD + sizeof(struct pt_regs) + 16); @@ -331,5 +327,30 @@ int main(void) DEFINE(PGD_TABLE_SIZE, PGD_TABLE_SIZE); +#ifdef CONFIG_KVM + DEFINE(TLBE_BYTES, sizeof(struct tlbe)); + + DEFINE(VCPU_HOST_STACK, offsetof(struct kvm_vcpu, arch.host_stack)); + DEFINE(VCPU_HOST_PID, offsetof(struct kvm_vcpu, arch.host_pid)); + DEFINE(VCPU_HOST_TLB, offsetof(struct kvm_vcpu, arch.host_tlb)); + DEFINE(VCPU_SHADOW_TLB, offsetof(struct kvm_vcpu, arch.shadow_tlb)); + DEFINE(VCPU_GPRS, offsetof(struct kvm_vcpu, arch.gpr)); + DEFINE(VCPU_LR, offsetof(struct kvm_vcpu, arch.lr)); + DEFINE(VCPU_CR, offsetof(struct kvm_vcpu, arch.cr)); + DEFINE(VCPU_XER, offsetof(struct kvm_vcpu, arch.xer)); + DEFINE(VCPU_CTR, offsetof(struct kvm_vcpu, arch.ctr)); + DEFINE(VCPU_PC, offsetof(struct kvm_vcpu, arch.pc)); + DEFINE(VCPU_MSR, offsetof(struct kvm_vcpu, arch.msr)); + DEFINE(VCPU_SPRG4, offsetof(struct kvm_vcpu, arch.sprg4)); + DEFINE(VCPU_SPRG5, offsetof(struct kvm_vcpu, arch.sprg5)); + DEFINE(VCPU_SPRG6, offsetof(struct kvm_vcpu, arch.sprg6)); + DEFINE(VCPU_SPRG7, offsetof(struct kvm_vcpu, arch.sprg7)); + DEFINE(VCPU_PID, offsetof(struct kvm_vcpu, arch.pid)); + + DEFINE(VCPU_LAST_INST, offsetof(struct kvm_vcpu, arch.last_inst)); + DEFINE(VCPU_FAULT_DEAR, offsetof(struct kvm_vcpu, arch.fault_dear)); + DEFINE(VCPU_FAULT_ESR, offsetof(struct kvm_vcpu, arch.fault_esr)); +#endif + return 0; } diff --git a/arch/powerpc/kernel/cpu_setup_44x.S b/arch/powerpc/kernel/cpu_setup_44x.S index 5465e8de0e61..e3623e3e3451 100644 --- a/arch/powerpc/kernel/cpu_setup_44x.S +++ b/arch/powerpc/kernel/cpu_setup_44x.S @@ -33,7 +33,6 @@ _GLOBAL(__setup_cpu_440grx) mtlr r4 blr _GLOBAL(__setup_cpu_460ex) -_GLOBAL(__setup_cpu_460gt) b __init_fpu_44x _GLOBAL(__setup_cpu_440gx) _GLOBAL(__setup_cpu_440spe) diff --git a/arch/powerpc/kernel/cpu_setup_6xx.S b/arch/powerpc/kernel/cpu_setup_6xx.S index f1ee0b3f78f2..72d1d7395254 100644 --- a/arch/powerpc/kernel/cpu_setup_6xx.S +++ b/arch/powerpc/kernel/cpu_setup_6xx.S @@ -17,7 +17,13 @@ #include <asm/cache.h> _GLOBAL(__setup_cpu_603) - b setup_common_caches + mflr r4 +BEGIN_FTR_SECTION + bl __init_fpu_registers +END_FTR_SECTION_IFCLR(CPU_FTR_FPU_UNAVAILABLE) + bl setup_common_caches + mtlr r4 + blr _GLOBAL(__setup_cpu_604) mflr r4 bl setup_common_caches diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index 26ffb44e2701..36080d4d1922 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -37,7 +37,6 @@ extern void __setup_cpu_440gx(unsigned long offset, struct cpu_spec* spec); extern void __setup_cpu_440grx(unsigned long offset, struct cpu_spec* spec); extern void __setup_cpu_440spe(unsigned long offset, struct cpu_spec* spec); extern void __setup_cpu_460ex(unsigned long offset, struct cpu_spec* spec); -extern void __setup_cpu_460gt(unsigned long offset, struct cpu_spec* spec); extern void __setup_cpu_603(unsigned long offset, struct cpu_spec* spec); extern void __setup_cpu_604(unsigned long offset, struct cpu_spec* spec); extern void __setup_cpu_750(unsigned long offset, struct cpu_spec* spec); @@ -1416,10 +1415,9 @@ static struct cpu_spec __initdata cpu_specs[] = { .pvr_value = 0x13020000, .cpu_name = "460GT", .cpu_features = CPU_FTRS_44X, - .cpu_user_features = COMMON_USER_BOOKE | PPC_FEATURE_HAS_FPU, + .cpu_user_features = COMMON_USER_BOOKE, .icache_bsize = 32, .dcache_bsize = 32, - .cpu_setup = __setup_cpu_460gt, .machine_check = machine_check_440A, .platform = "ppc440", }, diff --git a/arch/powerpc/kernel/head_fsl_booke.S b/arch/powerpc/kernel/head_fsl_booke.S index 4ff744143566..e581524d85bc 100644 --- a/arch/powerpc/kernel/head_fsl_booke.S +++ b/arch/powerpc/kernel/head_fsl_booke.S @@ -371,6 +371,17 @@ skpinv: addi r6,r6,1 /* Increment */ bl early_init +#ifdef CONFIG_RELOCATABLE + lis r3,kernstart_addr@ha + la r3,kernstart_addr@l(r3) +#ifdef CONFIG_PHYS_64BIT + stw r23,0(r3) + stw r25,4(r3) +#else + stw r25,0(r3) +#endif +#endif + mfspr r3,SPRN_TLB1CFG andi. r3,r3,0xfff lis r4,num_tlbcam_entries@ha diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S index 9d2c56621f1e..92ccc6fcc5b0 100644 --- a/arch/powerpc/kernel/misc_32.S +++ b/arch/powerpc/kernel/misc_32.S @@ -152,7 +152,7 @@ _GLOBAL(low_choose_750fx_pll) mtspr SPRN_HID1,r4 /* Store new HID1 image */ - rlwinm r6,r1,0,0,18 + rlwinm r6,r1,0,0,(31-THREAD_SHIFT) lwz r6,TI_CPU(r6) slwi r6,r6,2 addis r6,r6,nap_save_hid1@ha @@ -281,7 +281,7 @@ _GLOBAL(_tlbia) #endif /* CONFIG_SMP */ #else /* !(CONFIG_40x || CONFIG_44x || CONFIG_FSL_BOOKE) */ #if defined(CONFIG_SMP) - rlwinm r8,r1,0,0,18 + rlwinm r8,r1,0,0,(31-THREAD_SHIFT) lwz r8,TI_CPU(r8) oris r8,r8,10 mfmsr r10 @@ -377,7 +377,7 @@ _GLOBAL(_tlbie) #endif /* CONFIG_SMP */ #else /* !(CONFIG_40x || CONFIG_44x || CONFIG_FSL_BOOKE) */ #if defined(CONFIG_SMP) - rlwinm r8,r1,0,0,18 + rlwinm r8,r1,0,0,(31-THREAD_SHIFT) lwz r8,TI_CPU(r8) oris r8,r8,11 mfmsr r10 diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S index a3c491e88a72..942951e76586 100644 --- a/arch/powerpc/kernel/misc_64.S +++ b/arch/powerpc/kernel/misc_64.S @@ -27,23 +27,11 @@ .text -_GLOBAL(get_msr) - mfmsr r3 - blr - -_GLOBAL(get_srr0) - mfsrr0 r3 - blr - -_GLOBAL(get_srr1) - mfsrr1 r3 - blr - #ifdef CONFIG_IRQSTACKS _GLOBAL(call_do_softirq) mflr r0 std r0,16(r1) - stdu r1,THREAD_SIZE-112(r3) + stdu r1,THREAD_SIZE-STACK_FRAME_OVERHEAD(r3) mr r1,r3 bl .__do_softirq ld r1,0(r1) @@ -56,7 +44,7 @@ _GLOBAL(call_handle_irq) mflr r0 std r0,16(r1) mtctr r8 - stdu r1,THREAD_SIZE-112(r5) + stdu r1,THREAD_SIZE-STACK_FRAME_OVERHEAD(r5) mr r1,r5 bctrl ld r1,0(r1) @@ -599,7 +587,7 @@ _GLOBAL(kexec_sequence) std r0,16(r1) /* switch stacks to newstack -- &kexec_stack.stack */ - stdu r1,THREAD_SIZE-112(r3) + stdu r1,THREAD_SIZE-STACK_FRAME_OVERHEAD(r3) mr r1,r3 li r0,0 @@ -616,7 +604,7 @@ _GLOBAL(kexec_sequence) std r26,-48(r1) std r25,-56(r1) - stdu r1,-112-64(r1) + stdu r1,-STACK_FRAME_OVERHEAD-64(r1) /* save args into preserved regs */ mr r31,r3 /* newstack (both) */ diff --git a/arch/powerpc/kernel/of_platform.c b/arch/powerpc/kernel/of_platform.c index fb698d47082d..e79ad8afda07 100644 --- a/arch/powerpc/kernel/of_platform.c +++ b/arch/powerpc/kernel/of_platform.c @@ -275,6 +275,8 @@ static int __devinit of_pci_phb_probe(struct of_device *dev, /* Scan the bus */ scan_phb(phb); + if (phb->bus == NULL) + return -ENXIO; /* Claim resources. This might need some rework as well depending * wether we are doing probe-only or not, like assigning unassigned diff --git a/arch/powerpc/kernel/paca.c b/arch/powerpc/kernel/paca.c index ac163bd46cfd..c9bf17eec31b 100644 --- a/arch/powerpc/kernel/paca.c +++ b/arch/powerpc/kernel/paca.c @@ -7,17 +7,11 @@ * 2 of the License, or (at your option) any later version. */ -#include <linux/types.h> #include <linux/threads.h> #include <linux/module.h> -#include <asm/processor.h> -#include <asm/ptrace.h> -#include <asm/page.h> #include <asm/lppaca.h> #include <asm/paca.h> -#include <asm/mmu.h> - /* This symbol is provided by the linker - let it fill in the paca * field correctly */ @@ -65,60 +59,29 @@ struct slb_shadow slb_shadow[] __cacheline_aligned = { * processors. The processor VPD array needs one entry per physical * processor (not thread). */ -#define PACA_INIT(number) \ -{ \ - .lppaca_ptr = &lppaca[number], \ - .lock_token = 0x8000, \ - .paca_index = (number), /* Paca Index */ \ - .kernel_toc = (unsigned long)(&__toc_start) + 0x8000UL, \ - .hw_cpu_id = 0xffff, \ - .slb_shadow_ptr = &slb_shadow[number], \ - .__current = &init_task, \ -} - -struct paca_struct paca[] = { - PACA_INIT(0), -#if NR_CPUS > 1 - PACA_INIT( 1), PACA_INIT( 2), PACA_INIT( 3), -#if NR_CPUS > 4 - PACA_INIT( 4), PACA_INIT( 5), PACA_INIT( 6), PACA_INIT( 7), -#if NR_CPUS > 8 - PACA_INIT( 8), PACA_INIT( 9), PACA_INIT( 10), PACA_INIT( 11), - PACA_INIT( 12), PACA_INIT( 13), PACA_INIT( 14), PACA_INIT( 15), - PACA_INIT( 16), PACA_INIT( 17), PACA_INIT( 18), PACA_INIT( 19), - PACA_INIT( 20), PACA_INIT( 21), PACA_INIT( 22), PACA_INIT( 23), - PACA_INIT( 24), PACA_INIT( 25), PACA_INIT( 26), PACA_INIT( 27), - PACA_INIT( 28), PACA_INIT( 29), PACA_INIT( 30), PACA_INIT( 31), -#if NR_CPUS > 32 - PACA_INIT( 32), PACA_INIT( 33), PACA_INIT( 34), PACA_INIT( 35), - PACA_INIT( 36), PACA_INIT( 37), PACA_INIT( 38), PACA_INIT( 39), - PACA_INIT( 40), PACA_INIT( 41), PACA_INIT( 42), PACA_INIT( 43), - PACA_INIT( 44), PACA_INIT( 45), PACA_INIT( 46), PACA_INIT( 47), - PACA_INIT( 48), PACA_INIT( 49), PACA_INIT( 50), PACA_INIT( 51), - PACA_INIT( 52), PACA_INIT( 53), PACA_INIT( 54), PACA_INIT( 55), - PACA_INIT( 56), PACA_INIT( 57), PACA_INIT( 58), PACA_INIT( 59), - PACA_INIT( 60), PACA_INIT( 61), PACA_INIT( 62), PACA_INIT( 63), -#if NR_CPUS > 64 - PACA_INIT( 64), PACA_INIT( 65), PACA_INIT( 66), PACA_INIT( 67), - PACA_INIT( 68), PACA_INIT( 69), PACA_INIT( 70), PACA_INIT( 71), - PACA_INIT( 72), PACA_INIT( 73), PACA_INIT( 74), PACA_INIT( 75), - PACA_INIT( 76), PACA_INIT( 77), PACA_INIT( 78), PACA_INIT( 79), - PACA_INIT( 80), PACA_INIT( 81), PACA_INIT( 82), PACA_INIT( 83), - PACA_INIT( 84), PACA_INIT( 85), PACA_INIT( 86), PACA_INIT( 87), - PACA_INIT( 88), PACA_INIT( 89), PACA_INIT( 90), PACA_INIT( 91), - PACA_INIT( 92), PACA_INIT( 93), PACA_INIT( 94), PACA_INIT( 95), - PACA_INIT( 96), PACA_INIT( 97), PACA_INIT( 98), PACA_INIT( 99), - PACA_INIT(100), PACA_INIT(101), PACA_INIT(102), PACA_INIT(103), - PACA_INIT(104), PACA_INIT(105), PACA_INIT(106), PACA_INIT(107), - PACA_INIT(108), PACA_INIT(109), PACA_INIT(110), PACA_INIT(111), - PACA_INIT(112), PACA_INIT(113), PACA_INIT(114), PACA_INIT(115), - PACA_INIT(116), PACA_INIT(117), PACA_INIT(118), PACA_INIT(119), - PACA_INIT(120), PACA_INIT(121), PACA_INIT(122), PACA_INIT(123), - PACA_INIT(124), PACA_INIT(125), PACA_INIT(126), PACA_INIT(127), -#endif -#endif -#endif -#endif -#endif -}; +struct paca_struct paca[NR_CPUS]; EXPORT_SYMBOL(paca); + +void __init initialise_pacas(void) +{ + int cpu; + + /* The TOC register (GPR2) points 32kB into the TOC, so that 64kB + * of the TOC can be addressed using a single machine instruction. + */ + unsigned long kernel_toc = (unsigned long)(&__toc_start) + 0x8000UL; + + /* Can't use for_each_*_cpu, as they aren't functional yet */ + for (cpu = 0; cpu < NR_CPUS; cpu++) { + struct paca_struct *new_paca = &paca[cpu]; + + new_paca->lppaca_ptr = &lppaca[cpu]; + new_paca->lock_token = 0x8000; + new_paca->paca_index = cpu; + new_paca->kernel_toc = kernel_toc; + new_paca->hw_cpu_id = 0xffff; + new_paca->slb_shadow_ptr = &slb_shadow[cpu]; + new_paca->__current = &init_task; + + } +} diff --git a/arch/powerpc/kernel/ppc32.h b/arch/powerpc/kernel/ppc32.h index fda05e2211d6..90e562771791 100644 --- a/arch/powerpc/kernel/ppc32.h +++ b/arch/powerpc/kernel/ppc32.h @@ -135,6 +135,4 @@ struct ucontext32 { struct mcontext32 uc_mcontext; }; -extern int copy_siginfo_to_user32(struct compat_siginfo __user *d, siginfo_t *s); - #endif /* _PPC64_PPC32_H */ diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 703100d5e458..6caad17ea72e 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -1033,3 +1033,34 @@ void ppc64_runlatch_off(void) } } #endif + +#if THREAD_SHIFT < PAGE_SHIFT + +static struct kmem_cache *thread_info_cache; + +struct thread_info *alloc_thread_info(struct task_struct *tsk) +{ + struct thread_info *ti; + + ti = kmem_cache_alloc(thread_info_cache, GFP_KERNEL); + if (unlikely(ti == NULL)) + return NULL; +#ifdef CONFIG_DEBUG_STACK_USAGE + memset(ti, 0, THREAD_SIZE); +#endif + return ti; +} + +void free_thread_info(struct thread_info *ti) +{ + kmem_cache_free(thread_info_cache, ti); +} + +void thread_info_cache_init(void) +{ + thread_info_cache = kmem_cache_create("thread_info", THREAD_SIZE, + THREAD_SIZE, 0, NULL); + BUG_ON(thread_info_cache == NULL); +} + +#endif /* THREAD_SHIFT < PAGE_SHIFT */ diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index 3bfe7837e820..2aefe2a4129a 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -53,6 +53,7 @@ #include <asm/pci-bridge.h> #include <asm/phyp_dump.h> #include <asm/kexec.h> +#include <mm/mmu_decl.h> #ifdef DEBUG #define DBG(fmt...) printk(KERN_ERR fmt) @@ -978,7 +979,10 @@ static int __init early_init_dt_scan_memory(unsigned long node, } #endif lmb_add(base, size); + + memstart_addr = min((u64)memstart_addr, base); } + return 0; } diff --git a/arch/powerpc/kernel/prom_init_check.sh b/arch/powerpc/kernel/prom_init_check.sh new file mode 100644 index 000000000000..8e24fc1821e8 --- /dev/null +++ b/arch/powerpc/kernel/prom_init_check.sh @@ -0,0 +1,58 @@ +#!/bin/sh +# +# Copyright © 2008 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 script checks prom_init.o to see what external symbols it +# is using, if it finds symbols not in the whitelist it returns +# an error. The point of this is to discourage people from +# intentionally or accidentally adding new code to prom_init.c +# which has side effects on other parts of the kernel. + +# If you really need to reference something from prom_init.o add +# it to the list below: + +WHITELIST="add_reloc_offset __bss_start __bss_stop copy_and_flush +_end enter_prom memcpy memset reloc_offset __secondary_hold +__secondary_hold_acknowledge __secondary_hold_spinloop __start +strcmp strcpy strlcpy strlen strncmp strstr logo_linux_clut224 +reloc_got2" + +NM="$1" +OBJ="$2" + +ERROR=0 + +for UNDEF in $($NM -u $OBJ | awk '{print $2}') +do + # On 64-bit nm gives us the function descriptors, which have + # a leading . on the name, so strip it off here. + UNDEF="${UNDEF#.}" + + if [ $KBUILD_VERBOSE ]; then + if [ $KBUILD_VERBOSE -ne 0 ]; then + echo "Checking prom_init.o symbol '$UNDEF'" + fi + fi + + OK=0 + for WHITE in $WHITELIST + do + if [ "$UNDEF" = "$WHITE" ]; then + OK=1 + break + fi + done + + if [ $OK -eq 0 ]; then + ERROR=1 + echo "Error: External symbol '$UNDEF' referenced" \ + "from prom_init.c" >&2 + fi +done + +exit $ERROR diff --git a/arch/powerpc/kernel/ptrace32.c b/arch/powerpc/kernel/ptrace32.c index 9d30e10970ac..4c1de6af4c09 100644 --- a/arch/powerpc/kernel/ptrace32.c +++ b/arch/powerpc/kernel/ptrace32.c @@ -29,15 +29,12 @@ #include <linux/security.h> #include <linux/signal.h> #include <linux/compat.h> -#include <linux/elf.h> #include <asm/uaccess.h> #include <asm/page.h> #include <asm/pgtable.h> #include <asm/system.h> -#include "ppc32.h" - /* * does not yet catch signals sent when the child dies. * in exit.c or in signal.c. @@ -67,27 +64,6 @@ static long compat_ptrace_old(struct task_struct *child, long request, return -EPERM; } -static int compat_ptrace_getsiginfo(struct task_struct *child, compat_siginfo_t __user *data) -{ - siginfo_t lastinfo; - int error = -ESRCH; - - read_lock(&tasklist_lock); - if (likely(child->sighand != NULL)) { - error = -EINVAL; - spin_lock_irq(&child->sighand->siglock); - if (likely(child->last_siginfo != NULL)) { - lastinfo = *child->last_siginfo; - error = 0; - } - spin_unlock_irq(&child->sighand->siglock); - } - read_unlock(&tasklist_lock); - if (!error) - return copy_siginfo_to_user32(data, &lastinfo); - return error; -} - long compat_arch_ptrace(struct task_struct *child, compat_long_t request, compat_ulong_t caddr, compat_ulong_t cdata) { @@ -306,9 +282,6 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, 0, PT_REGS_COUNT * sizeof(compat_long_t), compat_ptr(data)); - case PTRACE_GETSIGINFO: - return compat_ptrace_getsiginfo(child, compat_ptr(data)); - case PTRACE_GETFPREGS: case PTRACE_SETFPREGS: case PTRACE_GETVRREGS: diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 31ada9fdfc5c..dff6308d1b5e 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -170,6 +170,9 @@ void __init setup_paca(int cpu) void __init early_setup(unsigned long dt_ptr) { + /* Fill in any unititialised pacas */ + initialise_pacas(); + /* Identify CPU type */ identify_cpu(0, mfspr(SPRN_PVR)); @@ -435,7 +438,7 @@ void __init setup_system(void) printk("htab_address = 0x%p\n", htab_address); printk("htab_hash_mask = 0x%lx\n", htab_hash_mask); #if PHYSICAL_START > 0 - printk("physical_start = 0x%x\n", PHYSICAL_START); + printk("physical_start = 0x%lx\n", PHYSICAL_START); #endif printk("-----------------------------------------------------\n"); diff --git a/arch/powerpc/kernel/stacktrace.c b/arch/powerpc/kernel/stacktrace.c index e3638eeaaae7..962944038430 100644 --- a/arch/powerpc/kernel/stacktrace.c +++ b/arch/powerpc/kernel/stacktrace.c @@ -13,7 +13,6 @@ #include <linux/sched.h> #include <linux/stacktrace.h> #include <asm/ptrace.h> -#include <asm/asm-offsets.h> /* * Save stack-backtrace addresses into a stack_trace buffer. diff --git a/arch/powerpc/kernel/udbg.c b/arch/powerpc/kernel/udbg.c index 7aad6203e411..7d6c9bb8c77f 100644 --- a/arch/powerpc/kernel/udbg.c +++ b/arch/powerpc/kernel/udbg.c @@ -154,8 +154,8 @@ static void udbg_console_write(struct console *con, const char *s, static struct console udbg_console = { .name = "udbg", .write = udbg_console_write, - .flags = CON_PRINTBUFFER | CON_ENABLED | CON_BOOT, - .index = -1, + .flags = CON_PRINTBUFFER | CON_ENABLED | CON_BOOT | CON_ANYTIME, + .index = 0, }; static int early_console_initialized; diff --git a/arch/powerpc/kvm/44x_tlb.c b/arch/powerpc/kvm/44x_tlb.c new file mode 100644 index 000000000000..f5d7a5eab96e --- /dev/null +++ b/arch/powerpc/kvm/44x_tlb.c @@ -0,0 +1,224 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright IBM Corp. 2007 + * + * Authors: Hollis Blanchard <hollisb@us.ibm.com> + */ + +#include <linux/types.h> +#include <linux/string.h> +#include <linux/kvm_host.h> +#include <linux/highmem.h> +#include <asm/mmu-44x.h> +#include <asm/kvm_ppc.h> + +#include "44x_tlb.h" + +#define PPC44x_TLB_USER_PERM_MASK (PPC44x_TLB_UX|PPC44x_TLB_UR|PPC44x_TLB_UW) +#define PPC44x_TLB_SUPER_PERM_MASK (PPC44x_TLB_SX|PPC44x_TLB_SR|PPC44x_TLB_SW) + +static unsigned int kvmppc_tlb_44x_pos; + +static u32 kvmppc_44x_tlb_shadow_attrib(u32 attrib, int usermode) +{ + /* Mask off reserved bits. */ + attrib &= PPC44x_TLB_PERM_MASK|PPC44x_TLB_ATTR_MASK; + + if (!usermode) { + /* Guest is in supervisor mode, so we need to translate guest + * supervisor permissions into user permissions. */ + attrib &= ~PPC44x_TLB_USER_PERM_MASK; + attrib |= (attrib & PPC44x_TLB_SUPER_PERM_MASK) << 3; + } + + /* Make sure host can always access this memory. */ + attrib |= PPC44x_TLB_SX|PPC44x_TLB_SR|PPC44x_TLB_SW; + + return attrib; +} + +/* Search the guest TLB for a matching entry. */ +int kvmppc_44x_tlb_index(struct kvm_vcpu *vcpu, gva_t eaddr, unsigned int pid, + unsigned int as) +{ + int i; + + /* XXX Replace loop with fancy data structures. */ + for (i = 0; i < PPC44x_TLB_SIZE; i++) { + struct tlbe *tlbe = &vcpu->arch.guest_tlb[i]; + unsigned int tid; + + if (eaddr < get_tlb_eaddr(tlbe)) + continue; + + if (eaddr > get_tlb_end(tlbe)) + continue; + + tid = get_tlb_tid(tlbe); + if (tid && (tid != pid)) + continue; + + if (!get_tlb_v(tlbe)) + continue; + + if (get_tlb_ts(tlbe) != as) + continue; + + return i; + } + + return -1; +} + +struct tlbe *kvmppc_44x_itlb_search(struct kvm_vcpu *vcpu, gva_t eaddr) +{ + unsigned int as = !!(vcpu->arch.msr & MSR_IS); + unsigned int index; + + index = kvmppc_44x_tlb_index(vcpu, eaddr, vcpu->arch.pid, as); + if (index == -1) + return NULL; + return &vcpu->arch.guest_tlb[index]; +} + +struct tlbe *kvmppc_44x_dtlb_search(struct kvm_vcpu *vcpu, gva_t eaddr) +{ + unsigned int as = !!(vcpu->arch.msr & MSR_DS); + unsigned int index; + + index = kvmppc_44x_tlb_index(vcpu, eaddr, vcpu->arch.pid, as); + if (index == -1) + return NULL; + return &vcpu->arch.guest_tlb[index]; +} + +static int kvmppc_44x_tlbe_is_writable(struct tlbe *tlbe) +{ + return tlbe->word2 & (PPC44x_TLB_SW|PPC44x_TLB_UW); +} + +/* Must be called with mmap_sem locked for writing. */ +static void kvmppc_44x_shadow_release(struct kvm_vcpu *vcpu, + unsigned int index) +{ + struct tlbe *stlbe = &vcpu->arch.shadow_tlb[index]; + struct page *page = vcpu->arch.shadow_pages[index]; + + kunmap(vcpu->arch.shadow_pages[index]); + + if (get_tlb_v(stlbe)) { + if (kvmppc_44x_tlbe_is_writable(stlbe)) + kvm_release_page_dirty(page); + else + kvm_release_page_clean(page); + } +} + +/* Caller must ensure that the specified guest TLB entry is safe to insert into + * the shadow TLB. */ +void kvmppc_mmu_map(struct kvm_vcpu *vcpu, u64 gvaddr, gfn_t gfn, u64 asid, + u32 flags) +{ + struct page *new_page; + struct tlbe *stlbe; + hpa_t hpaddr; + unsigned int victim; + + /* Future optimization: don't overwrite the TLB entry containing the + * current PC (or stack?). */ + victim = kvmppc_tlb_44x_pos++; + if (kvmppc_tlb_44x_pos > tlb_44x_hwater) + kvmppc_tlb_44x_pos = 0; + stlbe = &vcpu->arch.shadow_tlb[victim]; + + /* Get reference to new page. */ + down_write(¤t->mm->mmap_sem); + new_page = gfn_to_page(vcpu->kvm, gfn); + if (is_error_page(new_page)) { + printk(KERN_ERR "Couldn't get guest page!\n"); + kvm_release_page_clean(new_page); + return; + } + hpaddr = page_to_phys(new_page); + + /* Drop reference to old page. */ + kvmppc_44x_shadow_release(vcpu, victim); + up_write(¤t->mm->mmap_sem); + + vcpu->arch.shadow_pages[victim] = new_page; + + /* XXX Make sure (va, size) doesn't overlap any other + * entries. 