Messages in this thread Patch in this message | | | Date | Fri, 10 Oct 2008 02:08:17 +0200 | From | Ingo Molnar <> | Subject | [git pull] x86 updates for v2.6.28, phase #6, misc |
| |
Linus,
Please pull the latest x86-v28-for-linus-phase6 git tree from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git x86-v28-for-linus-phase6
tip/x86/core: various x86 updates independent of the sparse irqs support tree.
out-of-topic modifications in x86-v28-for-linus-phase6: ------------------------------------------------------- include/linux/kernel.h # d974ae3: generic, memparse(): constify arg include/linux/mm.h # f7d0b92: mm: define USE_SPLIT_PTLOCKS rath # 59ea746: MM: virtual address debug include/linux/mm_types.h # f7d0b92: mm: define USE_SPLIT_PTLOCKS rath include/linux/mmdebug.h # 7aa413d: x86, MM: virtual address debug, c # 59ea746: MM: virtual address debug lib/cmdline.c # d974ae3: generic, memparse(): constify arg mm/vmalloc.c # 7aa413d: x86, MM: virtual address debug, c # 59ea746: MM: virtual address debug
Thanks,
Ingo
------------------> Alex Nixon (10): x86: add cpu hotplug hooks into smp_ops x86_32: clean up play_dead x86: unify x86_32 and x86_64 play_dead into one function x86: separate generic cpu disabling code from APIC writes in cpu_disable xen: implement CPU hotplugging Xen: fix cpu_hotplug.c build by replacing is_running_on_xen() with xen_pv_domain() Xen: fix cpu_hotplug build when !CONFIG_SMP x86: build fix for !CONFIG_SMP x86, xen: fix build when !CONFIG_HOTPLUG_CPU xen: make CPU hotplug functions static
Alexander van Heukelum (47): i386: remove kprobes' restore_interrupts in favour of conditional_sti i386: prepare to convert exceptions to interrupts i386: convert hardware exception 0 to an interrupt gate i386: expand exception 3 DO_TRAP macro i386: convert hardware exception 4 to an interrupt gate i386: convert hardware exception 5 to an interrupt gate i386: convert hardware exception 6 to an interrupt gate i386: convert hardware exception 7 to an interrupt gate i386: convert hardware exception 9 to an interrupt gate i386: convert hardware exception 10 to an interrupt gate i386: convert hardware exception 11 to an interrupt gate i386: convert hardware exception 12 to an interrupt gate i386: convert hardware exception 13 to an interrupt gate i386: convert hardware exception 15 to an interrupt gate i386: convert hardware exception 16 to an interrupt gate i386: convert hardware exception 17 to an interrupt gate i386: convert hardware exception 18 to an interrupt gate i386: convert hardware exception 19 to an interrupt gate i386: remove temporary DO_TRAP macros, expanding the last one used i386: add TRACE_IRQS_OFF to entry_32.S in 'error_code' i386: add TRACE_IRQS_OFF for exception 1 (debug) i386: add TRACE_IRQS_OFF for the nmi i386: add TRACE_IRQS_OFF for the exception 3 (int3) i386: trace_hardirqs_fixup should now not be necessary: irqs are off. traps: x86_64: add TRACE_IRQS_OFF in error_entry traps: x86_64: add TRACE_IRQS_OFF in paranoidentry macro traps: x86_64: remove trace_hardirqs_fixup from DO_ERROR_INFO macro traps: x86_64: remove trace_hardirqs_fixup from int3 handler traps: x86_64: remove trace_hardirqs_fixup from debug handler traps: x86: remove trace_hardirqs_fixup from pagefault handler traps: i386: make do_trap more like x86_64 i386: split out dumpstack code from traps_32.c x86_64: split out dumpstack code from traps_64.c fix: x86: remove cpu_vendor_dev x86, traps: split out math_error and simd_math_error x86, traps, i386: factor out lazy io-bitmap copy x86, traps: introduce dotraplinkage x86, traps: converge do_debug handlers traps: x86: converge trap_init functions traps: x86_64: make math_state_restore more like i386 traps: i386: use preempt_conditional_sti/cli in do_int3 traps: x86_64: make io_check_error equal to the one on i386 traps: i386: expand clear_mem_error and remove from mach_traps.h traps: x86_64: use task_pid_nr(tsk) instead of tsk->pid in do_general_protection traps: x86: various noop-changes preparing for unification of traps_xx.c traps: x86: make traps_32.c and traps_64.c equal traps: x86: finalize unification of traps.c
Chuck Ebbert (2): x86: move prefill_possible_map calling early, fix, V2 x86: allow number of additional hotplug CPUs to be set at compile time, V2
Cyrill Gorcunov (1): x86: smpboot - check if we have ESR register in wakeup_secondary_cpu
Eduardo Habkost (2): x86, xen: Use native_pte_flags instead of native_pte_val for .pte_flags xen_alloc_ptpage: cast PFN_PHYS() argument to unsigned long
Ingo Molnar (10): x86, MM: virtual address debug, cleanups x86: x86_{phys,virt}_bits field also for i386, fix Revert "x86: x86_{phys,virt}_bits field also for i386, fix" Revert "x86: x86_{phys,virt}_bits field also for i386 (v2)" x86: print out EBDA/lowmem address x86: add PCI IDs for AMD Barcelona PCI devices x86: __show_registers() and __show_regs() API unification, fixlet x86: remove additional_cpus configurability Revert "acpi/x86: introduce __apci_map_table, v4" Revert "acpi: remove final __acpi_map_table mapping before setting acpi_gbl_permanent_mmap"
Jack Steiner (1): x86, UV: new UV genapic functions for x2apic
Jan Beulich (7): x86: adjust vmalloc_sync_all() for Xen (2nd try) x86-64: reduce boot fixmap space x86-64: slightly stream-line 32-bit syscall entry code x86: x86_{phys,virt}_bits field also for i386 (v2) x86-64: fix combining of regions in init_memory_mapping() x86: make mm/gup.c more virtualization friendly x86: adjust dependencies for CONFIG_X86_CMOV
Jeremy Fitzhardinge (30): x86/paravirt/xen: properly fill out the ldt ops x86: split spinlock implementations out into their own files x86: fix initialization of 'l' bit in ldt descriptors xen: fix allocation and use of large ldts generic, memparse(): constify argument xen-balloon: fix up sysfs issues xen-balloon: clean up unused functions xen: suppress known wrmsrs xen: compile irq functions without -pg for ftrace xen: fix allocation and use of large ldts, cleanup xen: clean up domain mode predicates x86/paravirt: add spin_lock_flags lock op xen: clarify locking used when pinning a pagetable. xen: add xen_ prefixes to make tracing with ftrace easier xen: save previous spinlock when blocking xen: add debugfs support xen: allow interrupts to be enabled while doing a blocking spin xen: measure how long spinlocks spend blocking x86: build fix in "xen spinlock updates and performance measurements" x86: export pv_lock_ops non-GPL x86: add _PAGE_IOMAP pte flag for IO mappings x86: remove duplicate early_ioremap declarations x86: add early_memremap() x86: use early_memremap() in setup.c x86-64: don't check for map replacement x86: use early_ioremap in __acpi_map_table x86: always explicitly map acpi memory mm: define USE_SPLIT_PTLOCKS rather than repeating expression xen: fix pinning when not using split pte locks acpi: remove final __acpi_map_table mapping before setting acpi_gbl_permanent_mmap
Jiri Slaby (2): MM: virtual address debug x86, MM: virtual address debug, v2
Krzysztof Helt (2): x86: merge winchip-2 and winchip-2a cpu choices x86: do not allow to optimize flag_is_changeable_p() (rev. 2)
Manfred Spraul (1): arch/x86/kernel/smpboot.c: Clarify when irq processing begins.
Pekka Enberg (1): x86: __show_registers() and __show_regs() API unification
Thomas Gleixner (2): x86: improve UP kernel when CPU-hotplug and SMP is enabled x86: remove additional_cpus
Vegard Nossum (2): x86: add memory clobber in switch_to() x86: fix virt_addr_valid() with CONFIG_DEBUG_VIRTUAL=y, v2
Yinghai Lu (5): x86: rename discontig_32.c to numa_32.c x86: change early_ioremap to use slots instead of nesting acpi/x86: introduce __apci_map_table, v4 x86: check dsdt before find oem table for es7000, v2 x86: cpu don't print duplicated vendor string
arch/x86/Kconfig.cpu | 29 +- arch/x86/Makefile_32.cpu | 1 - arch/x86/configs/i386_defconfig | 1 - arch/x86/configs/x86_64_defconfig | 1 - arch/x86/ia32/ia32entry.S | 26 +- arch/x86/kernel/Makefile | 6 +- arch/x86/kernel/acpi/boot.c | 34 +- arch/x86/kernel/alternative.c | 2 +- arch/x86/kernel/cpu/common.c | 48 +- arch/x86/kernel/doublefault_32.c | 2 +- arch/x86/kernel/dumpstack_32.c | 422 ++++++++++ arch/x86/kernel/dumpstack_64.c | 565 ++++++++++++++ arch/x86/kernel/entry_32.S | 22 +- arch/x86/kernel/entry_64.S | 8 +- arch/x86/kernel/es7000_32.c | 28 +- arch/x86/kernel/genx2apic_uv_x.c | 10 +- arch/x86/kernel/head.c | 1 + arch/x86/kernel/ldt.c | 9 +- arch/x86/kernel/paravirt-spinlocks.c | 37 + arch/x86/kernel/paravirt.c | 27 +- arch/x86/kernel/process_32.c | 43 +- arch/x86/kernel/process_64.c | 37 +- arch/x86/kernel/setup.c | 8 +- arch/x86/kernel/smp.c | 6 +- arch/x86/kernel/smpboot.c | 164 +++-- arch/x86/kernel/tlb_32.c | 8 + arch/x86/kernel/{traps_32.c => traps.c} | 860 ++++++++------------- arch/x86/kernel/traps_64.c | 1214 ----------------------------- arch/x86/kernel/vmlinux_64.lds.S | 2 +- arch/x86/mach-generic/es7000.c | 20 +- arch/x86/mm/Makefile | 6 +- arch/x86/mm/fault.c | 19 +- arch/x86/mm/gup.c | 10 +- arch/x86/mm/init_32.c | 2 +- arch/x86/mm/init_64.c | 9 +- arch/x86/mm/ioremap.c | 168 ++++- arch/x86/mm/{discontig_32.c => numa_32.c} | 0 arch/x86/xen/Kconfig | 10 +- arch/x86/xen/Makefile | 12 +- arch/x86/xen/debugfs.c | 123 +++ arch/x86/xen/debugfs.h | 10 + arch/x86/xen/enlighten.c | 189 ++--- arch/x86/xen/irq.c | 143 ++++ arch/x86/xen/mmu.c | 270 ++++++- arch/x86/xen/multicalls.c | 115 +++- arch/x86/xen/smp.c | 245 ++----- arch/x86/xen/spinlock.c | 428 ++++++++++ arch/x86/xen/time.c | 12 +- arch/x86/xen/xen-asm_32.S | 2 +- arch/x86/xen/xen-asm_64.S | 2 +- arch/x86/xen/xen-ops.h | 8 + drivers/block/xen-blkfront.c | 2 +- drivers/char/hvc_xen.c | 6 +- drivers/input/xen-kbdfront.c | 4 +- drivers/net/xen-netfront.c | 6 +- drivers/video/xen-fbfront.c | 4 +- drivers/xen/Makefile | 1 + drivers/xen/balloon.c | 174 +---- drivers/xen/cpu_hotplug.c | 90 +++ drivers/xen/events.c | 40 +- drivers/xen/grant-table.c | 2 +- drivers/xen/xenbus/xenbus_probe.c | 8 +- include/asm-x86/acpi.h | 3 - include/asm-x86/desc.h | 29 +- include/asm-x86/es7000/mpparse.h | 1 + include/asm-x86/fixmap_32.h | 8 +- include/asm-x86/fixmap_64.h | 16 +- include/asm-x86/io.h | 15 +- include/asm-x86/io_64.h | 3 - include/asm-x86/irqflags.h | 21 - include/asm-x86/kdebug.h | 3 +- include/asm-x86/kprobes.h | 9 - include/asm-x86/mach-default/mach_traps.h | 6 - include/asm-x86/mmzone_64.h | 3 +- include/asm-x86/module.h | 2 - include/asm-x86/nmi.h | 4 - include/asm-x86/page.h | 8 +- include/asm-x86/page_32.h | 13 +- include/asm-x86/paravirt.h | 20 + include/asm-x86/pgtable.h | 16 +- include/asm-x86/ptrace.h | 4 - include/asm-x86/smp.h | 40 +- include/asm-x86/spinlock.h | 9 +- include/asm-x86/system.h | 5 +- include/asm-x86/tlbflush.h | 10 + include/asm-x86/traps.h | 73 +- include/asm-x86/xen/hypervisor.h | 14 +- include/linux/kernel.h | 2 +- include/linux/mm.h | 13 +- include/linux/mm_types.h | 10 +- include/linux/mmdebug.h | 18 + include/linux/sched.h | 6 +- include/xen/events.h | 2 + lib/Kconfig.debug | 9 + lib/cmdline.c | 2 +- mm/vmalloc.c | 7 + 96 files changed, 3373 insertions(+), 2792 deletions(-) create mode 100644 arch/x86/kernel/dumpstack_32.c create mode 100644 arch/x86/kernel/dumpstack_64.c create mode 100644 arch/x86/kernel/paravirt-spinlocks.c rename arch/x86/kernel/{traps_32.c => traps.c} (60%) delete mode 100644 arch/x86/kernel/traps_64.c rename arch/x86/mm/{discontig_32.c => numa_32.c} (100%) create mode 100644 arch/x86/xen/debugfs.c create mode 100644 arch/x86/xen/debugfs.h create mode 100644 arch/x86/xen/irq.c create mode 100644 arch/x86/xen/spinlock.c create mode 100644 drivers/xen/cpu_hotplug.c create mode 100644 include/linux/mmdebug.h
diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index f8843c3..52763b7 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -38,8 +38,7 @@ config M386 - "Crusoe" for the Transmeta Crusoe series. - "Efficeon" for the Transmeta Efficeon series. - "Winchip-C6" for original IDT Winchip. - - "Winchip-2" for IDT Winchip 2. - - "Winchip-2A" for IDT Winchips with 3dNow! capabilities. + - "Winchip-2" for IDT Winchips with 3dNow! capabilities. - "GeodeGX1" for Geode GX1 (Cyrix MediaGX). - "Geode GX/LX" For AMD Geode GX and LX processors. - "CyrixIII/VIA C3" for VIA Cyrix III or VIA C3. @@ -194,19 +193,11 @@ config MWINCHIPC6 treat this chip as a 586TSC with some extended instructions and alignment requirements. -config MWINCHIP2 - bool "Winchip-2" - depends on X86_32 - help - Select this for an IDT Winchip-2. Linux and GCC - treat this chip as a 586TSC with some extended instructions - and alignment requirements. - config MWINCHIP3D - bool "Winchip-2A/Winchip-3" + bool "Winchip-2/Winchip-2A/Winchip-3" depends on X86_32 help - Select this for an IDT Winchip-2A or 3. Linux and GCC + Select this for an IDT Winchip-2, 2A or 3. Linux and GCC treat this chip as a 586TSC with some extended instructions and alignment requirements. Also enable out of order memory stores for this CPU, which can increase performance of some @@ -318,7 +309,7 @@ config X86_L1_CACHE_SHIFT int default "7" if MPENTIUM4 || X86_GENERIC || GENERIC_CPU || MPSC default "4" if X86_ELAN || M486 || M386 || MGEODEGX1 - default "5" if MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODE_LX + default "5" if MWINCHIP3D || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODE_LX default "6" if MK7 || MK8 || MPENTIUMM || MCORE2 || MVIAC7 config X86_XADD @@ -360,7 +351,7 @@ config X86_POPAD_OK config X86_ALIGNMENT_16 def_bool y - depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 + depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 config X86_INTEL_USERCOPY def_bool y @@ -368,7 +359,7 @@ config X86_INTEL_USERCOPY config X86_USE_PPRO_CHECKSUM def_bool y - depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX || MCORE2 + depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX || MCORE2 config X86_USE_3DNOW def_bool y @@ -376,7 +367,7 @@ config X86_USE_3DNOW config X86_OOSTORE def_bool y - depends on (MWINCHIP3D || MWINCHIP2 || MWINCHIPC6) && MTRR + depends on (MWINCHIP3D || MWINCHIPC6) && MTRR # # P6_NOPs are a relatively minor optimization that require a family >= @@ -396,7 +387,7 @@ config X86_P6_NOP config X86_TSC def_bool y - depends on ((MWINCHIP3D || MWINCHIP2 || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MVIAC7 || MGEODEGX1 || MGEODE_LX || MCORE2) && !X86_NUMAQ) || X86_64 + depends on ((MWINCHIP3D || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MVIAC7 || MGEODEGX1 || MGEODE_LX || MCORE2) && !X86_NUMAQ) || X86_64 config X86_CMPXCHG64 def_bool y @@ -406,7 +397,7 @@ config X86_CMPXCHG64 # generates cmov. config X86_CMOV def_bool y - depends on (MK7 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || X86_64) + depends on (MK8 || MK7 || MCORE2 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64) config X86_MINIMUM_CPU_FAMILY int @@ -417,7 +408,7 @@ config X86_MINIMUM_CPU_FAMILY config X86_DEBUGCTLMSR def_bool y - depends on !(MK6 || MWINCHIPC6 || MWINCHIP2 || MWINCHIP3D || MCYRIXIII || M586MMX || M586TSC || M586 || M486 || M386) + depends on !