440x6 user manual says the result would be + * "undefined." */ + + /* XXX what about AS? */ + + stlbe->tid = asid & 0xff; + + /* Force TS=1 for all guest mappings. */ + /* For now we hardcode 4KB mappings, but it will be important to + * use host large pages in the future. */ + stlbe->word0 = (gvaddr & PAGE_MASK) | PPC44x_TLB_VALID | PPC44x_TLB_TS + | PPC44x_TLB_4K; + + stlbe->word1 = (hpaddr & 0xfffffc00) | ((hpaddr >> 32) & 0xf); + stlbe->word2 = kvmppc_44x_tlb_shadow_attrib(flags, + vcpu->arch.msr & MSR_PR); +} + +void kvmppc_mmu_invalidate(struct kvm_vcpu *vcpu, u64 eaddr, u64 asid) +{ + unsigned int pid = asid & 0xff; + int i; + + /* XXX Replace loop with fancy data structures. */ + down_write(¤t->mm->mmap_sem); + for (i = 0; i <= tlb_44x_hwater; i++) { + struct tlbe *stlbe = &vcpu->arch.shadow_tlb[i]; + unsigned int tid; + + if (!get_tlb_v(stlbe)) + continue; + + if (eaddr < get_tlb_eaddr(stlbe)) + continue; + + if (eaddr > get_tlb_end(stlbe)) + continue; + + tid = get_tlb_tid(stlbe); + if (tid && (tid != pid)) + continue; + + kvmppc_44x_shadow_release(vcpu, i); + stlbe->word0 = 0; + } + up_write(¤t->mm->mmap_sem); +} + +/* Invalidate all mappings, so that when they fault back in they will get the + * proper permission bits. */ +void kvmppc_mmu_priv_switch(struct kvm_vcpu *vcpu, int usermode) +{ + int i; + + /* XXX Replace loop with fancy data structures. */ + down_write(¤t->mm->mmap_sem); + for (i = 0; i <= tlb_44x_hwater; i++) { + kvmppc_44x_shadow_release(vcpu, i); + vcpu->arch.shadow_tlb[i].word0 = 0; + } + up_write(¤t->mm->mmap_sem); +} diff --git a/arch/powerpc/kvm/44x_tlb.h b/arch/powerpc/kvm/44x_tlb.h new file mode 100644 index 000000000000..2ccd46b6f6b7 --- /dev/null +++ b/arch/powerpc/kvm/44x_tlb.h @@ -0,0 +1,91 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright IBM Corp. 2007 + * + * Authors: Hollis Blanchard <hollisb@us.ibm.com> + */ + +#ifndef __KVM_POWERPC_TLB_H__ +#define __KVM_POWERPC_TLB_H__ + +#include <linux/kvm_host.h> +#include <asm/mmu-44x.h> + +extern int kvmppc_44x_tlb_index(struct kvm_vcpu *vcpu, gva_t eaddr, + unsigned int pid, unsigned int as); +extern struct tlbe *kvmppc_44x_dtlb_search(struct kvm_vcpu *vcpu, gva_t eaddr); +extern struct tlbe *kvmppc_44x_itlb_search(struct kvm_vcpu *vcpu, gva_t eaddr); + +/* TLB helper functions */ +static inline unsigned int get_tlb_size(const struct tlbe *tlbe) +{ + return (tlbe->word0 >> 4) & 0xf; +} + +static inline gva_t get_tlb_eaddr(const struct tlbe *tlbe) +{ + return tlbe->word0 & 0xfffffc00; +} + +static inline gva_t get_tlb_bytes(const struct tlbe *tlbe) +{ + unsigned int pgsize = get_tlb_size(tlbe); + return 1 << 10 << (pgsize << 1); +} + +static inline gva_t get_tlb_end(const struct tlbe *tlbe) +{ + return get_tlb_eaddr(tlbe) + get_tlb_bytes(tlbe) - 1; +} + +static inline u64 get_tlb_raddr(const struct tlbe *tlbe) +{ + u64 word1 = tlbe->word1; + return ((word1 & 0xf) << 32) | (word1 & 0xfffffc00); +} + +static inline unsigned int get_tlb_tid(const struct tlbe *tlbe) +{ + return tlbe->tid & 0xff; +} + +static inline unsigned int get_tlb_ts(const struct tlbe *tlbe) +{ + return (tlbe->word0 >> 8) & 0x1; +} + +static inline unsigned int get_tlb_v(const struct tlbe *tlbe) +{ + return (tlbe->word0 >> 9) & 0x1; +} + +static inline unsigned int get_mmucr_stid(const struct kvm_vcpu *vcpu) +{ + return vcpu->arch.mmucr & 0xff; +} + +static inline unsigned int get_mmucr_sts(const struct kvm_vcpu *vcpu) +{ + return (vcpu->arch.mmucr >> 16) & 0x1; +} + +static inline gpa_t tlb_xlate(struct tlbe *tlbe, gva_t eaddr) +{ + unsigned int pgmask = get_tlb_bytes(tlbe) - 1; + + return get_tlb_raddr(tlbe) | (eaddr & pgmask); +} + +#endif /* __KVM_POWERPC_TLB_H__ */ diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig new file mode 100644 index 000000000000..6b076010213b --- /dev/null +++ b/arch/powerpc/kvm/Kconfig @@ -0,0 +1,42 @@ +# +# KVM configuration +# + +menuconfig VIRTUALIZATION + bool "Virtualization" + ---help--- + Say Y here to get to see options for using your Linux host to run + other operating systems inside virtual machines (guests). + This option alone does not add any kernel code. + + If you say N, all options in this submenu will be skipped and + disabled. + +if VIRTUALIZATION + +config KVM + bool "Kernel-based Virtual Machine (KVM) support" + depends on 44x && EXPERIMENTAL + select PREEMPT_NOTIFIERS + select ANON_INODES + # We can only run on Book E hosts so far + select KVM_BOOKE_HOST + ---help--- + Support hosting virtualized guest machines. You will also + need to select one or more of the processor modules below. + + This module provides access to the hardware capabilities through + a character device node named /dev/kvm. + + If unsure, say N. + +config KVM_BOOKE_HOST + bool "KVM host support for Book E PowerPC processors" + depends on KVM && 44x + ---help--- + Provides host support for KVM on Book E PowerPC processors. Currently + this works on 440 processors only. + +source drivers/virtio/Kconfig + +endif # VIRTUALIZATION diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile new file mode 100644 index 000000000000..d0d358d367ec --- /dev/null +++ b/arch/powerpc/kvm/Makefile @@ -0,0 +1,15 @@ +# +# Makefile for Kernel-based Virtual Machine module +# + +EXTRA_CFLAGS += -Ivirt/kvm -Iarch/powerpc/kvm + +common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o) + +kvm-objs := $(common-objs) powerpc.o emulate.o booke_guest.o +obj-$(CONFIG_KVM) += kvm.o + +AFLAGS_booke_interrupts.o := -I$(obj) + +kvm-booke-host-objs := booke_host.o booke_interrupts.o 44x_tlb.o +obj-$(CONFIG_KVM_BOOKE_HOST) += kvm-booke-host.o diff --git a/arch/powerpc/kvm/booke_guest.c b/arch/powerpc/kvm/booke_guest.c new file mode 100644 index 000000000000..6d9884a6884a --- /dev/null +++ b/arch/powerpc/kvm/booke_guest.c @@ -0,0 +1,615 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright IBM Corp. 2007 + * + * Authors: Hollis Blanchard <hollisb@us.ibm.com> + * Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com> + */ + +#include <linux/errno.h> +#include <linux/err.h> +#include <linux/kvm_host.h> +#include <linux/module.h> +#include <linux/vmalloc.h> +#include <linux/fs.h> +#include <asm/cputable.h> +#include <asm/uaccess.h> +#include <asm/kvm_ppc.h> + +#include "44x_tlb.h" + +#define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM +#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU + +struct kvm_stats_debugfs_item debugfs_entries[] = { + { "exits", VCPU_STAT(sum_exits) }, + { "mmio", VCPU_STAT(mmio_exits) }, + { "dcr", VCPU_STAT(dcr_exits) }, + { "sig", VCPU_STAT(signal_exits) }, + { "light", VCPU_STAT(light_exits) }, + { "itlb_r", VCPU_STAT(itlb_real_miss_exits) }, + { "itlb_v", VCPU_STAT(itlb_virt_miss_exits) }, + { "dtlb_r", VCPU_STAT(dtlb_real_miss_exits) }, + { "dtlb_v", VCPU_STAT(dtlb_virt_miss_exits) }, + { "sysc", VCPU_STAT(syscall_exits) }, + { "isi", VCPU_STAT(isi_exits) }, + { "dsi", VCPU_STAT(dsi_exits) }, + { "inst_emu", VCPU_STAT(emulated_inst_exits) }, + { "dec", VCPU_STAT(dec_exits) }, + { "ext_intr", VCPU_STAT(ext_intr_exits) }, + { NULL } +}; + +static const u32 interrupt_msr_mask[16] = { + [BOOKE_INTERRUPT_CRITICAL] = MSR_ME, + [BOOKE_INTERRUPT_MACHINE_CHECK] = 0, + [BOOKE_INTERRUPT_DATA_STORAGE] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_INST_STORAGE] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_EXTERNAL] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_ALIGNMENT] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_PROGRAM] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_FP_UNAVAIL] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_SYSCALL] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_AP_UNAVAIL] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_DECREMENTER] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_FIT] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_WATCHDOG] = MSR_ME, + [BOOKE_INTERRUPT_DTLB_MISS] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_ITLB_MISS] = MSR_CE|MSR_ME|MSR_DE, + [BOOKE_INTERRUPT_DEBUG] = MSR_ME, +}; + +const unsigned char exception_priority[] = { + [BOOKE_INTERRUPT_DATA_STORAGE] = 0, + [BOOKE_INTERRUPT_INST_STORAGE] = 1, + [BOOKE_INTERRUPT_ALIGNMENT] = 2, + [BOOKE_INTERRUPT_PROGRAM] = 3, + [BOOKE_INTERRUPT_FP_UNAVAIL] = 4, + [BOOKE_INTERRUPT_SYSCALL] = 5, + [BOOKE_INTERRUPT_AP_UNAVAIL] = 6, + [BOOKE_INTERRUPT_DTLB_MISS] = 7, + [BOOKE_INTERRUPT_ITLB_MISS] = 8, + [BOOKE_INTERRUPT_MACHINE_CHECK] = 9, + [BOOKE_INTERRUPT_DEBUG] = 10, + [BOOKE_INTERRUPT_CRITICAL] = 11, + [BOOKE_INTERRUPT_WATCHDOG] = 12, + [BOOKE_INTERRUPT_EXTERNAL] = 13, + [BOOKE_INTERRUPT_FIT] = 14, + [BOOKE_INTERRUPT_DECREMENTER] = 15, +}; + +const unsigned char priority_exception[] = { + BOOKE_INTERRUPT_DATA_STORAGE, + BOOKE_INTERRUPT_INST_STORAGE, + BOOKE_INTERRUPT_ALIGNMENT, + BOOKE_INTERRUPT_PROGRAM, + BOOKE_INTERRUPT_FP_UNAVAIL, + BOOKE_INTERRUPT_SYSCALL, + BOOKE_INTERRUPT_AP_UNAVAIL, + BOOKE_INTERRUPT_DTLB_MISS, + BOOKE_INTERRUPT_ITLB_MISS, + BOOKE_INTERRUPT_MACHINE_CHECK, + BOOKE_INTERRUPT_DEBUG, + BOOKE_INTERRUPT_CRITICAL, + BOOKE_INTERRUPT_WATCHDOG, + BOOKE_INTERRUPT_EXTERNAL, + BOOKE_INTERRUPT_FIT, + BOOKE_INTERRUPT_DECREMENTER, +}; + + +void kvmppc_dump_tlbs(struct kvm_vcpu *vcpu) +{ + struct tlbe *tlbe; + int i; + + printk("vcpu %d TLB dump:\n", vcpu->vcpu_id); + printk("| %2s | %3s | %8s | %8s | %8s |\n", + "nr", "tid", "word0", "word1", "word2"); + + for (i = 0; i < PPC44x_TLB_SIZE; i++) { + tlbe = &vcpu->arch.guest_tlb[i]; + if (tlbe->word0 & PPC44x_TLB_VALID) + printk(" G%2d | %02X | %08X | %08X | %08X |\n", + i, tlbe->tid, tlbe->word0, tlbe->word1, + tlbe->word2); + } + + for (i = 0; i < PPC44x_TLB_SIZE; i++) { + tlbe = &vcpu->arch.shadow_tlb[i]; + if (tlbe->word0 & PPC44x_TLB_VALID) + printk(" S%2d | %02X | %08X | %08X | %08X |\n", + i, tlbe->tid, tlbe->word0, tlbe->word1, + tlbe->word2); + } +} + +/* TODO: use vcpu_printf() */ +void kvmppc_dump_vcpu(struct kvm_vcpu *vcpu) +{ + int i; + + printk("pc: %08x msr: %08x\n", vcpu->arch.pc, vcpu->arch.msr); + printk("lr: %08x ctr: %08x\n", vcpu->arch.lr, vcpu->arch.ctr); + printk("srr0: %08x srr1: %08x\n", vcpu->arch.srr0, vcpu->arch.srr1); + + printk("exceptions: %08lx\n", vcpu->arch.pending_exceptions); + + for (i = 0; i < 32; i += 4) { + printk("gpr%02d: %08x %08x %08x %08x\n", i, + vcpu->arch.gpr[i], + vcpu->arch.gpr[i+1], + vcpu->arch.gpr[i+2], + vcpu->arch.gpr[i+3]); + } +} + +/* Check if we are ready to deliver the interrupt */ +static int kvmppc_can_deliver_interrupt(struct kvm_vcpu *vcpu, int interrupt) +{ + int r; + + switch (interrupt) { + case BOOKE_INTERRUPT_CRITICAL: + r = vcpu->arch.msr & MSR_CE; + break; + case BOOKE_INTERRUPT_MACHINE_CHECK: + r = vcpu->arch.msr & MSR_ME; + break; + case BOOKE_INTERRUPT_EXTERNAL: + r = vcpu->arch.msr & MSR_EE; + break; + case BOOKE_INTERRUPT_DECREMENTER: + r = vcpu->arch.msr & MSR_EE; + break; + case BOOKE_INTERRUPT_FIT: + r = vcpu->arch.msr & MSR_EE; + break; + case BOOKE_INTERRUPT_WATCHDOG: + r = vcpu->arch.msr & MSR_CE; + break; + case BOOKE_INTERRUPT_DEBUG: + r = vcpu->arch.msr & MSR_DE; + break; + default: + r = 1; + } + + return r; +} + +static void kvmppc_deliver_interrupt(struct kvm_vcpu *vcpu, int interrupt) +{ + switch (interrupt) { + case BOOKE_INTERRUPT_DECREMENTER: + vcpu->arch.tsr |= TSR_DIS; + break; + } + + vcpu->arch.srr0 = vcpu->arch.pc; + vcpu->arch.srr1 = vcpu->arch.msr; + vcpu->arch.pc = vcpu->arch.ivpr | vcpu->arch.ivor[interrupt]; + kvmppc_set_msr(vcpu, vcpu->arch.msr & interrupt_msr_mask[interrupt]); +} + +/* Check pending exceptions and deliver one, if possible. */ +void kvmppc_check_and_deliver_interrupts(struct kvm_vcpu *vcpu) +{ + unsigned long *pending = &vcpu->arch.pending_exceptions; + unsigned int exception; + unsigned int priority; + + priority = find_first_bit(pending, BITS_PER_BYTE * sizeof(*pending)); + while (priority <= BOOKE_MAX_INTERRUPT) { + exception = priority_exception[priority]; + if (kvmppc_can_deliver_interrupt(vcpu, exception)) { + kvmppc_clear_exception(vcpu, exception); + kvmppc_deliver_interrupt(vcpu, exception); + break; + } + + priority = find_next_bit(pending, + BITS_PER_BYTE * sizeof(*pending), + priority + 1); + } +} + +static int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu) +{ + enum emulation_result er; + int r; + + er = kvmppc_emulate_instruction(run, vcpu); + switch (er) { + case EMULATE_DONE: + /* Future optimization: only reload non-volatiles if they were + * actually modified. */ + r = RESUME_GUEST_NV; + break; + case EMULATE_DO_MMIO: + run->exit_reason = KVM_EXIT_MMIO; + /* We must reload nonvolatiles because "update" load/store + * instructions modify register state. */ + /* Future optimization: only reload non-volatiles if they were + * actually modified. */ + r = RESUME_HOST_NV; + break; + case EMULATE_FAIL: + /* XXX Deliver Program interrupt to guest. */ + printk(KERN_EMERG "%s: emulation failed (%08x)\n", __func__, + vcpu->arch.last_inst); + r = RESUME_HOST; + break; + default: + BUG(); + } + + return r; +} + +/** + * kvmppc_handle_exit + * + * Return value is in the form (errcode<<2 | RESUME_FLAG_HOST | RESUME_FLAG_NV) + */ +int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu, + unsigned int exit_nr) +{ + enum emulation_result er; + int r = RESUME_HOST; + + local_irq_enable(); + + run->exit_reason = KVM_EXIT_UNKNOWN; + run->ready_for_interrupt_injection = 1; + + switch (exit_nr) { + case BOOKE_INTERRUPT_MACHINE_CHECK: + printk("MACHINE CHECK: %lx\n", mfspr(SPRN_MCSR)); + kvmppc_dump_vcpu(vcpu); + r = RESUME_HOST; + break; + + case BOOKE_INTERRUPT_EXTERNAL: + case BOOKE_INTERRUPT_DECREMENTER: + /* Since we switched IVPR back to the host's value, the host + * handled this interrupt the moment we enabled interrupts. + * Now we just offer it a chance to reschedule the guest. */ + + /* XXX At this point the TLB still holds our shadow TLB, so if + * we do reschedule the host will fault over it. Perhaps we + * should politely restore the host's entries to minimize + * misses before ceding control. */ + if (need_resched()) + cond_resched(); + if (exit_nr == BOOKE_INTERRUPT_DECREMENTER) + vcpu->stat.dec_exits++; + else + vcpu->stat.ext_intr_exits++; + r = RESUME_GUEST; + break; + + case BOOKE_INTERRUPT_PROGRAM: + if (vcpu->arch.msr & MSR_PR) { + /* Program traps generated by user-level software must be handled + * by the guest kernel. */ + vcpu->arch.esr = vcpu->arch.fault_esr; + kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_PROGRAM); + r = RESUME_GUEST; + break; + } + + er = kvmppc_emulate_instruction(run, vcpu); + switch (er) { + case EMULATE_DONE: + /* Future optimization: only reload non-volatiles if + * they were actually modified by emulation. */ + vcpu->stat.emulated_inst_exits++; + r = RESUME_GUEST_NV; + break; + case EMULATE_DO_DCR: + run->exit_reason = KVM_EXIT_DCR; + r = RESUME_HOST; + break; + case EMULATE_FAIL: + /* XXX Deliver Program interrupt to guest. */ + printk(KERN_CRIT "%s: emulation at %x failed (%08x)\n", + __func__, vcpu->arch.pc, vcpu->arch.last_inst); + /* For debugging, encode the failing instruction and + * report it to userspace. */ + run->hw.hardware_exit_reason = ~0ULL << 32; + run->hw.hardware_exit_reason |= vcpu->arch.last_inst; + r = RESUME_HOST; + break; + default: + BUG(); + } + break; + + case BOOKE_INTERRUPT_DATA_STORAGE: + vcpu->arch.dear = vcpu->arch.fault_dear; + vcpu->arch.esr = vcpu->arch.fault_esr; + kvmppc_queue_exception(vcpu, exit_nr); + vcpu->stat.dsi_exits++; + r = RESUME_GUEST; + break; + + case BOOKE_INTERRUPT_INST_STORAGE: + vcpu->arch.esr = vcpu->arch.fault_esr; + kvmppc_queue_exception(vcpu, exit_nr); + vcpu->stat.isi_exits++; + r = RESUME_GUEST; + break; + + case BOOKE_INTERRUPT_SYSCALL: + kvmppc_queue_exception(vcpu, exit_nr); + vcpu->stat.syscall_exits++; + r = RESUME_GUEST; + break; + + case BOOKE_INTERRUPT_DTLB_MISS: { + struct tlbe *gtlbe; + unsigned long eaddr = vcpu->arch.fault_dear; + gfn_t gfn; + + /* Check the guest TLB. */ + gtlbe = kvmppc_44x_dtlb_search(vcpu, eaddr); + if (!gtlbe) { + /* The guest didn't have a mapping for it. */ + kvmppc_queue_exception(vcpu, exit_nr); + vcpu->arch.dear = vcpu->arch.fault_dear; + vcpu->arch.esr = vcpu->arch.fault_esr; + vcpu->stat.dtlb_real_miss_exits++; + r = RESUME_GUEST; + break; + } + + vcpu->arch.paddr_accessed = tlb_xlate(gtlbe, eaddr); + gfn = vcpu->arch.paddr_accessed >> PAGE_SHIFT; + + if (kvm_is_visible_gfn(vcpu->kvm, gfn)) { + /* The guest TLB had a mapping, but the shadow TLB + * didn't, and it is RAM. This could be because: + * a) the entry is mapping the host kernel, or + * b) the guest used a large mapping which we're faking + * Either way, we need to satisfy the fault without + * invoking the guest. */ + kvmppc_mmu_map(vcpu, eaddr, gfn, gtlbe->tid, + gtlbe->word2); + vcpu->stat.dtlb_virt_miss_exits++; + r = RESUME_GUEST; + } else { + /* Guest has mapped and accessed a page which is not + * actually RAM. */ + r = kvmppc_emulate_mmio(run, vcpu); + } + + break; + } + + case BOOKE_INTERRUPT_ITLB_MISS: { + struct tlbe *gtlbe; + unsigned long eaddr = vcpu->arch.pc; + gfn_t gfn; + + r = RESUME_GUEST; + + /* Check the guest TLB. */ + gtlbe = kvmppc_44x_itlb_search(vcpu, eaddr); + if (!gtlbe) { + /* The guest didn't have a mapping for it. */ + kvmppc_queue_exception(vcpu, exit_nr); + vcpu->stat.itlb_real_miss_exits++; + break; + } + + vcpu->stat.itlb_virt_miss_exits++; + + gfn = tlb_xlate(gtlbe, eaddr) >> PAGE_SHIFT; + + if (kvm_is_visible_gfn(vcpu->kvm, gfn)) { + /* The guest TLB had a mapping, but the shadow TLB + * didn't. This could be because: + * a) the entry is mapping the host kernel, or + * b) the guest used a large mapping which we're faking + * Either way, we need to satisfy the fault without + * invoking the guest. */ + kvmppc_mmu_map(vcpu, eaddr, gfn, gtlbe->tid, + gtlbe->word2); + } else { + /* Guest mapped and leaped at non-RAM! */ + kvmppc_queue_exception(vcpu, + BOOKE_INTERRUPT_MACHINE_CHECK); + } + + break; + } + + default: + printk(KERN_EMERG "exit_nr %d\n", exit_nr); + BUG(); + } + + local_irq_disable(); + + kvmppc_check_and_deliver_interrupts(vcpu); + + /* Do some exit accounting. */ + vcpu->stat.sum_exits++; + if (!(r & RESUME_HOST)) { + /* To avoid clobbering exit_reason, only check for signals if + * we aren't already exiting to userspace for some other + * reason. */ + if (signal_pending(current)) { + run->exit_reason = KVM_EXIT_INTR; + r = (-EINTR << 2) | RESUME_HOST | (r & RESUME_FLAG_NV); + + vcpu->stat.signal_exits++; + } else { + vcpu->stat.light_exits++; + } + } else { + switch (run->exit_reason) { + case KVM_EXIT_MMIO: + vcpu->stat.mmio_exits++; + break; + case KVM_EXIT_DCR: + vcpu->stat.dcr_exits++; + break; + case KVM_EXIT_INTR: + vcpu->stat.signal_exits++; + break; + } + } + + return r; +} + +/* Initial guest state: 16MB mapping 0 -> 0, PC = 0, MSR = 0, R1 = 16MB */ +int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) +{ + struct tlbe *tlbe = &vcpu->arch.guest_tlb[0]; + + tlbe->tid = 0; + tlbe->word0 = PPC44x_TLB_16M | PPC44x_TLB_VALID; + tlbe->word1 = 0; + tlbe->word2 = PPC44x_TLB_SX | PPC44x_TLB_SW | PPC44x_TLB_SR; + + tlbe++; + tlbe->tid = 0; + tlbe->word0 = 0xef600000 | PPC44x_TLB_4K | PPC44x_TLB_VALID; + tlbe->word1 = 0xef600000; + tlbe->word2 = PPC44x_TLB_SX | PPC44x_TLB_SW | PPC44x_TLB_SR + | PPC44x_TLB_I | PPC44x_TLB_G; + + vcpu->arch.pc = 0; + vcpu->arch.msr = 0; + vcpu->arch.gpr[1] = (16<<20) - 8; /* -8 for the callee-save LR slot */ + + /* Eye-catching number so we know if the guest takes an interrupt + * before it's programmed its own IVPR. */ + vcpu->arch.ivpr = 0x55550000; + + /* Since the guest can directly access the timebase, it must know the + * real timebase frequency. Accordingly, it must see the state of + * CCR1[TCS]. */ + vcpu->arch.ccr1 = mfspr(SPRN_CCR1); + + return 0; +} + +int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) +{ + int i; + + regs->pc = vcpu->arch.pc; + regs->cr = vcpu->arch.cr; + regs->ctr = vcpu->arch.ctr; + regs->lr = vcpu->arch.lr; + regs->xer = vcpu->arch.xer; + regs->msr = vcpu->arch.msr; + regs->srr0 = vcpu->arch.srr0; + regs->srr1 = vcpu->arch.srr1; + regs->pid = vcpu->arch.pid; + regs->sprg0 = vcpu->arch.sprg0; + regs->sprg1 = vcpu->arch.sprg1; + regs->sprg2 = vcpu->arch.sprg2; + regs->sprg3 = vcpu->arch.sprg3; + regs->sprg5 = vcpu->arch.sprg4; + regs->sprg6 = vcpu->arch.sprg5; + regs->sprg7 = vcpu->arch.sprg6; + + for (i = 0; i < ARRAY_SIZE(regs->gpr); i++) + regs->gpr[i] = vcpu->arch.gpr[i]; + + return 0; +} + +int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) +{ + int i; + + vcpu->arch.pc = regs->pc; + vcpu->arch.cr = regs->cr; + vcpu->arch.ctr = regs->ctr; + vcpu->arch.lr = regs->lr; + vcpu->arch.xer = regs->xer; + vcpu->arch.msr = regs->msr; + vcpu->arch.srr0 = regs->srr0; + vcpu->arch.srr1 = regs->srr1; + vcpu->arch.sprg0 = regs->sprg0; + vcpu->arch.sprg1 = regs->sprg1; + vcpu->arch.sprg2 = regs->sprg2; + vcpu->arch.sprg3 = regs->sprg3; + vcpu->arch.sprg5 = regs->sprg4; + vcpu->arch.sprg6 = regs->sprg5; + vcpu->arch.sprg7 = regs->sprg6; + + for (i = 0; i < ARRAY_SIZE(vcpu->arch.gpr); i++) + vcpu->arch.gpr[i] = regs->gpr[i]; + + return 0; +} + +int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu, + struct kvm_sregs *sregs) +{ + return -ENOTSUPP; +} + +int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu, + struct kvm_sregs *sregs) +{ + return -ENOTSUPP; +} + +int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) +{ + return -ENOTSUPP; +} + +int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) +{ + return -ENOTSUPP; +} + +/* 'linear_address' is actually an encoding of AS|PID|EADDR . */ +int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu, + struct kvm_translation *tr) +{ + struct tlbe *gtlbe; + int index; + gva_t eaddr; + u8 pid; + u8 as; + + eaddr = tr->linear_address; + pid = (tr->linear_address >> 32) & 0xff; + as = (tr->linear_address >> 40) & 0x1; + + index = kvmppc_44x_tlb_index(vcpu, eaddr, pid, as); + if (index == -1) { + tr->valid = 0; + return 0; + } + + gtlbe = &vcpu->arch.guest_tlb[index]; + + tr->physical_address = tlb_xlate(gtlbe, eaddr); + /* XXX what does "writeable" and "usermode" even mean? */ + tr->valid = 1; + + return 0; +} diff --git a/arch/powerpc/kvm/booke_host.c b/arch/powerpc/kvm/booke_host.c new file mode 100644 index 000000000000..b480341bc31e --- /dev/null +++ b/arch/powerpc/kvm/booke_host.c @@ -0,0 +1,83 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright IBM Corp. 2008 + * + * Authors: Hollis Blanchard <hollisb@us.ibm.com> + */ + +#include <linux/errno.h> +#include <linux/kvm_host.h> +#include <linux/module.h> +#include <asm/cacheflush.h> +#include <asm/kvm_ppc.h> + +unsigned long kvmppc_booke_handlers; + +static int kvmppc_booke_init(void) +{ + unsigned long ivor[16]; + unsigned long max_ivor = 0; + int i; + + /* We install our own exception handlers by hijacking IVPR. IVPR must + * be 16-bit aligned, so we need a 64KB allocation. */ + kvmppc_booke_handlers = __get_free_pages(GFP_KERNEL | __GFP_ZERO, + VCPU_SIZE_ORDER); + if (!kvmppc_booke_handlers) + return -ENOMEM; + + /* XXX make sure our handlers are smaller than Linux's */ + + /* Copy our interrupt handlers to match host IVORs. That way we don't + * have to swap the IVORs on every guest/host transition. */ + ivor[0] = mfspr(SPRN_IVOR0); + ivor[1] = mfspr(SPRN_IVOR1); + ivor[2] = mfspr(SPRN_IVOR2); + ivor[3] = mfspr(SPRN_IVOR3); + ivor[4] = mfspr(SPRN_IVOR4); + ivor[5] = mfspr(SPRN_IVOR5); + ivor[6] = mfspr(SPRN_IVOR6); + ivor[7] = mfspr(SPRN_IVOR7); + ivor[8] = mfspr(SPRN_IVOR8); + ivor[9] = mfspr(SPRN_IVOR9); + ivor[10] = mfspr(SPRN_IVOR10); + ivor[11] = mfspr(SPRN_IVOR11); + ivor[12] = mfspr(SPRN_IVOR12); + ivor[13] = mfspr(SPRN_IVOR13); + ivor[14] = mfspr(SPRN_IVOR14); + ivor[15] = mfspr(SPRN_IVOR15); + + for (i = 0; i < 16; i++) { + if (ivor[i] > max_ivor) + max_ivor = ivor[i]; + + memcpy((void *)kvmppc_booke_handlers + ivor[i], + kvmppc_handlers_start + i * kvmppc_handler_len, + kvmppc_handler_len); + } + flush_icache_range(kvmppc_booke_handlers, + kvmppc_booke_handlers + max_ivor + kvmppc_handler_len); + + return kvm_init(NULL, sizeof(struct kvm_vcpu), THIS_MODULE); +} + +static void __exit kvmppc_booke_exit(void) +{ + free_pages(kvmppc_booke_handlers, VCPU_SIZE_ORDER); + kvm_exit(); +} + +module_init(kvmppc_booke_init) +module_exit(kvmppc_booke_exit) diff --git a/arch/powerpc/kvm/booke_interrupts.S b/arch/powerpc/kvm/booke_interrupts.S new file mode 100644 index 000000000000..3b653b5309b8 --- /dev/null +++ b/arch/powerpc/kvm/booke_interrupts.S @@ -0,0 +1,436 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright IBM Corp. 2007 + * + * Authors: Hollis Blanchard <hollisb@us.ibm.com> + */ + +#include <asm/ppc_asm.h> +#include <asm/kvm_asm.h> +#include <asm/reg.h> +#include <asm/mmu-44x.h> +#include <asm/page.h> +#include <asm/asm-offsets.h> + +#define KVMPPC_MSR_MASK (MSR_CE|MSR_EE|MSR_PR|MSR_DE|MSR_ME|MSR_IS|MSR_DS) + +#define VCPU_GPR(n) (VCPU_GPRS + (n * 4)) + +/* The host stack layout: */ +#define HOST_R1 0 /* Implied by stwu. */ +#define HOST_CALLEE_LR 4 +#define HOST_RUN 8 +/* r2 is special: it holds 'current', and it made nonvolatile in the + * kernel with the -ffixed-r2 gcc option. */ +#define HOST_R2 12 +#define HOST_NV_GPRS 16 +#define HOST_NV_GPR(n) (HOST_NV_GPRS + ((n - 14) * 4)) +#define HOST_MIN_STACK_SIZE (HOST_NV_GPR(31) + 4) +#define HOST_STACK_SIZE (((HOST_MIN_STACK_SIZE + 15) / 16) * 16) /* Align. */ +#define HOST_STACK_LR (HOST_STACK_SIZE + 4) /* In caller stack frame. */ + +#define NEED_INST_MASK ((1<<BOOKE_INTERRUPT_PROGRAM) | \ + (1<<BOOKE_INTERRUPT_DTLB_MISS)) + +#define NEED_DEAR_MASK ((1<<BOOKE_INTERRUPT_DATA_STORAGE) | \ + (1<<BOOKE_INTERRUPT_DTLB_MISS)) + +#define NEED_ESR_MASK ((1<<BOOKE_INTERRUPT_DATA_STORAGE) | \ + (1<<BOOKE_INTERRUPT_INST_STORAGE) | \ + (1<<BOOKE_INTERRUPT_PROGRAM) | \ + (1<<BOOKE_INTERRUPT_DTLB_MISS)) + +.macro KVM_HANDLER ivor_nr +_GLOBAL(kvmppc_handler_\ivor_nr) + /* Get pointer to vcpu and record exit number. */ + mtspr SPRN_SPRG0, r4 + mfspr r4, SPRN_SPRG1 + stw r5, VCPU_GPR(r5)(r4) + stw r6, VCPU_GPR(r6)(r4) + mfctr r5 + lis r6, kvmppc_resume_host@h + stw r5, VCPU_CTR(r4) + li r5, \ivor_nr + ori r6, r6, kvmppc_resume_host@l + mtctr r6 + bctr +.endm + +_GLOBAL(kvmppc_handlers_start) +KVM_HANDLER BOOKE_INTERRUPT_CRITICAL +KVM_HANDLER BOOKE_INTERRUPT_MACHINE_CHECK +KVM_HANDLER BOOKE_INTERRUPT_DATA_STORAGE +KVM_HANDLER BOOKE_INTERRUPT_INST_STORAGE +KVM_HANDLER BOOKE_INTERRUPT_EXTERNAL +KVM_HANDLER BOOKE_INTERRUPT_ALIGNMENT +KVM_HANDLER BOOKE_INTERRUPT_PROGRAM +KVM_HANDLER BOOKE_INTERRUPT_FP_UNAVAIL +KVM_HANDLER BOOKE_INTERRUPT_SYSCALL +KVM_HANDLER BOOKE_INTERRUPT_AP_UNAVAIL +KVM_HANDLER BOOKE_INTERRUPT_DECREMENTER +KVM_HANDLER BOOKE_INTERRUPT_FIT +KVM_HANDLER BOOKE_INTERRUPT_WATCHDOG +KVM_HANDLER BOOKE_INTERRUPT_DTLB_MISS +KVM_HANDLER BOOKE_INTERRUPT_ITLB_MISS +KVM_HANDLER BOOKE_INTERRUPT_DEBUG + +_GLOBAL(kvmppc_handler_len) + .long kvmppc_handler_1 - kvmppc_handler_0 + + +/* Registers: + * SPRG0: guest r4 + * r4: vcpu pointer + * r5: KVM exit number + */ +_GLOBAL(kvmppc_resume_host) + stw r3, VCPU_GPR(r3)(r4) + mfcr r3 + stw r3, VCPU_CR(r4) + stw r7, VCPU_GPR(r7)(r4) + stw r8, VCPU_GPR(r8)(r4) + stw r9, VCPU_GPR(r9)(r4) + + li r6, 1 + slw r6, r6, r5 + + /* Save the faulting instruction and all GPRs for emulation. */ + andi. r7, r6, NEED_INST_MASK + beq ..skip_inst_copy + mfspr r9, SPRN_SRR0 + mfmsr r8 + ori r7, r8, MSR_DS + mtmsr r7 + isync + lwz r9, 0(r9) + mtmsr r8 + isync + stw r9, VCPU_LAST_INST(r4) + + stw r15, VCPU_GPR(r15)(r4) + stw r16, VCPU_GPR(r16)(r4) + stw r17, VCPU_GPR(r17)(r4) + stw r18, VCPU_GPR(r18)(r4) + stw r19, VCPU_GPR(r19)(r4) + stw r20, VCPU_GPR(r20)(r4) + stw r21, VCPU_GPR(r21)(r4) + stw r22, VCPU_GPR(r22)(r4) + stw r23, VCPU_GPR(r23)(r4) + stw r24, VCPU_GPR(r24)(r4) + stw r25, VCPU_GPR(r25)(r4) + stw r26, VCPU_GPR(r26)(r4) + stw r27, VCPU_GPR(r27)(r4) + stw r28, VCPU_GPR(r28)(r4) + stw r29, VCPU_GPR(r29)(r4) + stw r30, VCPU_GPR(r30)(r4) + stw r31, VCPU_GPR(r31)(r4) +..skip_inst_copy: + + /* Also grab DEAR and ESR before the host can clobber them. */ + + andi. r7, r6, NEED_DEAR_MASK + beq ..skip_dear + mfspr r9, SPRN_DEAR + stw r9, VCPU_FAULT_DEAR(r4) +..skip_dear: + + andi. r7, r6, NEED_ESR_MASK + beq ..skip_esr + mfspr r9, SPRN_ESR + stw r9, VCPU_FAULT_ESR(r4) +..skip_esr: + + /* Save remaining volatile guest register state to vcpu. */ + stw r0, VCPU_GPR(r0)(r4) + stw r1, VCPU_GPR(r1)(r4) + stw r2, VCPU_GPR(r2)(r4) + stw r10, VCPU_GPR(r10)(r4) + stw r11, VCPU_GPR(r11)(r4) + stw r12, VCPU_GPR(r12)(r4) + stw r13, VCPU_GPR(r13)(r4) + stw r14, VCPU_GPR(r14)(r4) /* We need a NV GPR below. */ + mflr r3 + stw r3, VCPU_LR(r4) + mfxer r3 + stw r3, VCPU_XER(r4) + mfspr r3, SPRN_SPRG0 + stw r3, VCPU_GPR(r4)(r4) + mfspr r3, SPRN_SRR0 + stw r3, VCPU_PC(r4) + + /* Restore host stack pointer and PID before IVPR, since the host + * exception handlers use them. */ + lwz r1, VCPU_HOST_STACK(r4) + lwz r3, VCPU_HOST_PID(r4) + mtspr SPRN_PID, r3 + + /* Restore host IVPR before re-enabling interrupts. We cheat and know + * that Linux IVPR is always 0xc0000000. */ + lis r3, 0xc000 + mtspr SPRN_IVPR, r3 + + /* Switch to kernel stack and jump to handler. */ + LOAD_REG_ADDR(r3, kvmppc_handle_exit) + mtctr r3 + lwz r3, HOST_RUN(r1) + lwz r2, HOST_R2(r1) + mr r14, r4 /* Save vcpu pointer. */ + + bctrl /* kvmppc_handle_exit() */ + + /* Restore vcpu pointer and the nonvolatiles we used. */ + mr r4, r14 + lwz r14, VCPU_GPR(r14)(r4) + + /* Sometimes instruction emulation must restore complete GPR state. */ + andi. r5, r3, RESUME_FLAG_NV + beq ..skip_nv_load + lwz r15, VCPU_GPR(r15)(r4) + lwz r16, VCPU_GPR(r16)(r4) + lwz r17, VCPU_GPR(r17)(r4) + lwz r18, VCPU_GPR(r18)(r4) + lwz r19, VCPU_GPR(r19)(r4) + lwz r20, VCPU_GPR(r20)(r4) + lwz r21, VCPU_GPR(r21)(r4) + lwz r22, VCPU_GPR(r22)(r4) + lwz r23, VCPU_GPR(r23)(r4) + lwz r24, VCPU_GPR(r24)(r4) + lwz r25, VCPU_GPR(r25)(r4) + lwz r26, VCPU_GPR(r26)(r4) + lwz r27, VCPU_GPR(r27)(r4) + lwz r28, VCPU_GPR(r28)(r4) + lwz r29, VCPU_GPR(r29)(r4) + lwz r30, VCPU_GPR(r30)(r4) + lwz r31, VCPU_GPR(r31)(r4) +..skip_nv_load: + + /* Should we return to the guest? */ + andi. r5, r3, RESUME_FLAG_HOST + beq lightweight_exit + + srawi r3, r3, 2 /* Shift -ERR back down. */ + +heavyweight_exit: + /* Not returning to guest. */ + + /* We already saved guest volatile register state; now save the + * non-volatiles. */ + stw r15, VCPU_GPR(r15)(r4) + stw r16, VCPU_GPR(r16)(r4) + stw r17, VCPU_GPR(r17)(r4) + stw r18, VCPU_GPR(r18)(r4) + stw r19, VCPU_GPR(r19)(r4) + stw r20, VCPU_GPR(r20)(r4) + stw r21, VCPU_GPR(r21)(r4) + stw r22, VCPU_GPR(r22)(r4) + stw r23, VCPU_GPR(r23)(r4) + stw r24, VCPU_GPR(r24)(r4) + stw r25, VCPU_GPR(r25)(r4) + stw r26, VCPU_GPR(r26)(r4) + stw r27, VCPU_GPR(r27)(r4) + stw r28, VCPU_GPR(r28)(r4) + stw r29, VCPU_GPR(r29)(r4) + stw r30, VCPU_GPR(r30)(r4) + stw r31, VCPU_GPR(r31)(r4) + + /* Load host non-volatile register state from host stack. */ + lwz r14, HOST_NV_GPR(r14)(r1) + lwz r15, HOST_NV_GPR(r15)(r1) + lwz r16, HOST_NV_GPR(r16)(r1) + lwz r17, HOST_NV_GPR(r17)(r1) + lwz r18, HOST_NV_GPR(r18)(r1) + lwz r19, HOST_NV_GPR(r19)(r1) + lwz r20, HOST_NV_GPR(r20)(r1) + lwz r21, HOST_NV_GPR(r21)(r1) + lwz r22, HOST_NV_GPR(r22)(r1) + lwz r23, HOST_NV_GPR(r23)(r1) + lwz r24, HOST_NV_GPR(r24)(r1) + lwz r25, HOST_NV_GPR(r25)(r1) + lwz r26, HOST_NV_GPR(r26)(r1) + lwz r27, HOST_NV_GPR(r27)(r1) + lwz r28, HOST_NV_GPR(r28)(r1) + lwz r29, HOST_NV_GPR(r29)(r1) + lwz r30, HOST_NV_GPR(r30)(r1) + lwz r31, HOST_NV_GPR(r31)(r1) + + /* Return to kvm_vcpu_run(). */ + lwz r4, HOST_STACK_LR(r1) + addi r1, r1, HOST_STACK_SIZE + mtlr r4 + /* r3 still contains the return code from kvmppc_handle_exit(). */ + blr + + +/* Registers: + * r3: kvm_run pointer + * r4: vcpu pointer + */ +_GLOBAL(__kvmppc_vcpu_run) + stwu r1, -HOST_STACK_SIZE(r1) + stw r1, VCPU_HOST_STACK(r4) /* Save stack pointer to vcpu. */ + + /* Save host state to stack. */ + stw r3, HOST_RUN(r1) + mflr r3 + stw r3, HOST_STACK_LR(r1) + + /* Save host non-volatile register state to stack. */ + stw r14, HOST_NV_GPR(r14)(r1) + stw r15, HOST_NV_GPR(r15)(r1) + stw r16, HOST_NV_GPR(r16)(r1) + stw r17, HOST_NV_GPR(r17)(r1) + stw r18, HOST_NV_GPR(r18)(r1) + stw r19, HOST_NV_GPR(r19)(r1) + stw r20, HOST_NV_GPR(r20)(r1) + stw r21, HOST_NV_GPR(r21)(r1) + stw r22, HOST_NV_GPR(r22)(r1) + stw r23, HOST_NV_GPR(r23)(r1) + stw r24, HOST_NV_GPR(r24)(r1) + stw r25, HOST_NV_GPR(r25)(r1) + stw r26, HOST_NV_GPR(r26)(r1) + stw r27, HOST_NV_GPR(r27)(r1) + stw r28, HOST_NV_GPR(r28)(r1) + stw r29, HOST_NV_GPR(r29)(r1) + stw r30, HOST_NV_GPR(r30)(r1) + stw r31, HOST_NV_GPR(r31)(r1) + + /* Load guest non-volatiles. */ + lwz r14, VCPU_GPR(r14)(r4) + lwz r15, VCPU_GPR(r15)(r4) + lwz r16, VCPU_GPR(r16)(r4) + lwz r17, VCPU_GPR(r17)(r4) + lwz r18, VCPU_GPR(r18)(r4) + lwz r19, VCPU_GPR(r19)(r4) + lwz r20, VCPU_GPR(r20)(r4) + lwz r21, VCPU_GPR(r21)(r4) + lwz r22, VCPU_GPR(r22)(r4) + lwz r23, VCPU_GPR(r23)(r4) + lwz r24, VCPU_GPR(r24)(r4) + lwz r25, VCPU_GPR(r25)(r4) + lwz r26, VCPU_GPR(r26)(r4) + lwz r27, VCPU_GPR(r27)(r4) + lwz r28, VCPU_GPR(r28)(r4) + lwz r29, VCPU_GPR(r29)(r4) + lwz r30, VCPU_GPR(r30)(r4) + lwz r31, VCPU_GPR(r31)(r4) + +lightweight_exit: + stw r2, HOST_R2(r1) + + mfspr r3, SPRN_PID + stw r3, VCPU_HOST_PID(r4) + lwz r3, VCPU_PID(r4) + mtspr SPRN_PID, r3 + + /* Prevent all TLB updates. */ + mfmsr r5 + lis r6, (MSR_EE|MSR_CE|MSR_ME|MSR_DE)@h + ori r6, r6, (MSR_EE|MSR_CE|MSR_ME|MSR_DE)@l + andc r6, r5, r6 + mtmsr r6 + + /* Save the host's non-pinned TLB mappings, and load the guest mappings + * over them. Leave the host's "pinned" kernel mappings in place. */ + /* XXX optimization: use generation count to avoid swapping unmodified + * entries. */ + mfspr r10, SPRN_MMUCR /* Save host MMUCR. */ + lis r8, tlb_44x_hwater@ha + lwz r8, tlb_44x_hwater@l(r8) + addi r3, r4, VCPU_HOST_TLB - 4 + addi r9, r4, VCPU_SHADOW_TLB - 4 + li r6, 0 +1: + /* Save host entry. */ + tlbre r7, r6, PPC44x_TLB_PAGEID + mfspr r5, SPRN_MMUCR + stwu r5, 4(r3) + stwu r7, 4(r3) + tlbre r7, r6, PPC44x_TLB_XLAT + stwu r7, 4(r3) + tlbre r7, r6, PPC44x_TLB_ATTRIB + stwu r7, 4(r3) + /* Load guest entry. */ + lwzu r7, 4(r9) + mtspr SPRN_MMUCR, r7 + lwzu r7, 4(r9) + tlbwe r7, r6, PPC44x_TLB_PAGEID + lwzu r7, 4(r9) + tlbwe r7, r6, PPC44x_TLB_XLAT + lwzu r7, 4(r9) + tlbwe r7, r6, PPC44x_TLB_ATTRIB + /* Increment index. */ + addi r6, r6, 1 + cmpw r6, r8 + blt 1b + mtspr SPRN_MMUCR, r10 /* Restore host MMUCR. */ + + iccci 0, 0 /* XXX hack */ + + /* Load some guest volatiles. */ + lwz r0, VCPU_GPR(r0)(r4) + lwz r2, VCPU_GPR(r2)(r4) + lwz r9, VCPU_GPR(r9)(r4) + lwz r10, VCPU_GPR(r10)(r4) + lwz r11, VCPU_GPR(r11)(r4) + lwz r12, VCPU_GPR(r12)(r4) + lwz r13, VCPU_GPR(r13)(r4) + lwz r3, VCPU_LR(r4) + mtlr r3 + lwz r3, VCPU_XER(r4) + mtxer r3 + + /* Switch the IVPR. XXX If we take a TLB miss after this we're screwed, + * so how do we make sure vcpu won't fault? */ + lis r8, kvmppc_booke_handlers@ha + lwz r8, kvmppc_booke_handlers@l(r8) + mtspr SPRN_IVPR, r8 + + /* Save vcpu pointer for the exception handlers. */ + mtspr SPRN_SPRG1, r4 + + /* Can't switch the stack pointer until after IVPR is switched, + * because host interrupt handlers would get confused. */ + lwz r1, VCPU_GPR(r1)(r4) + + /* XXX handle USPRG0 */ + /* Host interrupt handlers may have clobbered these guest-readable + * SPRGs, so we need to reload them here with the guest's values. */ + lwz r3, VCPU_SPRG4(r4) + mtspr SPRN_SPRG4, r3 + lwz r3, VCPU_SPRG5(r4) + mtspr SPRN_SPRG5, r3 + lwz r3, VCPU_SPRG6(r4) + mtspr SPRN_SPRG6, r3 + lwz r3, VCPU_SPRG7(r4) + mtspr SPRN_SPRG7, r3 + + /* Finish loading guest volatiles and jump to guest. */ + lwz r3, VCPU_CTR(r4) + mtctr r3 + lwz r3, VCPU_CR(r4) + mtcr r3 + lwz r5, VCPU_GPR(r5)(r4) + lwz r6, VCPU_GPR(r6)(r4) + lwz r7, VCPU_GPR(r7)(r4) + lwz r8, VCPU_GPR(r8)(r4) + lwz r3, VCPU_PC(r4) + mtsrr0 r3 + lwz r3, VCPU_MSR(r4) + oris r3, r3, KVMPPC_MSR_MASK@h + ori r3, r3, KVMPPC_MSR_MASK@l + mtsrr1 r3 + lwz r3, VCPU_GPR(r3)(r4) + lwz r4, VCPU_GPR(r4)(r4) + rfi diff --git a/arch/powerpc/kvm/emulate.c b/arch/powerpc/kvm/emulate.c new file mode 100644 index 000000000000..a03fe0c80698 --- /dev/null +++ b/arch/powerpc/kvm/emulate.c @@ -0,0 +1,760 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright IBM Corp. 2007 + * + * Authors: Hollis Blanchard <hollisb@us.ibm.com> + */ + +#include <linux/jiffies.h> +#include <linux/timer.h> +#include <linux/types.h> +#include <linux/string.h> +#include <linux/kvm_host.h> + +#include <asm/dcr.h> +#include <asm/dcr-regs.h> +#include <asm/time.h> +#include <asm/byteorder.h> +#include <asm/kvm_ppc.h> + +#include "44x_tlb.h" + +/* Instruction decoding */ +static inline unsigned int get_op(u32 inst) +{ + return inst >> 26; +} + +static inline unsigned int get_xop(u32 inst) +{ + return (inst >> 1) & 0x3ff; +} + +static inline unsigned int get_sprn(u32 inst) +{ + return ((inst >> 16) & 0x1f) | ((inst >> 6) & 0x3e0); +} + +static inline unsigned int get_dcrn(u32 inst) +{ + return ((inst >> 16) & 0x1f) | ((inst >> 6) & 0x3e0); +} + +static inline unsigned int get_rt(u32 inst) +{ + return (inst >> 21) & 0x1f; +} + +static inline unsigned int get_rs(u32 inst) +{ + return (inst >> 21) & 0x1f; +} + +static inline unsigned int get_ra(u32 inst) +{ + return (inst >> 16) & 0x1f; +} + +static inline unsigned int get_rb(u32 inst) +{ + return (inst >> 11) & 0x1f; +} + +static inline unsigned int get_rc(u32 inst) +{ + return inst & 0x1; +} + +static inline unsigned int get_ws(u32 inst) +{ + return (inst >> 11) & 0x1f; +} + +static inline unsigned int get_d(u32 inst) +{ + return inst & 0xffff; +} + +static int tlbe_is_host_safe(const struct kvm_vcpu *vcpu, + const struct tlbe *tlbe) +{ + gpa_t gpa; + + if (!get_tlb_v(tlbe)) + return 0; + + /* Does it match current guest AS? */ + /* XXX what about IS != DS? */ + if (get_tlb_ts(tlbe) != !!(vcpu->arch.msr & MSR_IS)) + return 0; + + gpa = get_tlb_raddr(tlbe); + if (!gfn_to_memslot(vcpu->kvm, gpa >> PAGE_SHIFT)) + /* Mapping is not for RAM. */ + return 0; + + return 1; +} + +static int kvmppc_emul_tlbwe(struct kvm_vcpu *vcpu, u32 inst) +{ + u64 eaddr; + u64 raddr; + u64 asid; + u32 flags; + struct tlbe *tlbe; + unsigned int ra; + unsigned int rs; + unsigned int ws; + unsigned int index; + + ra = get_ra(inst); + rs = get_rs(inst); + ws = get_ws(inst); + + index = vcpu->arch.gpr[ra]; + if (index > PPC44x_TLB_SIZE) { + printk("%s: index %d\n", __func__, index); + kvmppc_dump_vcpu(vcpu); + return EMULATE_FAIL; + } + + tlbe = &vcpu->arch.guest_tlb[index]; + + /* Invalidate shadow mappings for the about-to-be-clobbered TLBE. */ + if (tlbe->word0 & PPC44x_TLB_VALID) { + eaddr = get_tlb_eaddr(tlbe); + asid = (tlbe->word0 & PPC44x_TLB_TS) | tlbe->tid; + kvmppc_mmu_invalidate(vcpu, eaddr, asid); + } + + switch (ws) { + case PPC44x_TLB_PAGEID: + tlbe->tid = vcpu->arch.mmucr & 0xff; + tlbe->word0 = vcpu->arch.gpr[rs]; + break; + + case PPC44x_TLB_XLAT: + tlbe->word1 = vcpu->arch.gpr[rs]; + break; + + case PPC44x_TLB_ATTRIB: + tlbe->word2 = vcpu->arch.gpr[rs]; + break; + + default: + return EMULATE_FAIL; + } + + if (tlbe_is_host_safe(vcpu, tlbe)) { + eaddr = get_tlb_eaddr(tlbe); + raddr = get_tlb_raddr(tlbe); + asid = (tlbe->word0 & PPC44x_TLB_TS) | tlbe->tid; + flags = tlbe->word2 & 0xffff; + + /* Create a 4KB mapping on the host. If the guest wanted a + * large page, only the first 4KB is mapped here and the rest + * are mapped on the fly. */ + kvmppc_mmu_map(vcpu, eaddr, raddr >> PAGE_SHIFT, asid, flags); + } + + return EMULATE_DONE; +} + +static void kvmppc_emulate_dec(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.tcr & TCR_DIE) { + /* The decrementer ticks at the same rate as the timebase, so + * that's how we convert the guest DEC value to the number of + * host ticks. */ + unsigned long nr_jiffies; + + nr_jiffies = vcpu->arch.dec / tb_ticks_per_jiffy; + mod_timer(&vcpu->arch.dec_timer, + get_jiffies_64() + nr_jiffies); + } else { + del_timer(&vcpu->arch.dec_timer); + } +} + +static void kvmppc_emul_rfi(struct kvm_vcpu *vcpu) +{ + vcpu->arch.pc = vcpu->arch.srr0; + kvmppc_set_msr(vcpu, vcpu->arch.srr1); +} + +/* XXX to do: + * lhax + * lhaux + * lswx + * lswi + * stswx + * stswi + * lha + * lhau + * lmw + * stmw + * + * XXX is_bigendian should depend on MMU mapping or MSR[LE] + */ +int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu) +{ + u32 inst = vcpu->arch.last_inst; + u32 ea; + int ra; + int rb; + int rc; + int rs; + int rt; + int sprn; + int dcrn; + enum emulation_result emulated = EMULATE_DONE; + int advance = 1; + + switch (get_op(inst)) { + case 3: /* trap */ + printk("trap!\n"); + kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_PROGRAM); + advance = 0; + break; + + case 19: + switch (get_xop(inst)) { + case 50: /* rfi */ + kvmppc_emul_rfi(vcpu); + advance = 0; + break; + + default: + emulated = EMULATE_FAIL; + break; + } + break; + + case 31: + switch (get_xop(inst)) { + + case 83: /* mfmsr */ + rt = get_rt(inst); + vcpu->arch.gpr[rt] = vcpu->arch.msr; + break; + + case 87: /* lbzx */ + rt = get_rt(inst); + emulated = kvmppc_handle_load(run, vcpu, rt, 1, 1); + break; + + case 131: /* wrtee */ + rs = get_rs(inst); + vcpu->arch.msr = (vcpu->arch.msr & ~MSR_EE) + | (vcpu->arch.gpr[rs] & MSR_EE); + break; + + case 146: /* mtmsr */ + rs = get_rs(inst); + kvmppc_set_msr(vcpu, vcpu->arch.gpr[rs]); + break; + + case 163: /* wrteei */ + vcpu->arch.msr = (vcpu->arch.msr & ~MSR_EE) + | (inst & MSR_EE); + break; + + case 215: /* stbx */ + rs = get_rs(inst); + emulated = kvmppc_handle_store(run, vcpu, + vcpu->arch.gpr[rs], + 1, 1); + break; + + case 247: /* stbux */ + rs = get_rs(inst); + ra = get_ra(inst); + rb = get_rb(inst); + + ea = vcpu->arch.gpr[rb]; + if (ra) + ea += vcpu->arch.gpr[ra]; + + emulated = kvmppc_handle_store(run, vcpu, + vcpu->arch.gpr[rs], + 1, 1); + vcpu->arch.gpr[rs] = ea; + break; + + case 279: /* lhzx */ + rt = get_rt(inst); + emulated = kvmppc_handle_load(run, vcpu, rt, 2, 1); + break; + + case 311: /* lhzux */ + rt = get_rt(inst); + ra = get_ra(inst); + rb = get_rb(inst); + + ea = vcpu->arch.gpr[rb]; + if (ra) + ea += vcpu->arch.gpr[ra]; + + emulated = kvmppc_handle_load(run, vcpu, rt, 2, 1); + vcpu->arch.gpr[ra] = ea; + break; + + case 323: /* mfdcr */ + dcrn = get_dcrn(inst); + rt = get_rt(inst); + + /* The guest may access CPR0 registers to determine the timebase + * frequency, and it must know the real host frequency because it + * can directly access the timebase registers. + * + * It would be possible to emulate those accesses in userspace, + * but userspace can really only figure out the end frequency. + * We could decompose that into the factors that compute it, but + * that's tricky math, and it's easier to just report the real + * CPR0 values. + */ + switch (dcrn) { + case DCRN_CPR0_CONFIG_ADDR: + vcpu->arch.gpr[rt] = vcpu->arch.cpr0_cfgaddr; + break; + case DCRN_CPR0_CONFIG_DATA: + local_irq_disable(); + mtdcr(DCRN_CPR0_CONFIG_ADDR, + vcpu->arch.cpr0_cfgaddr); + vcpu->arch.gpr[rt] = mfdcr(DCRN_CPR0_CONFIG_DATA); + local_irq_enable(); + break; + default: + run->dcr.dcrn = dcrn; + run->dcr.data = 0; + run->dcr.is_write = 0; + vcpu->arch.io_gpr = rt; + vcpu->arch.dcr_needed = 1; + emulated = EMULATE_DO_DCR; + } + + break; + + case 339: /* mfspr */ + sprn = get_sprn(inst); + rt = get_rt(inst); + + switch (sprn) { + case SPRN_SRR0: + vcpu->arch.gpr[rt] = vcpu->arch.srr0; break; + case SPRN_SRR1: + vcpu->arch.gpr[rt] = vcpu->arch.srr1; break; + case SPRN_MMUCR: + vcpu->arch.gpr[rt] = vcpu->arch.mmucr; break; + case SPRN_PID: + vcpu->arch.gpr[rt] = vcpu->arch.pid; break; + case SPRN_IVPR: + vcpu->arch.gpr[rt] = vcpu->arch.ivpr; break; + case SPRN_CCR0: + vcpu->arch.gpr[rt] = vcpu->arch.ccr0; break; + case SPRN_CCR1: + vcpu->arch.gpr[rt] = vcpu->arch.ccr1; break; + case SPRN_PVR: + vcpu->arch.gpr[rt] = vcpu->arch.pvr; break; + case SPRN_DEAR: + vcpu->arch.gpr[rt] = vcpu->arch.dear; break; + case SPRN_ESR: + vcpu->arch.gpr[rt] = vcpu->arch.esr; break; + case SPRN_DBCR0: + vcpu->arch.gpr[rt] = vcpu->arch.dbcr0; break; + case SPRN_DBCR1: + vcpu->arch.gpr[rt] = vcpu->arch.dbcr1; break; + + /* Note: mftb and TBRL/TBWL are user-accessible, so + * the guest can always access the real TB anyways. + * In fact, we probably will never see these traps. */ + case SPRN_TBWL: + vcpu->arch.gpr[rt] = mftbl(); break; + case SPRN_TBWU: + vcpu->arch.gpr[rt] = mftbu(); break; + + case SPRN_SPRG0: + vcpu->arch.gpr[rt] = vcpu->arch.sprg0; break; + case SPRN_SPRG1: + vcpu->arch.gpr[rt] = vcpu->arch.sprg1; break; + case SPRN_SPRG2: + vcpu->arch.gpr[rt] = vcpu->arch.sprg2; break; + case SPRN_SPRG3: + vcpu->arch.gpr[rt] = vcpu->arch.sprg3; break; + /* Note: SPRG4-7 are user-readable, so we don't get + * a trap. */ + + case SPRN_IVOR0: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[0]; break; + case SPRN_IVOR1: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[1]; break; + case SPRN_IVOR2: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[2]; break; + case SPRN_IVOR3: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[3]; break; + case SPRN_IVOR4: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[4]; break; + case SPRN_IVOR5: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[5]; break; + case SPRN_IVOR6: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[6]; break; + case SPRN_IVOR7: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[7]; break; + case SPRN_IVOR8: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[8]; break; + case SPRN_IVOR9: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[9]; break; + case SPRN_IVOR10: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[10]; break; + case SPRN_IVOR11: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[11]; break; + case SPRN_IVOR12: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[12]; break; + case SPRN_IVOR13: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[13]; break; + case SPRN_IVOR14: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[14]; break; + case SPRN_IVOR15: + vcpu->arch.gpr[rt] = vcpu->arch.ivor[15]; break; + + default: + printk("mfspr: unknown spr %x\n", sprn); + vcpu->arch.gpr[rt] = 0; + break; + } + break; + + case 407: /* sthx */ + rs = get_rs(inst); + ra = get_ra(inst); + rb = get_rb(inst); + + emulated = kvmppc_handle_store(run, vcpu, + vcpu->arch.gpr[rs], + 2, 1); + break; + + case 439: /* sthux */ + rs = get_rs(inst); + ra = get_ra(inst); + rb = get_rb(inst); + + ea = vcpu->arch.gpr[rb]; + if (ra) + ea += vcpu->arch.gpr[ra]; + + emulated = kvmppc_handle_store(run, vcpu, + vcpu->arch.gpr[rs], + 2, 1); + vcpu->arch.gpr[ra] = ea; + break; + + case 451: /* mtdcr */ + dcrn = get_dcrn(inst); + rs = get_rs(inst); + + /* emulate some access in kernel */ + switch (dcrn) { + case DCRN_CPR0_CONFIG_ADDR: + vcpu->arch.cpr0_cfgaddr = vcpu->arch.gpr[rs]; + break; + default: + run->dcr.dcrn = dcrn; + run->dcr.data = vcpu->arch.gpr[rs]; + run->dcr.is_write = 1; + vcpu->arch.dcr_needed = 1; + emulated = EMULATE_DO_DCR; + } + + break; + + case 467: /* mtspr */ + sprn = get_sprn(inst); + rs = get_rs(inst); + switch (sprn) { + case SPRN_SRR0: + vcpu->arch.srr0 = vcpu->arch.gpr[rs]; break; + case SPRN_SRR1: + vcpu->arch.srr1 = vcpu->arch.gpr[rs]; break; + case SPRN_MMUCR: + vcpu->arch.mmucr = vcpu->arch.gpr[rs]; break; + case SPRN_PID: + vcpu->arch.pid = vcpu->arch.gpr[rs]; break; + case SPRN_CCR0: + vcpu->arch.ccr0 = vcpu->arch.gpr[rs]; break; + case SPRN_CCR1: + vcpu->arch.ccr1 = vcpu->arch.gpr[rs]; break; + case SPRN_DEAR: + vcpu->arch.dear = vcpu->arch.gpr[rs]; break; + case SPRN_ESR: + vcpu->arch.esr = vcpu->arch.gpr[rs]; break; + case SPRN_DBCR0: + vcpu->arch.dbcr0 = vcpu->arch.gpr[rs]; break; + case SPRN_DBCR1: + vcpu->arch.dbcr1 = vcpu->arch.gpr[rs]; break; + + /* XXX We need to context-switch the timebase for + * watchdog and FIT. */ + case SPRN_TBWL: break; + case SPRN_TBWU: break; + + case SPRN_DEC: + vcpu->arch.dec = vcpu->arch.gpr[rs]; + kvmppc_emulate_dec(vcpu); + break; + + case SPRN_TSR: + vcpu->arch.tsr &= ~vcpu->arch.gpr[rs]; break; + + case SPRN_TCR: + vcpu->arch.tcr = vcpu->arch.gpr[rs]; + kvmppc_emulate_dec(vcpu); + break; + + case SPRN_SPRG0: + vcpu->arch.sprg0 = vcpu->arch.gpr[rs]; break; + case SPRN_SPRG1: + vcpu->arch.sprg1 = vcpu->arch.gpr[rs]; break; + case SPRN_SPRG2: + vcpu->arch.sprg2 = vcpu->arch.gpr[rs]; break; + case SPRN_SPRG3: + vcpu->arch.sprg3 = vcpu->arch.gpr[rs]; break; + + /* Note: SPRG4-7 are user-readable. These values are + * loaded into the real SPRGs when resuming the + * guest. */ + case SPRN_SPRG4: + vcpu->arch.sprg4 = vcpu->arch.gpr[rs]; break; + case SPRN_SPRG5: + vcpu->arch.sprg5 = vcpu->arch.gpr[rs]; break; + case SPRN_SPRG6: + vcpu->arch.sprg6 = vcpu->arch.gpr[rs]; break; + case SPRN_SPRG7: + vcpu->arch.sprg7 = vcpu->arch.gpr[rs]; break; + + case SPRN_IVPR: + vcpu->arch.ivpr = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR0: + vcpu->arch.ivor[0] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR1: + vcpu->arch.ivor[1] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR2: + vcpu->arch.ivor[2] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR3: + vcpu->arch.ivor[3] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR4: + vcpu->arch.ivor[4] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR5: + vcpu->arch.ivor[5] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR6: + vcpu->arch.ivor[6] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR7: + vcpu->arch.ivor[7] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR8: + vcpu->arch.ivor[8] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR9: + vcpu->arch.ivor[9] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR10: + vcpu->arch.ivor[10] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR11: + vcpu->arch.ivor[11] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR12: + vcpu->arch.ivor[12] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR13: + vcpu->arch.ivor[13] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR14: + vcpu->arch.ivor[14] = vcpu->arch.gpr[rs]; break; + case SPRN_IVOR15: + vcpu->arch.ivor[15] = vcpu->arch.gpr[rs]; break; + + default: + printk("mtspr: unknown spr %x\n", sprn); + emulated = EMULATE_FAIL; + break; + } + break; + + case 470: /* dcbi */ + /* Do nothing. The guest is performing dcbi because + * hardware DMA is not snooped by the dcache, but + * emulated DMA either goes through the dcache as + * normal writes, or the host kernel has handled dcache + * coherence. */ + break; + + case 534: /* lwbrx */ + rt = get_rt(inst); + emulated = kvmppc_handle_load(run, vcpu, rt, 4, 0); + break; + + case 566: /* tlbsync */ + break; + + case 662: /* stwbrx */ + rs = get_rs(inst); + ra = get_ra(inst); + rb = get_rb(inst); + + emulated = kvmppc_handle_store(run, vcpu, + vcpu->arch.gpr[rs], + 4, 0); + break; + + case 978: /* tlbwe */ + emulated = kvmppc_emul_tlbwe(vcpu, inst); + break; + + case 914: { /* tlbsx */ + int index; + unsigned int as = get_mmucr_sts(vcpu); + unsigned int pid = get_mmucr_stid(vcpu); + + rt = get_rt(inst); + ra = get_ra(inst); + rb = get_rb(inst); + rc = get_rc(inst); + + ea = vcpu->arch.gpr[rb]; + if (ra) + ea += vcpu->arch.gpr[ra]; + + index = kvmppc_44x_tlb_index(vcpu, ea, pid, as); + if (rc) { + if (index < 0) + vcpu->arch.cr &= ~0x20000000; + else + vcpu->arch.cr |= 0x20000000; + } + vcpu->arch.gpr[rt] = index; + + } + break; + + case 790: /* lhbrx */ + rt = get_rt(inst); + emulated = kvmppc_handle_load(run, vcpu, rt, 2, 0); + break; + + case 918: /* sthbrx */ + rs = get_rs(inst); + ra = get_ra(inst); + rb = get_rb(inst); + + emulated = kvmppc_handle_store(run, vcpu, + vcpu->arch.gpr[rs], + 2, 0); + break; + + case 966: /* iccci */ + break; + + default: + printk("unknown: op %d xop %d\n", get_op(inst), + get_xop(inst)); + emulated = EMULATE_FAIL; + break; + } + break; + + case 32: /* lwz */ + rt = get_rt(inst); + emulated = kvmppc_handle_load(run, vcpu, rt, 4, 1); + break; + + case 33: /* lwzu */ + ra = get_ra(inst); + rt = get_rt(inst); + emulated = kvmppc_handle_load(run, vcpu, rt, 4, 1); + vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed; + break; + + case 34: /* lbz */ + rt = get_rt(inst); + emulated = kvmppc_handle_load(run, vcpu, rt, 1, 1); + break; + + case 35: /* lbzu */ + ra = get_ra(inst); + rt = get_rt(inst); + emulated = kvmppc_handle_load(run, vcpu, rt, 1, 1); + vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed; + break; + + case 36: /* stw */ + rs = get_rs(inst); + emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs], + 4, 1); + break; + + case 37: /* stwu */ + ra = get_ra(inst); + rs = get_rs(inst); + emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs], + 4, 1); + vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed; + break; + + case 38: /* stb */ + rs = get_rs(inst); + emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs], + 1, 1); + break; + + case 39: /* stbu */ + ra = get_ra(inst); + rs = get_rs(inst); + emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs], + 1, 1); + vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed; + break; + + case 40: /* lhz */ + rt = get_rt(inst); + emulated = kvmppc_handle_load(run, vcpu, rt, 2, 1); + break; + + case 41: /* lhzu */ + ra = get_ra(inst); + rt = get_rt(inst); + emulated = kvmppc_handle_load(run, vcpu, rt, 2, 1); + vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed; + break; + + case 44: /* sth */ + rs = get_rs(inst); + emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs], + 2, 1); + break; + + case 45: /* sthu */ + ra = get_ra(inst); + rs = get_rs(inst); + emulated = kvmppc_handle_store(run, vcpu, vcpu->arch.gpr[rs], + 2, 1); + vcpu->arch.gpr[ra] = vcpu->arch.paddr_accessed; + break; + + default: + printk("unknown op %d\n", get_op(inst)); + emulated = EMULATE_FAIL; + break; + } + + if (advance) + vcpu->arch.pc += 4; /* Advance past emulated instruction. */ + + return emulated; +} diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c new file mode 100644 index 000000000000..bad40bd2d3ac --- /dev/null +++ b/arch/powerpc/kvm/powerpc.c @@ -0,0 +1,436 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + * + * Copyright IBM Corp. 2007 + * + * Authors: Hollis Blanchard <hollisb@us.ibm.com> + * Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com> + */ + +#include <linux/errno.h> +#include <linux/err.h> +#include <linux/kvm_host.h> +#include <linux/module.h> +#include <linux/vmalloc.h> +#include <linux/fs.h> +#include <asm/cputable.h> +#include <asm/uaccess.h> +#include <asm/kvm_ppc.h> + + +gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn) +{ + return gfn; +} + +int kvm_cpu_has_interrupt(struct kvm_vcpu *v) +{ + /* XXX implement me */ + return 0; +} + +int kvm_arch_vcpu_runnable(struct kvm_vcpu *v) +{ + return 1; +} + + +int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu) +{ + enum emulation_result er; + int r; + + er = kvmppc_emulate_instruction(run, vcpu); + switch (er) { + case EMULATE_DONE: + /* Future optimization: only reload non-volatiles if they were + * actually modified. */ + r = RESUME_GUEST_NV; + break; + case EMULATE_DO_MMIO: + run->exit_reason = KVM_EXIT_MMIO; + /* We must reload nonvolatiles because "update" load/store + * instructions modify register state. */ + /* Future optimization: only reload non-volatiles if they were + * actually modified. */ + r = RESUME_HOST_NV; + break; + case EMULATE_FAIL: + /* XXX Deliver Program interrupt to guest. */ + printk(KERN_EMERG "%s: emulation failed (%08x)\n", __func__, + vcpu->arch.last_inst); + r = RESUME_HOST; + break; + default: + BUG(); + } + + return r; +} + +void kvm_arch_hardware_enable(void *garbage) +{ +} + +void kvm_arch_hardware_disable(void *garbage) +{ +} + +int kvm_arch_hardware_setup(void) +{ + return 0; +} + +void kvm_arch_hardware_unsetup(void) +{ +} + +void kvm_arch_check_processor_compat(void *rtn) +{ + int r; + + if (strcmp(cur_cpu_spec->platform, "ppc440") == 0) + r = 0; + else + r = -ENOTSUPP; + + *(int *)rtn = r; +} + +struct kvm *kvm_arch_create_vm(void) +{ + struct kvm *kvm; + + kvm = kzalloc(sizeof(struct kvm), GFP_KERNEL); + if (!kvm) + return ERR_PTR(-ENOMEM); + + return kvm; +} + +static void kvmppc_free_vcpus(struct kvm *kvm) +{ + unsigned int i; + + for (i = 0; i < KVM_MAX_VCPUS; ++i) { + if (kvm->vcpus[i]) { + kvm_arch_vcpu_free(kvm->vcpus[i]); + kvm->vcpus[i] = NULL; + } + } +} + +void kvm_arch_destroy_vm(struct kvm *kvm) +{ + kvmppc_free_vcpus(kvm); + kvm_free_physmem(kvm); + kfree(kvm); +} + +int kvm_dev_ioctl_check_extension(long ext) +{ + int r; + + switch (ext) { + case KVM_CAP_USER_MEMORY: + r = 1; + break; + default: + r = 0; + break; + } + return r; + +} + +long kvm_arch_dev_ioctl(struct file *filp, + unsigned int ioctl, unsigned long arg) +{ + return -EINVAL; +} + +int kvm_arch_set_memory_region(struct kvm *kvm, + struct kvm_userspace_memory_region *mem, + struct kvm_memory_slot old, + int user_alloc) +{ + return 0; +} + +struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id) +{ + struct kvm_vcpu *vcpu; + int err; + + vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL); + if (!vcpu) { + err = -ENOMEM; + goto out; + } + + err = kvm_vcpu_init(vcpu, kvm, id); + if (err) + goto free_vcpu; + + return vcpu; + +free_vcpu: + kmem_cache_free(kvm_vcpu_cache, vcpu); +out: + return ERR_PTR(err); +} + +void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu) +{ + kvm_vcpu_uninit(vcpu); + kmem_cache_free(kvm_vcpu_cache, vcpu); +} + +void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) +{ + kvm_arch_vcpu_free(vcpu); +} + +int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu) +{ + unsigned int priority = exception_priority[BOOKE_INTERRUPT_DECREMENTER]; + + return test_bit(priority, &vcpu->arch.pending_exceptions); +} + +static void kvmppc_decrementer_func(unsigned long data) +{ + struct kvm_vcpu *vcpu = (struct kvm_vcpu *)data; + + kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_DECREMENTER); +} + +int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) +{ + setup_timer(&vcpu->arch.dec_timer, kvmppc_decrementer_func, + (unsigned long)vcpu); + + return 0; +} + +void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) +{ +} + +void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) +{ +} + +void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) +{ +} + +void decache_vcpus_on_cpu(int cpu) +{ +} + +int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu, + struct kvm_debug_guest *dbg) +{ + return -ENOTSUPP; +} + +static void kvmppc_complete_dcr_load(struct kvm_vcpu *vcpu, + struct kvm_run *run) +{ + u32 *gpr = &vcpu->arch.gpr[vcpu->arch.io_gpr]; + *gpr = run->dcr.data; +} + +static void kvmppc_complete_mmio_load(struct kvm_vcpu *vcpu, + struct kvm_run *run) +{ + u32 *gpr = &vcpu->arch.gpr[vcpu->arch.io_gpr]; + + if (run->mmio.len > sizeof(*gpr)) { + printk(KERN_ERR "bad MMIO length: %d\n", run->mmio.len); + return; + } + + if (vcpu->arch.mmio_is_bigendian) { + switch (run->mmio.len) { + case 4: *gpr = *(u32 *)run->mmio.data; break; + case 2: *gpr = *(u16 *)run->mmio.data; break; + case 1: *gpr = *(u8 *)run->mmio.data; break; + } + } else { + /* Convert BE data from userland back to LE. */ + switch (run->mmio.len) { + case 4: *gpr = ld_le32((u32 *)run->mmio.data); break; + case 2: *gpr = ld_le16((u16 *)run->mmio.data); break; + case 1: *gpr = *(u8 *)run->mmio.data; break; + } + } +} + +int kvmppc_handle_load(struct kvm_run *run, struct kvm_vcpu *vcpu, + unsigned int rt, unsigned int bytes, int is_bigendian) +{ + if (bytes > sizeof(run->mmio.data)) { + printk(KERN_ERR "%s: bad MMIO length: %d\n", __func__, + run->mmio.len); + } + + run->mmio.phys_addr = vcpu->arch.paddr_accessed; + run->mmio.len = bytes; + run->mmio.is_write = 0; + + vcpu->arch.io_gpr = rt; + vcpu->arch.mmio_is_bigendian = is_bigendian; + vcpu->mmio_needed = 1; + vcpu->mmio_is_write = 0; + + return EMULATE_DO_MMIO; +} + +int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu, + u32 val, unsigned int bytes, int is_bigendian) +{ + void *data = run->mmio.data; + + if (bytes > sizeof(run->mmio.data)) { + printk(KERN_ERR "%s: bad MMIO length: %d\n", __func__, + run->mmio.len); + } + + run->mmio.phys_addr = vcpu->arch.paddr_accessed; + run->mmio.len = bytes; + run->mmio.is_write = 1; + vcpu->mmio_needed = 1; + vcpu->mmio_is_write = 1; + + /* Store the value at the lowest bytes in 'data'. */ + if (is_bigendian) { + switch (bytes) { + case 4: *(u32 *)data = val; break; + case 2: *(u16 *)data = val; break; + case 1: *(u8 *)data = val; break; + } + } else { + /* Store LE value into 'data'. */ + switch (bytes) { + case 4: st_le32(data, val); break; + case 2: st_le16(data, val); break; + case 1: *(u8 *)data = val; break; + } + } + + return EMULATE_DO_MMIO; +} + +int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) +{ + int r; + sigset_t sigsaved; + + if (vcpu->sigset_active) + sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved); + + if (vcpu->mmio_needed) { + if (!vcpu->mmio_is_write) + kvmppc_complete_mmio_load(vcpu, run); + vcpu->mmio_needed = 0; + } else if (vcpu->arch.dcr_needed) { + if (!vcpu->arch.dcr_is_write) + kvmppc_complete_dcr_load(vcpu, run); + vcpu->arch.dcr_needed = 0; + } + + kvmppc_check_and_deliver_interrupts(vcpu); + + local_irq_disable(); + kvm_guest_enter(); + r = __kvmppc_vcpu_run(run, vcpu); + kvm_guest_exit(); + local_irq_enable(); + + if (vcpu->sigset_active) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + + return r; +} + +int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq) +{ + kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_EXTERNAL); + return 0; +} + +int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state) +{ + return -EINVAL; +} + +int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state) +{ + return -EINVAL; +} + +long kvm_arch_vcpu_ioctl(struct file *filp, + unsigned int ioctl, unsigned long arg) +{ + struct kvm_vcpu *vcpu = filp->private_data; + void __user *argp = (void __user *)arg; + long r; + + switch (ioctl) { + case KVM_INTERRUPT: { + struct kvm_interrupt irq; + r = -EFAULT; + if (copy_from_user(&irq, argp, sizeof(irq))) + goto out; + r = kvm_vcpu_ioctl_interrupt(vcpu, &irq); + break; + } + default: + r = -EINVAL; + } + +out: + return r; +} + +int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log) +{ + return -ENOTSUPP; +} + +long kvm_arch_vm_ioctl(struct file *filp, + unsigned int ioctl, unsigned long arg) +{ + long r; + + switch (ioctl) { + default: + r = -EINVAL; + } + + return r; +} + +int kvm_arch_init(void *opaque) +{ + return 0; +} + +void kvm_arch_exit(void) +{ +} diff --git a/arch/powerpc/mm/fsl_booke_mmu.c b/arch/powerpc/mm/fsl_booke_mmu.c index ada249bf9779..ce10e2b1b902 100644 --- a/arch/powerpc/mm/fsl_booke_mmu.c +++ b/arch/powerpc/mm/fsl_booke_mmu.c @@ -202,7 +202,7 @@ adjust_total_lowmem(void) cam_max_size = max_lowmem_size; /* adjust lowmem size to max_lowmem_size */ - ram = min(max_lowmem_size, total_lowmem); + ram = min(max_lowmem_size, (phys_addr_t)total_lowmem); /* Calculate CAM values */ __cam0 = 1UL << 2 * (__ilog2(ram) / 2); diff --git a/arch/powerpc/mm/hash_low_32.S b/arch/powerpc/mm/hash_low_32.S index e10d76a860d3..ddeaf9e38ad5 100644 --- a/arch/powerpc/mm/hash_low_32.S +++ b/arch/powerpc/mm/hash_low_32.S @@ -191,7 +191,7 @@ _GLOBAL(add_hash_page) add r3,r3,r0 /* note create_hpte trims to 24 bits */ #ifdef CONFIG_SMP - rlwinm r8,r1,0,0,18 /* use cpu number to make tag */ + rlwinm r8,r1,0,0,(31-THREAD_SHIFT) /* use cpu number to make tag */ lwz r8,TI_CPU(r8) /* to go in mmu_hash_lock */ oris r8,r8,12 #endif /* CONFIG_SMP */ @@ -526,7 +526,7 @@ _GLOBAL(flush_hash_pages) #ifdef CONFIG_SMP addis r9,r7,mmu_hash_lock@ha addi r9,r9,mmu_hash_lock@l - rlwinm r8,r1,0,0,18 + rlwinm r8,r1,0,0,(31-THREAD_SHIFT) add r8,r8,r7 lwz r8,TI_CPU(r8) oris r8,r8,9 diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c index 47325f23c51f..1952b4d3fa7f 100644 --- a/arch/powerpc/mm/init_32.c +++ b/arch/powerpc/mm/init_32.c @@ -59,7 +59,10 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); unsigned long total_memory; unsigned long total_lowmem; -phys_addr_t memstart_addr; +phys_addr_t memstart_addr = (phys_addr_t)~0ull; +EXPORT_SYMBOL(memstart_addr); +phys_addr_t kernstart_addr; +EXPORT_SYMBOL(kernstart_addr); phys_addr_t lowmem_end_addr; int boot_mapsize; @@ -68,14 +71,6 @@ unsigned long agp_special_page; EXPORT_SYMBOL(agp_special_page); #endif -#ifdef CONFIG_HIGHMEM -pte_t *kmap_pte; -pgprot_t kmap_prot; - -EXPORT_SYMBOL(kmap_prot); -EXPORT_SYMBOL(kmap_pte); -#endif - void MMU_init(void); /* XXX should be in current.h -- paulus */ diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c index 698bd000f98b..c5ac532a0161 100644 --- a/arch/powerpc/mm/init_64.c +++ b/arch/powerpc/mm/init_64.c @@ -72,7 +72,8 @@ #warning TASK_SIZE is smaller than it needs to be. #endif -phys_addr_t memstart_addr; +phys_addr_t memstart_addr = ~0; +phys_addr_t kernstart_addr; void free_initmem(void) { diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index 16def4dcff6d..d9e37f365b54 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -45,6 +45,7 @@ #include <asm/tlb.h> #include <asm/sections.h> #include <asm/vdso.h> +#include <asm/fixmap.h> #include "mmu_decl.h" @@ -57,6 +58,20 @@ int init_bootmem_done; int mem_init_done; unsigned long memory_limit; +#ifdef CONFIG_HIGHMEM +pte_t *kmap_pte; +pgprot_t kmap_prot; + +EXPORT_SYMBOL(kmap_prot); +EXPORT_SYMBOL(kmap_pte); + +static inline pte_t *virt_to_kpte(unsigned long vaddr) +{ + return pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr), + vaddr), vaddr), vaddr); +} +#endif + int page_is_ram(unsigned long pfn) { unsigned long paddr = (pfn << PAGE_SHIFT); @@ -95,15 +110,6 @@ EXPORT_SYMBOL(phys_mem_access_prot); #ifdef CONFIG_MEMORY_HOTPLUG -void online_page(struct page *page) -{ - ClearPageReserved(page); - init_page_count(page); - __free_page(page); - totalram_pages++; - num_physpages++; -} - #ifdef CONFIG_NUMA int memory_add_physaddr_to_nid(u64 start) { @@ -216,7 +222,7 @@ void __init do_init_bootmem(void) unsigned long total_pages; int boot_mapsize; - max_pfn = lmb_end_of_DRAM() >> PAGE_SHIFT; + max_low_pfn = max_pfn = lmb_end_of_DRAM() >> PAGE_SHIFT; total_pages = (lmb_end_of_DRAM() - memstart_addr) >> PAGE_SHIFT; #ifdef CONFIG_HIGHMEM total_pages = total_lowmem >> PAGE_SHIFT; @@ -232,7 +238,8 @@ void __init do_init_bootmem(void) start = lmb_alloc(bootmap_pages << PAGE_SHIFT, PAGE_SIZE); - boot_mapsize = init_bootmem(start >> PAGE_SHIFT, total_pages); + min_low_pfn = MEMORY_START >> PAGE_SHIFT; + boot_mapsize = init_bootmem_node(NODE_DATA(0), start >> PAGE_SHIFT, min_low_pfn, max_low_pfn); /* Add active regions with valid PFNs */ for (i = 0; i < lmb.memory.cnt; i++) { @@ -310,14 +317,19 @@ void __init paging_init(void) unsigned long top_of_ram = lmb_end_of_DRAM(); unsigned long max_zone_pfns[MAX_NR_ZONES]; +#ifdef CONFIG_PPC32 + unsigned long v = __fix_to_virt(__end_of_fixed_addresses - 1); + unsigned long end = __fix_to_virt(FIX_HOLE); + + for (; v < end; v += PAGE_SIZE) + map_page(v, 0, 0); /* XXX gross */ +#endif + #ifdef CONFIG_HIGHMEM map_page(PKMAP_BASE, 0, 0); /* XXX gross */ - pkmap_page_table = pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k - (PKMAP_BASE), PKMAP_BASE), PKMAP_BASE), PKMAP_BASE); - map_page(KMAP_FIX_BEGIN, 0, 0); /* XXX gross */ - kmap_pte = pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k - (KMAP_FIX_BEGIN), KMAP_FIX_BEGIN), KMAP_FIX_BEGIN), - KMAP_FIX_BEGIN); + pkmap_page_table = virt_to_kpte(PKMAP_BASE); + + kmap_pte = virt_to_kpte(__fix_to_virt(FIX_KMAP_BEGIN)); kmap_prot = PAGE_KERNEL; #endif /* CONFIG_HIGHMEM */ diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c index 1efd631211ef..dc704da363eb 100644 --- a/arch/powerpc/mm/numa.c +++ b/arch/powerpc/mm/numa.c @@ -18,6 +18,7 @@ #include <linux/cpu.h> #include <linux/notifier.h> #include <linux/lmb.h> +#include <linux/of.h> #include <asm/sparsemem.h> #include <asm/prom.h> #include <asm/system.h> diff --git a/arch/powerpc/mm/pgtable_32.c b/arch/powerpc/mm/pgtable_32.c index 64c44bcc68de..80d1babb230d 100644 --- a/arch/powerpc/mm/pgtable_32.c +++ b/arch/powerpc/mm/pgtable_32.c @@ -29,6 +29,7 @@ #include <asm/pgtable.h> #include <asm/pgalloc.h> +#include <asm/fixmap.h> #include <asm/io.h> #include "mmu_decl.h" @@ -387,3 +388,25 @@ void kernel_map_pages(struct page *page, int numpages, int enable) change_page_attr(page, numpages, enable ? PAGE_KERNEL : __pgprot(0)); } #endif /* CONFIG_DEBUG_PAGEALLOC */ + +static int fixmaps; +unsigned long FIXADDR_TOP = 0xfffff000; +EXPORT_SYMBOL(FIXADDR_TOP); + +void __set_fixmap (enum fixed_addresses idx, phys_addr_t phys, pgprot_t flags) +{ + unsigned long address = __fix_to_virt(idx); + + if (idx >= __end_of_fixed_addresses) { + BUG(); + return; + } + + map_page(address, phys, flags); + fixmaps++; +} + +void __this_fixmap_does_not_exist(void) +{ + WARN_ON(1); +} diff --git a/arch/powerpc/platforms/86xx/mpc8610_hpcd.c b/arch/powerpc/platforms/86xx/mpc8610_hpcd.c index 18b8ebe930d5..5e1e8cf14e75 100644 --- a/arch/powerpc/platforms/86xx/mpc8610_hpcd.c +++ b/arch/powerpc/platforms/86xx/mpc8610_hpcd.c @@ -3,11 +3,12 @@ * * Initial author: Xianghua Xiao <x.xiao@freescale.com> * Recode: Jason Jin <jason.jin@freescale.com> + * York Sun <yorksun@freescale.com> * * Rewrite the interrupt routing. remove the 8259PIC support, * All the integrated device in ULI use sideband interrupt. * - * Copyright 2007 Freescale Semiconductor Inc. + * Copyright 2008 Freescale Semiconductor Inc. * * 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 @@ -38,6 +39,8 @@ #include <sysdev/fsl_pci.h> #include <sysdev/fsl_soc.h> +static unsigned char *pixis_bdcfg0, *pixis_arch; + static struct of_device_id __initdata mpc8610_ids[] = { { .compatible = "fsl,mpc8610-immr", }, {} @@ -52,8 +55,7 @@ static int __init mpc8610_declare_of_platform_devices(void) } machine_device_initcall(mpc86xx_hpcd, mpc8610_declare_of_platform_devices); -static void __init -mpc86xx_hpcd_init_irq(void) +static void __init mpc86xx_hpcd_init_irq(void) { struct mpic *mpic1; struct device_node *np; @@ -161,12 +163,159 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AL, 0x5229, quirk_uli5229); DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AL, 0x5288, final_uli5288); #endif /* CONFIG_PCI */ -static void __init -mpc86xx_hpcd_setup_arch(void) +#if defined(CONFIG_FB_FSL_DIU) || defined(CONFIG_FB_FSL_DIU_MODULE) + +static u32 get_busfreq(void) { -#ifdef CONFIG_PCI - struct device_node *np; + struct device_node *node; + + u32 fs_busfreq = 0; + node = of_find_node_by_type(NULL, "cpu"); + if (node) { + unsigned int size; + const unsigned int *prop = + of_get_property(node, "bus-frequency", &size); + if (prop) + fs_busfreq = *prop; + of_node_put(node); + }; + return fs_busfreq; +} + +unsigned int mpc8610hpcd_get_pixel_format(unsigned int bits_per_pixel, + int monitor_port) +{ + static const unsigned long pixelformat[][3] = { + {0x88882317, 0x88083218, 0x65052119}, + {0x88883316, 0x88082219, 0x65053118}, + }; + unsigned int pix_fmt, arch_monitor; + + arch_monitor = ((*pixis_arch == 0x01) && (monitor_port == 0))? 0 : 1; + /* DVI port for board version 0x01 */ + + if (bits_per_pixel == 32) + pix_fmt = pixelformat[arch_monitor][0]; + else if (bits_per_pixel == 24) + pix_fmt = pixelformat[arch_monitor][1]; + else if (bits_per_pixel == 16) + pix_fmt = pixelformat[arch_monitor][2]; + else + pix_fmt = pixelformat[1][0]; + + return pix_fmt; +} + +void mpc8610hpcd_set_gamma_table(int monitor_port, char *gamma_table_base) +{ + int i; + if (monitor_port == 2) { /* dual link LVDS */ + for (i = 0; i < 256*3; i++) + gamma_table_base[i] = (gamma_table_base[i] << 2) | + ((gamma_table_base[i] >> 6) & 0x03); + } +} + +void mpc8610hpcd_set_monitor_port(int monitor_port) +{ + static const u8 bdcfg[] = {0xBD, 0xB5, 0xA5}; + if (monitor_port < 3) + *pixis_bdcfg0 = bdcfg[monitor_port]; +} + +void mpc8610hpcd_set_pixel_clock(unsigned int pixclock) +{ + u32 __iomem *clkdvdr; + u32 temp; + /* variables for pixel clock calcs */ + ulong bestval, bestfreq, speed_ccb, minpixclock, maxpixclock; + ulong pixval; + long err; + int i; + + clkdvdr = ioremap(get_immrbase() + 0xe0800, sizeof(u32)); + if (!clkdvdr) { + printk(KERN_ERR "Err: can't map clock divider register!\n"); + return; + } + + /* Pixel Clock configuration */ + pr_debug("DIU: Bus Frequency = %d\n", get_busfreq()); + speed_ccb = get_busfreq(); + + /* Calculate the pixel clock with the smallest error */ + /* calculate the following in steps to avoid overflow */ + pr_debug("DIU pixclock in ps - %d\n", pixclock); + temp = 1000000000/pixclock; + temp *= 1000; + pixclock = temp; + pr_debug("DIU pixclock freq - %u\n", pixclock); + + temp = pixclock * 5 / 100; + pr_debug("deviation = %d\n", temp); + minpixclock = pixclock - temp; + maxpixclock = pixclock + temp; + pr_debug("DIU minpixclock - %lu\n", minpixclock); + pr_debug("DIU maxpixclock - %lu\n", maxpixclock); + pixval = speed_ccb/pixclock; + pr_debug("DIU pixval = %lu\n", pixval); + + err = 100000000; + bestval = pixval; + pr_debug("DIU bestval = %lu\n", bestval); + + bestfreq = 0; + for (i = -1; i <= 1; i++) { + temp = speed_ccb / ((pixval+i) + 1); + pr_debug("DIU test pixval i= %d, pixval=%lu, temp freq. = %u\n", + i, pixval, temp); + if ((temp < minpixclock) || (temp > maxpixclock)) + pr_debug("DIU exceeds monitor range (%lu to %lu)\n", + minpixclock, maxpixclock); + else if (abs(temp - pixclock) < err) { + pr_debug("Entered the else if block %d\n", i); + err = abs(temp - pixclock); + bestval = pixval+i; + bestfreq = temp; + } + } + + pr_debug("DIU chose = %lx\n", bestval); + pr_debug("DIU error = %ld\n NomPixClk ", err); + pr_debug("DIU: Best Freq = %lx\n", bestfreq); + /* Modify PXCLK in GUTS CLKDVDR */ + pr_debug("DIU: Current value of CLKDVDR = 0x%08x\n", (*clkdvdr)); + temp = (*clkdvdr) & 0x2000FFFF; + *clkdvdr = temp; /* turn off clock */ + *clkdvdr = temp | 0x80000000 | (((bestval) & 0x1F) << 16); + pr_debug("DIU: Modified value of CLKDVDR = 0x%08x\n", (*clkdvdr)); + iounmap(clkdvdr); +} + +ssize_t mpc8610hpcd_show_monitor_port(int monitor_port, char *buf) +{ + return snprintf(buf, PAGE_SIZE, + "%c0 - DVI\n" + "%c1 - Single link LVDS\n" + "%c2 - Dual link LVDS\n", + monitor_port == 0 ? '*' : ' ', + monitor_port == 1 ? '*' : ' ', + monitor_port == 2 ? '*' : ' '); +} + +int mpc8610hpcd_set_sysfs_monitor_port(int val) +{ + return val < 3 ? val : 0; +} + #endif + +static void __init mpc86xx_hpcd_setup_arch(void) +{ + struct resource r; + struct device_node *np; + unsigned char *pixis; + if (ppc_md.progress) ppc_md.progress("mpc86xx_hpcd_setup_arch()", 0); @@ -183,6 +332,30 @@ mpc86xx_hpcd_setup_arch(void) } } #endif +#if defined(CONFIG_FB_FSL_DIU) || defined(CONFIG_FB_FSL_DIU_MODULE) + preallocate_diu_videomemory(); + diu_ops.get_pixel_format = mpc8610hpcd_get_pixel_format; + diu_ops.set_gamma_table = mpc8610hpcd_set_gamma_table; + diu_ops.set_monitor_port = mpc8610hpcd_set_monitor_port; + diu_ops.set_pixel_clock = mpc8610hpcd_set_pixel_clock; + diu_ops.show_monitor_port = mpc8610hpcd_show_monitor_port; + diu_ops.set_sysfs_monitor_port = mpc8610hpcd_set_sysfs_monitor_port; +#endif + + np = of_find_compatible_node(NULL, NULL, "fsl,fpga-pixis"); + if (np) { + of_address_to_resource(np, 0, &r); + of_node_put(np); + pixis = ioremap(r.start, 32); + if (!pixis) { + printk(KERN_ERR "Err: can't map FPGA cfg register!\n"); + return; + } + pixis_bdcfg0 = pixis + 8; + pixis_arch = pixis + 1; + } else + printk(KERN_ERR "Err: " + "can't find device node 'fsl,fpga-pixis'\n"); printk("MPC86xx HPCD board from Freescale Semiconductor\n"); } @@ -200,8 +373,7 @@ static int __init mpc86xx_hpcd_probe(void) return 0; } -static long __init -mpc86xx_time_init(void) +static long __init mpc86xx_time_init(void) { unsigned int temp; diff --git a/arch/powerpc/platforms/Kconfig b/arch/powerpc/platforms/Kconfig index f38c50b4ce56..87454c526973 100644 --- a/arch/powerpc/platforms/Kconfig +++ b/arch/powerpc/platforms/Kconfig @@ -45,7 +45,6 @@ source "arch/powerpc/platforms/powermac/Kconfig" source "arch/powerpc/platforms/prep/Kconfig" source "arch/powerpc/platforms/maple/Kconfig" source "arch/powerpc/platforms/pasemi/Kconfig" -source "arch/powerpc/platforms/celleb/Kconfig" source "arch/powerpc/platforms/ps3/Kconfig" source "arch/powerpc/platforms/cell/Kconfig" source "arch/powerpc/platforms/8xx/Kconfig" diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype index 5fc7fac10e93..f7efaa925a13 100644 --- a/arch/powerpc/platforms/Kconfig.cputype +++ b/arch/powerpc/platforms/Kconfig.cputype @@ -220,8 +220,8 @@ config SMP If you don't know what to do here, say N. config NR_CPUS - int "Maximum number of CPUs (2-128)" - range 2 128 + int "Maximum number of CPUs (2-1024)" + range 2 1024 depends on SMP default "32" if PPC64 default "4" diff --git a/arch/powerpc/platforms/Makefile b/arch/powerpc/platforms/Makefile index a984894466d9..423a0234dc31 100644 --- a/arch/powerpc/platforms/Makefile +++ b/arch/powerpc/platforms/Makefile @@ -24,5 +24,4 @@ obj-$(CONFIG_PPC_MAPLE) += maple/ obj-$(CONFIG_PPC_PASEMI) += pasemi/ obj-$(CONFIG_PPC_CELL) += cell/ obj-$(CONFIG_PPC_PS3) += ps3/ -obj-$(CONFIG_PPC_CELLEB) += celleb/ obj-$(CONFIG_EMBEDDED6xx) += embedded6xx/ diff --git a/arch/powerpc/platforms/cell/Kconfig b/arch/powerpc/platforms/cell/Kconfig index 2f169991896d..3959fcfe731c 100644 --- a/arch/powerpc/platforms/cell/Kconfig +++ b/arch/powerpc/platforms/cell/Kconfig @@ -25,6 +25,19 @@ config PPC_IBM_CELL_BLADE select PPC_UDBG_16550 select UDBG_RTAS_CONSOLE +config PPC_CELLEB + bool "Toshiba's Cell Reference Set 'Celleb' Architecture" + depends on PPC_MULTIPLATFORM && PPC64 + select PPC_CELL + select PPC_CELL_NATIVE + select PPC_RTAS + select PPC_INDIRECT_IO + select PPC_OF_PLATFORM_PCI + select HAS_TXX9_SERIAL + select PPC_UDBG_BEAT + select USB_OHCI_BIG_ENDIAN_MMIO + select USB_EHCI_BIG_ENDIAN_MMIO + menu "Cell Broadband Engine options" depends on PPC_CELL diff --git a/arch/powerpc/platforms/cell/Makefile b/arch/powerpc/platforms/cell/Makefile index c89964c6fb1f..c2a7e4e5ddf9 100644 --- a/arch/powerpc/platforms/cell/Makefile +++ b/arch/powerpc/platforms/cell/Makefile @@ -1,6 +1,7 @@ obj-$(CONFIG_PPC_CELL_NATIVE) += interrupt.o iommu.o setup.o \ cbe_regs.o spider-pic.o \ - pervasive.o pmu.o io-workarounds.o + pervasive.o pmu.o io-workarounds.o \ + spider-pci.o obj-$(CONFIG_CBE_RAS) += ras.o obj-$(CONFIG_CBE_THERM) += cbe_thermal.o @@ -26,3 +27,20 @@ obj-$(CONFIG_SPU_BASE) += spu_callbacks.o spu_base.o \ spufs/ obj-$(CONFIG_PCI_MSI) += axon_msi.o + + +# celleb stuff +ifeq ($(CONFIG_PPC_CELLEB),y) +obj-y += celleb_setup.o \ + celleb_pci.o celleb_scc_epci.o \ + celleb_scc_pciex.o \ + celleb_scc_uhc.o \ + io-workarounds.o spider-pci.o \ + beat.o beat_htab.o beat_hvCall.o \ + beat_interrupt.o beat_iommu.o + +obj-$(CONFIG_SMP) += beat_smp.o +obj-$(CONFIG_PPC_UDBG_BEAT) += beat_udbg.o +obj-$(CONFIG_SERIAL_TXX9) += celleb_scc_sio.o +obj-$(CONFIG_SPU_BASE) += beat_spu_priv1.o +endif diff --git a/arch/powerpc/platforms/cell/axon_msi.c b/arch/powerpc/platforms/cell/axon_msi.c index d95e71dee91f..c39f5c225f2e 100644 --- a/arch/powerpc/platforms/cell/axon_msi.c +++ b/arch/powerpc/platforms/cell/axon_msi.c @@ -123,7 +123,7 @@ static struct axon_msic *find_msi_translator(struct pci_dev *dev) return NULL; } - for (; dn; tmp = of_get_parent(dn), of_node_put(dn), dn = tmp) { + for (; dn; dn = of_get_next_parent(dn)) { ph = of_get_property(dn, "msi-translator", NULL); if (ph) break; @@ -169,7 +169,7 @@ static int axon_msi_check_device(struct pci_dev *dev, int nvec, int type) static int setup_msi_msg_address(struct pci_dev *dev, struct msi_msg *msg) { - struct device_node *dn, *tmp; + struct device_node *dn; struct msi_desc *entry; int len; const u32 *prop; @@ -182,7 +182,7 @@ static int setup_msi_msg_address(struct pci_dev *dev, struct msi_msg *msg) entry = list_first_entry(&dev->msi_list, struct msi_desc, list); - for (; dn; tmp = of_get_parent(dn), of_node_put(dn), dn = tmp) { + for (; dn; dn = of_get_next_parent(dn)) { if (entry->msi_attrib.is_64) { prop = of_get_property(dn, "msi-address-64", &len); if (prop) diff --git a/arch/powerpc/platforms/celleb/beat.c b/arch/powerpc/platforms/cell/beat.c index b64b171f245b..48c690ea65da 100644 --- a/arch/powerpc/platforms/celleb/beat.c +++ b/arch/powerpc/platforms/cell/beat.c @@ -33,7 +33,7 @@ #include "beat_wrapper.h" #include "beat.h" -#include "interrupt.h" +#include "beat_interrupt.h" static int beat_pm_poweroff_flag; diff --git a/arch/powerpc/platforms/celleb/beat.h b/arch/powerpc/platforms/cell/beat.h index 32c8efcedc80..32c8efcedc80 100644 --- a/arch/powerpc/platforms/celleb/beat.h +++ b/arch/powerpc/platforms/cell/beat.h diff --git a/arch/powerpc/platforms/celleb/htab.c b/arch/powerpc/platforms/cell/beat_htab.c index 81467ff055c8..81467ff055c8 100644 --- a/arch/powerpc/platforms/celleb/htab.c +++ b/arch/powerpc/platforms/cell/beat_htab.c diff --git a/arch/powerpc/platforms/celleb/hvCall.S b/arch/powerpc/platforms/cell/beat_hvCall.S index 74c817448948..74c817448948 100644 --- a/arch/powerpc/platforms/celleb/hvCall.S +++ b/arch/powerpc/platforms/cell/beat_hvCall.S diff --git a/arch/powerpc/platforms/celleb/interrupt.c b/arch/powerpc/platforms/cell/beat_interrupt.c index 69562a867876..192a93509372 100644 --- a/arch/powerpc/platforms/celleb/interrupt.c +++ b/arch/powerpc/platforms/cell/beat_interrupt.c @@ -26,7 +26,7 @@ #include <asm/machdep.h> -#include "interrupt.h" +#include "beat_interrupt.h" #include "beat_wrapper.h" #define MAX_IRQS NR_IRQS diff --git a/arch/powerpc/platforms/celleb/interrupt.h b/arch/powerpc/platforms/cell/beat_interrupt.h index b470fd0051f1..b470fd0051f1 100644 --- a/arch/powerpc/platforms/celleb/interrupt.h +++ b/arch/powerpc/platforms/cell/beat_interrupt.h diff --git a/arch/powerpc/platforms/celleb/iommu.c b/arch/powerpc/platforms/cell/beat_iommu.c index 93b0efddd658..93b0efddd658 100644 --- a/arch/powerpc/platforms/celleb/iommu.c +++ b/arch/powerpc/platforms/cell/beat_iommu.c diff --git a/arch/powerpc/platforms/celleb/smp.c b/arch/powerpc/platforms/cell/beat_smp.c index a7631250aeb4..26efc204c47f 100644 --- a/arch/powerpc/platforms/celleb/smp.c +++ b/arch/powerpc/platforms/cell/beat_smp.c @@ -37,7 +37,7 @@ #include <asm/machdep.h> #include <asm/udbg.h> -#include "interrupt.h" +#include "beat_interrupt.h" #ifdef DEBUG #define DBG(fmt...) udbg_printf(fmt) diff --git a/arch/powerpc/platforms/celleb/spu_priv1.c b/arch/powerpc/platforms/cell/beat_spu_priv1.c index bcc17f7fe8ad..bcc17f7fe8ad 100644 --- a/arch/powerpc/platforms/celleb/spu_priv1.c +++ b/arch/powerpc/platforms/cell/beat_spu_priv1.c diff --git a/arch/powerpc/platforms/celleb/beat_syscall.h b/arch/powerpc/platforms/cell/beat_syscall.h index 8580dc7e1798..8580dc7e1798 100644 --- a/arch/powerpc/platforms/celleb/beat_syscall.h +++ b/arch/powerpc/platforms/cell/beat_syscall.h diff --git a/arch/powerpc/platforms/celleb/udbg_beat.c b/arch/powerpc/platforms/cell/beat_udbg.c index 6b418f6b6175..6b418f6b6175 100644 --- a/arch/powerpc/platforms/celleb/udbg_beat.c +++ b/arch/powerpc/platforms/cell/beat_udbg.c diff --git a/arch/powerpc/platforms/celleb/beat_wrapper.h b/arch/powerpc/platforms/cell/beat_wrapper.h index b47dfda48d06..b47dfda48d06 100644 --- a/arch/powerpc/platforms/celleb/beat_wrapper.h +++ b/arch/powerpc/platforms/cell/beat_wrapper.h diff --git a/arch/powerpc/platforms/celleb/pci.c b/arch/powerpc/platforms/cell/celleb_pci.c index 51b390d34e4d..f39a3b2a1667 100644 --- a/arch/powerpc/platforms/celleb/pci.c +++ b/arch/powerpc/platforms/cell/celleb_pci.c @@ -37,12 +37,11 @@ #include <asm/io.h> #include <asm/irq.h> #include <asm/prom.h> -#include <asm/machdep.h> #include <asm/pci-bridge.h> #include <asm/ppc-pci.h> -#include "pci.h" -#include "interrupt.h" +#include "io-workarounds.h" +#include "celleb_pci.h" #define MAX_PCI_DEVICES 32 #define MAX_PCI_FUNCTIONS 8 @@ -190,7 +189,7 @@ static int celleb_fake_pci_read_config(struct pci_bus *bus, static int celleb_fake_pci_write_config(struct pci_bus *bus, - unsigned int devfn, int where, int size, u32 val) + unsigned int devfn, int where, int size, u32 val) { char *config; struct device_node *node; @@ -457,33 +456,42 @@ static int __init celleb_setup_fake_pci(struct device_node *dev, return 0; } -void __init fake_pci_workaround_init(struct pci_controller *phb) -{ - /** - * We will add fake pci bus to scc_pci_bus for the purpose to improve - * I/O Macro performance. But device-tree and device drivers - * are not ready to use address with a token. - */ - - /* celleb_pci_add_one(phb, NULL); */ -} +static struct celleb_phb_spec celleb_fake_pci_spec __initdata = { + .setup = celleb_setup_fake_pci, +}; static struct of_device_id celleb_phb_match[] __initdata = { { .name = "pci-pseudo", - .data = celleb_setup_fake_pci, + .data = &celleb_fake_pci_spec, }, { .name = "epci", - .data = celleb_setup_epci, + .data = &celleb_epci_spec, + }, { + .name = "pcie", + .data = &celleb_pciex_spec, }, { }, }; +static int __init celleb_io_workaround_init(struct pci_controller *phb, + struct celleb_phb_spec *phb_spec) +{ + if (phb_spec->ops) { + iowa_register_bus(phb, phb_spec->ops, phb_spec->iowa_init, + phb_spec->iowa_data); + io_workaround_init(); + } + + return 0; +} + int __init celleb_setup_phb(struct pci_controller *phb) { struct device_node *dev = phb->dn; const struct of_device_id *match; - int (*setup_func)(struct device_node *, struct pci_controller *); + struct celleb_phb_spec *phb_spec; + int rc; match = of_match_node(celleb_phb_match, dev); if (!match) @@ -492,8 +500,12 @@ int __init celleb_setup_phb(struct pci_controller *phb) phb_set_bus_ranges(dev, phb); phb->buid = 1; - setup_func = match->data; - return (*setup_func)(dev, phb); + phb_spec = match->data; + rc = (*phb_spec->setup)(dev, phb); + if (rc) + return 1; + + return celleb_io_workaround_init(phb, phb_spec); } int celleb_pci_probe_mode(struct pci_bus *bus) diff --git a/arch/powerpc/platforms/celleb/pci.h b/arch/powerpc/platforms/cell/celleb_pci.h index 5d5544ffeddb..4cba1523ec50 100644 --- a/arch/powerpc/platforms/celleb/pci.h +++ b/arch/powerpc/platforms/cell/celleb_pci.h @@ -27,16 +27,19 @@ #include <asm/prom.h> #include <asm/ppc-pci.h> +#include "io-workarounds.h" + +struct celleb_phb_spec { + int (*setup)(struct device_node *, struct pci_controller *); + struct ppc_pci_io *ops; + int (*iowa_init)(struct iowa_bus *, void *); + void *iowa_data; +}; + extern int celleb_setup_phb(struct pci_controller *); extern int celleb_pci_probe_mode(struct pci_bus *); -extern int celleb_setup_epci(struct device_node *, struct pci_controller *); - -extern void *celleb_dummy_page_va; -extern int __init celleb_pci_workaround_init(void); -extern void __init celleb_pci_add_one(struct pci_controller *, - void (*)(struct pci_controller *)); -extern void fake_pci_workaround_init(struct pci_controller *); -extern void epci_workaround_init(struct pci_controller *); +extern struct celleb_phb_spec celleb_epci_spec; +extern struct celleb_phb_spec celleb_pciex_spec; #endif /* _CELLEB_PCI_H */ diff --git a/arch/powerpc/platforms/celleb/scc.h b/arch/powerpc/platforms/cell/celleb_scc.h index 6be1542a6e66..b596a711c348 100644 --- a/arch/powerpc/platforms/celleb/scc.h +++ b/arch/powerpc/platforms/cell/celleb_scc.h @@ -125,6 +125,93 @@ /* bits for SCC_EPCI_CNTOPT */ #define SCC_EPCI_CNTOPT_O2PMB 0x00000002 +/* SCC PCIEXC SMMIO registers */ +#define PEXCADRS 0x000 +#define PEXCWDATA 0x004 +#define PEXCRDATA 0x008 +#define PEXDADRS 0x010 +#define PEXDCMND 0x014 +#define PEXDWDATA 0x018 +#define PEXDRDATA 0x01c +#define PEXREQID 0x020 +#define PEXTIDMAP 0x024 +#define PEXINTMASK 0x028 +#define PEXINTSTS 0x02c +#define PEXAERRMASK 0x030 +#define PEXAERRSTS 0x034 +#define PEXPRERRMASK 0x040 +#define PEXPRERRSTS 0x044 +#define PEXPRERRID01 0x048 +#define PEXPRERRID23 0x04c +#define PEXVDMASK 0x050 +#define PEXVDSTS 0x054 +#define PEXRCVCPLIDA 0x060 +#define PEXLENERRIDA 0x068 +#define PEXPHYPLLST 0x070 +#define PEXDMRDEN0 0x100 +#define PEXDMRDADR0 0x104 +#define PEXDMRDENX 0x110 +#define PEXDMRDADRX 0x114 +#define PEXECMODE 0xf00 +#define PEXMAEA(n) (0xf50 + (8 * n)) +#define PEXMAEC(n) (0xf54 + (8 * n)) +#define PEXCCRCTRL 0xff0 + +/* SCC PCIEXC bits and shifts for PEXCADRS */ +#define PEXCADRS_BYTE_EN_SHIFT 20 +#define PEXCADRS_CMD_SHIFT 16 +#define PEXCADRS_CMD_READ (0xa << PEXCADRS_CMD_SHIFT) +#define PEXCADRS_CMD_WRITE (0xb << PEXCADRS_CMD_SHIFT) + +/* SCC PCIEXC shifts for PEXDADRS */ +#define PEXDADRS_BUSNO_SHIFT 20 +#define PEXDADRS_DEVNO_SHIFT 15 +#define PEXDADRS_FUNCNO_SHIFT 12 + +/* SCC PCIEXC bits and shifts for PEXDCMND */ +#define PEXDCMND_BYTE_EN_SHIFT 4 +#define PEXDCMND_IO_READ 0x2 +#define PEXDCMND_IO_WRITE 0x3 +#define PEXDCMND_CONFIG_READ 0xa +#define PEXDCMND_CONFIG_WRITE 0xb + +/* SCC PCIEXC bits for PEXPHYPLLST */ +#define PEXPHYPLLST_PEXPHYAPLLST 0x00000001 + +/* SCC PCIEXC bits for PEXECMODE */ +#define PEXECMODE_ALL_THROUGH 0x00000000 +#define PEXECMODE_ALL_8BIT 0x00550155 +#define PEXECMODE_ALL_16BIT 0x00aa02aa + +/* SCC PCIEXC bits for PEXCCRCTRL */ +#define PEXCCRCTRL_PEXIPCOREEN 0x00040000 +#define PEXCCRCTRL_PEXIPCONTEN 0x00020000 +#define PEXCCRCTRL_PEXPHYPLLEN 0x00010000 +#define PEXCCRCTRL_PCIEXCAOCKEN 0x00000100 + +/* SCC PCIEXC port configuration registers */ +#define PEXTCERRCHK 0x21c +#define PEXTAMAPB0 0x220 +#define PEXTAMAPL0 0x224 +#define PEXTAMAPB(n) (PEXTAMAPB0 + 8 * (n)) +#define PEXTAMAPL(n) (PEXTAMAPL0 + 8 * (n)) +#define PEXCHVC0P 0x500 +#define PEXCHVC0NP 0x504 +#define PEXCHVC0C 0x508 +#define PEXCDVC0P 0x50c +#define PEXCDVC0NP 0x510 +#define PEXCDVC0C 0x514 +#define PEXCHVCXP 0x518 +#define PEXCHVCXNP 0x51c +#define PEXCHVCXC 0x520 +#define PEXCDVCXP 0x524 +#define PEXCDVCXNP 0x528 +#define PEXCDVCXC 0x52c +#define PEXCTTRG 0x530 +#define PEXTSCTRL 0x700 +#define PEXTSSTS 0x704 +#define PEXSKPCTRL 0x708 + /* UHC registers */ #define SCC_UHC_CKRCTRL 0xff0 #define SCC_UHC_ECMODE 0xf00 diff --git a/arch/powerpc/platforms/celleb/scc_epci.c b/arch/powerpc/platforms/cell/celleb_scc_epci.c index a999b393f6f6..08c285b10e30 100644 --- a/arch/powerpc/platforms/celleb/scc_epci.c +++ b/arch/powerpc/platforms/cell/celleb_scc_epci.c @@ -30,23 +30,17 @@ #include <asm/io.h> #include <asm/irq.h> #include <asm/prom.h> -#include <asm/machdep.h> #include <asm/pci-bridge.h> #include <asm/ppc-pci.h> -#include "scc.h" -#include "pci.h" -#include "interrupt.h" +#include "celleb_scc.h" +#include "celleb_pci.h" #define MAX_PCI_DEVICES 32 #define MAX_PCI_FUNCTIONS 8 #define iob() __asm__ __volatile__("eieio; sync":::"memory") -struct epci_private { - dma_addr_t dummy_page_da; -}; - static inline PCI_IO_ADDR celleb_epci_get_epci_base( struct pci_controller *hose) { @@ -71,42 +65,6 @@ static inline PCI_IO_ADDR celleb_epci_get_epci_cfg( return hose->cfg_data; } -static void scc_epci_dummy_read(struct pci_controller *hose) -{ - PCI_IO_ADDR epci_base; - u32 val; - - epci_base = celleb_epci_get_epci_base(hose); - - val = in_be32(epci_base + SCC_EPCI_WATRP); - iosync(); - - return; -} - -void __init epci_workaround_init(struct pci_controller *hose) -{ - PCI_IO_ADDR epci_base; - PCI_IO_ADDR reg; - struct epci_private *private = hose->private_data; - - BUG_ON(!private); - - private->dummy_page_da = dma_map_single(hose->parent, - celleb_dummy_page_va, PAGE_SIZE, DMA_FROM_DEVICE); - if (private->dummy_page_da == DMA_ERROR_CODE) { - printk(KERN_ERR "EPCI: dummy read disabled. " - "Map dummy page failed.\n"); - return; - } - - celleb_pci_add_one(hose, scc_epci_dummy_read); - epci_base = celleb_epci_get_epci_base(hose); - - reg = epci_base + SCC_EPCI_DUMYRADR; - out_be32(reg, private->dummy_page_da); -} - static inline void clear_and_disable_master_abort_interrupt( struct pci_controller *hose) { @@ -151,10 +109,8 @@ static int celleb_epci_check_abort(struct pci_controller *hose, return PCIBIOS_SUCCESSFUL; } -static PCI_IO_ADDR celleb_epci_make_config_addr( - struct pci_bus *bus, - struct pci_controller *hose, - unsigned int devfn, int where) +static PCI_IO_ADDR celleb_epci_make_config_addr(struct pci_bus *bus, + struct pci_controller *hose, unsigned int devfn, int where) { PCI_IO_ADDR addr; @@ -425,8 +381,8 @@ static int __init celleb_epci_init(struct pci_controller *hose) return 0; } -int __init celleb_setup_epci(struct device_node *node, - struct pci_controller *hose) +static int __init celleb_setup_epci(struct device_node *node, + struct pci_controller *hose) { struct resource r; @@ -450,8 +406,7 @@ int __init celleb_setup_epci(struct device_node *node, if (!hose->cfg_addr) goto error; pr_debug("EPCI: cfg_addr map 0x%016lx->0x%016lx + 0x%016lx\n", - r.start, (unsigned long)hose->cfg_addr, - (r.end - r.start + 1)); + r.start, (unsigned long)hose->cfg_addr, (r.end - r.start + 1)); if (of_address_to_resource(node, 2, &r)) goto error; @@ -459,14 +414,7 @@ int __init celleb_setup_epci(struct device_node *node, if (!hose->cfg_data) goto error; pr_debug("EPCI: cfg_data map 0x%016lx->0x%016lx + 0x%016lx\n", - r.start, (unsigned long)hose->cfg_data, - (r.end - r.start + 1)); - - hose->private_data = kzalloc(sizeof(struct epci_private), GFP_KERNEL); - if (hose->private_data == NULL) { - printk(KERN_ERR "EPCI: no memory for private data.