(MK6 || MWINCHIPC6 || MWINCHIP3D || MCYRIXIII || M586MMX || M586TSC || M586 || M486 || M386) menuconfig PROCESSOR_SELECT default y diff --git a/arch/x86/Makefile_32.cpu b/arch/x86/Makefile_32.cpu index e372b58..12bdd85 100644 --- a/arch/x86/Makefile_32.cpu +++ b/arch/x86/Makefile_32.cpu @@ -28,7 +28,6 @@ cflags-$(CONFIG_MK8) += $(call cc-option,-march=k8,-march=athlon) cflags-$(CONFIG_MCRUSOE) += -march=i686 $(align)-functions=0 $(align)-jumps=0 $(align)-loops=0 cflags-$(CONFIG_MEFFICEON) += -march=i686 $(call tune,pentium3) $(align)-functions=0 $(align)-jumps=0 $(align)-loops=0 cflags-$(CONFIG_MWINCHIPC6) += $(call cc-option,-march=winchip-c6,-march=i586) -cflags-$(CONFIG_MWINCHIP2) += $(call cc-option,-march=winchip2,-march=i586) cflags-$(CONFIG_MWINCHIP3D) += $(call cc-option,-march=winchip2,-march=i586) cflags-$(CONFIG_MCYRIXIII) += $(call cc-option,-march=c3,-march=i486) $(align)-functions=0 $(align)-jumps=0 $(align)-loops=0 cflags-$(CONFIG_MVIAC3_2) += $(call cc-option,-march=c3-2,-march=i686) diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig index ef9a520..9269ef4 100644 --- a/arch/x86/configs/i386_defconfig +++ b/arch/x86/configs/i386_defconfig @@ -213,7 +213,6 @@ CONFIG_M686=y # CONFIG_MCRUSOE is not set # CONFIG_MEFFICEON is not set # CONFIG_MWINCHIPC6 is not set -# CONFIG_MWINCHIP2 is not set # CONFIG_MWINCHIP3D is not set # CONFIG_MGEODEGX1 is not set # CONFIG_MGEODE_LX is not set diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig index e620ea6..638bc85 100644 --- a/arch/x86/configs/x86_64_defconfig +++ b/arch/x86/configs/x86_64_defconfig @@ -210,7 +210,6 @@ CONFIG_X86_PC=y # CONFIG_MCRUSOE is not set # CONFIG_MEFFICEON is not set # CONFIG_MWINCHIPC6 is not set -# CONFIG_MWINCHIP2 is not set # CONFIG_MWINCHIP3D is not set # CONFIG_MGEODEGX1 is not set # CONFIG_MGEODE_LX is not set diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index ffc1bb4..eb43147 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -39,11 +39,11 @@ .endm /* clobbers %eax */ - .macro CLEAR_RREGS + .macro CLEAR_RREGS _r9=rax xorl %eax,%eax movq %rax,R11(%rsp) movq %rax,R10(%rsp) - movq %rax,R9(%rsp) + movq %\_r9,R9(%rsp) movq %rax,R8(%rsp) .endm @@ -52,11 +52,10 @@ * We don't reload %eax because syscall_trace_enter() returned * the value it wants us to use in the table lookup. */ - .macro LOAD_ARGS32 offset - movl \offset(%rsp),%r11d - movl \offset+8(%rsp),%r10d + .macro LOAD_ARGS32 offset, _r9=0 + .if \_r9 movl \offset+16(%rsp),%r9d - movl \offset+24(%rsp),%r8d + .endif movl \offset+40(%rsp),%ecx movl \offset+48(%rsp),%edx movl \offset+56(%rsp),%esi @@ -145,7 +144,7 @@ ENTRY(ia32_sysenter_target) SAVE_ARGS 0,0,1 /* no need to do an access_ok check here because rbp has been 32bit zero extended */ -1: movl (%rbp),%r9d +1: movl (%rbp),%ebp .section __ex_table,"a" .quad 1b,ia32_badarg .previous @@ -157,7 +156,7 @@ ENTRY(ia32_sysenter_target) cmpl $(IA32_NR_syscalls-1),%eax ja ia32_badsys sysenter_do_call: - IA32_ARG_FIXUP 1 + IA32_ARG_FIXUP sysenter_dispatch: call *ia32_sys_call_table(,%rax,8) movq %rax,RAX-ARGOFFSET(%rsp) @@ -234,20 +233,17 @@ sysexit_audit: #endif sysenter_tracesys: - xchgl %r9d,%ebp #ifdef CONFIG_AUDITSYSCALL testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10) jz sysenter_auditsys #endif SAVE_REST CLEAR_RREGS - movq %r9,R9(%rsp) movq $-ENOSYS,RAX(%rsp)/* ptrace can change this for a bad syscall */ movq %rsp,%rdi /* &pt_regs -> arg1 */ call syscall_trace_enter LOAD_ARGS32 ARGOFFSET /* reload args from stack in case ptrace changed it */ RESTORE_REST - xchgl %ebp,%r9d cmpl $(IA32_NR_syscalls-1),%eax ja int_ret_from_sys_call /* sysenter_tracesys has set RAX(%rsp) */ jmp sysenter_do_call @@ -314,9 +310,9 @@ ENTRY(ia32_cstar_target) testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10) CFI_REMEMBER_STATE jnz cstar_tracesys -cstar_do_call: cmpl $IA32_NR_syscalls-1,%eax ja ia32_badsys +cstar_do_call: IA32_ARG_FIXUP 1 cstar_dispatch: call *ia32_sys_call_table(,%rax,8) @@ -357,15 +353,13 @@ cstar_tracesys: #endif xchgl %r9d,%ebp SAVE_REST - CLEAR_RREGS - movq %r9,R9(%rsp) + CLEAR_RREGS r9 movq $-ENOSYS,RAX(%rsp) /* ptrace can change this for a bad syscall */ movq %rsp,%rdi /* &pt_regs -> arg1 */ call syscall_trace_enter - LOAD_ARGS32 ARGOFFSET /* reload args from stack in case ptrace changed it */ + LOAD_ARGS32 ARGOFFSET, 1 /* reload args from stack in case ptrace changed it */ RESTORE_REST xchgl %ebp,%r9d - movl RSP-ARGOFFSET(%rsp), %r8d cmpl $(IA32_NR_syscalls-1),%eax ja int_ret_from_sys_call /* cstar_tracesys has set RAX(%rsp) */ jmp cstar_do_call diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index c9be69f..f6160cc 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -10,7 +10,7 @@ ifdef CONFIG_FTRACE # Do not profile debug and lowlevel utilities CFLAGS_REMOVE_tsc.o = -pg CFLAGS_REMOVE_rtc.o = -pg -CFLAGS_REMOVE_paravirt.o = -pg +CFLAGS_REMOVE_paravirt-spinlocks.o = -pg endif # @@ -23,7 +23,7 @@ CFLAGS_hpet.o := $(nostackp) CFLAGS_tsc.o := $(nostackp) obj-y := process_$(BITS).o signal_$(BITS).o entry_$(BITS).o -obj-y += traps_$(BITS).o irq_$(BITS).o +obj-y += traps.o irq_$(BITS).o dumpstack_$(BITS).o obj-y += time_$(BITS).o ioport.o ldt.o obj-y += setup.o i8259.o irqinit_$(BITS).o setup_percpu.o obj-$(CONFIG_X86_VISWS) += visws_quirks.o @@ -90,7 +90,7 @@ obj-$(CONFIG_DEBUG_NX_TEST) += test_nx.o obj-$(CONFIG_VMI) += vmi_32.o vmiclock_32.o obj-$(CONFIG_KVM_GUEST) += kvm.o obj-$(CONFIG_KVM_CLOCK) += kvmclock.o -obj-$(CONFIG_PARAVIRT) += paravirt.o paravirt_patch_$(BITS).o +obj-$(CONFIG_PARAVIRT) += paravirt.o paravirt_patch_$(BITS).o paravirt-spinlocks.o obj-$(CONFIG_PARAVIRT_CLOCK) += pvclock.o obj-$(CONFIG_PCSPKR_PLATFORM) += pcspeaker.o diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 0ff30ae..e8ddc8b 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -121,35 +121,19 @@ enum acpi_irq_model_id acpi_irq_model = ACPI_IRQ_MODEL_PIC; */ char *__init __acpi_map_table(unsigned long phys, unsigned long size) { - unsigned long base, offset, mapped_size; - int idx; + static char *prev_map; + static unsigned long prev_size; if (!phys || !size) return NULL; - if (phys+size <= (max_low_pfn_mapped << PAGE_SHIFT)) - return __va(phys); + if (prev_map) + early_iounmap(prev_map, prev_size); - offset = phys & (PAGE_SIZE - 1); - mapped_size = PAGE_SIZE - offset; - clear_fixmap(FIX_ACPI_END); - set_fixmap(FIX_ACPI_END, phys); - base = fix_to_virt(FIX_ACPI_END); + prev_size = size; + prev_map = early_ioremap(phys, size); - /* - * Most cases can be covered by the below. - */ - idx = FIX_ACPI_END; - while (mapped_size < size) { - if (--idx < FIX_ACPI_BEGIN) - return NULL; /* cannot handle this */ - phys += PAGE_SIZE; - clear_fixmap(idx); - set_fixmap(idx, phys); - mapped_size += PAGE_SIZE; - } - - return ((unsigned char *)base + offset); + return prev_map; } #ifdef CONFIG_PCI_MMCONFIG diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index fb04e49..a84ac7b 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -444,7 +444,7 @@ void __init alternative_instructions(void) _text, _etext); /* Only switch to UP mode if we don't immediately boot others */ - if (num_possible_cpus() == 1 || setup_max_cpus <= 1) + if (num_present_cpus() == 1 || setup_max_cpus <= 1) alternatives_smp_switch(0); } #endif diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index d0abfd8..e378a16 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -124,18 +124,25 @@ static inline int flag_is_changeable_p(u32 flag) { u32 f1, f2; - asm("pushfl\n\t" - "pushfl\n\t" - "popl %0\n\t" - "movl %0,%1\n\t" - "xorl %2,%0\n\t" - "pushl %0\n\t" - "popfl\n\t" - "pushfl\n\t" - "popl %0\n\t" - "popfl\n\t" - : "=&r" (f1), "=&r" (f2) - : "ir" (flag)); + /* + * Cyrix and IDT cpus allow disabling of CPUID + * so the code below may return different results + * when it is executed before and after enabling + * the CPUID. Add "volatile" to not allow gcc to + * optimize the subsequent calls to this function. + */ + asm volatile ("pushfl\n\t" + "pushfl\n\t" + "popl %0\n\t" + "movl %0,%1\n\t" + "xorl %2,%0\n\t" + "pushl %0\n\t" + "popfl\n\t" + "pushfl\n\t" + "popl %0\n\t" + "popfl\n\t" + : "=&r" (f1), "=&r" (f2) + : "ir" (flag)); return ((f1^f2) & flag) != 0; } @@ -574,6 +581,10 @@ void __init early_cpu_init(void) * The NOPL instruction is supposed to exist on all CPUs with * family >= 6; unfortunately, that's not true in practice because * of early VIA chips and (more importantly) broken virtualizers that + * + * Note: no 64-bit chip is known to lack these, but put the code here + * for consistency with 32 bits, and to make it utterly trivial to + * diagnose the problem should it ever surface. * are not easy to detect. In the latter case it doesn't even *fail* * reliably, so probing for it doesn't even work. Disable it completely * unless we can find a reliable way to detect all the broken cases. @@ -797,7 +808,7 @@ void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) else if (c->cpuid_level >= 0) vendor = c->x86_vendor_id; - if (vendor && strncmp(c->x86_model_id, vendor, strlen(vendor))) + if (vendor && !strstr(c->x86_model_id, vendor)) printk(KERN_CONT "%s ", vendor); if (c->x86_model_id[0]) @@ -1121,16 +1132,5 @@ void __cpuinit cpu_init(void) xsave_init(); } -#ifdef CONFIG_HOTPLUG_CPU -void __cpuinit cpu_uninit(void) -{ - int cpu = raw_smp_processor_id(); - cpu_clear(cpu, cpu_initialized); - - /* lazy TLB state */ - per_cpu(cpu_tlbstate, cpu).state = 0; - per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm; -} -#endif #endif diff --git a/arch/x86/kernel/doublefault_32.c b/arch/x86/kernel/doublefault_32.c index a47798b..b4f14c6 100644 --- a/arch/x86/kernel/doublefault_32.c +++ b/arch/x86/kernel/doublefault_32.c @@ -66,6 +66,6 @@ struct tss_struct doublefault_tss __cacheline_aligned = { .ds = __USER_DS, .fs = __KERNEL_PERCPU, - .__cr3 = __pa(swapper_pg_dir) + .__cr3 = __pa_nodebug(swapper_pg_dir), } }; diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c new file mode 100644 index 0000000..c398b27 --- /dev/null +++ b/arch/x86/kernel/dumpstack_32.c @@ -0,0 +1,422 @@ +/* + * Copyright (C) 1991, 1992 Linus Torvalds + * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs + */ +#include <linux/kallsyms.h> +#include <linux/kprobes.h> +#include <linux/uaccess.h> +#include <linux/utsname.h> +#include <linux/hardirq.h> +#include <linux/kdebug.h> +#include <linux/module.h> +#include <linux/ptrace.h> +#include <linux/kexec.h> +#include <linux/bug.h> +#include <linux/nmi.h> + +#include <asm/stacktrace.h> + +int panic_on_unrecovered_nmi; +int kstack_depth_to_print = 24; +static unsigned int code_bytes = 64; +static int die_counter; + +void printk_address(unsigned long address, int reliable) +{ +#ifdef CONFIG_KALLSYMS + unsigned long offset = 0; + unsigned long symsize; + const char *symname; + char *modname; + char *delim = ":"; + char namebuf[KSYM_NAME_LEN]; + char reliab[4] = ""; + + symname = kallsyms_lookup(address, &symsize, &offset, + &modname, namebuf); + if (!symname) { + printk(" [<%08lx>]\n", address); + return; + } + if (!reliable) + strcpy(reliab, "? "); + + if (!modname) + modname = delim = ""; + printk(" [<%08lx>] %s%s%s%s%s+0x%lx/0x%lx\n", + address, reliab, delim, modname, delim, symname, offset, symsize); +#else + printk(" [<%08lx>]\n", address); +#endif +} + +static inline int valid_stack_ptr(struct thread_info *tinfo, + void *p, unsigned int size) +{ + void *t = tinfo; + return p > t && p <= t + THREAD_SIZE - size; +} + +/* The form of the top of the frame on the stack */ +struct stack_frame { + struct stack_frame *next_frame; + unsigned long return_address; +}; + +static inline unsigned long +print_context_stack(struct thread_info *tinfo, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data) +{ + struct stack_frame *frame = (struct stack_frame *)bp; + + while (valid_stack_ptr(tinfo, stack, sizeof(*stack))) { + unsigned long addr; + + addr = *stack; + if (__kernel_text_address(addr)) { + if ((unsigned long) stack == bp + 4) { + ops->address(data, addr, 1); + frame = frame->next_frame; + bp = (unsigned long) frame; + } else { + ops->address(data, addr, bp == 0); + } + } + stack++; + } + return bp; +} + +void dump_trace(struct task_struct *task, struct pt_regs *regs, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data) +{ + if (!task) + task = current; + + if (!stack) { + unsigned long dummy; + stack = &dummy; + if (task != current) + stack = (unsigned long *)task->thread.sp; + } + +#ifdef CONFIG_FRAME_POINTER + if (!bp) { + if (task == current) { + /* Grab bp right from our regs */ + asm("movl %%ebp, %0" : "=r" (bp) :); + } else { + /* bp is the last reg pushed by switch_to */ + bp = *(unsigned long *) task->thread.sp; + } + } +#endif + + for (;;) { + struct thread_info *context; + + context = (struct thread_info *) + ((unsigned long)stack & (~(THREAD_SIZE - 1))); + bp = print_context_stack(context, stack, bp, ops, data); + /* + * Should be after the line below, but somewhere + * in early boot context comes out corrupted and we + * can't reference it: + */ + if (ops->stack(data, "IRQ") < 0) + break; + stack = (unsigned long *)context->previous_esp; + if (!stack) + break; + touch_nmi_watchdog(); + } +} +EXPORT_SYMBOL(dump_trace); + +static void +print_trace_warning_symbol(void *data, char *msg, unsigned long symbol) +{ + printk(data); + print_symbol(msg, symbol); + printk("\n"); +} + +static void print_trace_warning(void *data, char *msg) +{ + printk("%s%s\n", (char *)data, msg); +} + +static int print_trace_stack(void *data, char *name) +{ + return 0; +} + +/* + * Print one address/symbol entries per line. + */ +static void print_trace_address(void *data, unsigned long addr, int reliable) +{ + printk("%s [<%08lx>] ", (char *)data, addr); + if (!reliable) + printk("? "); + print_symbol("%s\n", addr); + touch_nmi_watchdog(); +} + +static const struct stacktrace_ops print_trace_ops = { + .warning = print_trace_warning, + .warning_symbol = print_trace_warning_symbol, + .stack = print_trace_stack, + .address = print_trace_address, +}; + +static void +show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, + unsigned long *stack, unsigned long bp, char *log_lvl) +{ + dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl); + printk("%s =======================\n", log_lvl); +} + +void show_trace(struct task_struct *task, struct pt_regs *regs, + unsigned long *stack, unsigned long bp) +{ + show_trace_log_lvl(task, regs, stack, bp, ""); +} + +static void +show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, + unsigned long *sp, unsigned long bp, char *log_lvl) +{ + unsigned long *stack; + int i; + + if (sp == NULL) { + if (task) + sp = (unsigned long *)task->thread.sp; + else + sp = (unsigned long *)&sp; + } + + stack = sp; + for (i = 0; i < kstack_depth_to_print; i++) { + if (kstack_end(stack)) + break; + if (i && ((i % 8) == 0)) + printk("\n%s ", log_lvl); + printk("%08lx ", *stack++); + } + printk("\n%sCall Trace:\n", log_lvl); + + show_trace_log_lvl(task, regs, sp, bp, log_lvl); +} + +void show_stack(struct task_struct *task, unsigned long *sp) +{ + printk(" "); + show_stack_log_lvl(task, NULL, sp, 0, ""); +} + +/* + * The architecture-independent dump_stack generator + */ +void dump_stack(void) +{ + unsigned long bp = 0; + unsigned long stack; + +#ifdef CONFIG_FRAME_POINTER + if (!bp) + asm("movl %%ebp, %0" : "=r" (bp):); +#endif + + printk("Pid: %d, comm: %.20s %s %s %.*s\n", + current->pid, current->comm, print_tainted(), + init_utsname()->release, + (int)strcspn(init_utsname()->version, " "), + init_utsname()->version); + + show_trace(current, NULL, &stack, bp); +} + +EXPORT_SYMBOL(dump_stack); + +void show_registers(struct pt_regs *regs) +{ + int i; + + print_modules(); + __show_regs(regs, 0); + + printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)", + TASK_COMM_LEN, current->comm, task_pid_nr(current), + current_thread_info(), current, task_thread_info(current)); + /* + * When in-kernel, we also print out the stack and code at the + * time of the fault.. + */ + if (!user_mode_vm(regs)) { + unsigned int code_prologue = code_bytes * 43 / 64; + unsigned int code_len = code_bytes; + unsigned char c; + u8 *ip; + + printk("\n" KERN_EMERG "Stack: "); + show_stack_log_lvl(NULL, regs, ®s->sp, 0, KERN_EMERG); + + printk(KERN_EMERG "Code: "); + + ip = (u8 *)regs->ip - code_prologue; + if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) { + /* try starting at EIP */ + ip = (u8 *)regs->ip; + code_len = code_len - code_prologue + 1; + } + for (i = 0; i < code_len; i++, ip++) { + if (ip < (u8 *)PAGE_OFFSET || + probe_kernel_address(ip, c)) { + printk(" Bad EIP value."); + break; + } + if (ip == (u8 *)regs->ip) + printk("<%02x> ", c); + else + printk("%02x ", c); + } + } + printk("\n"); +} + +int is_valid_bugaddr(unsigned long ip) +{ + unsigned short ud2; + + if (ip < PAGE_OFFSET) + return 0; + if (probe_kernel_address((unsigned short *)ip, ud2)) + return 0; + + return ud2 == 0x0b0f; +} + +static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; +static int die_owner = -1; +static unsigned int die_nest_count; + +unsigned __kprobes long oops_begin(void) +{ + unsigned long flags; + + oops_enter(); + + if (die_owner != raw_smp_processor_id()) { + console_verbose(); + raw_local_irq_save(flags); + __raw_spin_lock(&die_lock); + die_owner = smp_processor_id(); + die_nest_count = 0; + bust_spinlocks(1); + } else { + raw_local_irq_save(flags); + } + die_nest_count++; + return flags; +} + +void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) +{ + bust_spinlocks(0); + die_owner = -1; + add_taint(TAINT_DIE); + __raw_spin_unlock(&die_lock); + raw_local_irq_restore(flags); + + if (!regs) + return; + + if (kexec_should_crash(current)) + crash_kexec(regs); + + if (in_interrupt()) + panic("Fatal exception in interrupt"); + + if (panic_on_oops) + panic("Fatal exception"); + + oops_exit(); + do_exit(signr); +} + +int __kprobes __die(const char *str, struct pt_regs *regs, long err) +{ + unsigned short ss; + unsigned long sp; + + printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter); +#ifdef CONFIG_PREEMPT + printk("PREEMPT "); +#endif +#ifdef CONFIG_SMP + printk("SMP "); +#endif +#ifdef CONFIG_DEBUG_PAGEALLOC + printk("DEBUG_PAGEALLOC"); +#endif + printk("\n"); + if (notify_die(DIE_OOPS, str, regs, err, + current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) + return 1; + + show_registers(regs); + /* Executive summary in case the oops scrolled away */ + sp = (unsigned long) (®s->sp); + savesegment(ss, ss); + if (user_mode(regs)) { + sp = regs->sp; + ss = regs->ss & 0xffff; + } + printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip); + print_symbol("%s", regs->ip); + printk(" SS:ESP %04x:%08lx\n", ss, sp); + return 0; +} + +/* + * This is gone through when something in the kernel has done something bad + * and is about to be terminated: + */ +void die(const char *str, struct pt_regs *regs, long err) +{ + unsigned long flags = oops_begin(); + + if (die_nest_count < 3) { + report_bug(regs->ip, regs); + + if (__die(str, regs, err)) + regs = NULL; + } else { + printk(KERN_EMERG "Recursive die() failure, output