\n"); - goto error; - } + r.start, (unsigned long)hose->cfg_data, (r.end - r.start + 1)); hose->ops = &celleb_epci_ops; celleb_epci_init(hose); @@ -474,8 +422,6 @@ int __init celleb_setup_epci(struct device_node *node, return 0; error: - kfree(hose->private_data); - if (hose->cfg_addr) iounmap(hose->cfg_addr); @@ -483,3 +429,10 @@ error: iounmap(hose->cfg_data); return 1; } + +struct celleb_phb_spec celleb_epci_spec __initdata = { + .setup = celleb_setup_epci, + .ops = &spiderpci_ops, + .iowa_init = &spiderpci_iowa_init, + .iowa_data = (void *)0, +}; diff --git a/arch/powerpc/platforms/cell/celleb_scc_pciex.c b/arch/powerpc/platforms/cell/celleb_scc_pciex.c new file mode 100644 index 000000000000..31da84c458d2 --- /dev/null +++ b/arch/powerpc/platforms/cell/celleb_scc_pciex.c @@ -0,0 +1,547 @@ +/* + * Support for Celleb PCI-Express. + * + * (C) Copyright 2007-2008 TOSHIBA 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., + * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. + */ + +#undef DEBUG + +#include <linux/kernel.h> +#include <linux/pci.h> +#include <linux/string.h> +#include <linux/init.h> +#include <linux/bootmem.h> +#include <linux/delay.h> +#include <linux/interrupt.h> + +#include <asm/io.h> +#include <asm/irq.h> +#include <asm/iommu.h> +#include <asm/byteorder.h> + +#include "celleb_scc.h" +#include "celleb_pci.h" + +#define PEX_IN(base, off) in_be32((void __iomem *)(base) + (off)) +#define PEX_OUT(base, off, data) out_be32((void __iomem *)(base) + (off), (data)) + +static void scc_pciex_io_flush(struct iowa_bus *bus) +{ + (void)PEX_IN(bus->phb->cfg_addr, PEXDMRDEN0); +} + +/* + * Memory space access to device on PCIEX + */ +#define PCIEX_MMIO_READ(name, ret) \ +static ret scc_pciex_##name(const PCI_IO_ADDR addr) \ +{ \ + ret val = __do_##name(addr); \ + scc_pciex_io_flush(iowa_mem_find_bus(addr)); \ + return val; \ +} + +#define PCIEX_MMIO_READ_STR(name) \ +static void scc_pciex_##name(const PCI_IO_ADDR addr, void *buf, \ + unsigned long count) \ +{ \ + __do_##name(addr, buf, count); \ + scc_pciex_io_flush(iowa_mem_find_bus(addr)); \ +} + +PCIEX_MMIO_READ(readb, u8) +PCIEX_MMIO_READ(readw, u16) +PCIEX_MMIO_READ(readl, u32) +PCIEX_MMIO_READ(readq, u64) +PCIEX_MMIO_READ(readw_be, u16) +PCIEX_MMIO_READ(readl_be, u32) +PCIEX_MMIO_READ(readq_be, u64) +PCIEX_MMIO_READ_STR(readsb) +PCIEX_MMIO_READ_STR(readsw) +PCIEX_MMIO_READ_STR(readsl) + +static void scc_pciex_memcpy_fromio(void *dest, const PCI_IO_ADDR src, + unsigned long n) +{ + __do_memcpy_fromio(dest, src, n); + scc_pciex_io_flush(iowa_mem_find_bus(src)); +} + +/* + * I/O port access to devices on PCIEX. + */ + +static inline unsigned long get_bus_address(struct pci_controller *phb, + unsigned long port) +{ + return port - ((unsigned long)(phb->io_base_virt) - _IO_BASE); +} + +static u32 scc_pciex_read_port(struct pci_controller *phb, + unsigned long port, int size) +{ + unsigned int byte_enable; + unsigned int cmd, shift; + unsigned long addr; + u32 data, ret; + + BUG_ON(((port & 0x3ul) + size) > 4); + + addr = get_bus_address(phb, port); + shift = addr & 0x3ul; + byte_enable = ((1 << size) - 1) << shift; + cmd = PEXDCMND_IO_READ | (byte_enable << PEXDCMND_BYTE_EN_SHIFT); + PEX_OUT(phb->cfg_addr, PEXDADRS, (addr & ~0x3ul)); + PEX_OUT(phb->cfg_addr, PEXDCMND, cmd); + data = PEX_IN(phb->cfg_addr, PEXDRDATA); + ret = (data >> (shift * 8)) & (0xFFFFFFFF >> ((4 - size) * 8)); + + pr_debug("PCIEX:PIO READ:port=0x%lx, addr=0x%lx, size=%d, be=%x," + " cmd=%x, data=%x, ret=%x\n", port, addr, size, byte_enable, + cmd, data, ret); + + return ret; +} + +static void scc_pciex_write_port(struct pci_controller *phb, + unsigned long port, int size, u32 val) +{ + unsigned int byte_enable; + unsigned int cmd, shift; + unsigned long addr; + u32 data; + + BUG_ON(((port & 0x3ul) + size) > 4); + + addr = get_bus_address(phb, port); + shift = addr & 0x3ul; + byte_enable = ((1 << size) - 1) << shift; + cmd = PEXDCMND_IO_WRITE | (byte_enable << PEXDCMND_BYTE_EN_SHIFT); + data = (val & (0xFFFFFFFF >> (4 - size) * 8)) << (shift * 8); + PEX_OUT(phb->cfg_addr, PEXDADRS, (addr & ~0x3ul)); + PEX_OUT(phb->cfg_addr, PEXDCMND, cmd); + PEX_OUT(phb->cfg_addr, PEXDWDATA, data); + + pr_debug("PCIEX:PIO WRITE:port=0x%lx, addr=%lx, size=%d, val=%x," + " be=%x, cmd=%x, data=%x\n", port, addr, size, val, + byte_enable, cmd, data); +} + +static u8 __scc_pciex_inb(struct pci_controller *phb, unsigned long port) +{ + return (u8)scc_pciex_read_port(phb, port, 1); +} + +static u16 __scc_pciex_inw(struct pci_controller *phb, unsigned long port) +{ + u32 data; + if ((port & 0x3ul) < 3) + data = scc_pciex_read_port(phb, port, 2); + else { + u32 d1 = scc_pciex_read_port(phb, port, 1); + u32 d2 = scc_pciex_read_port(phb, port + 1, 1); + data = d1 | (d2 << 8); + } + return (u16)data; +} + +static u32 __scc_pciex_inl(struct pci_controller *phb, unsigned long port) +{ + unsigned int mod = port & 0x3ul; + u32 data; + if (mod == 0) + data = scc_pciex_read_port(phb, port, 4); + else { + u32 d1 = scc_pciex_read_port(phb, port, 4 - mod); + u32 d2 = scc_pciex_read_port(phb, port + 1, mod); + data = d1 | (d2 << (mod * 8)); + } + return data; +} + +static void __scc_pciex_outb(struct pci_controller *phb, + u8 val, unsigned long port) +{ + scc_pciex_write_port(phb, port, 1, (u32)val); +} + +static void __scc_pciex_outw(struct pci_controller *phb, + u16 val, unsigned long port) +{ + if ((port & 0x3ul) < 3) + scc_pciex_write_port(phb, port, 2, (u32)val); + else { + u32 d1 = val & 0x000000FF; + u32 d2 = (val & 0x0000FF00) >> 8; + scc_pciex_write_port(phb, port, 1, d1); + scc_pciex_write_port(phb, port + 1, 1, d2); + } +} + +static void __scc_pciex_outl(struct pci_controller *phb, + u32 val, unsigned long port) +{ + unsigned int mod = port & 0x3ul; + if (mod == 0) + scc_pciex_write_port(phb, port, 4, val); + else { + u32 d1 = val & (0xFFFFFFFFul >> (mod * 8)); + u32 d2 = val >> ((4 - mod) * 8); + scc_pciex_write_port(phb, port, 4 - mod, d1); + scc_pciex_write_port(phb, port + 1, mod, d2); + } +} + +#define PCIEX_PIO_FUNC(size, name) \ +static u##size scc_pciex_in##name(unsigned long port) \ +{ \ + struct iowa_bus *bus = iowa_pio_find_bus(port); \ + u##size data = __scc_pciex_in##name(bus->phb, port); \ + scc_pciex_io_flush(bus); \ + return data; \ +} \ +static void scc_pciex_ins##name(unsigned long p, void *b, unsigned long c) \ +{ \ + struct iowa_bus *bus = iowa_pio_find_bus(p); \ + u##size *dst = b; \ + for (; c != 0; c--, dst++) \ + *dst = cpu_to_le##size(__scc_pciex_in##name(bus->phb, p)); \ + scc_pciex_io_flush(bus); \ +} \ +static void scc_pciex_out##name(u##size val, unsigned long port) \ +{ \ + struct iowa_bus *bus = iowa_pio_find_bus(port); \ + __scc_pciex_out##name(bus->phb, val, port); \ +} \ +static void scc_pciex_outs##name(unsigned long p, const void *b, \ + unsigned long c) \ +{ \ + struct iowa_bus *bus = iowa_pio_find_bus(p); \ + const u##size *src = b; \ + for (; c != 0; c--, src++) \ + __scc_pciex_out##name(bus->phb, le##size##_to_cpu(*src), p); \ +} +#define cpu_to_le8(x) (x) +#define le8_to_cpu(x) (x) +PCIEX_PIO_FUNC(8, b) +PCIEX_PIO_FUNC(16, w) +PCIEX_PIO_FUNC(32, l) + +static struct ppc_pci_io scc_pciex_ops = { + .readb = scc_pciex_readb, + .readw = scc_pciex_readw, + .readl = scc_pciex_readl, + .readq = scc_pciex_readq, + .readw_be = scc_pciex_readw_be, + .readl_be = scc_pciex_readl_be, + .readq_be = scc_pciex_readq_be, + .readsb = scc_pciex_readsb, + .readsw = scc_pciex_readsw, + .readsl = scc_pciex_readsl, + .memcpy_fromio = scc_pciex_memcpy_fromio, + .inb = scc_pciex_inb, + .inw = scc_pciex_inw, + .inl = scc_pciex_inl, + .outb = scc_pciex_outb, + .outw = scc_pciex_outw, + .outl = scc_pciex_outl, + .insb = scc_pciex_insb, + .insw = scc_pciex_insw, + .insl = scc_pciex_insl, + .outsb = scc_pciex_outsb, + .outsw = scc_pciex_outsw, + .outsl = scc_pciex_outsl, +}; + +static int __init scc_pciex_iowa_init(struct iowa_bus *bus, void *data) +{ + dma_addr_t dummy_page_da; + void *dummy_page_va; + + dummy_page_va = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!dummy_page_va) { + pr_err("PCIEX:Alloc dummy_page_va failed\n"); + return -1; + } + + dummy_page_da = dma_map_single(bus->phb->parent, dummy_page_va, + PAGE_SIZE, DMA_FROM_DEVICE); + if (dma_mapping_error(dummy_page_da)) { + pr_err("PCIEX:Map dummy page failed.\n"); + kfree(dummy_page_va); + return -1; + } + + PEX_OUT(bus->phb->cfg_addr, PEXDMRDADR0, dummy_page_da); + + return 0; +} + +/* + * config space access + */ +#define MK_PEXDADRS(bus_no, dev_no, func_no, addr) \ + ((uint32_t)(((addr) & ~0x3UL) | \ + ((bus_no) << PEXDADRS_BUSNO_SHIFT) | \ + ((dev_no) << PEXDADRS_DEVNO_SHIFT) | \ + ((func_no) << PEXDADRS_FUNCNO_SHIFT))) + +#define MK_PEXDCMND_BYTE_EN(addr, size) \ + ((((0x1 << (size))-1) << ((addr) & 0x3)) << PEXDCMND_BYTE_EN_SHIFT) +#define MK_PEXDCMND(cmd, addr, size) ((cmd) | MK_PEXDCMND_BYTE_EN(addr, size)) + +static uint32_t config_read_pciex_dev(unsigned int __iomem *base, + uint64_t bus_no, uint64_t dev_no, uint64_t func_no, + uint64_t off, uint64_t size) +{ + uint32_t ret; + uint32_t addr, cmd; + + addr = MK_PEXDADRS(bus_no, dev_no, func_no, off); + cmd = MK_PEXDCMND(PEXDCMND_CONFIG_READ, off, size); + PEX_OUT(base, PEXDADRS, addr); + PEX_OUT(base, PEXDCMND, cmd); + ret = (PEX_IN(base, PEXDRDATA) + >> ((off & (4-size)) * 8)) & ((0x1 << (size * 8)) - 1); + return ret; +} + +static void config_write_pciex_dev(unsigned int __iomem *base, uint64_t bus_no, + uint64_t dev_no, uint64_t func_no, uint64_t off, uint64_t size, + uint32_t data) +{ + uint32_t addr, cmd; + + addr = MK_PEXDADRS(bus_no, dev_no, func_no, off); + cmd = MK_PEXDCMND(PEXDCMND_CONFIG_WRITE, off, size); + PEX_OUT(base, PEXDADRS, addr); + PEX_OUT(base, PEXDCMND, cmd); + PEX_OUT(base, PEXDWDATA, + (data & ((0x1 << (size * 8)) - 1)) << ((off & (4-size)) * 8)); +} + +#define MK_PEXCADRS_BYTE_EN(off, len) \ + ((((0x1 << (len)) - 1) << ((off) & 0x3)) << PEXCADRS_BYTE_EN_SHIFT) +#define MK_PEXCADRS(cmd, addr, size) \ + ((cmd) | MK_PEXCADRS_BYTE_EN(addr, size) | ((addr) & ~0x3)) +static uint32_t config_read_pciex_rc(unsigned int __iomem *base, + uint32_t where, uint32_t size) +{ + PEX_OUT(base, PEXCADRS, MK_PEXCADRS(PEXCADRS_CMD_READ, where, size)); + return (PEX_IN(base, PEXCRDATA) + >> ((where & (4 - size)) * 8)) & ((0x1 << (size * 8)) - 1); +} + +static void config_write_pciex_rc(unsigned int __iomem *base, uint32_t where, + uint32_t size, uint32_t val) +{ + uint32_t data; + + data = (val & ((0x1 << (size * 8)) - 1)) << ((where & (4 - size)) * 8); + PEX_OUT(base, PEXCADRS, MK_PEXCADRS(PEXCADRS_CMD_WRITE, where, size)); + PEX_OUT(base, PEXCWDATA, data); +} + +/* Interfaces */ +/* Note: Work-around + * On SCC PCIEXC, one device is seen on all 32 dev_no. + * As SCC PCIEXC can have only one device on the bus, we look only one dev_no. + * (dev_no = 1) + */ +static int scc_pciex_read_config(struct pci_bus *bus, unsigned int devfn, + int where, int size, unsigned int *val) +{ + struct device_node *dn; + struct pci_controller *phb; + + dn = bus->sysdata; + phb = pci_find_hose_for_OF_device(dn); + + if (bus->number == phb->first_busno && PCI_SLOT(devfn) != 1) { + *val = ~0; + return PCIBIOS_DEVICE_NOT_FOUND; + } + + if (bus->number == 0 && PCI_SLOT(devfn) == 0) + *val = config_read_pciex_rc(phb->cfg_addr, where, size); + else + *val = config_read_pciex_dev(phb->cfg_addr, bus->number, + PCI_SLOT(devfn), PCI_FUNC(devfn), where, size); + + return PCIBIOS_SUCCESSFUL; +} + +static int scc_pciex_write_config(struct pci_bus *bus, unsigned int devfn, + int where, int size, unsigned int val) +{ + struct device_node *dn; + struct pci_controller *phb; + + dn = bus->sysdata; + phb = pci_find_hose_for_OF_device(dn); + + if (bus->number == phb->first_busno && PCI_SLOT(devfn) != 1) + return PCIBIOS_DEVICE_NOT_FOUND; + + if (bus->number == 0 && PCI_SLOT(devfn) == 0) + config_write_pciex_rc(phb->cfg_addr, where, size, val); + else + config_write_pciex_dev(phb->cfg_addr, bus->number, + PCI_SLOT(devfn), PCI_FUNC(devfn), where, size, val); + return PCIBIOS_SUCCESSFUL; +} + +static struct pci_ops scc_pciex_pci_ops = { + scc_pciex_read_config, + scc_pciex_write_config, +}; + +static void pciex_clear_intr_all(unsigned int __iomem *base) +{ + PEX_OUT(base, PEXAERRSTS, 0xffffffff); + PEX_OUT(base, PEXPRERRSTS, 0xffffffff); + PEX_OUT(base, PEXINTSTS, 0xffffffff); +} + +#if 0 +static void pciex_disable_intr_all(unsigned int *base) +{ + PEX_OUT(base, PEXINTMASK, 0x0); + PEX_OUT(base, PEXAERRMASK, 0x0); + PEX_OUT(base, PEXPRERRMASK, 0x0); + PEX_OUT(base, PEXVDMASK, 0x0); +} +#endif + +static void pciex_enable_intr_all(unsigned int __iomem *base) +{ + PEX_OUT(base, PEXINTMASK, 0x0000e7f1); + PEX_OUT(base, PEXAERRMASK, 0x03ff01ff); + PEX_OUT(base, PEXPRERRMASK, 0x0001010f); + PEX_OUT(base, PEXVDMASK, 0x00000001); +} + +static void pciex_check_status(unsigned int __iomem *base) +{ + uint32_t err = 0; + uint32_t intsts, aerr, prerr, rcvcp, lenerr; + uint32_t maea, maec; + + intsts = PEX_IN(base, PEXINTSTS); + aerr = PEX_IN(base, PEXAERRSTS); + prerr = PEX_IN(base, PEXPRERRSTS); + rcvcp = PEX_IN(base, PEXRCVCPLIDA); + lenerr = PEX_IN(base, PEXLENERRIDA); + + if (intsts || aerr || prerr || rcvcp || lenerr) + err = 1; + + pr_info("PCEXC interrupt!!\n"); + pr_info("PEXINTSTS :0x%08x\n", intsts); + pr_info("PEXAERRSTS :0x%08x\n", aerr); + pr_info("PEXPRERRSTS :0x%08x\n", prerr); + pr_info("PEXRCVCPLIDA :0x%08x\n", rcvcp); + pr_info("PEXLENERRIDA :0x%08x\n", lenerr); + + /* print detail of Protection Error */ + if (intsts & 0x00004000) { + uint32_t i, n; + for (i = 0; i < 4; i++) { + n = 1 << i; + if (prerr & n) { + maea = PEX_IN(base, PEXMAEA(i)); + maec = PEX_IN(base, PEXMAEC(i)); + pr_info("PEXMAEC%d :0x%08x\n", i, maec); + pr_info("PEXMAEA%d :0x%08x\n", i, maea); + } + } + } + + if (err) + pciex_clear_intr_all(base); +} + +static irqreturn_t pciex_handle_internal_irq(int irq, void *dev_id) +{ + struct pci_controller *phb = dev_id; + + pr_debug("PCIEX:pciex_handle_internal_irq(irq=%d)\n", irq); + + BUG_ON(phb->cfg_addr == NULL); + + pciex_check_status(phb->cfg_addr); + + return IRQ_HANDLED; +} + +static __init int celleb_setup_pciex(struct device_node *node, + struct pci_controller *phb) +{ + struct resource r; + struct of_irq oirq; + int virq; + + /* SMMIO registers; used inside this file */ + if (of_address_to_resource(node, 0, &r)) { + pr_err("PCIEXC:Failed to get config resource.\n"); + return 1; + } + phb->cfg_addr = ioremap(r.start, r.end - r.start + 1); + if (!phb->cfg_addr) { + pr_err("PCIEXC:Failed to remap SMMIO region.\n"); + return 1; + } + + /* Not use cfg_data, cmd and data regs are near address reg */ + phb->cfg_data = NULL; + + /* set pci_ops */ + phb->ops = &scc_pciex_pci_ops; + + /* internal interrupt handler */ + if (of_irq_map_one(node, 1, &oirq)) { + pr_err("PCIEXC:Failed to map irq\n"); + goto error; + } + virq = irq_create_of_mapping(oirq.controller, oirq.specifier, + oirq.size); + if (request_irq(virq, pciex_handle_internal_irq, + IRQF_DISABLED, "pciex", (void *)phb)) { + pr_err("PCIEXC:Failed to request irq\n"); + goto error; + } + + /* enable all interrupts */ + pciex_clear_intr_all(phb->cfg_addr); + pciex_enable_intr_all(phb->cfg_addr); + /* MSI: TBD */ + + return 0; + +error: + phb->cfg_data = NULL; + if (phb->cfg_addr) + iounmap(phb->cfg_addr); + phb->cfg_addr = NULL; + return 1; +} + +struct celleb_phb_spec celleb_pciex_spec __initdata = { + .setup = celleb_setup_pciex, + .ops = &scc_pciex_ops, + .iowa_init = &scc_pciex_iowa_init, +}; diff --git a/arch/powerpc/platforms/celleb/scc_sio.c b/arch/powerpc/platforms/cell/celleb_scc_sio.c index 3a16c5b3c464..3a16c5b3c464 100644 --- a/arch/powerpc/platforms/celleb/scc_sio.c +++ b/arch/powerpc/platforms/cell/celleb_scc_sio.c diff --git a/arch/powerpc/platforms/celleb/scc_uhc.c b/arch/powerpc/platforms/cell/celleb_scc_uhc.c index cb4307994087..d63b720bfe3a 100644 --- a/arch/powerpc/platforms/celleb/scc_uhc.c +++ b/arch/powerpc/platforms/cell/celleb_scc_uhc.c @@ -25,7 +25,7 @@ #include <asm/io.h> #include <asm/machdep.h> -#include "scc.h" +#include "celleb_scc.h" #define UHC_RESET_WAIT_MAX 10000 diff --git a/arch/powerpc/platforms/celleb/setup.c b/arch/powerpc/platforms/cell/celleb_setup.c index f27ae1e3fb58..b11cb30decb2 100644 --- a/arch/powerpc/platforms/celleb/setup.c +++ b/arch/powerpc/platforms/cell/celleb_setup.c @@ -56,13 +56,13 @@ #include <asm/rtas.h> #include <asm/cell-regs.h> -#include "interrupt.h" +#include "beat_interrupt.h" #include "beat_wrapper.h" #include "beat.h" -#include "pci.h" -#include "../cell/interrupt.h" -#include "../cell/pervasive.h" -#include "../cell/ras.h" +#include "celleb_pci.h" +#include "interrupt.h" +#include "pervasive.h" +#include "ras.h" static char celleb_machine_type[128] = "Celleb"; @@ -114,8 +114,6 @@ static int __init celleb_publish_devices(void) /* Publish OF platform devices for southbridge IOs */ of_platform_bus_probe(NULL, celleb_bus_ids, NULL); - celleb_pci_workaround_init(); - return 0; } machine_device_initcall(celleb_beat, celleb_publish_devices); diff --git a/arch/powerpc/platforms/cell/io-workarounds.c b/arch/powerpc/platforms/cell/io-workarounds.c index 979d4b67efb4..3b84e8be314c 100644 --- a/arch/powerpc/platforms/cell/io-workarounds.c +++ b/arch/powerpc/platforms/cell/io-workarounds.c @@ -1,6 +1,9 @@ /* + * Support PCI IO workaround + * * Copyright (C) 2006 Benjamin Herrenschmidt <benh@kernel.crashing.org> * IBM, Corp. + * (C) Copyright 2007-2008 TOSHIBA CORPORATION * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -9,335 +12,174 @@ #undef DEBUG #include <linux/kernel.h> -#include <linux/mm.h> -#include <linux/pci.h> + #include <asm/io.h> #include <asm/machdep.h> -#include <asm/pci-bridge.h> +#include <asm/pgtable.h> #include <asm/ppc-pci.h> +#include "io-workarounds.h" -#define SPIDER_PCI_REG_BASE 0xd000 -#define SPIDER_PCI_VCI_CNTL_STAT 0x0110 -#define SPIDER_PCI_DUMMY_READ 0x0810 -#define SPIDER_PCI_DUMMY_READ_BASE 0x0814 +#define IOWA_MAX_BUS 8 -/* Undefine that to re-enable bogus prefetch - * - * Without that workaround, the chip will do bogus prefetch past - * page boundary from system memory. This setting will disable that, - * though the documentation is unclear as to the consequences of doing - * so, either purely performances, or possible misbehaviour... It's not - * clear wether the chip can handle unaligned accesses at all without - * prefetching enabled. - * - * For now, things appear to be behaving properly with that prefetching - * disabled and IDE, possibly because IDE isn't doing any unaligned - * access. - */ -#define SPIDER_DISABLE_PREFETCH +static struct iowa_bus iowa_busses[IOWA_MAX_BUS]; +static unsigned int iowa_bus_count; -#define MAX_SPIDERS 3 +static struct iowa_bus *iowa_pci_find(unsigned long vaddr, unsigned long paddr) +{ + int i, j; + struct resource *res; + unsigned long vstart, vend; -static struct spider_pci_bus { - void __iomem *regs; - unsigned long mmio_start; - unsigned long mmio_end; - unsigned long pio_vstart; - unsigned long pio_vend; -} spider_pci_busses[MAX_SPIDERS]; -static int spider_pci_count; + for (i = 0; i < iowa_bus_count; i++) { + struct iowa_bus *bus = &iowa_busses[i]; + struct pci_controller *phb = bus->phb; -static struct spider_pci_bus *spider_pci_find(unsigned long vaddr, - unsigned long paddr) -{ - int i; - - for (i = 0; i < spider_pci_count; i++) { - struct spider_pci_bus *bus = &spider_pci_busses[i]; - if (paddr && paddr >= bus->mmio_start && paddr < bus->mmio_end) - return bus; - if (vaddr && vaddr >= bus->pio_vstart && vaddr < bus->pio_vend) - return bus; + if (vaddr) { + vstart = (unsigned long)phb->io_base_virt; + vend = vstart + phb->pci_io_size - 1; + if ((vaddr >= vstart) && (vaddr <= vend)) + return bus; + } + + if (paddr) + for (j = 0; j < 3; j++) { + res = &phb->mem_resources[j]; + if (paddr >= res->start && paddr <= res->end) + return bus; + } } + return NULL; } -static void spider_io_flush(const volatile void __iomem *addr) +struct iowa_bus *iowa_mem_find_bus(const PCI_IO_ADDR addr) { - struct spider_pci_bus *bus; + struct iowa_bus *bus; int token; - /* Get platform token (set by ioremap) from address */ token = PCI_GET_ADDR_TOKEN(addr); - /* Fast path if we have a non-0 token, it indicates which bus we - * are on. - * - * If the token is 0, that means either that the ioremap was done - * before we initialized this layer, or it's a PIO operation. We - * fallback to a low path in this case. Hopefully, internal devices - * which are ioremap'ed early should use in_XX/out_XX functions - * instead of the PCI ones and thus not suffer from the slowdown. - * - * Also note that currently, the workaround will not work for areas - * that are not mapped with PTEs (bolted in the hash table). This - * is the case for ioremaps done very early at boot (before - * mem_init_done) and includes the mapping of the ISA IO space. - * - * Fortunately, none of the affected devices is expected to do DMA - * and thus there should be no problem in practice. - * - * In order to improve performances, we only do the PTE search for - * addresses falling in the PHB IO space area. That means it will - * not work for hotplug'ed PHBs but those don't exist with Spider. - */ - if (token && token <= spider_pci_count) - bus = &spider_pci_busses[token - 1]; + if (token && token <= iowa_bus_count) + bus = &iowa_busses[token - 1]; else { unsigned long vaddr, paddr; pte_t *ptep; - /* Fixup physical address */ vaddr = (unsigned long)PCI_FIX_ADDR(addr); + if (vaddr < PHB_IO_BASE || vaddr >= PHB_IO_END) + return NULL; - /* Check if it's in allowed range for PIO */ - if (vaddr < PHB_IO_BASE || vaddr > PHB_IO_END) - return; - - /* Try to find a PTE. If not, clear the paddr, we'll do - * a vaddr only lookup (PIO only) - */ ptep = find_linux_pte(init_mm.pgd, vaddr); if (ptep == NULL) paddr = 0; else paddr = pte_pfn(*ptep) << PAGE_SHIFT; + bus = iowa_pci_find(vaddr, paddr); - bus = spider_pci_find(vaddr, paddr); if (bus == NULL) - return; + return NULL; } - /* Now do the workaround - */ - (void)in_be32(bus->regs + SPIDER_PCI_DUMMY_READ); + return bus; } -static u8 spider_readb(const volatile void __iomem *addr) +struct iowa_bus *iowa_pio_find_bus(unsigned long port) { - u8 val = __do_readb(addr); - spider_io_flush(addr); - return val; + unsigned long vaddr = (unsigned long)pci_io_base + port; + return iowa_pci_find(vaddr, 0); } -static u16 spider_readw(const volatile void __iomem *addr) -{ - u16 val = __do_readw(addr); - spider_io_flush(addr); - return val; -} -static u32 spider_readl(const volatile void __iomem *addr) -{ - u32 val = __do_readl(addr); - spider_io_flush(addr); - return val; +#define DEF_PCI_AC_RET(name, ret, at, al, space, aa) \ +static ret iowa_##name at \ +{ \ + struct iowa_bus *bus; \ + bus = iowa_##space##_find_bus(aa); \ + if (bus && bus->ops && bus->ops->name) \ + return bus->ops->name al; \ + return __do_##name al; \ } -static u64 spider_readq(const volatile void __iomem *addr) -{ - u64 val = __do_readq(addr); - spider_io_flush(addr); - return val; +#define DEF_PCI_AC_NORET(name, at, al, space, aa) \ +static void iowa_##name at \ +{ \ + struct iowa_bus *bus; \ + bus = iowa_##space##_find_bus(aa); \ + if (bus && bus->ops && bus->ops->name) { \ + bus->ops->name al; \ + return; \ + } \ + __do_##name al; \ } -static u16 spider_readw_be(const volatile void __iomem *addr) -{ - u16 val = __do_readw_be(addr); - spider_io_flush(addr); - return val; -} +#include <asm/io-defs.h> -static u32 spider_readl_be(const volatile void __iomem *addr) -{ - u32 val = __do_readl_be(addr); - spider_io_flush(addr); - return val; -} +#undef DEF_PCI_AC_RET +#undef DEF_PCI_AC_NORET -static u64 spider_readq_be(const volatile void __iomem *addr) -{ - u64 val = __do_readq_be(addr); - spider_io_flush(addr); - return val; -} +static struct ppc_pci_io __initdata iowa_pci_io = { -static void spider_readsb(const volatile void __iomem *addr, void *buf, - unsigned long count) -{ - __do_readsb(addr, buf, count); - spider_io_flush(addr); -} +#define DEF_PCI_AC_RET(name, ret, at, al, space, aa) .name = iowa_##name, +#define DEF_PCI_AC_NORET(name, at, al, space, aa) .name = iowa_##name, -static void spider_readsw(const volatile void __iomem *addr, void *buf, - unsigned long count) -{ - __do_readsw(addr, buf, count); - spider_io_flush(addr); -} +#include <asm/io-defs.h> -static void spider_readsl(const volatile void __iomem *addr, void *buf, - unsigned long count) -{ - __do_readsl(addr, buf, count); - spider_io_flush(addr); -} - -static void spider_memcpy_fromio(void *dest, const volatile void __iomem *src, - unsigned long n) -{ - __do_memcpy_fromio(dest, src, n); - spider_io_flush(src); -} +#undef DEF_PCI_AC_RET +#undef DEF_PCI_AC_NORET +}; -static void __iomem * spider_ioremap(unsigned long addr, unsigned long size, - unsigned long flags) +static void __iomem *iowa_ioremap(unsigned long addr, unsigned long size, + unsigned long flags) { - struct spider_pci_bus *bus; + struct iowa_bus *bus; void __iomem *res = __ioremap(addr, size, flags); int busno; - pr_debug("spider_ioremap(0x%lx, 0x%lx, 0x%lx) -> 0x%p\n", - addr, size, flags, res); - - bus = spider_pci_find(0, addr); + bus = iowa_pci_find(0, addr); if (bus != NULL) { - busno = bus - spider_pci_busses; - pr_debug(" found bus %d, setting token\n", busno); + busno = bus - iowa_busses; PCI_SET_ADDR_TOKEN(res, busno + 1); } - pr_debug(" result=0x%p\n", res); - return res; } -static void __init spider_pci_setup_chip(struct spider_pci_bus *bus) -{ -#ifdef SPIDER_DISABLE_PREFETCH - u32 val = in_be32(bus->regs + SPIDER_PCI_VCI_CNTL_STAT); - pr_debug(" PVCI_Control_Status was 0x%08x\n", val); - out_be32(bus->regs + SPIDER_PCI_VCI_CNTL_STAT, val | 0x8); -#endif - - /* Configure the dummy address for the workaround */ - out_be32(bus->regs + SPIDER_PCI_DUMMY_READ_BASE, 0x80000000); -} - -static void __init spider_pci_add_one(struct pci_controller *phb) +/* Regist new bus to support workaround */ +void __init iowa_register_bus(struct pci_controller *phb, + struct ppc_pci_io *ops, + int (*initfunc)(struct iowa_bus *, void *), void *data) { - struct spider_pci_bus *bus = &spider_pci_busses[spider_pci_count]; + struct iowa_bus *bus; struct device_node *np = phb->dn; - struct resource rsrc; - void __iomem *regs; - if (spider_pci_count >= MAX_SPIDERS) { - printk(KERN_ERR "Too many spider bridges, workarounds" - " disabled for %s\n", np->full_name); + if (iowa_bus_count >= IOWA_MAX_BUS) { + pr_err("IOWA:Too many pci bridges, " + "workarounds disabled for %s\n", np->full_name); return; } - /* Get the registers for the beast */ - if (of_address_to_resource(np, 0, &rsrc)) { - printk(KERN_ERR "Failed to get registers for spider %s" - " workarounds disabled\n", np->full_name); - return; - } + bus = &iowa_busses[iowa_bus_count]; + bus->phb = phb; + bus->ops = ops; - /* Mask out some useless bits in there to get to the base of the - * spider chip - */ - rsrc.start &= ~0xfffffffful; - - /* Map them */ - regs = ioremap(rsrc.start + SPIDER_PCI_REG_BASE, 0x1000); - if (regs == NULL) { - printk(KERN_ERR "Failed to map registers for spider %s" - " workarounds disabled\n", np->full_name); - return; - } - - spider_pci_count++; - - /* We assume spiders only have one MMIO resource */ - bus->mmio_start = phb->mem_resources[0].start; - bus->mmio_end = phb->mem_resources[0].end + 1; - - bus->pio_vstart = (unsigned long)phb->io_base_virt; - bus->pio_vend = bus->pio_vstart + phb->pci_io_size; - - bus->regs = regs; - - printk(KERN_INFO "PCI: Spider MMIO workaround for %s\n",np->full_name); + if (initfunc) + if ((*initfunc)(bus, data)) + return; - pr_debug(" mmio (P) = 0x%016lx..0x%016lx\n", - bus->mmio_start, bus->mmio_end); - pr_debug(" pio (V) = 0x%016lx..0x%016lx\n", - bus->pio_vstart, bus->pio_vend); - pr_debug(" regs (P) = 0x%016lx (V) = 0x%p\n", - rsrc.start + SPIDER_PCI_REG_BASE, bus->regs); + iowa_bus_count++; - spider_pci_setup_chip(bus); + pr_debug("IOWA:[%d]Add bus, %s.