suppressed\n"); + } + + oops_end(flags, regs, SIGSEGV); +} + +static int __init kstack_setup(char *s) +{ + kstack_depth_to_print = simple_strtoul(s, NULL, 0); + + return 1; +} +__setup("kstack=", kstack_setup); + +static int __init code_bytes_setup(char *s) +{ + code_bytes = simple_strtoul(s, NULL, 0); + if (code_bytes > 8192) + code_bytes = 8192; + + return 1; +} +__setup("code_bytes=", code_bytes_setup); diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c new file mode 100644 index 0000000..6f15050 --- /dev/null +++ b/arch/x86/kernel/dumpstack_64.c @@ -0,0 +1,565 @@ +/* + * Copyright (C) 1991, 1992 Linus Torvalds + * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs + */ +#include <linux/kallsyms.h> +#include <linux/kprobes.h> +#include <linux/uaccess.h> +#include <linux/utsname.h> +#include <linux/hardirq.h> +#include <linux/kdebug.h> +#include <linux/module.h> +#include <linux/ptrace.h> +#include <linux/kexec.h> +#include <linux/bug.h> +#include <linux/nmi.h> + +#include <asm/stacktrace.h> + +int panic_on_unrecovered_nmi; +int kstack_depth_to_print = 12; +static unsigned int code_bytes = 64; +static int die_counter; + +void printk_address(unsigned long address, int reliable) +{ + printk(" [<%016lx>] %s%pS\n", + address, reliable ? "" : "? ", (void *) address); +} + +static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, + unsigned *usedp, char **idp) +{ + static char ids[][8] = { + [DEBUG_STACK - 1] = "#DB", + [NMI_STACK - 1] = "NMI", + [DOUBLEFAULT_STACK - 1] = "#DF", + [STACKFAULT_STACK - 1] = "#SS", + [MCE_STACK - 1] = "#MC", +#if DEBUG_STKSZ > EXCEPTION_STKSZ + [N_EXCEPTION_STACKS ... + N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]" +#endif + }; + unsigned k; + + /* + * Iterate over all exception stacks, and figure out whether + * 'stack' is in one of them: + */ + for (k = 0; k < N_EXCEPTION_STACKS; k++) { + unsigned long end = per_cpu(orig_ist, cpu).ist[k]; + /* + * Is 'stack' above this exception frame's end? + * If yes then skip to the next frame. + */ + if (stack >= end) + continue; + /* + * Is 'stack' above this exception frame's start address? + * If yes then we found the right frame. + */ + if (stack >= end - EXCEPTION_STKSZ) { + /* + * Make sure we only iterate through an exception + * stack once. If it comes up for the second time + * then there's something wrong going on - just + * break out and return NULL: + */ + if (*usedp & (1U << k)) + break; + *usedp |= 1U << k; + *idp = ids[k]; + return (unsigned long *)end; + } + /* + * If this is a debug stack, and if it has a larger size than + * the usual exception stacks, then 'stack' might still + * be within the lower portion of the debug stack: + */ +#if DEBUG_STKSZ > EXCEPTION_STKSZ + if (k == DEBUG_STACK - 1 && stack >= end - DEBUG_STKSZ) { + unsigned j = N_EXCEPTION_STACKS - 1; + + /* + * Black magic. A large debug stack is composed of + * multiple exception stack entries, which we + * iterate through now. Dont look: + */ + do { + ++j; + end -= EXCEPTION_STKSZ; + ids[j][4] = '1' + (j - N_EXCEPTION_STACKS); + } while (stack < end - EXCEPTION_STKSZ); + if (*usedp & (1U << j)) + break; + *usedp |= 1U << j; + *idp = ids[j]; + return (unsigned long *)end; + } +#endif + } + return NULL; +} + +/* + * x86-64 can have up to three kernel stacks: + * process stack + * interrupt stack + * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack + */ + +static inline int valid_stack_ptr(struct thread_info *tinfo, + void *p, unsigned int size, void *end) +{ + void *t = tinfo; + if (end) { + if (p < end && p >= (end-THREAD_SIZE)) + return 1; + else + return 0; + } + return p > t && p < t + THREAD_SIZE - size; +} + +/* The form of the top of the frame on the stack */ +struct stack_frame { + struct stack_frame *next_frame; + unsigned long return_address; +}; + +static inline unsigned long +print_context_stack(struct thread_info *tinfo, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data, + unsigned long *end) +{ + struct stack_frame *frame = (struct stack_frame *)bp; + + while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) { + unsigned long addr; + + addr = *stack; + if (__kernel_text_address(addr)) { + if ((unsigned long) stack == bp + 8) { + ops->address(data, addr, 1); + frame = frame->next_frame; + bp = (unsigned long) frame; + } else { + ops->address(data, addr, bp == 0); + } + } + stack++; + } + return bp; +} + +void dump_trace(struct task_struct *task, struct pt_regs *regs, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data) +{ + const unsigned cpu = get_cpu(); + unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr; + unsigned used = 0; + struct thread_info *tinfo; + + if (!task) + task = current; + + if (!stack) { + unsigned long dummy; + stack = &dummy; + if (task && task != current) + stack = (unsigned long *)task->thread.sp; + } + +#ifdef CONFIG_FRAME_POINTER + if (!bp) { + if (task == current) { + /* Grab bp right from our regs */ + asm("movq %%rbp, %0" : "=r" (bp) : ); + } else { + /* bp is the last reg pushed by switch_to */ + bp = *(unsigned long *) task->thread.sp; + } + } +#endif + + /* + * Print function call entries in all stacks, starting at the + * current stack address. If the stacks consist of nested + * exceptions + */ + tinfo = task_thread_info(task); + for (;;) { + char *id; + unsigned long *estack_end; + estack_end = in_exception_stack(cpu, (unsigned long)stack, + &used, &id); + + if (estack_end) { + if (ops->stack(data, id) < 0) + break; + + bp = print_context_stack(tinfo, stack, bp, ops, + data, estack_end); + ops->stack(data, "<EOE>"); + /* + * We link to the next stack via the + * second-to-last pointer (index -2 to end) in the + * exception stack: + */ + stack = (unsigned long *) estack_end[-2]; + continue; + } + if (irqstack_end) { + unsigned long *irqstack; + irqstack = irqstack_end - + (IRQSTACKSIZE - 64) / sizeof(*irqstack); + + if (stack >= irqstack && stack < irqstack_end) { + if (ops->stack(data, "IRQ") < 0) + break; + bp = print_context_stack(tinfo, stack, bp, + ops, data, irqstack_end); + /* + * We link to the next stack (which would be + * the process stack normally) the last + * pointer (index -1 to end) in the IRQ stack: + */ + stack = (unsigned long *) (irqstack_end[-1]); + irqstack_end = NULL; + ops->stack(data, "EOI"); + continue; + } + } + break; + } + + /* + * This handles the process stack: + */ + bp = print_context_stack(tinfo, stack, bp, ops, data, NULL); + put_cpu(); +} +EXPORT_SYMBOL(dump_trace); + +static void +print_trace_warning_symbol(void *data, char *msg, unsigned long symbol) +{ + print_symbol(msg, symbol); + printk("\n"); +} + +static void print_trace_warning(void *data, char *msg) +{ + printk("%s\n", msg); +} + +static int print_trace_stack(void *data, char *name) +{ + printk(" <%s> ", name); + return 0; +} + +static void print_trace_address(void *data, unsigned long addr, int reliable) +{ + touch_nmi_watchdog(); + printk_address(addr, reliable); +} + +static const struct stacktrace_ops print_trace_ops = { + .warning = print_trace_warning, + .warning_symbol = print_trace_warning_symbol, + .stack = print_trace_stack, + .address = print_trace_address, +}; + +static void +show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, + unsigned long *stack, unsigned long bp, char *log_lvl) +{ + printk("Call Trace:\n"); + dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl); +} + +void show_trace(struct task_struct *task, struct pt_regs *regs, + unsigned long *stack, unsigned long bp) +{ + show_trace_log_lvl(task, regs, stack, bp, ""); +} + +static void +show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, + unsigned long *sp, unsigned long bp, char *log_lvl) +{ + unsigned long *stack; + int i; + const int cpu = smp_processor_id(); + unsigned long *irqstack_end = + (unsigned long *) (cpu_pda(cpu)->irqstackptr); + unsigned long *irqstack = + (unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE); + + /* + * debugging aid: "show_stack(NULL, NULL);" prints the + * back trace for this cpu. + */ + + if (sp == NULL) { + if (task) + sp = (unsigned long *)task->thread.sp; + else + sp = (unsigned long *)&sp; + } + + stack = sp; + for (i = 0; i < kstack_depth_to_print; i++) { + if (stack >= irqstack && stack <= irqstack_end) { + if (stack == irqstack_end) { + stack = (unsigned long *) (irqstack_end[-1]); + printk(" <EOI> "); + } + } else { + if (((long) stack & (THREAD_SIZE-1)) == 0) + break; + } + if (i && ((i % 4) == 0)) + printk("\n"); + printk(" %016lx", *stack++); + touch_nmi_watchdog(); + } + printk("\n"); + show_trace_log_lvl(task, regs, sp, bp, log_lvl); +} + +void show_stack(struct task_struct *task, unsigned long *sp) +{ + show_stack_log_lvl(task, NULL, sp, 0, ""); +} + +/* + * The architecture-independent dump_stack generator + */ +void dump_stack(void) +{ + unsigned long bp = 0; + unsigned long stack; + +#ifdef CONFIG_FRAME_POINTER + if (!bp) + asm("movq %%rbp, %0" : "=r" (bp) : ); +#endif + + printk("Pid: %d, comm: %.20s %s %s %.*s\n", + current->pid, current->comm, print_tainted(), + init_utsname()->release, + (int)strcspn(init_utsname()->version, " "), + init_utsname()->version); + show_trace(NULL, NULL, &stack, bp); +} +EXPORT_SYMBOL(dump_stack); + +void show_registers(struct pt_regs *regs) +{ + int i; + unsigned long sp; + const int cpu = smp_processor_id(); + struct task_struct *cur = cpu_pda(cpu)->pcurrent; + + sp = regs->sp; + printk("CPU %d ", cpu); + __show_regs(regs, 1); + printk("Process %s (pid: %d, threadinfo %p, task %p)\n", + cur->comm, cur->pid, task_thread_info(cur), cur); + + /* + * When in-kernel, we also print out the stack and code at the + * time of the fault.. + */ + if (!user_mode(regs)) { + unsigned int code_prologue = code_bytes * 43 / 64; + unsigned int code_len = code_bytes; + unsigned char c; + u8 *ip; + + printk("Stack: "); + show_stack_log_lvl(NULL, regs, (unsigned long *)sp, + regs->bp, ""); + + printk(KERN_EMERG "Code: "); + + ip = (u8 *)regs->ip - code_prologue; + if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) { + /* try starting at RIP */ + ip = (u8 *)regs->ip; + code_len = code_len - code_prologue + 1; + } + for (i = 0; i < code_len; i++, ip++) { + if (ip < (u8 *)PAGE_OFFSET || + probe_kernel_address(ip, c)) { + printk(" Bad RIP value."); + break; + } + if (ip == (u8 *)regs->ip) + printk("<%02x> ", c); + else + printk("%02x ", c); + } + } + printk("\n"); +} + +int is_valid_bugaddr(unsigned long ip) +{ + unsigned short ud2; + + if (__copy_from_user(&ud2, (const void __user *) ip, sizeof(ud2))) + return 0; + + return ud2 == 0x0b0f; +} + +static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; +static int die_owner = -1; +static unsigned int die_nest_count; + +unsigned __kprobes long oops_begin(void) +{ + int cpu; + unsigned long flags; + + oops_enter(); + + /* racy, but better than risking deadlock. */ + raw_local_irq_save(flags); + cpu = smp_processor_id(); + if (!__raw_spin_trylock(&die_lock)) { + if (cpu == die_owner) + /* nested oops. should stop eventually */; + else + __raw_spin_lock(&die_lock); + } + die_nest_count++; + die_owner = cpu; + console_verbose(); + bust_spinlocks(1); + return flags; +} + +void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) +{ + die_owner = -1; + bust_spinlocks(0); + die_nest_count--; + if (!die_nest_count) + /* Nest count reaches zero, release the lock. */ + __raw_spin_unlock(&die_lock); + raw_local_irq_restore(flags); + if (!regs) { + oops_exit(); + return; + } + if (in_interrupt()) + panic("Fatal exception in interrupt"); + if (panic_on_oops) + panic("Fatal exception"); + oops_exit(); + do_exit(signr); +} + +int __kprobes __die(const char *str, struct pt_regs *regs, long err) +{ + printk(KERN_EMERG "%s: %04lx [%u] ", str, err & 0xffff, ++die_counter); +#ifdef CONFIG_PREEMPT + printk("PREEMPT "); +#endif +#ifdef CONFIG_SMP + printk("SMP "); +#endif +#ifdef CONFIG_DEBUG_PAGEALLOC + printk("DEBUG_PAGEALLOC"); +#endif + printk("\n"); + if (notify_die(DIE_OOPS, str, regs, err, + current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) + return 1; + + show_registers(regs); + add_taint(TAINT_DIE); + /* Executive summary in case the oops scrolled away */ + printk(KERN_ALERT "RIP "); + printk_address(regs->ip, 1); + printk(" RSP <%016lx>\n", regs->sp); + if (kexec_should_crash(current)) + crash_kexec(regs); + return 0; +} + +void die(const char *str, struct pt_regs *regs, long err) +{ + unsigned long flags = oops_begin(); + + if (!user_mode(regs)) + report_bug(regs->ip, regs); + + if (__die(str, regs, err)) + regs = NULL; + oops_end(flags, regs, SIGSEGV); +} + +notrace __kprobes void +die_nmi(char *str, struct pt_regs *regs, int do_panic) +{ + unsigned long flags; + + if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP) + return; + + flags = oops_begin(); + /* + * We are in trouble anyway, lets at least try + * to get a message out. + */ + printk(KERN_EMERG "%s", str); + printk(" on CPU%d, ip %08lx, registers:\n", + smp_processor_id(), regs->ip); + show_registers(regs); + if (kexec_should_crash(current)) + crash_kexec(regs); + if (do_panic || panic_on_oops) + panic("Non maskable interrupt"); + oops_end(flags, NULL, SIGBUS); + nmi_exit(); + local_irq_enable(); + do_exit(SIGBUS); +} + +static int __init oops_setup(char *s) +{ + if (!s) + return -EINVAL; + if (!strcmp(s, "panic")) + panic_on_oops = 1; + return 0; +} +early_param("oops", oops_setup); + +static int __init kstack_setup(char *s) +{ + if (!s) + return -EINVAL; + kstack_depth_to_print = simple_strtoul(s, NULL, 0); + return 0; +} +early_param("kstack", kstack_setup); + +static int __init code_bytes_setup(char *s) +{ + code_bytes = simple_strtoul(s, NULL, 0); + if (code_bytes > 8192) + code_bytes = 8192; + + return 1; +} +__setup("code_bytes=", code_bytes_setup); diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index 109792b..b21fbfa 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -730,6 +730,7 @@ error_code: movl $(__USER_DS), %ecx movl %ecx, %ds movl %ecx, %es + TRACE_IRQS_OFF movl %esp,%eax # pt_regs pointer call *%edi jmp ret_from_exception @@ -760,20 +761,9 @@ ENTRY(device_not_available) RING0_INT_FRAME pushl $-1 # mark this as an int CFI_ADJUST_CFA_OFFSET 4 - SAVE_ALL - GET_CR0_INTO_EAX - testl $0x4, %eax # EM (math emulation bit) - jne device_not_available_emulate - preempt_stop(CLBR_ANY) - call math_state_restore - jmp ret_from_exception -device_not_available_emulate: - pushl $0 # temporary storage for ORIG_EIP + pushl $do_device_not_available CFI_ADJUST_CFA_OFFSET 4 - call math_emulate - addl $4, %esp - CFI_ADJUST_CFA_OFFSET -4 - jmp ret_from_exception + jmp error_code CFI_ENDPROC END(device_not_available) @@ -814,6 +804,7 @@ debug_stack_correct: pushl $-1 # mark this as an int CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL + TRACE_IRQS_OFF xorl %edx,%edx # error code 0 movl %esp,%eax # pt_regs pointer call do_debug @@ -858,6 +849,7 @@ nmi_stack_correct: pushl %eax CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL + TRACE_IRQS_OFF xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer call do_nmi @@ -898,6 +890,7 @@ nmi_espfix_stack: pushl %eax CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL + TRACE_IRQS_OFF FIXUP_ESPFIX_STACK # %eax == %esp xorl %edx,%edx # zero error code call do_nmi @@ -928,6 +921,7 @@ KPROBE_ENTRY(int3) pushl $-1 # mark this as an int CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL + TRACE_IRQS_OFF xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer call do_int3 @@ -1030,7 +1024,7 @@ ENTRY(machine_check) RING0_INT_FRAME pushl $0 CFI_ADJUST_CFA_OFFSET 4 - pushl machine_check_vector + pushl $do_machine_check CFI_ADJUST_CFA_OFFSET 4 jmp error_code CFI_ENDPROC diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index cf3a0b2..104fa65 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -932,6 +932,9 @@ END(spurious_interrupt) .if \ist movq %gs:pda_data_offset, %rbp .endif + .if \irqtrace + TRACE_IRQS_OFF + .endif movq %rsp,%rdi movq ORIG_RAX(%rsp),%rsi movq $-1,ORIG_RAX(%rsp) @@ -1058,7 +1061,8 @@ KPROBE_ENTRY(error_entry) je error_kernelspace error_swapgs: SWAPGS -error_sti: +error_sti: + TRACE_IRQS_OFF movq %rdi,RDI(%rsp) CFI_REL_OFFSET rdi,RDI movq %rsp,%rdi @@ -1232,7 +1236,7 @@ ENTRY(simd_coprocessor_error) END(simd_coprocessor_error) ENTRY(device_not_available) - zeroentry math_state_restore + zeroentry do_device_not_available END(device_not_available) /* runs on exception stack */ diff --git a/arch/x86/kernel/es7000_32.c b/arch/x86/kernel/es7000_32.c index 849e5cd..f454c78 100644 --- a/arch/x86/kernel/es7000_32.c +++ b/arch/x86/kernel/es7000_32.c @@ -109,6 +109,7 @@ struct oem_table { }; extern int find_unisys_acpi_oem_table(unsigned long *oem_addr); +extern void unmap_unisys_acpi_oem_table(unsigned long oem_addr); #endif struct mip_reg { @@ -243,21 +244,38 @@ parse_unisys_oem (char *oemptr) } #ifdef CONFIG_ACPI -int __init -find_unisys_acpi_oem_table(unsigned long *oem_addr) +static unsigned long oem_addrX; +static unsigned long oem_size; +int __init find_unisys_acpi_oem_table(unsigned long *oem_addr) { struct acpi_table_header *header = NULL; int i = 0; - while (ACPI_SUCCESS(acpi_get_table("OEM1", i++, &header))) { + acpi_size tbl_size; + + while (ACPI_SUCCESS(acpi_get_table_with_size("OEM1", i++, &header, &tbl_size))) { if (!memcmp((char *) &header->oem_id, "UNISYS", 6)) { struct oem_table *t = (struct oem_table *)header; - *oem_addr = (unsigned long)__acpi_map_table(t->OEMTableAddr, - t->OEMTableSize); + + oem_addrX = t->OEMTableAddr; + oem_size = t->OEMTableSize; + early_acpi_os_unmap_memory(header, tbl_size); + + *oem_addr = (unsigned long)__acpi_map_table(oem_addrX, + oem_size); return 0; } + early_acpi_os_unmap_memory(header, tbl_size); } return -1; } + +void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr) +{ + if (!oem_addr) + return; + + __acpi_unmap_table((char *)oem_addr, oem_size); +} #endif static void diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c index ae2ffc8..7b0ff40 100644 --- a/arch/x86/kernel/genx2apic_uv_x.c +++ b/arch/x86/kernel/genx2apic_uv_x.c @@ -114,7 +114,7 @@ static void uv_send_IPI_one(int cpu, int vector) unsigned long val, apicid, lapicid; int pnode; - apicid = per_cpu(x86_cpu_to_apicid, cpu); /* ZZZ - cache node-local ? */ + apicid = per_cpu(x86_cpu_to_apicid, cpu); lapicid = apicid & 0x3f; /* ZZZ macro needed */ pnode = uv_apicid_to_pnode(apicid); val = @@ -202,12 +202,10 @@ static unsigned int phys_pkg_id(int index_msb) return uv_read_apic_id() >> index_msb; } -#ifdef ZZZ /* Needs x2apic patch */ static void uv_send_IPI_self(int vector) { apic_write(APIC_SELF_IPI, vector); } -#endif struct genapic apic_x2apic_uv_x = { .name = "UV large system", @@ -215,15 +213,15 @@ struct genapic apic_x2apic_uv_x = { .int_delivery_mode = dest_Fixed, .int_dest_mode = (APIC_DEST_PHYSICAL != 0), .target_cpus = uv_target_cpus, - .vector_allocation_domain = uv_vector_allocation_domain,/* Fixme ZZZ */ + .vector_allocation_domain = uv_vector_allocation_domain, .apic_id_registered = uv_apic_id_registered, .init_apic_ldr = uv_init_apic_ldr, .send_IPI_all = uv_send_IPI_all, .send_IPI_allbutself = uv_send_IPI_allbutself, .send_IPI_mask = uv_send_IPI_mask, - /* ZZZ.send_IPI_self = uv_send_IPI_self, */ + .send_IPI_self = uv_send_IPI_self, .cpu_mask_to_apicid = uv_cpu_mask_to_apicid, - .phys_pkg_id = phys_pkg_id, /* Fixme ZZZ */ + .phys_pkg_id = phys_pkg_id, .get_apic_id = get_apic_id, .set_apic_id = set_apic_id, .apic_id_mask = (0xFFFFFFFFu), diff --git a/arch/x86/kernel/head.c b/arch/x86/kernel/head.c index 3e66bd3..1dcb0f1 100644 --- a/arch/x86/kernel/head.c +++ b/arch/x86/kernel/head.c @@ -35,6 +35,7 @@ void __init reserve_ebda_region(void) /* start of EBDA area */ ebda_addr = get_bios_ebda(); + printk(KERN_INFO "BIOS EBDA/lowmem at: %08x/%08x\n", ebda_addr, lowmem); /* Fixup: bios puts an EBDA in the top 64K segment */ /* of conventional memory, but does not adjust lowmem. */ diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c index 0ed5f93..eee32b4 100644 --- a/arch/x86/kernel/ldt.c +++ b/arch/x86/kernel/ldt.c @@ -52,6 +52,8 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload) memset(newldt + oldsize * LDT_ENTRY_SIZE, 0, (mincount - oldsize) * LDT_ENTRY_SIZE); + paravirt_alloc_ldt(newldt, mincount); + #ifdef CONFIG_X86_64 /* CHECKME: Do we really need this ? */ wmb(); @@ -74,6 +76,7 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload) #endif } if (oldsize) { + paravirt_free_ldt(oldldt, oldsize); if (oldsize * LDT_ENTRY_SIZE > PAGE_SIZE) vfree(oldldt); else @@ -85,10 +88,13 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload) static inline int copy_ldt(mm_context_t *new, mm_context_t *old) { int err = alloc_ldt(new, old->size, 0); + int i; if (err < 0) return err; - memcpy(new->ldt, old->ldt, old->size * LDT_ENTRY_SIZE); + + for(i = 0; i < old->size; i++) + write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE); return 0; } @@ -125,6 +131,7 @@ void destroy_context(struct mm_struct *mm) if (mm == current->active_mm) clear_LDT(); #endif + paravirt_free_ldt(mm->context.ldt, mm->context.size); if (mm->context.size * LDT_ENTRY_SIZE > PAGE_SIZE) vfree(mm->context.ldt); else diff --git a/arch/x86/kernel/paravirt-spinlocks.c b/arch/x86/kernel/paravirt-spinlocks.c new file mode 100644 index 0000000..0e9f198 --- /dev/null +++ b/arch/x86/kernel/paravirt-spinlocks.c @@ -0,0 +1,37 @@ +/* + * Split spinlock implementation out into its own file, so it can be + * compiled in a FTRACE-compatible way. + */ +#include <linux/spinlock.h> +#include <linux/module.h> + +#include <asm/paravirt.h> + +static void default_spin_lock_flags(struct raw_spinlock *lock, unsigned long flags) +{ + __raw_spin_lock(lock); +} + +struct pv_lock_ops pv_lock_ops = { +#ifdef CONFIG_SMP + .spin_is_locked = __ticket_spin_is_locked, + .spin_is_contended = __ticket_spin_is_contended, + + .spin_lock = __ticket_spin_lock, + .spin_lock_flags = default_spin_lock_flags, + .spin_trylock = __ticket_spin_trylock, + .spin_unlock = __ticket_spin_unlock, +#endif +}; +EXPORT_SYMBOL(pv_lock_ops); + +void __init paravirt_use_bytelocks(void) +{ +#ifdef CONFIG_SMP + pv_lock_ops.spin_is_locked = __byte_spin_is_locked; + pv_lock_ops.spin_is_contended = __byte_spin_is_contended; + pv_lock_ops.spin_lock = __byte_spin_lock; + pv_lock_ops.spin_trylock = __byte_spin_trylock; + pv_lock_ops.spin_unlock = __byte_spin_unlock; +#endif +} diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 6b0bb73..e4c8fb6 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -268,17 +268,6 @@ enum paravirt_lazy_mode paravirt_get_lazy_mode(void) return __get_cpu_var(paravirt_lazy_mode); } -void __init paravirt_use_bytelocks(void) -{ -#ifdef CONFIG_SMP - pv_lock_ops.spin_is_locked = __byte_spin_is_locked; - pv_lock_ops.spin_is_contended = __byte_spin_is_contended; - pv_lock_ops.spin_lock = __byte_spin_lock; - pv_lock_ops.spin_trylock = __byte_spin_trylock; - pv_lock_ops.spin_unlock = __byte_spin_unlock; -#endif -} - struct pv_info pv_info = { .name = "bare hardware", .paravirt_enabled = 0, @@ -349,6 +338,10 @@ struct pv_cpu_ops pv_cpu_ops = { .write_ldt_entry = native_write_ldt_entry, .write_gdt_entry = native_write_gdt_entry, .write_idt_entry = native_write_idt_entry, + + .alloc_ldt = paravirt_nop, + .free_ldt = paravirt_nop, + .load_sp0 = native_load_sp0, #if defined(CONFIG_X86_32) || defined(CONFIG_IA32_EMULATION) @@ -460,18 +453,6 @@ struct pv_mmu_ops pv_mmu_ops = { .set_fixmap = native_set_fixmap, }; -struct pv_lock_ops pv_lock_ops = { -#ifdef CONFIG_SMP - .spin_is_locked = __ticket_spin_is_locked, - .spin_is_contended = __ticket_spin_is_contended, - - .spin_lock = __ticket_spin_lock, - .spin_trylock = __ticket_spin_trylock, - .spin_unlock = __ticket_spin_unlock, -#endif -}; -EXPORT_SYMBOL(pv_lock_ops); - EXPORT_SYMBOL_GPL(pv_time_ops); EXPORT_SYMBOL (pv_cpu_ops); EXPORT_SYMBOL (pv_mmu_ops); diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 205188d..0a1302f 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -76,47 +76,12 @@ unsigned long thread_saved_pc(struct task_struct *tsk) return ((unsigned long *)tsk->thread.sp)[3]; } -#ifdef CONFIG_HOTPLUG_CPU -#include <asm/nmi.h> - -static void cpu_exit_clear(void) -{ - int cpu = raw_smp_processor_id(); - - idle_task_exit(); - - cpu_uninit(); - irq_ctx_exit(cpu); - - cpu_clear(cpu, cpu_callout_map); - cpu_clear(cpu, cpu_callin_map); - - numa_remove_cpu(cpu); - c1e_remove_cpu(cpu); -} - -/* We don't actually take CPU down, just spin without interrupts. */ -static inline void play_dead(void) -{ - /* This must be done before dead CPU ack */ - cpu_exit_clear(); - mb(); - /* Ack it */ - __get_cpu_var(cpu_state) = CPU_DEAD; - - /* - * With physical CPU hotplug, we should halt the cpu - */ - local_irq_disable(); - /* mask all interrupts, flush any and all caches, and halt */ - wbinvd_halt(); -} -#else +#ifndef CONFIG_SMP static inline void play_dead(void) { BUG(); } -#endif /* CONFIG_HOTPLUG_CPU */ +#endif /* * The idle thread. There's no useful work to be @@ -158,7 +123,7 @@ void cpu_idle(void) } } -void __show_registers(struct pt_regs *regs, int all) +void __show_regs(struct pt_regs *regs, int all) { unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L; unsigned long d0, d1, d2, d3, d6, d7; @@ -224,7 +189,7 @@ void __show_registers(struct pt_regs *regs, int all) void show_regs(struct pt_regs *regs) { - __show_registers(regs, 1); + __show_regs(regs, 1); show_trace(NULL, regs, ®s->sp, regs->bp); } diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 2a8ccb9..410cc5f 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -86,30 +86,12 @@ void exit_idle(void) __exit_idle(); } -#ifdef CONFIG_HOTPLUG_CPU -DECLARE_PER_CPU(int, cpu_state); - -#include <linux/nmi.h> -/* We halt the CPU with physical CPU hotplug */ -static inline void play_dead(void) -{ - idle_task_exit(); - c1e_remove_cpu(raw_smp_processor_id()); - - mb(); - /* Ack it */ - __get_cpu_var(cpu_state) = CPU_DEAD; - - local_irq_disable(); - /* mask all interrupts, flush any and all caches, and halt */ - wbinvd_halt(); -} -#else +#ifndef CONFIG_SMP static inline void play_dead(void) { BUG(); } -#endif /* CONFIG_HOTPLUG_CPU */ +#endif /* * The idle thread. There's no useful work to be @@ -154,7 +136,7 @@ void cpu_idle(void) } /* Prints also some state that isn't saved in the pt_regs */ -void __show_regs(struct pt_regs *regs) +void __show_regs(struct pt_regs *regs, int all) { unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L, fs, gs, shadowgs; unsigned long d0, d1, d2, d3, d6, d7; @@ -193,13 +175,20 @@ void __show_regs(struct pt_regs *regs) rdmsrl(MSR_GS_BASE, gs); rdmsrl(MSR_KERNEL_GS_BASE, shadowgs); + if (!all) + return; + + printk(KERN_INFO "FS: %016lx(%04x) GS:%016lx(%04x) knlGS:%016lx\n", + fs, fsindex, gs, gsindex, shadowgs); + + if (!all) + return; + cr0 = read_cr0(); cr2 = read_cr2(); cr3 = read_cr3(); cr4 = read_cr4(); - printk(KERN_INFO "FS: %016lx(%04x) GS:%016lx(%04x) knlGS:%016lx\n", - fs, fsindex, gs, gsindex, shadowgs); printk(KERN_INFO "CS: %04x DS: %04x ES: %04x CR0: %016lx\n", cs, ds, es, cr0); printk(KERN_INFO "CR2: %016lx CR3: %016lx CR4: %016lx\n", cr2, cr3, @@ -218,7 +207,7 @@ void __show_regs(struct pt_regs *regs) void show_regs(struct pt_regs *regs) { printk(KERN_INFO "CPU %d:", smp_processor_id()); - __show_regs(regs); + __show_regs(regs, 1); show_trace(NULL, regs, (void *)(regs + 1), regs->bp); } diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 46c98ef..09ee73a 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -302,7 +302,7 @@ static void __init relocate_initrd(void) if (clen > MAX_MAP_CHUNK-slop) clen = MAX_MAP_CHUNK-slop; mapaddr = ramdisk_image & PAGE_MASK; - p = early_ioremap(mapaddr, clen+slop); + p = early_memremap(mapaddr, clen+slop); memcpy(q, p+slop, clen); early_iounmap(p, clen+slop); q += clen; @@ -379,7 +379,7 @@ static void __init parse_setup_data(void) return; pa_data = boot_params.hdr.setup_data; while (pa_data) { - data = early_ioremap(pa_data, PAGE_SIZE); + data = early_memremap(pa_data, PAGE_SIZE); switch (data->type) { case SETUP_E820_EXT: parse_e820_ext(data, pa_data); @@ -402,7 +402,7 @@ static void __init e820_reserve_setup_data(void) return; pa_data = boot_params.hdr.setup_data; while (pa_data) { - data = early_ioremap(pa_data, sizeof(*data)); + data = early_memremap(pa_data, sizeof(*data)); e820_update_range(pa_data, sizeof(*data)+data->len, E820_RAM, E820_RESERVED_KERN); found = 1; @@ -428,7 +428,7 @@ static void __init reserve_early_setup_data(void) return; pa_data = boot_params.hdr.setup_data; while (pa_data) { - data = early_ioremap(pa_data, sizeof(*data)); + data = early_memremap(pa_data, sizeof(*data)); sprintf(buf, "setup data %x", data->type); reserve_early(pa_data, pa_data+sizeof(*data)+data->len, buf); pa_data = data->next; diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c index 361b7a4..18f9b19 100644 --- a/arch/x86/kernel/smp.c +++ b/arch/x86/kernel/smp.c @@ -214,12 +214,16 @@ void smp_call_function_single_interrupt(struct pt_regs *regs) struct smp_ops smp_ops = { .smp_prepare_boot_cpu = native_smp_prepare_boot_cpu, .smp_prepare_cpus = native_smp_prepare_cpus, - .cpu_up = native_cpu_up, .smp_cpus_done = native_smp_cpus_done, .smp_send_stop = native_smp_send_stop, .smp_send_reschedule = native_smp_send_reschedule, + .cpu_up = native_cpu_up, + .cpu_die = native_cpu_die, + .cpu_disable = native_cpu_disable, + .play_dead = native_play_dead, + .send_call_func_ipi = native_send_call_func_ipi, .send_call_func_single_ipi = native_send_call_func_single_ipi, }; diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 2ff0bbc..d6a4d95 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -52,6 +52,7 @@ #include <asm/desc.h> #include <asm/nmi.h> #include <asm/irq.h> +#include <asm/idle.h> #include <asm/smp.h> #include <asm/trampoline.h> #include <asm/cpu.h> @@ -332,14 +333,17 @@ static void __cpuinit start_secondary(void *unused) * does not change while we are assigning vectors to cpus. Holding * this lock ensures we don't half assign or remove an irq from a cpu. */ - ipi_call_lock_irq(); + ipi_call_lock(); lock_vector_lock(); __setup_vector_irq(smp_processor_id()); cpu_set(smp_processor_id(), cpu_online_map); unlock_vector_lock(); - ipi_call_unlock_irq(); + ipi_call_unlock(); per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE; + /* enable local interrupts */ + local_irq_enable(); + setup_secondary_clock(); wmb(); @@ -594,10 +598,12 @@ wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip) * Give the other CPU some time to accept the IPI. */ udelay(200); - maxlvt = lapic_get_maxlvt(); - if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ - apic_write(APIC_ESR, 0); - accept_status = (apic_read(APIC_ESR) & 0xEF); + if (APIC_INTEGRATED(apic_version[phys_apicid])) { + maxlvt = lapic_get_maxlvt(); + if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ + apic_write(APIC_ESR, 0); + accept_status = (apic_read(APIC_ESR) & 0xEF); + } pr_debug("NMI sent.\n"); if (send_status) @@ -1254,39 +1260,6 @@ void __init native_smp_cpus_done(unsigned int max_cpus) check_nmi_watchdog(); } -#ifdef CONFIG_HOTPLUG_CPU - -static void remove_siblinginfo(int cpu) -{ - int sibling; - struct cpuinfo_x86 *c = &cpu_data(cpu); - - for_each_cpu_mask_nr(sibling, per_cpu(cpu_core_map, cpu)) { - cpu_clear(cpu, per_cpu(cpu_core_map, sibling)); - /*/ - * last thread sibling in this cpu core going down - */ - if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1) - cpu_data(sibling).booted_cores--; - } - - for_each_cpu_mask_nr(sibling, per_cpu(cpu_sibling_map, cpu)) - cpu_clear(cpu, per_cpu(cpu_sibling_map, sibling)); - cpus_clear(per_cpu(cpu_sibling_map, cpu)); - cpus_clear(per_cpu(cpu_core_map, cpu)); - c->phys_proc_id = 0; - c->cpu_core_id = 0; - cpu_clear(cpu, cpu_sibling_setup_map); -} - -static int additional_cpus __initdata = -1; - -static __init int setup_additional_cpus(char *s) -{ - return s && get_option(&s, &additional_cpus) ? 0 : -EINVAL; -} -early_param("additional_cpus", setup_additional_cpus); - /* * cpu_possible_map should be static, it cannot change as cpu's * are onlined, or offlined. The reason is per-cpu data-structures @@ -1306,21 +1279,13 @@ early_param("additional_cpus", setup_additional_cpus); */ __init void prefill_possible_map(void) { - int i; - int possible; + int i, possible; /* no processor from mptable or madt */ if (!num_processors) num_processors = 1; - if (additional_cpus == -1) { - if (disabled_cpus > 0) - additional_cpus = disabled_cpus; - else - additional_cpus = 0; - } - - possible = num_processors + additional_cpus; + possible = num_processors + disabled_cpus; if (possible > NR_CPUS) possible = NR_CPUS; @@ -1333,6 +1298,31 @@ __init void prefill_possible_map(void) nr_cpu_ids = possible; } +#ifdef CONFIG_HOTPLUG_CPU + +static void remove_siblinginfo(int cpu) +{ + int sibling; + struct cpuinfo_x86 *c = &cpu_data(cpu); + + for_each_cpu_mask_nr(sibling, per_cpu(cpu_core_map, cpu)) { + cpu_clear(cpu, per_cpu(cpu_core_map, sibling)); + /*/ + * last thread sibling in this cpu core going down + */ + if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1) + cpu_data(sibling).booted_cores--; + } + + for_each_cpu_mask_nr(sibling, per_cpu(cpu_sibling_map, cpu)) + cpu_clear(cpu, per_cpu(cpu_sibling_map, sibling)); + cpus_clear(per_cpu(cpu_sibling_map, cpu)); + cpus_clear(per_cpu(cpu_core_map, cpu)); + c->phys_proc_id = 0; + c->cpu_core_id = 0; + cpu_clear(cpu, cpu_sibling_setup_map); +} + static void __ref remove_cpu_from_maps(int cpu) { cpu_clear(cpu, cpu_online_map); @@ -1343,25 +1333,9 @@ static void __ref remove_cpu_from_maps(int cpu) numa_remove_cpu(cpu); } -int __cpu_disable(void) +void cpu_disable_common(void) { int cpu = smp_processor_id(); - - /* - * Perhaps use cpufreq to drop frequency, but that could go - * into generic code. - * - * We won't take down the boot processor on i386 due to some - * interrupts only being able to be serviced by the BSP. - * Especially so if we're not using an IOAPIC -zwane - */ - if (cpu == 0) - return -EBUSY; - - if (nmi_watchdog == NMI_LOCAL_APIC) - stop_apic_nmi_watchdog(NULL); - clear_local_APIC(); - /* * HACK: * Allow any queued timer interrupts to get serviced @@ -1379,10 +1353,32 @@ int __cpu_disable(void) remove_cpu_from_maps(cpu); unlock_vector_lock(); fixup_irqs(cpu_online_map); +} + +int native_cpu_disable(void) +{ + int cpu = smp_processor_id(); + + /* + * Perhaps use cpufreq to drop frequency, but that could go + * into generic code. + * + * We won't take down the boot processor on i386 due to some + * interrupts only being able to be serviced by the BSP. + * Especially so if we're not using an IOAPIC -zwane + */ + if (cpu == 0) + return -EBUSY; + + if (nmi_watchdog == NMI_LOCAL_APIC) + stop_apic_nmi_watchdog(NULL); + clear_local_APIC(); + + cpu_disable_common(); return 0; } -void __cpu_die(unsigned int cpu) +void native_cpu_die(unsigned int cpu) { /* We don't do anything here: idle task is faking death itself. */ unsigned int i; @@ -1399,15 +1395,45 @@ void __cpu_die(unsigned int cpu) } printk(KERN_ERR "CPU %u didn't die...