\n", iowa_bus_count-1, np->full_name); } -static struct ppc_pci_io __initdata spider_pci_io = { - .readb = spider_readb, - .readw = spider_readw, - .readl = spider_readl, - .readq = spider_readq, - .readw_be = spider_readw_be, - .readl_be = spider_readl_be, - .readq_be = spider_readq_be, - .readsb = spider_readsb, - .readsw = spider_readsw, - .readsl = spider_readsl, - .memcpy_fromio = spider_memcpy_fromio, -}; - -static int __init spider_pci_workaround_init(void) +/* enable IO workaround */ +void __init io_workaround_init(void) { - struct pci_controller *phb; - - /* Find spider bridges. We assume they have been all probed - * in setup_arch(). If that was to change, we would need to - * update this code to cope with dynamically added busses - */ - list_for_each_entry(phb, &hose_list, list_node) { - struct device_node *np = phb->dn; - const char *model = of_get_property(np, "model", NULL); - - /* If no model property or name isn't exactly "pci", skip */ - if (model == NULL || strcmp(np->name, "pci")) - continue; - /* If model is not "Spider", skip */ - if (strcmp(model, "Spider")) - continue; - spider_pci_add_one(phb); - } - - /* No Spider PCI found, exit */ - if (spider_pci_count == 0) - return 0; + static int io_workaround_inited; - /* Setup IO callbacks. We only setup MMIO reads. PIO reads will - * fallback to MMIO reads (though without a token, thus slower) - */ - ppc_pci_io = spider_pci_io; - - /* Setup ioremap callback */ - ppc_md.ioremap = spider_ioremap; - - return 0; + if (io_workaround_inited) + return; + ppc_pci_io = iowa_pci_io; + ppc_md.ioremap = iowa_ioremap; + io_workaround_inited = 1; } -machine_arch_initcall(cell, spider_pci_workaround_init); diff --git a/arch/powerpc/platforms/cell/io-workarounds.h b/arch/powerpc/platforms/cell/io-workarounds.h new file mode 100644 index 000000000000..79d8ed3d510f --- /dev/null +++ b/arch/powerpc/platforms/cell/io-workarounds.h @@ -0,0 +1,49 @@ +/* + * Support PCI IO workaround + * + * (C) Copyright 2007-2008 TOSHIBA 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., + * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. + */ + +#ifndef _IO_WORKAROUNDS_H +#define _IO_WORKAROUNDS_H + +#include <linux/io.h> +#include <asm/pci-bridge.h> + +/* Bus info */ +struct iowa_bus { + struct pci_controller *phb; + struct ppc_pci_io *ops; + void *private; +}; + +void __init io_workaround_init(void); +void __init iowa_register_bus(struct pci_controller *, struct ppc_pci_io *, + int (*)(struct iowa_bus *, void *), void *); +struct iowa_bus *iowa_mem_find_bus(const PCI_IO_ADDR); +struct iowa_bus *iowa_pio_find_bus(unsigned long); + +extern struct ppc_pci_io spiderpci_ops; +extern int spiderpci_iowa_init(struct iowa_bus *, void *); + +#define SPIDER_PCI_REG_BASE 0xd000 +#define SPIDER_PCI_REG_SIZE 0x1000 +#define SPIDER_PCI_VCI_CNTL_STAT 0x0110 +#define SPIDER_PCI_DUMMY_READ 0x0810 +#define SPIDER_PCI_DUMMY_READ_BASE 0x0814 + +#endif /* _IO_WORKAROUNDS_H */ diff --git a/arch/powerpc/platforms/cell/setup.c b/arch/powerpc/platforms/cell/setup.c index 5c531e8f9f6f..ab721b50fbba 100644 --- a/arch/powerpc/platforms/cell/setup.c +++ b/arch/powerpc/platforms/cell/setup.c @@ -57,6 +57,7 @@ #include "interrupt.h" #include "pervasive.h" #include "ras.h" +#include "io-workarounds.h" #ifdef DEBUG #define DBG(fmt...) udbg_printf(fmt) @@ -117,13 +118,50 @@ static void cell_fixup_pcie_rootcomplex(struct pci_dev *dev) } DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, cell_fixup_pcie_rootcomplex); +static int __devinit cell_setup_phb(struct pci_controller *phb) +{ + const char *model; + struct device_node *np; + + int rc = rtas_setup_phb(phb); + if (rc) + return rc; + + np = phb->dn; + model = of_get_property(np, "model", NULL); + if (model == NULL || strcmp(np->name, "pci")) + return 0; + + /* Setup workarounds for spider */ + if (strcmp(model, "Spider")) + return 0; + + iowa_register_bus(phb, &spiderpci_ops, &spiderpci_iowa_init, + (void *)SPIDER_PCI_REG_BASE); + io_workaround_init(); + + return 0; +} + static int __init cell_publish_devices(void) { + struct device_node *root = of_find_node_by_path("/"); + struct device_node *np; int node; /* Publish OF platform devices for southbridge IOs */ of_platform_bus_probe(NULL, NULL, NULL); + /* On spider based blades, we need to manually create the OF + * platform devices for the PCI host bridges + */ + for_each_child_of_node(root, np) { + if (np->type == NULL || (strcmp(np->type, "pci") != 0 && + strcmp(np->type, "pciex") != 0)) + continue; + of_platform_device_create(np, NULL, NULL); + } + /* There is no device for the MIC memory controller, thus we create * a platform device for it to attach the EDAC driver to. */ @@ -132,6 +170,7 @@ static int __init cell_publish_devices(void) continue; platform_device_register_simple("cbe-mic", node, NULL, 0); } + return 0; } machine_subsys_initcall(cell, cell_publish_devices); @@ -213,7 +252,7 @@ static void __init cell_setup_arch(void) /* Find and initialize PCI host bridges */ init_pci_config_tokens(); - find_and_init_phbs(); + cbe_pervasive_init(); #ifdef CONFIG_DUMMY_CONSOLE conswitchp = &dummy_con; @@ -249,7 +288,7 @@ define_machine(cell) { .calibrate_decr = generic_calibrate_decr, .progress = cell_progress, .init_IRQ = cell_init_irq, - .pci_setup_phb = rtas_setup_phb, + .pci_setup_phb = cell_setup_phb, #ifdef CONFIG_KEXEC .machine_kexec = default_machine_kexec, .machine_kexec_prepare = default_machine_kexec_prepare, diff --git a/arch/powerpc/platforms/cell/spider-pci.c b/arch/powerpc/platforms/cell/spider-pci.c new file mode 100644 index 000000000000..418b605ac35a --- /dev/null +++ b/arch/powerpc/platforms/cell/spider-pci.c @@ -0,0 +1,184 @@ +/* + * IO workarounds for PCI on Celleb/Cell platform + * + * (C) Copyright 2006-2007 TOSHIBA 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., + * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. + */ + +#undef DEBUG + +#include <linux/kernel.h> +#include <linux/of_platform.h> +#include <linux/io.h> + +#include <asm/ppc-pci.h> +#include <asm/pci-bridge.h> + +#include "io-workarounds.h" + +#define SPIDER_PCI_DISABLE_PREFETCH + +struct spiderpci_iowa_private { + void __iomem *regs; +}; + +static void spiderpci_io_flush(struct iowa_bus *bus) +{ + struct spiderpci_iowa_private *priv; + u32 val; + + priv = bus->private; + val = in_be32(priv->regs + SPIDER_PCI_DUMMY_READ); + iosync(); +} + +#define SPIDER_PCI_MMIO_READ(name, ret) \ +static ret spiderpci_##name(const PCI_IO_ADDR addr) \ +{ \ + ret val = __do_##name(addr); \ + spiderpci_io_flush(iowa_mem_find_bus(addr)); \ + return val; \ +} + +#define SPIDER_PCI_MMIO_READ_STR(name) \ +static void spiderpci_##name(const PCI_IO_ADDR addr, void *buf, \ + unsigned long count) \ +{ \ + __do_##name(addr, buf, count); \ + spiderpci_io_flush(iowa_mem_find_bus(addr)); \ +} + +SPIDER_PCI_MMIO_READ(readb, u8) +SPIDER_PCI_MMIO_READ(readw, u16) +SPIDER_PCI_MMIO_READ(readl, u32) +SPIDER_PCI_MMIO_READ(readq, u64) +SPIDER_PCI_MMIO_READ(readw_be, u16) +SPIDER_PCI_MMIO_READ(readl_be, u32) +SPIDER_PCI_MMIO_READ(readq_be, u64) +SPIDER_PCI_MMIO_READ_STR(readsb) +SPIDER_PCI_MMIO_READ_STR(readsw) +SPIDER_PCI_MMIO_READ_STR(readsl) + +static void spiderpci_memcpy_fromio(void *dest, const PCI_IO_ADDR src, + unsigned long n) +{ + __do_memcpy_fromio(dest, src, n); + spiderpci_io_flush(iowa_mem_find_bus(src)); +} + +static int __init spiderpci_pci_setup_chip(struct pci_controller *phb, + void __iomem *regs) +{ + void *dummy_page_va; + dma_addr_t dummy_page_da; + +#ifdef SPIDER_PCI_DISABLE_PREFETCH + u32 val = in_be32(regs + SPIDER_PCI_VCI_CNTL_STAT); + pr_debug("SPIDER_IOWA:PVCI_Control_Status was 0x%08x\n", val); + out_be32(regs + SPIDER_PCI_VCI_CNTL_STAT, val | 0x8); +#endif /* SPIDER_PCI_DISABLE_PREFETCH */ + + /* setup dummy read */ + /* + * On CellBlade, we can't know that which XDR memory is used by + * kmalloc() to allocate dummy_page_va. + * In order to imporve the performance, the XDR which is used to + * allocate dummy_page_va is the nearest the spider-pci. + * We have to select the CBE which is the nearest the spider-pci + * to allocate memory from the best XDR, but I don't know that + * how to do. + * + * Celleb does not have this problem, because it has only one XDR. + */ + dummy_page_va = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!dummy_page_va) { + pr_err("SPIDERPCI-IOWA:Alloc dummy_page_va failed.\n"); + return -1; + } + + dummy_page_da = dma_map_single(phb->parent, dummy_page_va, + PAGE_SIZE, DMA_FROM_DEVICE); + if (dma_mapping_error(dummy_page_da)) { + pr_err("SPIDER-IOWA:Map dummy page filed.\n"); + kfree(dummy_page_va); + return -1; + } + + out_be32(regs + SPIDER_PCI_DUMMY_READ_BASE, dummy_page_da); + + return 0; +} + +int __init spiderpci_iowa_init(struct iowa_bus *bus, void *data) +{ + void __iomem *regs = NULL; + struct spiderpci_iowa_private *priv; + struct device_node *np = bus->phb->dn; + struct resource r; + unsigned long offset = (unsigned long)data; + + pr_debug("SPIDERPCI-IOWA:Bus initialize for spider(%s)\n", + np->full_name); + + priv = kzalloc(sizeof(struct spiderpci_iowa_private), GFP_KERNEL); + if (!priv) { + pr_err("SPIDERPCI-IOWA:" + "Can't allocate struct spiderpci_iowa_private"); + return -1; + } + + if (of_address_to_resource(np, 0, &r)) { + pr_err("SPIDERPCI-IOWA:Can't get resource.\n"); + goto error; + } + + regs = ioremap(r.start + offset, SPIDER_PCI_REG_SIZE); + if (!regs) { + pr_err("SPIDERPCI-IOWA:ioremap failed.\n"); + goto error; + } + priv->regs = regs; + bus->private = priv; + + if (spiderpci_pci_setup_chip(bus->phb, regs)) + goto error; + + return 0; + +error: + kfree(priv); + bus->private = NULL; + + if (regs) + iounmap(regs); + + return -1; +} + +struct ppc_pci_io spiderpci_ops = { + .readb = spiderpci_readb, + .readw = spiderpci_readw, + .readl = spiderpci_readl, + .readq = spiderpci_readq, + .readw_be = spiderpci_readw_be, + .readl_be = spiderpci_readl_be, + .readq_be = spiderpci_readq_be, + .readsb = spiderpci_readsb, + .readsw = spiderpci_readsw, + .readsl = spiderpci_readsl, + .memcpy_fromio = spiderpci_memcpy_fromio, +}; + diff --git a/arch/powerpc/platforms/celleb/Kconfig b/arch/powerpc/platforms/celleb/Kconfig deleted file mode 100644 index 372891edcdd2..000000000000 --- a/arch/powerpc/platforms/celleb/Kconfig +++ /dev/null @@ -1,12 +0,0 @@ -config PPC_CELLEB - bool "Toshiba's Cell Reference Set 'Celleb' Architecture" - depends on PPC_MULTIPLATFORM && PPC64 - select PPC_CELL - select PPC_CELL_NATIVE - select PPC_RTAS - select PPC_INDIRECT_IO - select PPC_OF_PLATFORM_PCI - select HAS_TXX9_SERIAL - select PPC_UDBG_BEAT - select USB_OHCI_BIG_ENDIAN_MMIO - select USB_EHCI_BIG_ENDIAN_MMIO diff --git a/arch/powerpc/platforms/celleb/Makefile b/arch/powerpc/platforms/celleb/Makefile deleted file mode 100644 index 889d43f715ea..000000000000 --- a/arch/powerpc/platforms/celleb/Makefile +++ /dev/null @@ -1,9 +0,0 @@ -obj-y += interrupt.o iommu.o setup.o \ - htab.o beat.o hvCall.o pci.o \ - scc_epci.o scc_uhc.o \ - io-workarounds.o - -obj-$(CONFIG_SMP) += smp.o -obj-$(CONFIG_PPC_UDBG_BEAT) += udbg_beat.o -obj-$(CONFIG_SERIAL_TXX9) += scc_sio.o -obj-$(CONFIG_SPU_BASE) += spu_priv1.o diff --git a/arch/powerpc/platforms/celleb/io-workarounds.c b/arch/powerpc/platforms/celleb/io-workarounds.c deleted file mode 100644 index 423339be1bac..000000000000 --- a/arch/powerpc/platforms/celleb/io-workarounds.c +++ /dev/null @@ -1,280 +0,0 @@ -/* - * Support for Celleb io workarounds - * - * (C) Copyright 2006-2007 TOSHIBA CORPORATION - * - * This file is based to arch/powerpc/platform/cell/io-workarounds.c - * - * 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., - * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. - */ - -#undef DEBUG - -#include <linux/of.h> -#include <linux/of_device.h> -#include <linux/irq.h> - -#include <asm/io.h> -#include <asm/prom.h> -#include <asm/machdep.h> -#include <asm/pci-bridge.h> -#include <asm/ppc-pci.h> - -#include "pci.h" - -#define MAX_CELLEB_PCI_BUS 4 - -void *celleb_dummy_page_va; - -static struct celleb_pci_bus { - struct pci_controller *phb; - void (*dummy_read)(struct pci_controller *); -} celleb_pci_busses[MAX_CELLEB_PCI_BUS]; - -static int celleb_pci_count = 0; - -static struct celleb_pci_bus *celleb_pci_find(unsigned long vaddr, - unsigned long paddr) -{ - int i, j; - struct resource *res; - - for (i = 0; i < celleb_pci_count; i++) { - struct celleb_pci_bus *bus = &celleb_pci_busses[i]; - struct pci_controller *phb = bus->phb; - if (paddr) - for (j = 0; j < 3; j++) { - res = &phb->mem_resources[j]; - if (paddr >= res->start && paddr <= res->end) - return bus; - } - res = &phb->io_resource; - if (vaddr && vaddr >= res->start && vaddr <= res->end) - return bus; - } - return NULL; -} - -static void celleb_io_flush(const PCI_IO_ADDR addr) -{ - struct celleb_pci_bus *bus; - int token; - - token = PCI_GET_ADDR_TOKEN(addr); - - if (token && token <= celleb_pci_count) - bus = &celleb_pci_busses[token - 1]; - else { - unsigned long vaddr, paddr; - pte_t *ptep; - - vaddr = (unsigned long)PCI_FIX_ADDR(addr); - if (vaddr < PHB_IO_BASE || vaddr >= PHB_IO_END) - return; - - ptep = find_linux_pte(init_mm.pgd, vaddr); - if (ptep == NULL) - paddr = 0; - else - paddr = pte_pfn(*ptep) << PAGE_SHIFT; - bus = celleb_pci_find(vaddr, paddr); - - if (bus == NULL) - return; - } - - if (bus->dummy_read) - bus->dummy_read(bus->phb); -} - -static u8 celleb_readb(const PCI_IO_ADDR addr) -{ - u8 val; - val = __do_readb(addr); - celleb_io_flush(addr); - return val; -} - -static u16 celleb_readw(const PCI_IO_ADDR addr) -{ - u16 val; - val = __do_readw(addr); - celleb_io_flush(addr); - return val; -} - -static u32 celleb_readl(const PCI_IO_ADDR addr) -{ - u32 val; - val = __do_readl(addr); - celleb_io_flush(addr); - return val; -} - -static u64 celleb_readq(const PCI_IO_ADDR addr) -{ - u64 val; - val = __do_readq(addr); - celleb_io_flush(addr); - return val; -} - -static u16 celleb_readw_be(const PCI_IO_ADDR addr) -{ - u16 val; - val = __do_readw_be(addr); - celleb_io_flush(addr); - return val; -} - -static u32 celleb_readl_be(const PCI_IO_ADDR addr) -{ - u32 val; - val = __do_readl_be(addr); - celleb_io_flush(addr); - return val; -} - -static u64 celleb_readq_be(const PCI_IO_ADDR addr) -{ - u64 val; - val = __do_readq_be(addr); - celleb_io_flush(addr); - return val; -} - -static void celleb_readsb(const PCI_IO_ADDR addr, - void *buf, unsigned long count) -{ - __do_readsb(addr, buf, count); - celleb_io_flush(addr); -} - -static void celleb_readsw(const PCI_IO_ADDR addr, - void *buf, unsigned long count) -{ - __do_readsw(addr, buf, count); - celleb_io_flush(addr); -} - -static void celleb_readsl(const PCI_IO_ADDR addr, - void *buf, unsigned long count) -{ - __do_readsl(addr, buf, count); - celleb_io_flush(addr); -} - -static void celleb_memcpy_fromio(void *dest, - const PCI_IO_ADDR src, - unsigned long n) -{ - __do_memcpy_fromio(dest, src, n); - celleb_io_flush(src); -} - -static void __iomem *celleb_ioremap(unsigned long addr, - unsigned long size, - unsigned long flags) -{ - struct celleb_pci_bus *bus; - void __iomem *res = __ioremap(addr, size, flags); - int busno; - - bus = celleb_pci_find(0, addr); - if (bus != NULL) { - busno = bus - celleb_pci_busses; - PCI_SET_ADDR_TOKEN(res, busno + 1); - } - return res; -} - -static void celleb_iounmap(volatile void __iomem *addr) -{ - return __iounmap(PCI_FIX_ADDR(addr)); -} - -static struct ppc_pci_io celleb_pci_io __initdata = { - .readb = celleb_readb, - .readw = celleb_readw, - .readl = celleb_readl, - .readq = celleb_readq, - .readw_be = celleb_readw_be, - .readl_be = celleb_readl_be, - .readq_be = celleb_readq_be, - .readsb = celleb_readsb, - .readsw = celleb_readsw, - .readsl = celleb_readsl, - .memcpy_fromio = celleb_memcpy_fromio, -}; - -void __init celleb_pci_add_one(struct pci_controller *phb, - void (*dummy_read)(struct pci_controller *)) -{ - struct celleb_pci_bus *bus = &celleb_pci_busses[celleb_pci_count]; - struct device_node *np = phb->dn; - - if (celleb_pci_count >= MAX_CELLEB_PCI_BUS) { - printk(KERN_ERR "Too many pci bridges, workarounds" - " disabled for %s\n", np->full_name); - return; - } - - celleb_pci_count++; - - bus->phb = phb; - bus->dummy_read = dummy_read; -} - -static struct of_device_id celleb_pci_workaround_match[] __initdata = { - { - .name = "pci-pseudo", - .data = fake_pci_workaround_init, - }, { - .name = "epci", - .data = epci_workaround_init, - }, { - }, -}; - -int __init celleb_pci_workaround_init(void) -{ - struct pci_controller *phb; - struct device_node *node; - const struct of_device_id *match; - void (*init_func)(struct pci_controller *); - - celleb_dummy_page_va = kmalloc(PAGE_SIZE, GFP_KERNEL); - if (!celleb_dummy_page_va) { - printk(KERN_ERR "Celleb: dummy read disabled. " - "Alloc celleb_dummy_page_va failed\n"); - return 1; - } - - list_for_each_entry(phb, &hose_list, list_node) { - node = phb->dn; - match = of_match_node(celleb_pci_workaround_match, node); - - if (match) { - init_func = match->data; - (*init_func)(phb); - } - } - - ppc_pci_io = celleb_pci_io; - ppc_md.ioremap = celleb_ioremap; - ppc_md.iounmap = celleb_iounmap; - - return 0; -} diff --git a/arch/powerpc/platforms/iseries/exception.S b/arch/powerpc/platforms/iseries/exception.S index c775cd4b3d6e..8ff330d026ca 100644 --- a/arch/powerpc/platforms/iseries/exception.S +++ b/arch/powerpc/platforms/iseries/exception.S @@ -59,8 +59,33 @@ system_reset_iSeries: andc r4,r4,r5 mtspr SPRN_CTRLT,r4 +/* Spin on __secondary_hold_spinloop until it is updated by the boot cpu. */ +/* In the UP case we'll yeild() later, and we will not access the paca anyway */ +#ifdef CONFIG_SMP 1: HMT_LOW + LOAD_REG_IMMEDIATE(r23, __secondary_hold_spinloop) + ld r23,0(r23) + sync + LOAD_REG_IMMEDIATE(r3,current_set) + sldi r28,r24,3 /* get current_set[cpu#] */ + ldx r3,r3,r28 + addi r1,r3,THREAD_SIZE + subi r1,r1,STACK_FRAME_OVERHEAD + + cmpwi 0,r23,0 /* Keep poking the Hypervisor until */ + bne 2f /* we're released */ + /* Let the Hypervisor know we are alive */ + /* 8002 is a call to HvCallCfg::getLps, a harmless Hypervisor function */ + lis r3,0x8002 + rldicr r3,r3,32,15 /* r0 = (r3 << 32) & 0xffff000000000000 */ + li r0,-1 /* r0=-1 indicates a Hypervisor call */ + sc /* Invoke the hypervisor via a system call */ + b 1b +#endif + +2: + HMT_LOW #ifdef CONFIG_SMP lbz r23,PACAPROCSTART(r13) /* Test if this processor * should start */ @@ -91,7 +116,7 @@ iSeries_secondary_smp_loop: li r0,-1 /* r0=-1 indicates a Hypervisor call */ sc /* Invoke the hypervisor via a system call */ mfspr r13,SPRN_SPRG3 /* Put r13 back ???? */ - b 1b /* If SMP not configured, secondaries + b 2b /* If SMP not configured, secondaries * loop forever */ /*** ISeries-LPAR interrupt handlers ***/ diff --git a/arch/powerpc/platforms/ps3/os-area.c b/arch/powerpc/platforms/ps3/os-area.c index c73379ec9141..1d201782d4e5 100644 --- a/arch/powerpc/platforms/ps3/os-area.c +++ b/arch/powerpc/platforms/ps3/os-area.c @@ -25,6 +25,7 @@ #include <linux/syscalls.h> #include <linux/ctype.h> #include <linux/lmb.h> +#include <linux/of.h> #include <asm/prom.h> diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig index 306a9d07491d..07fe5b69b9e2 100644 --- a/arch/powerpc/platforms/pseries/Kconfig +++ b/arch/powerpc/platforms/pseries/Kconfig @@ -34,3 +34,8 @@ config LPARCFG help Provide system capacity information via human readable <key word>=<value> pairs through a /proc/ppc64/lparcfg interface. + +config PPC_PSERIES_DEBUG + depends on PPC_PSERIES && PPC_EARLY_DEBUG + bool "Enable extra debug logging in platforms/pseries" + default y diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile index bdae04bb7a01..bd2593ed28dd 100644 --- a/arch/powerpc/platforms/pseries/Makefile +++ b/arch/powerpc/platforms/pseries/Makefile @@ -2,6 +2,10 @@ ifeq ($(CONFIG_PPC64),y) EXTRA_CFLAGS += -mno-minimal-toc endif +ifeq ($(CONFIG_PPC_PSERIES_DEBUG),y) +EXTRA_CFLAGS += -DDEBUG +endif + obj-y := lpar.o hvCall.o nvram.o reconfig.o \ setup.o iommu.o ras.o rtasd.o \ firmware.o power.o diff --git a/arch/powerpc/platforms/pseries/eeh.c b/arch/powerpc/platforms/pseries/eeh.c index 550b2f7d2cc1..a3fd56b186e6 100644 --- a/arch/powerpc/platforms/pseries/eeh.c +++ b/arch/powerpc/platforms/pseries/eeh.c @@ -39,7 +39,6 @@ #include <asm/ppc-pci.h> #include <asm/rtas.h> -#undef DEBUG /** Overview: * EEH, or "Extended Error Handling" is a PCI bridge technology for diff --git a/arch/powerpc/platforms/pseries/eeh_cache.c b/arch/powerpc/platforms/pseries/eeh_cache.c index 1e83fcd0df31..ce37040af870 100644 --- a/arch/powerpc/platforms/pseries/eeh_cache.c +++ b/arch/powerpc/platforms/pseries/eeh_cache.c @@ -28,7 +28,6 @@ #include <asm/pci-bridge.h> #include <asm/ppc-pci.h> -#undef DEBUG /** * The pci address cache subsystem. This subsystem places diff --git a/arch/powerpc/platforms/pseries/firmware.c b/arch/powerpc/platforms/pseries/firmware.c index b765b7c77b65..9d3a40f45974 100644 --- a/arch/powerpc/platforms/pseries/firmware.c +++ b/arch/powerpc/platforms/pseries/firmware.c @@ -21,17 +21,11 @@ * 2 of the License, or (at your option) any later version. */ -#undef DEBUG #include <asm/firmware.h> #include <asm/prom.h> #include <asm/udbg.h> -#ifdef DEBUG -#define DBG(fmt...) udbg_printf(fmt) -#else -#define DBG(fmt...) -#endif typedef struct { unsigned long val; @@ -72,7 +66,7 @@ void __init fw_feature_init(const char *hypertas, unsigned long len) const char *s; int i; - DBG(" -> fw_feature_init()\n"); + pr_debug(" -> fw_feature_init()\n"); for (s = hypertas; s < hypertas + len; s += strlen(s) + 1) { for (i = 0; i < FIRMWARE_MAX_FEATURES; i++) { @@ -88,5 +82,5 @@ void __init fw_feature_init(const char *hypertas, unsigned long len) } } - DBG(" <- fw_feature_init()\n"); + pr_debug(" <- fw_feature_init()\n"); } diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index a65c76308201..176f1f39d2d5 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -47,7 +47,6 @@ #include "plpar_wrappers.h" -#define DBG(fmt...) static void tce_build_pSeries(struct iommu_table *tbl, long index, long npages, unsigned long uaddr, @@ -322,7 +321,7 @@ static void pci_dma_bus_setup_pSeries(struct pci_bus *bus) dn = pci_bus_to_OF_node(bus); - DBG("pci_dma_bus_setup_pSeries: setting up bus %s\n", dn->full_name); + pr_debug("pci_dma_bus_setup_pSeries: setting up bus %s\n", dn->full_name); if (bus->self) { /* This is not a root bus, any setup will be done for the @@ -347,7 +346,7 @@ static void pci_dma_bus_setup_pSeries(struct pci_bus *bus) for (children = 0, tmp = dn->child; tmp; tmp = tmp->sibling) children++; - DBG("Children: %d\n", children); + pr_debug("Children: %d\n", children); /* Calculate amount of DMA window per slot. Each window must be * a power of two (due to pci_alloc_consistent requirements). @@ -361,8 +360,8 @@ static void pci_dma_bus_setup_pSeries(struct pci_bus *bus) while (pci->phb->dma_window_size * children > 0x80000000ul) pci->phb->dma_window_size >>= 1; - DBG("No ISA/IDE, window size is 0x%lx\n", - pci->phb->dma_window_size); + pr_debug("No ISA/IDE, window size is 0x%lx\n", + pci->phb->dma_window_size); pci->phb->dma_window_base_cur = 0; return; @@ -387,8 +386,7 @@ static void pci_dma_bus_setup_pSeries(struct pci_bus *bus) while (pci->phb->dma_window_size * children > 0x70000000ul) pci->phb->dma_window_size >>= 1; - DBG("ISA/IDE, window size is 0x%lx\n", pci->phb->dma_window_size); - + pr_debug("ISA/IDE, window size is 0x%lx\n", pci->phb->dma_window_size); } @@ -401,7 +399,8 @@ static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus) dn = pci_bus_to_OF_node(bus); - DBG("pci_dma_bus_setup_pSeriesLP: setting up bus %s\n", dn->full_name); + pr_debug("pci_dma_bus_setup_pSeriesLP: setting up bus %s\n", + dn->full_name); /* Find nearest ibm,dma-window, walking up the device tree */ for (pdn = dn; pdn != NULL; pdn = pdn->parent) { @@ -411,14 +410,14 @@ static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus) } if (dma_window == NULL) { - DBG(" no ibm,dma-window property !