\n", cpu); } + +void play_dead_common(void) +{ + idle_task_exit(); + reset_lazy_tlbstate(); + irq_ctx_exit(raw_smp_processor_id()); + c1e_remove_cpu(raw_smp_processor_id()); + + mb(); + /* Ack it */ + __get_cpu_var(cpu_state) = CPU_DEAD; + + /* + * With physical CPU hotplug, we should halt the cpu + */ + local_irq_disable(); +} + +void native_play_dead(void) +{ + play_dead_common(); + wbinvd_halt(); +} + #else /* ... !CONFIG_HOTPLUG_CPU */ -int __cpu_disable(void) +int native_cpu_disable(void) { return -ENOSYS; } -void __cpu_die(unsigned int cpu) +void native_cpu_die(unsigned int cpu) { /* We said "no" in __cpu_disable */ BUG(); } + +void native_play_dead(void) +{ + BUG(); +} + #endif diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c index fec1ece..e00534b 100644 --- a/arch/x86/kernel/tlb_32.c +++ b/arch/x86/kernel/tlb_32.c @@ -241,3 +241,11 @@ void flush_tlb_all(void) on_each_cpu(do_flush_tlb_all, NULL, 1); } +void reset_lazy_tlbstate(void) +{ + int cpu = raw_smp_processor_id(); + + per_cpu(cpu_tlbstate, cpu).state = 0; + per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm; +} + diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps.c similarity index 60% rename from arch/x86/kernel/traps_32.c rename to arch/x86/kernel/traps.c index 0429c5d..ffb131f 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps.c @@ -7,13 +7,11 @@ */ /* - * 'Traps.c' handles hardware traps and faults after we have saved some - * state in 'asm.s'. + * Handle hardware traps and faults. */ #include <linux/interrupt.h> #include <linux/kallsyms.h> #include <linux/spinlock.h> -#include <linux/highmem.h> #include <linux/kprobes.h> #include <linux/uaccess.h> #include <linux/utsname.h> @@ -32,6 +30,8 @@ #include <linux/bug.h> #include <linux/nmi.h> #include <linux/mm.h> +#include <linux/smp.h> +#include <linux/io.h> #ifdef CONFIG_EISA #include <linux/ioport.h> @@ -46,21 +46,31 @@ #include <linux/edac.h> #endif -#include <asm/arch_hooks.h> #include <asm/stacktrace.h> #include <asm/processor.h> #include <asm/debugreg.h> #include <asm/atomic.h> #include <asm/system.h> #include <asm/unwind.h> +#include <asm/traps.h> #include <asm/desc.h> #include <asm/i387.h> + +#include <mach_traps.h> + +#ifdef CONFIG_X86_64 +#include <asm/pgalloc.h> +#include <asm/proto.h> +#include <asm/pda.h> +#else +#include <asm/processor-flags.h> +#include <asm/arch_hooks.h> #include <asm/nmi.h> #include <asm/smp.h> #include <asm/io.h> #include <asm/traps.h> -#include "mach_traps.h" +#include "cpu/mcheck/mce.h" DECLARE_BITMAP(used_vectors, NR_VECTORS); EXPORT_SYMBOL_GPL(used_vectors); @@ -77,418 +87,104 @@ char ignore_fpu_irq; */ gate_desc idt_table[256] __attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, }; - -int panic_on_unrecovered_nmi; -int kstack_depth_to_print = 24; -static unsigned int code_bytes = 64; -static int ignore_nmis; -static int die_counter; - -void printk_address(unsigned long address, int reliable) -{ -#ifdef CONFIG_KALLSYMS - unsigned long offset = 0; - unsigned long symsize; - const char *symname; - char *modname; - char *delim = ":"; - char namebuf[KSYM_NAME_LEN]; - char reliab[4] = ""; - - symname = kallsyms_lookup(address, &symsize, &offset, - &modname, namebuf); - if (!symname) { - printk(" [<%08lx>]\n", address); - return; - } - if (!reliable) - strcpy(reliab, "? "); - - if (!modname) - modname = delim = ""; - printk(" [<%08lx>] %s%s%s%s%s+0x%lx/0x%lx\n", - address, reliab, delim, modname, delim, symname, offset, symsize); -#else - printk(" [<%08lx>]\n", address); -#endif -} - -static inline int valid_stack_ptr(struct thread_info *tinfo, - void *p, unsigned int size) -{ - void *t = tinfo; - return p > t && p <= t + THREAD_SIZE - size; -} - -/* The form of the top of the frame on the stack */ -struct stack_frame { - struct stack_frame *next_frame; - unsigned long return_address; -}; - -static inline unsigned long -print_context_stack(struct thread_info *tinfo, - unsigned long *stack, unsigned long bp, - const struct stacktrace_ops *ops, void *data) -{ - struct stack_frame *frame = (struct stack_frame *)bp; - - while (valid_stack_ptr(tinfo, stack, sizeof(*stack))) { - unsigned long addr; - - addr = *stack; - if (__kernel_text_address(addr)) { - if ((unsigned long) stack == bp + 4) { - ops->address(data, addr, 1); - frame = frame->next_frame; - bp = (unsigned long) frame; - } else { - ops->address(data, addr, bp == 0); - } - } - stack++; - } - return bp; -} - -void dump_trace(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, - const struct stacktrace_ops *ops, void *data) -{ - if (!task) - task = current; - - if (!stack) { - unsigned long dummy; - stack = &dummy; - if (task != current) - stack = (unsigned long *)task->thread.sp; - } - -#ifdef CONFIG_FRAME_POINTER - if (!bp) { - if (task == current) { - /* Grab bp right from our regs */ - asm("movl %%ebp, %0" : "=r" (bp) :); - } else { - /* bp is the last reg pushed by switch_to */ - bp = *(unsigned long *) task->thread.sp; - } - } #endif - for (;;) { - struct thread_info *context; - - context = (struct thread_info *) - ((unsigned long)stack & (~(THREAD_SIZE - 1))); - bp = print_context_stack(context, stack, bp, ops, data); - /* - * Should be after the line below, but somewhere - * in early boot context comes out corrupted and we - * can't reference it: - */ - if (ops->stack(data, "IRQ") < 0) - break; - stack = (unsigned long *)context->previous_esp; - if (!stack) - break; - touch_nmi_watchdog(); - } -} -EXPORT_SYMBOL(dump_trace); - -static void -print_trace_warning_symbol(void *data, char *msg, unsigned long symbol) -{ - printk(data); - print_symbol(msg, symbol); - printk("\n"); -} - -static void print_trace_warning(void *data, char *msg) -{ - printk("%s%s\n", (char *)data, msg); -} - -static int print_trace_stack(void *data, char *name) -{ - return 0; -} - -/* - * Print one address/symbol entries per line. - */ -static void print_trace_address(void *data, unsigned long addr, int reliable) -{ - printk("%s [<%08lx>] ", (char *)data, addr); - if (!reliable) - printk("? "); - print_symbol("%s\n", addr); - touch_nmi_watchdog(); -} - -static const struct stacktrace_ops print_trace_ops = { - .warning = print_trace_warning, - .warning_symbol = print_trace_warning_symbol, - .stack = print_trace_stack, - .address = print_trace_address, -}; +static int ignore_nmis; -static void -show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, char *log_lvl) +static inline void conditional_sti(struct pt_regs *regs) { - dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl); - printk("%s =======================\n", log_lvl); + if (regs->flags & X86_EFLAGS_IF) + local_irq_enable(); } -void show_trace(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp) +static inline void preempt_conditional_sti(struct pt_regs *regs) { - show_trace_log_lvl(task, regs, stack, bp, ""); + inc_preempt_count(); + if (regs->flags & X86_EFLAGS_IF) + local_irq_enable(); } -static void -show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *sp, unsigned long bp, char *log_lvl) +static inline void preempt_conditional_cli(struct pt_regs *regs) { - unsigned long *stack; - int i; - - if (sp == NULL) { - if (task) - sp = (unsigned long *)task->thread.sp; - else - sp = (unsigned long *)&sp; - } - - stack = sp; - for (i = 0; i < kstack_depth_to_print; i++) { - if (kstack_end(stack)) - break; - if (i && ((i % 8) == 0)) - printk("\n%s ", log_lvl); - printk("%08lx ", *stack++); - } - printk("\n%sCall Trace:\n", log_lvl); - - show_trace_log_lvl(task, regs, sp, bp, log_lvl); + if (regs->flags & X86_EFLAGS_IF) + local_irq_disable(); + dec_preempt_count(); } -void show_stack(struct task_struct *task, unsigned long *sp) +#ifdef CONFIG_X86_32 +static inline void +die_if_kernel(const char *str, struct pt_regs *regs, long err) { - printk(" "); - show_stack_log_lvl(task, NULL, sp, 0, ""); + if (!user_mode_vm(regs)) + die(str, regs, err); } /* - * The architecture-independent dump_stack generator + * Perform the lazy TSS's I/O bitmap copy. If the TSS has an + * invalid offset set (the LAZY one) and the faulting thread has + * a valid I/O bitmap pointer, we copy the I/O bitmap in the TSS, + * we set the offset field correctly and return 1. */ -void dump_stack(void) +static int lazy_iobitmap_copy(void) { - unsigned long bp = 0; - unsigned long stack; - -#ifdef CONFIG_FRAME_POINTER - if (!bp) - asm("movl %%ebp, %0" : "=r" (bp):); -#endif - - printk("Pid: %d, comm: %.20s %s %s %.*s\n", - current->pid, current->comm, print_tainted(), - init_utsname()->release, - (int)strcspn(init_utsname()->version, " "), - init_utsname()->version); - - show_trace(current, NULL, &stack, bp); -} - -EXPORT_SYMBOL(dump_stack); - -void show_registers(struct pt_regs *regs) -{ - int i; - - print_modules(); - __show_registers(regs, 0); - - printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)", - TASK_COMM_LEN, current->comm, task_pid_nr(current), - current_thread_info(), current, task_thread_info(current)); - /* - * When in-kernel, we also print out the stack and code at the - * time of the fault.. - */ - if (!user_mode_vm(regs)) { - unsigned int code_prologue = code_bytes * 43 / 64; - unsigned int code_len = code_bytes; - unsigned char c; - u8 *ip; - - printk("\n" KERN_EMERG "Stack: "); - show_stack_log_lvl(NULL, regs, ®s->sp, 0, KERN_EMERG); + struct thread_struct *thread; + struct tss_struct *tss; + int cpu; - printk(KERN_EMERG "Code: "); + cpu = get_cpu(); + tss = &per_cpu(init_tss, cpu); + thread = ¤t->thread; - ip = (u8 *)regs->ip - code_prologue; - if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) { - /* try starting at EIP */ - ip = (u8 *)regs->ip; - code_len = code_len - code_prologue + 1; - } - for (i = 0; i < code_len; i++, ip++) { - if (ip < (u8 *)PAGE_OFFSET || - probe_kernel_address(ip, c)) { - printk(" Bad EIP value."); - break; - } - if (ip == (u8 *)regs->ip) - printk("<%02x> ", c); - else - printk("%02x ", c); + if (tss->x86_tss.io_bitmap_base == INVALID_IO_BITMAP_OFFSET_LAZY && + thread->io_bitmap_ptr) { + memcpy(tss->io_bitmap, thread->io_bitmap_ptr, + thread->io_bitmap_max); + /* + * If the previously set map was extending to higher ports + * than the current one, pad extra space with 0xff (no access). + */ + if (thread->io_bitmap_max < tss->io_bitmap_max) { + memset((char *) tss->io_bitmap + + thread->io_bitmap_max, 0xff, + tss->io_bitmap_max - thread->io_bitmap_max); } - } - printk("\n"); -} - -int is_valid_bugaddr(unsigned long ip) -{ - unsigned short ud2; - - if (ip < PAGE_OFFSET) - return 0; - if (probe_kernel_address((unsigned short *)ip, ud2)) - return 0; - - return ud2 == 0x0b0f; -} - -static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; -static int die_owner = -1; -static unsigned int die_nest_count; - -unsigned __kprobes long oops_begin(void) -{ - unsigned long flags; - - oops_enter(); - - if (die_owner != raw_smp_processor_id()) { - console_verbose(); - raw_local_irq_save(flags); - __raw_spin_lock(&die_lock); - die_owner = smp_processor_id(); - die_nest_count = 0; - bust_spinlocks(1); - } else { - raw_local_irq_save(flags); - } - die_nest_count++; - return flags; -} - -void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) -{ - bust_spinlocks(0); - die_owner = -1; - add_taint(TAINT_DIE); - __raw_spin_unlock(&die_lock); - raw_local_irq_restore(flags); - - if (!regs) - return; - - if (kexec_should_crash(current)) - crash_kexec(regs); - - if (in_interrupt()) - panic("Fatal exception in interrupt"); - - if (panic_on_oops) - panic("Fatal exception"); - - oops_exit(); - do_exit(signr); -} - -int __kprobes __die(const char *str, struct pt_regs *regs, long err) -{ - unsigned short ss; - unsigned long sp; + tss->io_bitmap_max = thread->io_bitmap_max; + tss->x86_tss.io_bitmap_base = IO_BITMAP_OFFSET; + tss->io_bitmap_owner = thread; + put_cpu(); - printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter); -#ifdef CONFIG_PREEMPT - printk("PREEMPT "); -#endif -#ifdef CONFIG_SMP - printk("SMP "); -#endif -#ifdef CONFIG_DEBUG_PAGEALLOC - printk("DEBUG_PAGEALLOC"); -#endif - printk("\n"); - if (notify_die(DIE_OOPS, str, regs, err, - current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) return 1; - - show_registers(regs); - /* Executive summary in case the oops scrolled away */ - sp = (unsigned long) (®s->sp); - savesegment(ss, ss); - if (user_mode(regs)) { - sp = regs->sp; - ss = regs->ss & 0xffff; } - printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip); - print_symbol("%s", regs->ip); - printk(" SS:ESP %04x:%08lx\n", ss, sp); - return 0; -} - -/* - * This is gone through when something in the kernel has done something bad - * and is about to be terminated: - */ -void die(const char *str, struct pt_regs *regs, long err) -{ - unsigned long flags = oops_begin(); - - if (die_nest_count < 3) { - report_bug(regs->ip, regs); - - if (__die(str, regs, err)) - regs = NULL; - } else { - printk(KERN_EMERG "Recursive die() failure, output suppressed\n"); - } - - oops_end(flags, regs, SIGSEGV); -} + put_cpu(); -static inline void -die_if_kernel(const char *str, struct pt_regs *regs, long err) -{ - if (!user_mode_vm(regs)) - die(str, regs, err); + return 0; } +#endif static void __kprobes -do_trap(int trapnr, int signr, char *str, int vm86, struct pt_regs *regs, +do_trap(int trapnr, int signr, char *str, struct pt_regs *regs, long error_code, siginfo_t *info) { struct task_struct *tsk = current; +#ifdef CONFIG_X86_32 if (regs->flags & X86_VM_MASK) { - if (vm86) + /* + * traps 0, 1, 3, 4, and 5 should be forwarded to vm86. + * On nmi (interrupt 2), do_trap should not be called. + */ + if (trapnr < 6) goto vm86_trap; goto trap_signal; } +#endif if (!user_mode(regs)) goto kernel_trap; +#ifdef CONFIG_X86_32 trap_signal: +#endif /* * We want error_code and trap_no set for userspace faults and * kernelspace faults which result in die(), but not @@ -501,6 +197,18 @@ trap_signal: tsk->thread.error_code = error_code; tsk->thread.trap_no = trapnr; +#ifdef CONFIG_X86_64 + if (show_unhandled_signals && unhandled_signal(tsk, signr) && + printk_ratelimit()) { + printk(KERN_INFO + "%s[%d] trap %s ip:%lx sp:%lx error:%lx", + tsk->comm, tsk->pid, str, + regs->ip, regs->sp, error_code); + print_vma_addr(" in ", regs->ip); + printk("\n"); + } +#endif + if (info) force_sig_info(signr, info, tsk); else @@ -515,120 +223,98 @@ kernel_trap: } return; +#ifdef CONFIG_X86_32 vm86_trap: if (handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, trapnr)) goto trap_signal; return; +#endif } #define DO_ERROR(trapnr, signr, str, name) \ -void do_##name(struct pt_regs *regs, long error_code) \ -{ \ - trace_hardirqs_fixup(); \ - if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ - == NOTIFY_STOP) \ - return; \ - do_trap(trapnr, signr, str, 0, regs, error_code, NULL); \ -} - -#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr, irq) \ -void do_##name(struct pt_regs *regs, long error_code) \ -{ \ - siginfo_t info; \ - if (irq) \ - local_irq_enable(); \ - info.si_signo = signr; \ - info.si_errno = 0; \ - info.si_code = sicode; \ - info.si_addr = (void __user *)siaddr; \ - if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ - == NOTIFY_STOP) \ - return; \ - do_trap(trapnr, signr, str, 0, regs, error_code, &info); \ -} - -#define DO_VM86_ERROR(trapnr, signr, str, name) \ -void do_##name(struct pt_regs *regs, long error_code) \ +dotraplinkage void do_##name(struct pt_regs *regs, long error_code) \ { \ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ == NOTIFY_STOP) \ return; \ - do_trap(trapnr, signr, str, 1, regs, error_code, NULL); \ + conditional_sti(regs); \ + do_trap(trapnr, signr, str, regs, error_code, NULL); \ } -#define DO_VM86_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ -void do_##name(struct pt_regs *regs, long error_code) \ +#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ +dotraplinkage void do_##name(struct pt_regs *regs, long error_code) \ { \ siginfo_t info; \ info.si_signo = signr; \ info.si_errno = 0; \ info.si_code = sicode; \ info.si_addr = (void __user *)siaddr; \ - trace_hardirqs_fixup(); \ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ == NOTIFY_STOP) \ return; \ - do_trap(trapnr, signr, str, 1, regs, error_code, &info); \ + conditional_sti(regs); \ + do_trap(trapnr, signr, str, regs, error_code, &info); \ } -DO_VM86_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip) -#ifndef CONFIG_KPROBES -DO_VM86_ERROR(3, SIGTRAP, "int3", int3) -#endif -DO_VM86_ERROR(4, SIGSEGV, "overflow", overflow) -DO_VM86_ERROR(5, SIGSEGV, "bounds", bounds) -DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip, 0) +DO_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip) +DO_ERROR(4, SIGSEGV, "overflow", overflow) +DO_ERROR(5, SIGSEGV, "bounds", bounds) +DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip) DO_ERROR(9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun) DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS) DO_ERROR(11, SIGBUS, "segment not present", segment_not_present) +#ifdef CONFIG_X86_32 DO_ERROR(12, SIGBUS, "stack segment", stack_segment) -DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0, 0) -DO_ERROR_INFO(32, SIGILL, "iret exception", iret_error, ILL_BADSTK, 0, 1) +#endif +DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0) + +#ifdef CONFIG_X86_64 +/* Runs on IST stack */ +dotraplinkage void do_stack_segment(struct pt_regs *regs, long error_code) +{ + if (notify_die(DIE_TRAP, "stack segment", regs, error_code, + 12, SIGBUS) == NOTIFY_STOP) + return; + preempt_conditional_sti(regs); + do_trap(12, SIGBUS, "stack segment", regs, error_code, NULL); + preempt_conditional_cli(regs); +} -void __kprobes +dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code) +{ + static const char str[] = "double fault"; + struct task_struct *tsk = current; + + /* Return not checked because double check cannot be ignored */ + notify_die(DIE_TRAP, str, regs, error_code, 8, SIGSEGV); + + tsk->thread.error_code = error_code; + tsk->thread.trap_no = 8; + + /* This is always a kernel trap and never fixable (and thus must + never return). */ + for (;;) + die(str, regs, error_code); +} +#endif + +dotraplinkage void __kprobes do_general_protection(struct pt_regs *regs, long error_code) { struct task_struct *tsk; - struct thread_struct *thread; - struct tss_struct *tss; - int cpu; - cpu = get_cpu(); - tss = &per_cpu(init_tss, cpu); - thread = ¤t->thread; - - /* - * Perform the lazy TSS's I/O bitmap copy. If the TSS has an - * invalid offset set (the LAZY one) and the faulting thread has - * a valid I/O bitmap pointer, we copy the I/O bitmap in the TSS - * and we set the offset field correctly. Then we let the CPU to - * restart the faulting instruction. - */ - if (tss->x86_tss.io_bitmap_base == INVALID_IO_BITMAP_OFFSET_LAZY && - thread->io_bitmap_ptr) { - memcpy(tss->io_bitmap, thread->io_bitmap_ptr, - thread->io_bitmap_max); - /* - * If the previously set map was extending to higher ports - * than the current one, pad extra space with 0xff (no access). - */ - if (thread->io_bitmap_max < tss->io_bitmap_max) { - memset((char *) tss->io_bitmap + - thread->io_bitmap_max, 0xff, - tss->io_bitmap_max - thread->io_bitmap_max); - } - tss->io_bitmap_max = thread->io_bitmap_max; - tss->x86_tss.io_bitmap_base = IO_BITMAP_OFFSET; - tss->io_bitmap_owner = thread; - put_cpu(); + conditional_sti(regs); +#ifdef CONFIG_X86_32 + if (lazy_iobitmap_copy()) { + /* restart the faulting instruction */ return; } - put_cpu(); if (regs->flags & X86_VM_MASK) goto gp_in_vm86; +#endif tsk = current; if (!user_mode(regs)) @@ -650,10 +336,12 @@ do_general_protection(struct pt_regs *regs, long error_code) force_sig(SIGSEGV, tsk); return; +#ifdef CONFIG_X86_32 gp_in_vm86: local_irq_enable(); handle_vm86_fault((struct kernel_vm86_regs *) regs, error_code); return; +#endif gp_in_kernel: if (fixup_exception(regs)) @@ -690,7 +378,8 @@ mem_parity_error(unsigned char reason, struct pt_regs *regs) printk(KERN_EMERG "Dazed and confused, but trying to continue\n"); /* Clear and disable the memory parity error line. */ - clear_mem_error(reason); + reason = (reason & 0xf) | 4; + outb(reason, 0x61); } static notrace __kprobes void @@ -716,7 +405,8 @@ io_check_error(unsigned char reason, struct pt_regs *regs) static notrace __kprobes void unknown_nmi_error(unsigned char reason, struct pt_regs *regs) { - if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) + if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == + NOTIFY_STOP) return; #ifdef CONFIG_MCA /* @@ -739,6 +429,7 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs) printk(KERN_EMERG "Dazed and confused, but trying to continue\n"); } +#ifdef CONFIG_X86_32 static DEFINE_SPINLOCK(nmi_print_lock); void notrace __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic) @@ -773,6 +464,7 @@ void notrace __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic) do_exit(SIGSEGV); } +#endif static notrace __kprobes void default_do_nmi(struct pt_regs *regs) { @@ -812,22 +504,25 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) mem_parity_error(reason, regs); if (reason & 0x40) io_check_error(reason, regs); +#ifdef CONFIG_X86_32 /* * Reassert NMI in case it became active meanwhile * as it's edge-triggered: */ reassert_nmi(); +#endif } -notrace __kprobes void do_nmi(struct pt_regs *regs, long error_code) +dotraplinkage notrace __kprobes void +do_nmi(struct pt_regs *regs, long error_code) { - int cpu; - nmi_enter(); - cpu = smp_processor_id(); - - ++nmi_count(cpu); +#ifdef CONFIG_X86_32 + { int cpu; cpu = smp_processor_id(); ++nmi_count(cpu); } +#else + add_pda(__nmi_count, 1); +#endif if (!ignore_nmis) default_do_nmi(regs); @@ -847,21 +542,44 @@ void restart_nmi(void) acpi_nmi_enable(); } -#ifdef CONFIG_KPROBES -void __kprobes do_int3(struct pt_regs *regs, long error_code) +/* May run on IST stack. */ +dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) { - trace_hardirqs_fixup(); - +#ifdef CONFIG_KPROBES if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) == NOTIFY_STOP) return; - /* - * This is an interrupt gate, because kprobes wants interrupts - * disabled. Normal trap handlers don't. - */ - restore_interrupts(regs); +#else + if (notify_die(DIE_TRAP, "int3", regs, error_code, 3, SIGTRAP) + == NOTIFY_STOP) + return; +#endif + + preempt_conditional_sti(regs); + do_trap(3, SIGTRAP, "int3", regs, error_code, NULL); + preempt_conditional_cli(regs); +} - do_trap(3, SIGTRAP, "int3", 1, regs, error_code, NULL); +#ifdef CONFIG_X86_64 +/* Help handler running on IST stack to switch back to user stack + for scheduling or signal handling. The actual stack switch is done in + entry.S */ +asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs) +{ + struct pt_regs *regs = eregs; + /* Did already sync */ + if (eregs == (struct pt_regs *)eregs->sp) + ; + /* Exception from user space */ + else if (user_mode(eregs)) + regs = task_pt_regs(current); + /* Exception from kernel and interrupts are enabled. Move to + kernel process stack. */ + else if (eregs->flags & X86_EFLAGS_IF) + regs = (struct pt_regs *)(eregs->sp -= sizeof(struct pt_regs)); + if (eregs != regs) + *regs = *eregs; + return regs; } #endif @@ -886,15 +604,15 @@ void __kprobes do_int3(struct pt_regs *regs, long error_code) * about restoring all the debug state, and ptrace doesn't have to * find every occurrence of the TF bit that could be saved away even * by user code) + * + * May run on IST stack. */ -void __kprobes do_debug(struct pt_regs *regs, long error_code) +dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code) { struct task_struct *tsk = current; - unsigned int condition; + unsigned long condition; int si_code; - trace_hardirqs_fixup(); - get_debugreg(condition, 6); /* @@ -906,9 +624,9 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code) if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code, SIGTRAP) == NOTIFY_STOP) return; + /* It's safe to allow irq's after DR6 has been saved */ - if (regs->flags & X86_EFLAGS_IF) - local_irq_enable(); + preempt_conditional_sti(regs); /* Mask out spurious debug traps due to lazy DR7 setting */ if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) { @@ -916,8 +634,10 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code) goto clear_dr7; } +#ifdef CONFIG_X86_32 if (regs->flags & X86_VM_MASK) goto debug_vm86; +#endif /* Save debug status register where ptrace can see it */ tsk->thread.debugreg6 = condition; @@ -927,16 +647,11 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code) * kernel space (but re-enable TF when returning to user mode). */ if (condition & DR_STEP) { - /* - * We already checked v86 mode above, so we can - * check for kernel mode by just checking the CPL - * of CS. - */ if (!user_mode(regs)) goto clear_TF_reenable; } - si_code = get_si_code((unsigned long)condition); + si_code = get_si_code(condition); /* Ok, finally something we can handle */ send_sigtrap(tsk, regs, error_code, si_code); @@ -946,18 +661,37 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code) */ clear_dr7: set_debugreg(0, 7); + preempt_conditional_cli(regs); return; +#ifdef CONFIG_X86_32 debug_vm86: handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1); + preempt_conditional_cli(regs); return; +#endif clear_TF_reenable: set_tsk_thread_flag(tsk, TIF_SINGLESTEP); regs->flags &= ~X86_EFLAGS_TF; + preempt_conditional_cli(regs); return; } +#ifdef CONFIG_X86_64 +static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr) +{ + if (fixup_exception(regs)) + return 1; + + notify_die(DIE_GPF, str, regs, 0, trapnr, SIGFPE); + /* Illegal floating point operation in the kernel */ + current->thread.trap_no = trapnr; + die(str, regs, 0); + return 0; +} +#endif + /* * Note that we play around with the 'TS' bit in an attempt to get * the correct behaviour even in the presence of the asynchronous @@ -994,7 +728,9 @@ void math_error(void __user *ip) swd = get_fpu_swd(task); switch (swd & ~cwd & 0x3f) { case 0x000: /* No unmasked exception */ +#ifdef CONFIG_X86_32 return; +#endif default: /* Multiple exceptions */ break; case 0x001: /* Invalid Op */ @@ -1022,9 +758,18 @@ void math_error(void __user *ip) force_sig_info(SIGFPE, &info, task); } -void do_coprocessor_error(struct pt_regs *regs, long error_code) +dotraplinkage void do_coprocessor_error(struct pt_regs *regs, long error_code) { + conditional_sti(regs); + +#ifdef CONFIG_X86_32 ignore_fpu_irq = 1; +#else + if (!user_mode(regs) && + kernel_math_error(regs, "kernel x87 math error", 16)) + return; +#endif + math_error((void __user *)regs->ip); } @@ -1076,8 +821,12 @@ static void simd_math_error(void __user *ip) force_sig_info(SIGFPE, &info, task); } -void do_simd_coprocessor_error(struct pt_regs *regs, long error_code) +dotraplinkage void +do_simd_coprocessor_error(struct pt_regs *regs, long error_code) { + conditional_sti(regs); + +#ifdef CONFIG_X86_32 if (cpu_has_xmm) { /* Handle SIMD FPU exceptions on PIII+ processors. */ ignore_fpu_irq = 1; @@ -1096,16 +845,25 @@ void do_simd_coprocessor_error(struct pt_regs *regs, long error_code) current->thread.error_code = error_code; die_if_kernel("cache flush denied", regs, error_code); force_sig(SIGSEGV, current); +#else + if (!user_mode(regs) && + kernel_math_error(regs, "kernel simd math error", 19)) + return; + simd_math_error((void __user *)regs->ip); +#endif } -void do_spurious_interrupt_bug(struct pt_regs *regs, long error_code) +dotraplinkage void +do_spurious_interrupt_bug(struct pt_regs *regs, long error_code) { + conditional_sti(regs); #if 0 /* No need to warn about this any longer. */ printk(KERN_INFO "Ignoring P6 Local APIC Spurious Interrupt Bug...\n"); #endif } +#ifdef CONFIG_X86_32 unsigned long patch_espfix_desc(unsigned long uesp, unsigned long kesp) { struct desc_struct *gdt = get_cpu_gdt_table(smp_processor_id()); @@ -1124,6 +882,15 @@ unsigned long patch_espfix_desc(unsigned long uesp, unsigned long kesp) return new_kesp; } +#else +asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void) +{ +} + +asmlinkage void __attribute__((weak)) mce_threshold_interrupt(void) +{ +} +#endif /* * 'math_state_restore()' saves the current math information in the @@ -1156,14 +923,24 @@ asmlinkage void math_state_restore(void) } clts(); /* Allow maths ops (or we recurse) */ +#ifdef CONFIG_X86_32 restore_fpu(tsk); +#else + /* + * Paranoid restore. send a SIGSEGV if we fail to restore the state. + */ + if (unlikely(restore_fpu_checking(tsk))) { + stts(); + force_sig(SIGSEGV, tsk); + return; + } +#endif thread->status |= TS_USEDFPU; /* So we fnsave on switch_to() */ tsk->fpu_counter++; } EXPORT_SYMBOL_GPL(math_state_restore); #ifndef CONFIG_MATH_EMULATION - asmlinkage void math_emulate(long arg) { printk(KERN_EMERG @@ -1172,12 +949,54 @@ asmlinkage void math_emulate(long arg) force_sig(SIGFPE, current); schedule(); } - #endif /* CONFIG_MATH_EMULATION */ +dotraplinkage void __kprobes +do_device_not_available(struct pt_regs *regs, long error) +{ +#ifdef CONFIG_X86_32 + if (read_cr0() & X86_CR0_EM) { + conditional_sti(regs); + math_emulate(0); + } else { + math_state_restore(); /* interrupts still off */ + conditional_sti(regs); + } +#else + math_state_restore(); +#endif +} + +#ifdef CONFIG_X86_32 +#ifdef CONFIG_X86_MCE +dotraplinkage void __kprobes do_machine_check(struct pt_regs *regs, long error) +{ + conditional_sti(regs); + machine_check_vector(regs, error); +} +#endif + +dotraplinkage void do_iret_error(struct pt_regs *regs, long error_code) +{ + siginfo_t info; + local_irq_enable(); + + info.si_signo = SIGILL; + info.si_errno = 0; + info.si_code = ILL_BADSTK; + info.si_addr = 0; + if (notify_die(DIE_TRAP, "iret exception", + regs, error_code, 32, SIGILL) == NOTIFY_STOP) + return; + do_trap(32, SIGILL, "iret exception", regs, error_code, &info); +} +#endif + void __init trap_init(void) { +#ifdef CONFIG_X86_32 int i; +#endif #ifdef CONFIG_EISA void __iomem *p = early_ioremap(0x0FFFD9, 4); @@ -1187,29 +1006,40 @@ void __init trap_init(void) early_iounmap(p, 4); #endif - set_trap_gate(0, ÷_error); - set_intr_gate(1, &debug); - set_intr_gate(2, &nmi); - set_system_intr_gate(3, &int3); /* int3 can be called from all */ - set_system_gate(4, &overflow); /* int4 can be called from all */ - set_trap_gate(5, &bounds); - set_trap_gate(6, &invalid_op); - set_trap_gate(7, &device_not_available); + set_intr_gate(0, ÷_error); + set_intr_gate_ist(1, &debug, DEBUG_STACK); + set_intr_gate_ist(2, &nmi, NMI_STACK); + /* int3 can be called from all */ + set_system_intr_gate_ist(3, &int3, DEBUG_STACK); + /* int4 can be called from all */ + set_system_intr_gate(4, &overflow); + set_intr_gate(5, &bounds); + set_intr_gate(6, &invalid_op); + set_intr_gate(7, &device_not_available); +#ifdef CONFIG_X86_32 set_task_gate(8, GDT_ENTRY_DOUBLEFAULT_TSS); - set_trap_gate(9, &coprocessor_segment_overrun); - set_trap_gate(10, &invalid_TSS); - set_trap_gate(11, &segment_not_present); - set_trap_gate(12, &stack_segment); - set_trap_gate(13, &general_protection); +#else + set_intr_gate_ist(8, &double_fault, DOUBLEFAULT_STACK); +#endif + set_intr_gate(9, &coprocessor_segment_overrun); + set_intr_gate(10, &invalid_TSS); + set_intr_gate(11, &segment_not_present); + set_intr_gate_ist(12, &stack_segment, STACKFAULT_STACK); + set_intr_gate(13, &general_protection); set_intr_gate(14, &page_fault); - set_trap_gate(15, &spurious_interrupt_bug); - set_trap_gate(16, &coprocessor_error); - set_trap_gate(17, &alignment_check); + set_intr_gate(15, &spurious_interrupt_bug); + set_intr_gate(16, &coprocessor_error); + set_intr_gate(17, &alignment_check); #ifdef CONFIG_X86_MCE - set_trap_gate(18, &machine_check); + set_intr_gate_ist(18, &machine_check, MCE_STACK); #endif - set_trap_gate(19, &simd_coprocessor_error); + set_intr_gate(19, &simd_coprocessor_error); +#ifdef CONFIG_IA32_EMULATION + set_system_intr_gate(IA32_SYSCALL_VECTOR, ia32_syscall); +#endif + +#ifdef CONFIG_X86_32 if (cpu_has_fxsr) { printk(KERN_INFO "Enabling fast FPU save and restore... "); set_in_cr4(X86_CR4_OSFXSR); @@ -1222,36 +1052,20 @@ void __init trap_init(void) printk("done.\n"); } - set_system_gate(SYSCALL_VECTOR, &system_call); + set_system_trap_gate(SYSCALL_VECTOR, &system_call); /* Reserve all the builtin and the syscall vector: */ for (i = 0; i < FIRST_EXTERNAL_VECTOR; i++) set_bit(i, used_vectors); set_bit(SYSCALL_VECTOR, used_vectors); - +#endif /* * Should be a barrier for any external CPU state: */ cpu_init(); +#ifdef CONFIG_X86_32 trap_init_hook(); +#endif } - -static int __init kstack_setup(char *s) -{ - kstack_depth_to_print = simple_strtoul(s, NULL, 0); - - return 1; -} -__setup("kstack=", kstack_setup); - -static int __init code_bytes_setup(char *s) -{ - code_bytes = simple_strtoul(s, NULL, 0); - if (code_bytes > 8192) - code_bytes = 8192; - - return 1; -} -__setup("code_bytes=", code_bytes_setup); diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c deleted file mode 100644 index 9c0ac0c..0000000 --- a/arch/x86/kernel/traps_64.c +++ /dev/null @@ -1,1214 +0,0 @@ -/* - * Copyright (C) 1991, 1992 Linus Torvalds - * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs - * - * Pentium III FXSR, SSE support - * Gareth Hughes <gareth@valinux.com>, May 2000 - */ - -/* - * 'Traps.c' handles hardware traps and faults after we have saved some - * state in 'entry.S'. - */ -#include <linux/moduleparam.h> -#include <linux/interrupt.h> -#include <linux/kallsyms.h> -#include <linux/spinlock.h> -#include <linux/kprobes.h> -#include <linux/uaccess.h> -#include <linux/utsname.h> -#include <linux/kdebug.h> -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/ptrace.h> -#include <linux/string.h> -#include <linux/unwind.h> -#include <linux/delay.h> -#include <linux/errno.h> -#include <linux/kexec.h> -#include <linux/sched.h> -#include <linux/timer.h> -#include <linux/init.h> -#include <linux/bug.h> -#include <linux/nmi.h> -#include <linux/mm.h> -#include <linux/smp.h> -#include <linux/io.h> - -#if defined(CONFIG_EDAC) -#include <linux/edac.h> -#endif - -#include <asm/stacktrace.h> -#include <asm/processor.h> -#include <asm/debugreg.h> -#include <asm/atomic.h> -#include <asm/system.h> -#include <asm/unwind.h> -#include <asm/desc.h> -#include <asm/i387.h> -#include <asm/pgalloc.h> -#include <asm/proto.h> -#include <asm/pda.h> -#include <asm/traps.h> - -#include <mach_traps.h> - -int panic_on_unrecovered_nmi; -int kstack_depth_to_print = 12; -static unsigned int code_bytes = 64; -static int ignore_nmis; -static int die_counter; - -static inline void conditional_sti(struct pt_regs *regs) -{ - if (regs->flags & X86_EFLAGS_IF) - local_irq_enable(); -} - -static inline void preempt_conditional_sti(struct pt_regs *regs) -{ - inc_preempt_count(); - if (regs->flags & X86_EFLAGS_IF) - local_irq_enable(); -} - -static inline void preempt_conditional_cli(struct pt_regs *regs) -{ - if (regs->flags & X86_EFLAGS_IF) - local_irq_disable(); - /* Make sure to not schedule here because we could be running - on an exception stack. */ - dec_preempt_count(); -} - -void printk_address(unsigned long address, int reliable) -{ - printk(" [<%016lx>] %s%pS\n", - address, reliable ? "" : "? ", (void *) address); -} - -static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, - unsigned *usedp, char **idp) -{ - static char ids[][8] = { - [DEBUG_STACK - 1] = "#DB", - [NMI_STACK - 1] = "NMI", - [DOUBLEFAULT_STACK - 1] = "#DF", - [STACKFAULT_STACK - 1] = "#SS", - [MCE_STACK - 1] = "#MC", -#if DEBUG_STKSZ > EXCEPTION_STKSZ - [N_EXCEPTION_STACKS ... - N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]" -#endif - }; - unsigned k; - - /* - * Iterate over all exception stacks, and figure out whether - * 'stack' is in one of them: - */ - for (k = 0; k < N_EXCEPTION_STACKS; k++) { - unsigned long end = per_cpu(orig_ist, cpu).ist[k]; - /* - * Is 'stack' above this exception frame's end? - * If yes then skip to the next frame. - */ - if (stack >= end) - continue; - /* - * Is 'stack' above this exception frame's start address? - * If yes then we found the right frame. - */ - if (stack >= end - EXCEPTION_STKSZ) { - /* - * Make sure we only iterate through an exception - * stack once. If it comes up for the second time - * then there's something wrong going on - just - * break out and return NULL: - */ - if (*usedp & (1U << k)) - break; - *usedp |= 1U << k; - *idp = ids[k]; - return (unsigned long *)end; - } - /* - * If this is a debug stack, and if it has a larger size than - * the usual exception stacks, then 'stack' might still - * be within the lower portion of the debug stack: - */ -#if DEBUG_STKSZ > EXCEPTION_STKSZ - if (k == DEBUG_STACK - 1 && stack >= end - DEBUG_STKSZ) { - unsigned j = N_EXCEPTION_STACKS - 1; - - /* - * Black magic. A large debug stack is composed of - * multiple exception stack entries, which we - * iterate through now. Dont look: - */ - do { - ++j; - end -= EXCEPTION_STKSZ; - ids[j][4] = '1' + (j - N_EXCEPTION_STACKS); - } while (stack < end - EXCEPTION_STKSZ); - if (*usedp & (1U << j)) - break; - *usedp |= 1U << j; - *idp = ids[j]; - return (unsigned long *)end; - } -#endif - } - return NULL; -} - -/* - * x86-64 can have up to three kernel stacks: - * process stack - * interrupt stack - * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack - */ - -static inline int valid_stack_ptr(struct thread_info *tinfo, - void *p, unsigned int size, void *end) -{ - void *t = tinfo; - if (end) { - if (p < end && p >= (end-THREAD_SIZE)) - return 1; - else - return 0; - } - return p > t && p < t + THREAD_SIZE - size; -} - -/* The form of the top of the frame on the stack */ -struct stack_frame { - struct stack_frame *next_frame; - unsigned long return_address; -}; - -static inline unsigned long -print_context_stack(struct thread_info *tinfo, - unsigned long *stack, unsigned long bp, - const struct stacktrace_ops *ops, void *data, - unsigned long *end) -{ - struct stack_frame *frame = (struct stack_frame *)bp; - - while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) { - unsigned long addr; - - addr = *stack; - if (__kernel_text_address(addr)) { - if ((unsigned long) stack == bp + 8) { - ops->address(data, addr, 1); - frame = frame->next_frame; - bp = (unsigned long) frame; - } else { - ops->address(data, addr, bp == 0); - } - } - stack++; - } - return bp; -} - -void dump_trace(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, - const struct stacktrace_ops *ops, void *data) -{ - const unsigned cpu = get_cpu(); - unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr; - unsigned used = 0; - struct thread_info *tinfo; - - if (!task) - task = current; - - if (!stack) { - unsigned long dummy; - stack = &dummy; - if (task && task != current) - stack = (unsigned long *)task->thread.sp; - } - -#ifdef CONFIG_FRAME_POINTER - if (!bp) { - if (task == current) { - /* Grab bp right from our regs */ - asm("movq %%rbp, %0" : "=r" (bp) : ); - } else { - /* bp is the last reg pushed by switch_to */ - bp = *(unsigned long *) task->thread.sp; - } - } -#endif - - /* - * Print function call entries in all stacks, starting at the - * current stack address. If the stacks consist of nested - * exceptions - */ - tinfo = task_thread_info(task); - for (;;) { - char *id; - unsigned long *estack_end; - estack_end = in_exception_stack(cpu, (unsigned long)stack, - &used, &id); - - if (estack_end) { - if (ops->stack(data, id) < 0) - break; - - bp = print_context_stack(tinfo, stack, bp, ops, - data, estack_end); - ops->stack(data, "<EOE>"); - /* - * We link to the next stack via the - * second-to-last pointer (index -2 to end) in the - * exception stack: - */ - stack = (unsigned long *) estack_end[-2]; - continue; - } - if (irqstack_end) { - unsigned long *irqstack; - irqstack = irqstack_end - - (IRQSTACKSIZE - 64) / sizeof(*irqstack); - - if (stack >= irqstack && stack < irqstack_end) { - if (ops->stack(data, "IRQ") < 0) - break; - bp = print_context_stack(tinfo, stack, bp, - ops, data, irqstack_end); - /* - * We link to the next stack (which would be - * the process stack normally) the last - * pointer (index -1 to end) in the IRQ stack: - */ - stack = (unsigned long *) (irqstack_end[-1]); - irqstack_end = NULL; - ops->stack(data, "EOI"); - continue; - } - } - break; - } - - /* - * This handles the process stack: - */ - bp = print_context_stack(tinfo, stack, bp, ops, data, NULL); - put_cpu(); -} -EXPORT_SYMBOL(dump_trace); - -static void -print_trace_warning_symbol(void *data, char *msg, unsigned long symbol) -{ - print_symbol(msg, symbol); - printk("\n"); -} - -static void print_trace_warning(void *data, char *msg) -{ - printk("%s\n", msg); -} - -static int print_trace_stack(void *data, char *name) -{ - printk(" <%s> ", name); - return 0; -} - -static void print_trace_address(void *data, unsigned long addr, int reliable) -{ - touch_nmi_watchdog(); - printk_address(addr, reliable); -} - -static const struct stacktrace_ops print_trace_ops = { - .warning = print_trace_warning, - .warning_symbol = print_trace_warning_symbol, - .stack = print_trace_stack, - .address = print_trace_address, -}; - -static void -show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, char *log_lvl) -{ - printk("Call Trace:\n"); - dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl); -} - -void show_trace(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp) -{ - show_trace_log_lvl(task, regs, stack, bp, ""); -} - -static void -show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *sp, unsigned long bp, char *log_lvl) -{ - unsigned long *stack; - int i; - const int cpu = smp_processor_id(); - unsigned long *irqstack_end = - (unsigned long *) (cpu_pda(cpu)->irqstackptr); - unsigned long *irqstack = - (unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE); - - /* - * debugging aid: "show_stack(NULL, NULL);" prints the - * back trace for this cpu. - */ - - if (sp == NULL) { - if (task) - sp = (unsigned long *)task->thread.sp; - else - sp = (unsigned long *)&sp; - } - - stack = sp; - for (i = 0; i < kstack_depth_to_print; i++) { - if (stack >= irqstack && stack <= irqstack_end) { - if (stack == irqstack_end) { - stack = (unsigned long *) (irqstack_end[-1]); - printk(" <EOI> "); - } - } else { - if (((long) stack & (THREAD_SIZE-1)) == 0) - break; - } - if (i && ((i % 4) == 0)) - printk("\n"); - printk(" %016lx", *stack++); - touch_nmi_watchdog(); - } - printk("\n"); - show_trace_log_lvl(task, regs, sp, bp, log_lvl); -} - -void show_stack(struct task_struct *task, unsigned long *sp) -{ - show_stack_log_lvl(task, NULL, sp, 0, ""); -} - -/* - * The architecture-independent dump_stack generator - */ -void dump_stack(void) -{ - unsigned long bp = 0; - unsigned long stack; - -#ifdef CONFIG_FRAME_POINTER - if (!bp) - asm("movq %%rbp, %0" : "=r" (bp) : ); -#endif - - printk("Pid: %d, comm: %.20s %s %s %.*s\n", - current->pid, current->comm, print_tainted(), - init_utsname()->release, - (int)strcspn(init_utsname()->version, " "), - init_utsname()->version); - show_trace(NULL, NULL, &stack, bp); -} -EXPORT_SYMBOL(dump_stack); - -void show_registers(struct pt_regs *regs) -{ - int i; - unsigned long sp; - const int cpu = smp_processor_id(); - struct task_struct *cur = cpu_pda(cpu)->pcurrent; - - sp = regs->sp; - printk("CPU %d ", cpu); - __show_regs(regs); - printk("Process %s (pid: %d, threadinfo %p, task %p)\n", - cur->comm, cur->pid, task_thread_info(cur), cur); - - /* - * When in-kernel, we also print out the stack and code at the - * time of the fault.. - */ - if (!user_mode(regs)) { - unsigned int code_prologue = code_bytes * 43 / 64; - unsigned int code_len = code_bytes; - unsigned char c; - u8 *ip; - - printk("Stack: "); - show_stack_log_lvl(NULL, regs, (unsigned long *)sp, - regs->bp, ""); - - printk(KERN_EMERG "Code: "); - - ip = (u8 *)regs->ip - code_prologue; - if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) { - /* try starting at RIP */ - ip = (u8 *)regs->ip; - code_len = code_len - code_prologue + 1; - } - for (i = 0; i < code_len; i++, ip++) { - if (ip < (u8 *)PAGE_OFFSET || - probe_kernel_address(ip, c)) { - printk(" Bad RIP value."); - break; - } - if (ip == (u8 *)regs->ip) - printk("<%02x> ", c); - else - printk("%02x ", c); - } - } - printk("\n"); -} - -int is_valid_bugaddr(unsigned long ip) -{ - unsigned short ud2; - - if (__copy_from_user(&ud2, (const void __user *) ip, sizeof(ud2))) - return 0; - - return ud2 == 0x0b0f; -} - -static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED; -static int die_owner = -1; -static unsigned int die_nest_count; - -unsigned __kprobes long oops_begin(void) -{ - int cpu; - unsigned long flags; - - oops_enter(); - - /* racy, but better than risking deadlock. */ - raw_local_irq_save(flags); - cpu = smp_processor_id(); - if (!__raw_spin_trylock(&die_lock)) { - if (cpu == die_owner) - /* nested oops. should stop eventually */; - else - __raw_spin_lock(&die_lock); - } - die_nest_count++; - die_owner = cpu; - console_verbose(); - bust_spinlocks(1); - return flags; -} - -void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) -{ - die_owner = -1; - bust_spinlocks(0); - die_nest_count--; - if (!die_nest_count) - /* Nest count reaches zero, release the lock. */ - __raw_spin_unlock(&die_lock); - raw_local_irq_restore(flags); - if (!regs) { - oops_exit(); - return; - } - if (panic_on_oops) - panic("Fatal exception"); - oops_exit(); - do_exit(signr); -} - -int __kprobes __die(const char *str, struct pt_regs *regs, long err) -{ - printk(KERN_EMERG "%s: %04lx [%u] ", str, err & 0xffff, ++die_counter); -#ifdef CONFIG_PREEMPT - printk("PREEMPT "); -#endif -#ifdef CONFIG_SMP - printk("SMP "); -#endif -#ifdef CONFIG_DEBUG_PAGEALLOC - printk("DEBUG_PAGEALLOC"); -#endif - printk("\n"); - if (notify_die(DIE_OOPS, str, regs, err, - current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) - return 1; - - show_registers(regs); - add_taint(TAINT_DIE); - /* Executive summary in case the oops scrolled away */ - printk(KERN_ALERT "RIP "); - printk_address(regs->ip, 1); - printk(" RSP <%016lx>\n", regs->sp); - if (kexec_should_crash(current)) - crash_kexec(regs); - return 0; -} - -void die(const char *str, struct pt_regs *regs, long err) -{ - unsigned long flags = oops_begin(); - - if (!user_mode(regs)) - report_bug(regs->ip, regs); - - if (__die(str, regs, err)) - regs = NULL; - oops_end(flags, regs, SIGSEGV); -} - -notrace __kprobes void -die_nmi(char *str, struct pt_regs *regs, int do_panic) -{ - unsigned long flags; - - if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP) - return; - - flags = oops_begin(); - /* - * We are in trouble anyway, lets at least try - * to get a message out. - */ - printk(KERN_EMERG "%s", str); - printk(" on CPU%d, ip %08lx, registers:\n", - smp_processor_id(), regs->ip); - show_registers(regs); - if (kexec_should_crash(current)) - crash_kexec(regs); - if (do_panic || panic_on_oops) - panic("Non maskable interrupt"); - oops_end(flags, NULL, SIGBUS); - nmi_exit(); - local_irq_enable(); - do_exit(SIGBUS); -} - -static void __kprobes -do_trap(int trapnr, int signr, char *str, struct pt_regs *regs, - long error_code, siginfo_t *info) -{ - struct task_struct *tsk = current; - - if (!user_mode(regs)) - goto kernel_trap; - - /* - * We want error_code and trap_no set for userspace faults and - * kernelspace faults which result in die(), but not - * kernelspace faults which are fixed up. die() gives the - * process no chance to handle the signal and notice the - * kernel fault information, so that won't result in polluting - * the information about previously queued, but not yet - * delivered, faults. See also do_general_protection below. - */ - tsk->thread.error_code = error_code; - tsk->thread.trap_no = trapnr; - - if (show_unhandled_signals && unhandled_signal(tsk, signr) && - printk_ratelimit()) { - printk(KERN_INFO - "%s[%d] trap %s ip:%lx sp:%lx error:%lx", - tsk->comm, tsk->pid, str, - regs->ip, regs->sp, error_code); - print_vma_addr(" in ", regs->ip); - printk("\n"); - } - - if (info) - force_sig_info(signr, info, tsk); - else - force_sig(signr, tsk); - return; - -kernel_trap: - if (!fixup_exception(regs)) { - tsk->thread.error_code = error_code; - tsk->thread.trap_no = trapnr; - die(str, regs, error_code); - } - return; -} - -#define DO_ERROR(trapnr, signr, str, name) \ -asmlinkage void do_##name(struct pt_regs *regs, long error_code) \ -{ \ - if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ - == NOTIFY_STOP) \ - return; \ - conditional_sti(regs); \ - do_trap(trapnr, signr, str, regs, error_code, NULL); \ -} - -#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ -asmlinkage void do_##name(struct pt_regs *regs, long error_code) \ -{ \ - siginfo_t info; \ - info.si_signo = signr; \ - info.si_errno = 0; \ - info.si_code = sicode; \ - info.si_addr = (void __user *)siaddr; \ - trace_hardirqs_fixup(); \ - if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ - == NOTIFY_STOP) \ - return; \ - conditional_sti(regs); \ - do_trap(trapnr, signr, str, regs, error_code, &info); \ -} - -DO_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip) -DO_ERROR(4, SIGSEGV, "overflow", overflow) -DO_ERROR(5, SIGSEGV, "bounds", bounds) -DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip) -DO_ERROR(9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun) -DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS) -DO_ERROR(11, SIGBUS, "segment not present", segment_not_present) -DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0) - -/* Runs on IST stack */ -asmlinkage void do_stack_segment(struct pt_regs *regs, long error_code) -{ - if (notify_die(DIE_TRAP, "stack segment", regs, error_code, - 12, SIGBUS) == NOTIFY_STOP) - return; - preempt_conditional_sti(regs); - do_trap(12, SIGBUS, "stack segment", regs, error_code, NULL); - preempt_conditional_cli(regs); -} - -asmlinkage void do_double_fault(struct pt_regs *regs, long error_code) -{ - static const char str[] = "double fault"; - struct task_struct *tsk = current; - - /* Return not checked because double check cannot be ignored */ - notify_die(DIE_TRAP, str, regs, error_code, 8, SIGSEGV); - - tsk->thread.error_code = error_code; - tsk->thread.trap_no = 8; - - /* This is always a kernel trap and never fixable (and thus must - never return). */ - for (;;) - die(str, regs, error_code); -} - -asmlinkage void __kprobes -do_general_protection(struct pt_regs *regs, long error_code) -{ - struct task_struct *tsk; - - conditional_sti(regs); - - tsk = current; - if (!user_mode(regs)) - goto gp_in_kernel; - - tsk->thread.error_code = error_code; - tsk->thread.trap_no = 13; - - if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) && - printk_ratelimit()) { - printk(KERN_INFO - "%s[%d] general protection ip:%lx sp:%lx error:%lx", - tsk->comm, tsk->pid, - regs->ip, regs->sp, error_code); - print_vma_addr(" in ", regs->ip); - printk("\n"); - } - - force_sig(SIGSEGV, tsk); - return; - -gp_in_kernel: - if (fixup_exception(regs)) - return; - - tsk->thread.error_code = error_code; - tsk->thread.trap_no = 13; - if (notify_die(DIE_GPF, "general protection fault", regs, - error_code, 13, SIGSEGV) == NOTIFY_STOP) - return; - die("general protection fault", regs, error_code); -} - -static notrace __kprobes void -mem_parity_error(unsigned char reason, struct pt_regs *regs) -{ - printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n", - reason); - printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n"); - -#if defined(CONFIG_EDAC) - if (edac_handler_set()) { - edac_atomic_assert_error(); - return; - } -#endif - - if (panic_on_unrecovered_nmi) - panic("NMI: Not continuing"); - - printk(KERN_EMERG "Dazed and confused, but trying to continue\n"); - - /* Clear and disable the memory parity error line. */ - reason = (reason & 0xf) | 4; - outb(reason, 0x61); -} - -static notrace __kprobes void -io_check_error(unsigned char reason, struct pt_regs *regs) -{ - printk("NMI: IOCK error (debug interrupt?)