\n"); + pr_debug(" no ibm,dma-window property !\n"); return; } ppci = PCI_DN(pdn); - DBG(" parent is %s, iommu_table: 0x%p\n", - pdn->full_name, ppci->iommu_table); + pr_debug(" parent is %s, iommu_table: 0x%p\n", + pdn->full_name, ppci->iommu_table); if (!ppci->iommu_table) { tbl = kmalloc_node(sizeof(struct iommu_table), GFP_KERNEL, @@ -426,7 +425,7 @@ static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus) iommu_table_setparms_lpar(ppci->phb, pdn, tbl, dma_window, bus->number); ppci->iommu_table = iommu_init_table(tbl, ppci->phb->node); - DBG(" created table: %p\n", ppci->iommu_table); + pr_debug(" created table: %p\n", ppci->iommu_table); } if (pdn != dn) @@ -439,7 +438,7 @@ static void pci_dma_dev_setup_pSeries(struct pci_dev *dev) struct device_node *dn; struct iommu_table *tbl; - DBG("pci_dma_dev_setup_pSeries: %s\n", pci_name(dev)); + pr_debug("pci_dma_dev_setup_pSeries: %s\n", pci_name(dev)); dn = dev->dev.archdata.of_node; @@ -450,7 +449,7 @@ static void pci_dma_dev_setup_pSeries(struct pci_dev *dev) if (!dev->bus->self) { struct pci_controller *phb = PCI_DN(dn)->phb; - DBG(" --> first child, no bridge. Allocating iommu table.\n"); + pr_debug(" --> first child, no bridge. Allocating iommu table.\n"); tbl = kmalloc_node(sizeof(struct iommu_table), GFP_KERNEL, phb->node); iommu_table_setparms(phb, dn, tbl); @@ -480,7 +479,7 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev) const void *dma_window = NULL; struct pci_dn *pci; - DBG("pci_dma_dev_setup_pSeriesLP: %s\n", pci_name(dev)); + pr_debug("pci_dma_dev_setup_pSeriesLP: %s\n", pci_name(dev)); /* dev setup for LPAR is a little tricky, since the device tree might * contain the dma-window properties per-device and not neccesarily @@ -489,7 +488,7 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev) * already allocated. */ dn = pci_device_to_OF_node(dev); - DBG(" node is %s\n", dn->full_name); + pr_debug(" node is %s\n", dn->full_name); for (pdn = dn; pdn && PCI_DN(pdn) && !PCI_DN(pdn)->iommu_table; pdn = pdn->parent) { @@ -504,13 +503,13 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev) pci_name(dev), dn? dn->full_name : "<null>"); return; } - DBG(" parent is %s\n", pdn->full_name); + pr_debug(" parent is %s\n", pdn->full_name); /* Check for parent == NULL so we don't try to setup the empty EADS * slots on POWER4 machines. */ if (dma_window == NULL || pdn->parent == NULL) { - DBG(" no dma window for device, linking to parent\n"); + pr_debug(" no dma window for device, linking to parent\n"); dev->dev.archdata.dma_data = PCI_DN(pdn)->iommu_table; return; } @@ -522,9 +521,9 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev) iommu_table_setparms_lpar(pci->phb, pdn, tbl, dma_window, pci->phb->bus->number); pci->iommu_table = iommu_init_table(tbl, pci->phb->node); - DBG(" created table: %p\n", pci->iommu_table); + pr_debug(" created table: %p\n", pci->iommu_table); } else { - DBG(" found DMA window, table: %p\n", pci->iommu_table); + pr_debug(" found DMA window, table: %p\n", pci->iommu_table); } dev->dev.archdata.dma_data = pci->iommu_table; diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c index 9235c469449e..2cbaedb17f3e 100644 --- a/arch/powerpc/platforms/pseries/lpar.c +++ b/arch/powerpc/platforms/pseries/lpar.c @@ -19,7 +19,8 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#undef DEBUG_LOW +/* Enables debugging of low-level hash table routines - careful! */ +#undef DEBUG #include <linux/kernel.h> #include <linux/dma-mapping.h> @@ -42,11 +43,6 @@ #include "plpar_wrappers.h" #include "pseries.h" -#ifdef DEBUG_LOW -#define DBG_LOW(fmt...) do { udbg_printf(fmt); } while(0) -#else -#define DBG_LOW(fmt...) do { } while(0) -#endif /* in hvCall.S */ EXPORT_SYMBOL(plpar_hcall); @@ -196,6 +192,8 @@ void __init udbg_init_debug_lpar(void) udbg_putc = udbg_putcLP; udbg_getc = udbg_getcLP; udbg_getc_poll = udbg_getc_pollLP; + + register_early_udbg_console(); } /* returns 0 if couldn't find or use /chosen/stdout as console */ @@ -288,15 +286,15 @@ static long pSeries_lpar_hpte_insert(unsigned long hpte_group, unsigned long hpte_v, hpte_r; if (!(vflags & HPTE_V_BOLTED)) - DBG_LOW("hpte_insert(group=%lx, va=%016lx, pa=%016lx, " - "rflags=%lx, vflags=%lx, psize=%d)\n", - hpte_group, va, pa, rflags, vflags, psize); + pr_debug("hpte_insert(group=%lx, va=%016lx, pa=%016lx, " + "rflags=%lx, vflags=%lx, psize=%d)\n", + hpte_group, va, pa, rflags, vflags, psize); hpte_v = hpte_encode_v(va, psize, ssize) | vflags | HPTE_V_VALID; hpte_r = hpte_encode_r(pa, psize) | rflags; if (!(vflags & HPTE_V_BOLTED)) - DBG_LOW(" hpte_v=%016lx, hpte_r=%016lx\n", hpte_v, hpte_r); + pr_debug(" hpte_v=%016lx, hpte_r=%016lx\n", hpte_v, hpte_r); /* Now fill in the actual HPTE */ /* Set CEC cookie to 0 */ @@ -313,7 +311,7 @@ static long pSeries_lpar_hpte_insert(unsigned long hpte_group, lpar_rc = plpar_pte_enter(flags, hpte_group, hpte_v, hpte_r, &slot); if (unlikely(lpar_rc == H_PTEG_FULL)) { if (!(vflags & HPTE_V_BOLTED)) - DBG_LOW(" full\n"); + pr_debug(" full\n"); return -1; } @@ -324,11 +322,11 @@ static long pSeries_lpar_hpte_insert(unsigned long hpte_group, */ if (unlikely(lpar_rc != H_SUCCESS)) { if (!(vflags & HPTE_V_BOLTED)) - DBG_LOW(" lpar err %d\n", lpar_rc); + pr_debug(" lpar err %lu\n", lpar_rc); return -2; } if (!(vflags & HPTE_V_BOLTED)) - DBG_LOW(" -> slot: %d\n", slot & 7); + pr_debug(" -> slot: %lu\n", slot & 7); /* Because of iSeries, we have to pass down the secondary * bucket bit here as well @@ -420,17 +418,17 @@ static long pSeries_lpar_hpte_updatepp(unsigned long slot, want_v = hpte_encode_avpn(va, psize, ssize); - DBG_LOW(" update: avpnv=%016lx, hash=%016lx, f=%x, psize: %d ... ", - want_v, slot, flags, psize); + pr_debug(" update: avpnv=%016lx, hash=%016lx, f=%lx, psize: %d ...", + want_v, slot, flags, psize); lpar_rc = plpar_pte_protect(flags, slot, want_v); if (lpar_rc == H_NOT_FOUND) { - DBG_LOW("not found !\n"); + pr_debug("not found !\n"); return -1; } - DBG_LOW("ok\n"); + pr_debug("ok\n"); BUG_ON(lpar_rc != H_SUCCESS); @@ -505,8 +503,8 @@ static void pSeries_lpar_hpte_invalidate(unsigned long slot, unsigned long va, unsigned long lpar_rc; unsigned long dummy1, dummy2; - DBG_LOW(" inval : slot=%lx, va=%016lx, psize: %d, local: %d", - slot, va, psize, local); + pr_debug(" inval : slot=%lx, va=%016lx, psize: %d, local: %d\n", + slot, va, psize, local); want_v = hpte_encode_avpn(va, psize, ssize); lpar_rc = plpar_pte_remove(H_AVPN, slot, want_v, &dummy1, &dummy2); diff --git a/arch/powerpc/platforms/pseries/ras.c b/arch/powerpc/platforms/pseries/ras.c index a1ab25c7082f..2b548afd1003 100644 --- a/arch/powerpc/platforms/pseries/ras.c +++ b/arch/powerpc/platforms/pseries/ras.c @@ -67,8 +67,6 @@ static int ras_check_exception_token; static irqreturn_t ras_epow_interrupt(int irq, void *dev_id); static irqreturn_t ras_error_interrupt(int irq, void *dev_id); -/* #define DEBUG */ - static void request_ras_irqs(struct device_node *np, irq_handler_t handler, @@ -237,7 +235,7 @@ static irqreturn_t ras_error_interrupt(int irq, void *dev_id) printk(KERN_EMERG "Error: Fatal hardware error <0x%lx 0x%x>\n", *((unsigned long *)&ras_log_buf), status); -#ifndef DEBUG +#ifndef DEBUG_RTAS_POWER_OFF /* Don't actually power off when debugging so we can test * without actually failing while injecting errors. * Error data will not be logged to syslog. diff --git a/arch/powerpc/platforms/pseries/rtasd.c b/arch/powerpc/platforms/pseries/rtasd.c index e3078ce41518..befadd4f9524 100644 --- a/arch/powerpc/platforms/pseries/rtasd.c +++ b/arch/powerpc/platforms/pseries/rtasd.c @@ -29,11 +29,6 @@ #include <asm/atomic.h> #include <asm/machdep.h> -#if 0 -#define DEBUG(A...) printk(KERN_ERR A) -#else -#define DEBUG(A...) -#endif static DEFINE_SPINLOCK(rtasd_log_lock); @@ -198,7 +193,7 @@ void pSeries_log_error(char *buf, unsigned int err_type, int fatal) unsigned long s; int len = 0; - DEBUG("logging event\n"); + pr_debug("rtasd: logging event\n"); if (buf == NULL) return; @@ -409,7 +404,8 @@ static int rtasd(void *unused) daemonize("rtasd"); printk(KERN_DEBUG "RTAS daemon started\n"); - DEBUG("will sleep for %d milliseconds\n", (30000/rtas_event_scan_rate)); + pr_debug("rtasd: will sleep for %d milliseconds\n", + (30000 / rtas_event_scan_rate)); /* See if we have any error stored in NVRAM */ memset(logdata, 0, rtas_error_log_max); @@ -428,9 +424,9 @@ static int rtasd(void *unused) do_event_scan_all_cpus(1000); if (surveillance_timeout != -1) { - DEBUG("enabling surveillance\n"); + pr_debug("rtasd: enabling surveillance\n"); enable_surveillance(surveillance_timeout); - DEBUG("surveillance enabled\n"); + pr_debug("rtasd: surveillance enabled\n"); } /* Delay should be at least one second since some diff --git a/arch/powerpc/platforms/pseries/scanlog.c b/arch/powerpc/platforms/pseries/scanlog.c index e5b0ea870164..bec3803f0618 100644 --- a/arch/powerpc/platforms/pseries/scanlog.c +++ b/arch/powerpc/platforms/pseries/scanlog.c @@ -38,9 +38,7 @@ #define SCANLOG_HWERROR -1 #define SCANLOG_CONTINUE 1 -#define DEBUG(A...) do { if (scanlog_debug) printk(KERN_ERR "scanlog: " A); } while (0) -static int scanlog_debug; static unsigned int ibm_scan_log_dump; /* RTAS token */ static struct proc_dir_entry *proc_ppc64_scan_log_dump; /* The proc file */ @@ -86,14 +84,14 @@ static ssize_t scanlog_read(struct file *file, char __user *buf, memcpy(data, rtas_data_buf, RTAS_DATA_BUF_SIZE); spin_unlock(&rtas_data_buf_lock); - DEBUG("status=%d, data[0]=%x, data[1]=%x, data[2]=%x\n", - status, data[0], data[1], data[2]); + pr_debug("scanlog: status=%d, data[0]=%x, data[1]=%x, " \ + "data[2]=%x\n", status, data[0], data[1], data[2]); switch (status) { case SCANLOG_COMPLETE: - DEBUG("hit eof\n"); + pr_debug("scanlog: hit eof\n"); return 0; case SCANLOG_HWERROR: - DEBUG("hardware error reading scan log data\n"); + pr_debug("scanlog: hardware error reading data\n"); return -EIO; case SCANLOG_CONTINUE: /* We may or may not have data yet */ @@ -110,7 +108,8 @@ static ssize_t scanlog_read(struct file *file, char __user *buf, /* Assume extended busy */ wait_time = rtas_busy_delay_time(status); if (!wait_time) { - printk(KERN_ERR "scanlog: unknown error from rtas: %d\n", status); + printk(KERN_ERR "scanlog: unknown error " \ + "from rtas: %d\n", status); return -EIO; } } @@ -134,15 +133,9 @@ static ssize_t scanlog_write(struct file * file, const char __user * buf, if (buf) { if (strncmp(stkbuf, "reset", 5) == 0) { - DEBUG("reset scanlog\n"); + pr_debug("scanlog: reset scanlog\n"); status = rtas_call(ibm_scan_log_dump, 2, 1, NULL, 0, 0); - DEBUG("rtas returns %d\n", status); - } else if (strncmp(stkbuf, "debugon", 7) == 0) { - printk(KERN_ERR "scanlog: debug on\n"); - scanlog_debug = 1; - } else if (strncmp(stkbuf, "debugoff", 8) == 0) { - printk(KERN_ERR "scanlog: debug off\n"); - scanlog_debug = 0; + pr_debug("scanlog: rtas returns %d\n", status); } } return count; diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c index f66aa9c3b135..f5d29f5b13c1 100644 --- a/arch/powerpc/platforms/pseries/setup.c +++ b/arch/powerpc/platforms/pseries/setup.c @@ -16,8 +16,6 @@ * bootup setup stuff.. */ -#undef DEBUG - #include <linux/cpu.h> #include <linux/errno.h> #include <linux/sched.h> @@ -70,11 +68,6 @@ #include "plpar_wrappers.h" #include "pseries.h" -#ifdef DEBUG -#define DBG(fmt...) udbg_printf(fmt) -#else -#define DBG(fmt...) -#endif int fwnmi_active; /* TRUE if an FWNMI handler is present */ @@ -326,7 +319,7 @@ static int pseries_set_xdabr(unsigned long dabr) */ static void __init pSeries_init_early(void) { - DBG(" -> pSeries_init_early()\n"); + pr_debug(" -> pSeries_init_early()\n"); if (firmware_has_feature(FW_FEATURE_LPAR)) find_udbg_vterm(); @@ -338,7 +331,7 @@ static void __init pSeries_init_early(void) iommu_init_early_pSeries(); - DBG(" <- pSeries_init_early()\n"); + pr_debug(" <- pSeries_init_early()\n"); } /* @@ -383,7 +376,7 @@ static int __init pSeries_probe(void) of_flat_dt_is_compatible(root, "IBM,CBEA")) return 0; - DBG("pSeries detected, looking for LPAR capability...\n"); + pr_debug("pSeries detected, looking for LPAR capability...\n"); /* Now try to figure out if we are running on LPAR */ of_scan_flat_dt(pSeries_probe_hypertas, NULL); @@ -393,8 +386,8 @@ static int __init pSeries_probe(void) else hpte_init_native(); - DBG("Machine is%s LPAR !\n", - (powerpc_firmware_features & FW_FEATURE_LPAR) ? "" : " not"); + pr_debug("Machine is%s LPAR !\n", + (powerpc_firmware_features & FW_FEATURE_LPAR) ? "" : " not"); return 1; } diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c index ea4c65917a64..9d8f8c84ab89 100644 --- a/arch/powerpc/platforms/pseries/smp.c +++ b/arch/powerpc/platforms/pseries/smp.c @@ -12,7 +12,6 @@ * 2 of the License, or (at your option) any later version. */ -#undef DEBUG #include <linux/kernel.h> #include <linux/module.h> @@ -51,12 +50,6 @@ #include "plpar_wrappers.h" #include "pseries.h" -#ifdef DEBUG -#include <asm/udbg.h> -#define DBG(fmt...) udbg_printf(fmt) -#else -#define DBG(fmt...) -#endif /* * The primary thread of each non-boot processor is recorded here before @@ -231,7 +224,7 @@ static void __init smp_init_pseries(void) { int i; - DBG(" -> smp_init_pSeries()\n"); + pr_debug(" -> smp_init_pSeries()\n"); /* Mark threads which are still spinning in hold loops. */ if (cpu_has_feature(CPU_FTR_SMT)) { @@ -255,7 +248,7 @@ static void __init smp_init_pseries(void) smp_ops->take_timebase = pSeries_take_timebase; } - DBG(" <- smp_init_pSeries()\n"); + pr_debug(" <- smp_init_pSeries()\n"); } #ifdef CONFIG_MPIC diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c index 43df53c30aa0..ebebc28fe895 100644 --- a/arch/powerpc/platforms/pseries/xics.c +++ b/arch/powerpc/platforms/pseries/xics.c @@ -9,7 +9,6 @@ * 2 of the License, or (at your option) any later version. */ -#undef DEBUG #include <linux/types.h> #include <linux/threads.h> diff --git a/arch/powerpc/sysdev/axonram.c b/arch/powerpc/sysdev/axonram.c index d359d6e92975..7f59188cd9a1 100644 --- a/arch/powerpc/sysdev/axonram.c +++ b/arch/powerpc/sysdev/axonram.c @@ -143,7 +143,7 @@ axon_ram_make_request(struct request_queue *queue, struct bio *bio) */ static int axon_ram_direct_access(struct block_device *device, sector_t sector, - unsigned long *data) + void **kaddr, unsigned long *pfn) { struct axon_ram_bank *bank = device->bd_disk->private_data; loff_t offset; @@ -154,7 +154,8 @@ axon_ram_direct_access(struct block_device *device, sector_t sector, return -ERANGE; } - *data = bank->ph_addr + offset; + *kaddr = (void *)(bank->ph_addr + offset); + *pfn = virt_to_phys(kaddr) >> PAGE_SHIFT; return 0; } diff --git a/arch/powerpc/sysdev/fsl_soc.c b/arch/powerpc/sysdev/fsl_soc.c index 5c1b246aaccc..7b45670c7af3 100644 --- a/arch/powerpc/sysdev/fsl_soc.c +++ b/arch/powerpc/sysdev/fsl_soc.c @@ -892,3 +892,44 @@ void fsl_rstcr_restart(char *cmd) while (1) ; } #endif + +#if defined(CONFIG_FB_FSL_DIU) || defined(CONFIG_FB_FSL_DIU_MODULE) +struct platform_diu_data_ops diu_ops = { + .diu_size = 1280 * 1024 * 4, /* default one 1280x1024 buffer */ +}; +EXPORT_SYMBOL(diu_ops); + +int __init preallocate_diu_videomemory(void) +{ + pr_debug("diu_size=%lu\n", diu_ops.diu_size); + + diu_ops.diu_mem = __alloc_bootmem(diu_ops.diu_size, 8, 0); + if (!diu_ops.diu_mem) { + printk(KERN_ERR "fsl-diu: cannot allocate %lu bytes\n", + diu_ops.diu_size); + return -ENOMEM; + } + + pr_debug("diu_mem=%p\n", diu_ops.diu_mem); + + rh_init(&diu_ops.diu_rh_info, 4096, ARRAY_SIZE(diu_ops.diu_rh_block), + diu_ops.diu_rh_block); + return rh_attach_region(&diu_ops.diu_rh_info, + (unsigned long) diu_ops.diu_mem, + diu_ops.diu_size); +} + +static int __init early_parse_diufb(char *p) +{ + if (!p) + return 1; + + diu_ops.diu_size = _ALIGN_UP(memparse(p, &p), 8); + + pr_debug("diu_size=%lu\n", diu_ops.diu_size); + + return 0; +} +early_param("diufb", early_parse_diufb); + +#endif diff --git a/arch/powerpc/sysdev/fsl_soc.h b/arch/powerpc/sysdev/fsl_soc.h index 74c4a9657b33..52c831fa1886 100644 --- a/arch/powerpc/sysdev/fsl_soc.h +++ b/arch/powerpc/sysdev/fsl_soc.h @@ -17,5 +17,28 @@ extern int fsl_spi_init(struct spi_board_info *board_infos, void (*deactivate_cs)(u8 cs, u8 polarity)); extern void fsl_rstcr_restart(char *cmd); + +#if defined(CONFIG_FB_FSL_DIU) || defined(CONFIG_FB_FSL_DIU_MODULE) +#include <linux/bootmem.h> +#include <asm/rheap.h> +struct platform_diu_data_ops { + rh_block_t diu_rh_block[16]; + rh_info_t diu_rh_info; + unsigned long diu_size; + void *diu_mem; + + unsigned int (*get_pixel_format) (unsigned int bits_per_pixel, + int monitor_port); + void (*set_gamma_table) (int monitor_port, char *gamma_table_base); + void (*set_monitor_port) (int monitor_port); + void (*set_pixel_clock) (unsigned int pixclock); + ssize_t (*show_monitor_port) (int monitor_port, char *buf); + int (*set_sysfs_monitor_port) (int val); +}; + +extern struct platform_diu_data_ops diu_ops; +int __init preallocate_diu_videomemory(void); +#endif + #endif #endif diff --git a/arch/powerpc/sysdev/mv64x60_dev.c b/arch/powerpc/sysdev/mv64x60_dev.c index 047b31027fa6..41af1223e2a0 100644 --- a/arch/powerpc/sysdev/mv64x60_dev.c +++ b/arch/powerpc/sysdev/mv64x60_dev.c @@ -338,15 +338,13 @@ static int __init mv64x60_i2c_device_setup(struct device_node *np, int id) pdata.freq_m = 8; /* default */ prop = of_get_property(np, "freq_m", NULL); - if (!prop) - return -ENODEV; - pdata.freq_m = *prop; + if (prop) + pdata.freq_m = *prop; pdata.freq_m = 3; /* default */ prop = of_get_property(np, "freq_n", NULL); - if (!prop) - return -ENODEV; - pdata.freq_n = *prop; + if (prop) + pdata.freq_n = *prop; pdata.timeout = 1000; /* default: 1 second */ @@ -433,9 +431,13 @@ static int __init mv64x60_device_setup(void) int err; id = 0; - for_each_compatible_node(np, "serial", "marvell,mv64360-mpsc") - if ((err = mv64x60_mpsc_device_setup(np, id++))) - goto error; + for_each_compatible_node(np, "serial", "marvell,mv64360-mpsc") { + err = mv64x60_mpsc_device_setup(np, id++); + if (err) + printk(KERN_ERR "Failed to initialize MV64x60 " + "serial device %s: error %d.\n", + np->full_name, err); + } id = 0; id2 = 0; @@ -443,38 +445,44 @@ static int __init mv64x60_device_setup(void) pdev = mv64x60_eth_register_shared_pdev(np, id++); if (IS_ERR(pdev)) { err = PTR_ERR(pdev); - goto error; + printk(KERN_ERR "Failed to initialize MV64x60 " + "network block %s: error %d.\n", + np->full_name, err); + continue; } for_each_child_of_node(np, np2) { if (!of_device_is_compatible(np2, "marvell,mv64360-eth")) continue; err = mv64x60_eth_device_setup(np2, id2++, pdev); - if (err) { - of_node_put(np2); - goto error; - } + if (err) + printk(KERN_ERR "Failed to initialize " + "MV64x60 network device %s: " + "error %d.\n", + np2->full_name, err); } } id = 0; - for_each_compatible_node(np, "i2c", "marvell,mv64360-i2c") - if ((err = mv64x60_i2c_device_setup(np, id++))) - goto error; + for_each_compatible_node(np, "i2c", "marvell,mv64360-i2c") { + err = mv64x60_i2c_device_setup(np, id++); + if (err) + printk(KERN_ERR "Failed to initialize MV64x60 I2C " + "bus %s: error %d.\n", + np->full_name, err); + } /* support up to one watchdog timer */ np = of_find_compatible_node(np, NULL, "marvell,mv64360-wdt"); if (np) { if ((err = mv64x60_wdt_device_setup(np, id))) - goto error; + printk(KERN_ERR "Failed to initialize MV64x60 " + "Watchdog %s: error %d.\n", + np->full_name, err); of_node_put(np); } return 0; - -error: - of_node_put(np); - return err; } arch_initcall(mv64x60_device_setup); diff --git a/arch/powerpc/sysdev/mv64x60_udbg.c b/arch/powerpc/sysdev/mv64x60_udbg.c index ccdb3b0418fc..2792dc8b038c 100644 --- a/arch/powerpc/sysdev/mv64x60_udbg.c +++ b/arch/powerpc/sysdev/mv64x60_udbg.c @@ -94,7 +94,7 @@ static void mv64x60_udbg_init(void) if (!np) return; - block_index = of_get_property(np, "block-index", NULL); + block_index = of_get_property(np, "cell-index", NULL); if (!block_index) goto error; |