\n"); - show_registers(regs); - - /* Re-enable the IOCK line, wait for a few seconds */ - reason = (reason & 0xf) | 8; - outb(reason, 0x61); - mdelay(2000); - reason &= ~8; - outb(reason, 0x61); -} - -static notrace __kprobes void -unknown_nmi_error(unsigned char reason, struct pt_regs *regs) -{ - if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == - NOTIFY_STOP) - return; - printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n", - reason); - printk(KERN_EMERG "Do you have a strange power saving mode enabled?\n"); - - if (panic_on_unrecovered_nmi) - panic("NMI: Not continuing"); - - printk(KERN_EMERG "Dazed and confused, but trying to continue\n"); -} - -/* Runs on IST stack. This code must keep interrupts off all the time. - Nested NMIs are prevented by the CPU. */ -asmlinkage notrace __kprobes void default_do_nmi(struct pt_regs *regs) -{ - unsigned char reason = 0; - int cpu; - - cpu = smp_processor_id(); - - /* Only the BSP gets external NMIs from the system. */ - if (!cpu) - reason = get_nmi_reason(); - - if (!(reason & 0xc0)) { - if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT) - == NOTIFY_STOP) - return; - /* - * Ok, so this is none of the documented NMI sources, - * so it must be the NMI watchdog. - */ - if (nmi_watchdog_tick(regs, reason)) - return; - if (!do_nmi_callback(regs, cpu)) - unknown_nmi_error(reason, regs); - - return; - } - if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) - return; - - /* AK: following checks seem to be broken on modern chipsets. FIXME */ - if (reason & 0x80) - mem_parity_error(reason, regs); - if (reason & 0x40) - io_check_error(reason, regs); -} - -asmlinkage notrace __kprobes void -do_nmi(struct pt_regs *regs, long error_code) -{ - nmi_enter(); - - add_pda(__nmi_count, 1); - - if (!ignore_nmis) - default_do_nmi(regs); - - nmi_exit(); -} - -void stop_nmi(void) -{ - acpi_nmi_disable(); - ignore_nmis++; -} - -void restart_nmi(void) -{ - ignore_nmis--; - acpi_nmi_enable(); -} - -/* runs on IST stack. */ -asmlinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) -{ - trace_hardirqs_fixup(); - - if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) - == NOTIFY_STOP) - return; - - preempt_conditional_sti(regs); - do_trap(3, SIGTRAP, "int3", regs, error_code, NULL); - preempt_conditional_cli(regs); -} - -/* Help handler running on IST stack to switch back to user stack - for scheduling or signal handling. The actual stack switch is done in - entry.S */ -asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs) -{ - struct pt_regs *regs = eregs; - /* Did already sync */ - if (eregs == (struct pt_regs *)eregs->sp) - ; - /* Exception from user space */ - else if (user_mode(eregs)) - regs = task_pt_regs(current); - /* Exception from kernel and interrupts are enabled. Move to - kernel process stack. */ - else if (eregs->flags & X86_EFLAGS_IF) - regs = (struct pt_regs *)(eregs->sp -= sizeof(struct pt_regs)); - if (eregs != regs) - *regs = *eregs; - return regs; -} - -/* runs on IST stack. */ -asmlinkage void __kprobes do_debug(struct pt_regs *regs, - unsigned long error_code) -{ - struct task_struct *tsk = current; - unsigned long condition; - siginfo_t info; - - trace_hardirqs_fixup(); - - get_debugreg(condition, 6); - - /* - * The processor cleared BTF, so don't mark that we need it set. - */ - clear_tsk_thread_flag(tsk, TIF_DEBUGCTLMSR); - tsk->thread.debugctlmsr = 0; - - if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code, - SIGTRAP) == NOTIFY_STOP) - return; - - preempt_conditional_sti(regs); - - /* Mask out spurious debug traps due to lazy DR7 setting */ - if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) { - if (!tsk->thread.debugreg7) - goto clear_dr7; - } - - tsk->thread.debugreg6 = condition; - - /* - * Single-stepping through TF: make sure we ignore any events in - * kernel space (but re-enable TF when returning to user mode). - */ - if (condition & DR_STEP) { - if (!user_mode(regs)) - goto clear_TF_reenable; - } - - /* Ok, finally something we can handle */ - tsk->thread.trap_no = 1; - tsk->thread.error_code = error_code; - info.si_signo = SIGTRAP; - info.si_errno = 0; - info.si_code = get_si_code(condition); - info.si_addr = user_mode(regs) ? (void __user *)regs->ip : NULL; - force_sig_info(SIGTRAP, &info, tsk); - -clear_dr7: - set_debugreg(0, 7); - preempt_conditional_cli(regs); - return; - -clear_TF_reenable: - set_tsk_thread_flag(tsk, TIF_SINGLESTEP); - regs->flags &= ~X86_EFLAGS_TF; - preempt_conditional_cli(regs); - return; -} - -static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr) -{ - if (fixup_exception(regs)) - return 1; - - notify_die(DIE_GPF, str, regs, 0, trapnr, SIGFPE); - /* Illegal floating point operation in the kernel */ - current->thread.trap_no = trapnr; - die(str, regs, 0); - return 0; -} - -/* - * Note that we play around with the 'TS' bit in an attempt to get - * the correct behaviour even in the presence of the asynchronous - * IRQ13 behaviour - */ -asmlinkage void do_coprocessor_error(struct pt_regs *regs) -{ - void __user *ip = (void __user *)(regs->ip); - struct task_struct *task; - siginfo_t info; - unsigned short cwd, swd; - - conditional_sti(regs); - if (!user_mode(regs) && - kernel_math_error(regs, "kernel x87 math error", 16)) - return; - - /* - * Save the info for the exception handler and clear the error. - */ - task = current; - save_init_fpu(task); - task->thread.trap_no = 16; - task->thread.error_code = 0; - info.si_signo = SIGFPE; - info.si_errno = 0; - info.si_code = __SI_FAULT; - info.si_addr = ip; - /* - * (~cwd & swd) will mask out exceptions that are not set to unmasked - * status. 0x3f is the exception bits in these regs, 0x200 is the - * C1 reg you need in case of a stack fault, 0x040 is the stack - * fault bit. We should only be taking one exception at a time, - * so if this combination doesn't produce any single exception, - * then we have a bad program that isn't synchronizing its FPU usage - * and it will suffer the consequences since we won't be able to - * fully reproduce the context of the exception - */ - cwd = get_fpu_cwd(task); - swd = get_fpu_swd(task); - switch (swd & ~cwd & 0x3f) { - case 0x000: /* No unmasked exception */ - default: /* Multiple exceptions */ - break; - case 0x001: /* Invalid Op */ - /* - * swd & 0x240 == 0x040: Stack Underflow - * swd & 0x240 == 0x240: Stack Overflow - * User must clear the SF bit (0x40) if set - */ - info.si_code = FPE_FLTINV; - break; - case 0x002: /* Denormalize */ - case 0x010: /* Underflow */ - info.si_code = FPE_FLTUND; - break; - case 0x004: /* Zero Divide */ - info.si_code = FPE_FLTDIV; - break; - case 0x008: /* Overflow */ - info.si_code = FPE_FLTOVF; - break; - case 0x020: /* Precision */ - info.si_code = FPE_FLTRES; - break; - } - force_sig_info(SIGFPE, &info, task); -} - -asmlinkage void bad_intr(void) -{ - printk("bad interrupt"); -} - -asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs) -{ - void __user *ip = (void __user *)(regs->ip); - struct task_struct *task; - siginfo_t info; - unsigned short mxcsr; - - conditional_sti(regs); - if (!user_mode(regs) && - kernel_math_error(regs, "kernel simd math error", 19)) - return; - - /* - * Save the info for the exception handler and clear the error. - */ - task = current; - save_init_fpu(task); - task->thread.trap_no = 19; - task->thread.error_code = 0; - info.si_signo = SIGFPE; - info.si_errno = 0; - info.si_code = __SI_FAULT; - info.si_addr = ip; - /* - * The SIMD FPU exceptions are handled a little differently, as there - * is only a single status/control register. Thus, to determine which - * unmasked exception was caught we must mask the exception mask bits - * at 0x1f80, and then use these to mask the exception bits at 0x3f. - */ - mxcsr = get_fpu_mxcsr(task); - switch (~((mxcsr & 0x1f80) >> 7) & (mxcsr & 0x3f)) { - case 0x000: - default: - break; - case 0x001: /* Invalid Op */ - info.si_code = FPE_FLTINV; - break; - case 0x002: /* Denormalize */ - case 0x010: /* Underflow */ - info.si_code = FPE_FLTUND; - break; - case 0x004: /* Zero Divide */ - info.si_code = FPE_FLTDIV; - break; - case 0x008: /* Overflow */ - info.si_code = FPE_FLTOVF; - break; - case 0x020: /* Precision */ - info.si_code = FPE_FLTRES; - break; - } - force_sig_info(SIGFPE, &info, task); -} - -asmlinkage void do_spurious_interrupt_bug(struct pt_regs *regs) -{ -} - -asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void) -{ -} - -asmlinkage void __attribute__((weak)) mce_threshold_interrupt(void) -{ -} - -/* - * 'math_state_restore()' saves the current math information in the - * old math state array, and gets the new ones from the current task - * - * Careful.. There are problems with IBM-designed IRQ13 behaviour. - * Don't touch unless you *really* know how it works. - */ -asmlinkage void math_state_restore(void) -{ - struct task_struct *me = current; - - if (!used_math()) { - local_irq_enable(); - /* - * does a slab alloc which can sleep - */ - if (init_fpu(me)) { - /* - * ran out of memory! - */ - do_group_exit(SIGKILL); - return; - } - local_irq_disable(); - } - - clts(); /* Allow maths ops (or we recurse) */ - /* - * Paranoid restore. send a SIGSEGV if we fail to restore the state. - */ - if (unlikely(restore_fpu_checking(me))) { - stts(); - force_sig(SIGSEGV, me); - return; - } - task_thread_info(me)->status |= TS_USEDFPU; - me->fpu_counter++; -} -EXPORT_SYMBOL_GPL(math_state_restore); - -void __init trap_init(void) -{ - set_intr_gate(0, ÷_error); - set_intr_gate_ist(1, &debug, DEBUG_STACK); - set_intr_gate_ist(2, &nmi, NMI_STACK); - /* int3 can be called from all */ - set_system_gate_ist(3, &int3, DEBUG_STACK); - /* int4 can be called from all */ - set_system_gate(4, &overflow); - set_intr_gate(5, &bounds); - set_intr_gate(6, &invalid_op); - set_intr_gate(7, &device_not_available); - set_intr_gate_ist(8, &double_fault, DOUBLEFAULT_STACK); - set_intr_gate(9, &coprocessor_segment_overrun); - set_intr_gate(10, &invalid_TSS); - set_intr_gate(11, &segment_not_present); - set_intr_gate_ist(12, &stack_segment, STACKFAULT_STACK); - set_intr_gate(13, &general_protection); - set_intr_gate(14, &page_fault); - set_intr_gate(15, &spurious_interrupt_bug); - set_intr_gate(16, &coprocessor_error); - set_intr_gate(17, &alignment_check); -#ifdef CONFIG_X86_MCE - set_intr_gate_ist(18, &machine_check, MCE_STACK); -#endif - set_intr_gate(19, &simd_coprocessor_error); - -#ifdef CONFIG_IA32_EMULATION - set_system_gate(IA32_SYSCALL_VECTOR, ia32_syscall); -#endif - /* - * Should be a barrier for any external CPU state: - */ - cpu_init(); -} - -static int __init oops_setup(char *s) -{ - if (!s) - return -EINVAL; - if (!strcmp(s, "panic")) - panic_on_oops = 1; - return 0; -} -early_param("oops", oops_setup); - -static int __init kstack_setup(char *s) -{ - if (!s) - return -EINVAL; - kstack_depth_to_print = simple_strtoul(s, NULL, 0); - return 0; -} -early_param("kstack", kstack_setup); - -static int __init code_bytes_setup(char *s) -{ - code_bytes = simple_strtoul(s, NULL, 0); - if (code_bytes > 8192) - code_bytes = 8192; - - return 1; -} -__setup("code_bytes=", code_bytes_setup); diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index 201e81a..46e0544 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -172,8 +172,8 @@ SECTIONS .x86_cpu_dev.init : AT(ADDR(.x86_cpu_dev.init) - LOAD_OFFSET) { *(.x86_cpu_dev.init) } - SECURITY_INIT __x86_cpu_dev_end = .; + SECURITY_INIT . = ALIGN(8); .parainstructions : AT(ADDR(.parainstructions) - LOAD_OFFSET) { diff --git a/arch/x86/mach-generic/es7000.c b/arch/x86/mach-generic/es7000.c index 520cca0..6513d41 100644 --- a/arch/x86/mach-generic/es7000.c +++ b/arch/x86/mach-generic/es7000.c @@ -47,16 +47,26 @@ static __init int mps_oem_check(struct mp_config_table *mpc, char *oem, /* Hook from generic ACPI tables.c */ static int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) { - unsigned long oem_addr; + unsigned long oem_addr = 0; + int check_dsdt; + int ret = 0; + + /* check dsdt at first to avoid clear fix_map for oem_addr */ + check_dsdt = es7000_check_dsdt(); + if (!find_unisys_acpi_oem_table(&oem_addr)) { - if (es7000_check_dsdt()) - return parse_unisys_oem((char *)oem_addr); + if (check_dsdt) + ret = parse_unisys_oem((char *)oem_addr); else { setup_unisys(); - return 1; + ret = 1; } + /* + * we need to unmap it + */ + unmap_unisys_acpi_oem_table(oem_addr); } - return 0; + return ret; } #else static int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index dfb932d..59f89b4 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -13,12 +13,8 @@ obj-$(CONFIG_MMIOTRACE) += mmiotrace.o mmiotrace-y := pf_in.o mmio-mod.o obj-$(CONFIG_MMIOTRACE_TEST) += testmmiotrace.o -ifeq ($(CONFIG_X86_32),y) -obj-$(CONFIG_NUMA) += discontig_32.o -else -obj-$(CONFIG_NUMA) += numa_64.o +obj-$(CONFIG_NUMA) += numa_$(BITS).o obj-$(CONFIG_K8_NUMA) += k8topology_64.o -endif obj-$(CONFIG_ACPI_NUMA) += srat_$(BITS).o obj-$(CONFIG_MEMTEST) += memtest.o diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 8f92cac..3f2b896 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -592,11 +592,6 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code) unsigned long flags; #endif - /* - * We can fault from pretty much anywhere, with unknown IRQ state. - */ - trace_hardirqs_fixup(); - tsk = current; mm = tsk->mm; prefetchw(&mm->mmap_sem); @@ -914,15 +909,15 @@ LIST_HEAD(pgd_list); void vmalloc_sync_all(void) { -#ifdef CONFIG_X86_32 - unsigned long start = VMALLOC_START & PGDIR_MASK; unsigned long address; +#ifdef CONFIG_X86_32 if (SHARED_KERNEL_PMD) return; - BUILD_BUG_ON(TASK_SIZE & ~PGDIR_MASK); - for (address = start; address >= TASK_SIZE; address += PGDIR_SIZE) { + for (address = VMALLOC_START & PMD_MASK; + address >= TASK_SIZE && address < FIXADDR_TOP; + address += PMD_SIZE) { unsigned long flags; struct page *page; @@ -935,10 +930,8 @@ void vmalloc_sync_all(void) spin_unlock_irqrestore(&pgd_lock, flags); } #else /* CONFIG_X86_64 */ - unsigned long start = VMALLOC_START & PGDIR_MASK; - unsigned long address; - - for (address = start; address <= VMALLOC_END; address += PGDIR_SIZE) { + for (address = VMALLOC_START & PGDIR_MASK; address <= VMALLOC_END; + address += PGDIR_SIZE) { const pgd_t *pgd_ref = pgd_offset_k(address); unsigned long flags; struct page *page; diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c index 007bb06..4ba373c 100644 --- a/arch/x86/mm/gup.c +++ b/arch/x86/mm/gup.c @@ -82,7 +82,7 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr, pte_t pte = gup_get_pte(ptep); struct page *page; - if ((pte_val(pte) & (mask | _PAGE_SPECIAL)) != mask) { + if ((pte_flags(pte) & (mask | _PAGE_SPECIAL)) != mask) { pte_unmap(ptep); return 0; } @@ -116,10 +116,10 @@ static noinline int gup_huge_pmd(pmd_t pmd, unsigned long addr, mask = _PAGE_PRESENT|_PAGE_USER; if (write) mask |= _PAGE_RW; - if ((pte_val(pte) & mask) != mask) + if ((pte_flags(pte) & mask) != mask) return 0; /* hugepages are never "special" */ - VM_BUG_ON(pte_val(pte) & _PAGE_SPECIAL); + VM_BUG_ON(pte_flags(pte) & _PAGE_SPECIAL); VM_BUG_ON(!pfn_valid(pte_pfn(pte))); refs = 0; @@ -173,10 +173,10 @@ static noinline int gup_huge_pud(pud_t pud, unsigned long addr, mask = _PAGE_PRESENT|_PAGE_USER; if (write) mask |= _PAGE_RW; - if ((pte_val(pte) & mask) != mask) + if ((pte_flags(pte) & mask) != mask) return 0; /* hugepages are never "special" */ - VM_BUG_ON(pte_val(pte) & _PAGE_SPECIAL); + VM_BUG_ON(pte_flags(pte) & _PAGE_SPECIAL); VM_BUG_ON(!pfn_valid(pte_pfn(pte))); refs = 0; diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index c3789bb..c67d78d 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -557,7 +557,7 @@ void zap_low_mappings(void) int nx_enabled; -pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL); +pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL | _PAGE_IOMAP); EXPORT_SYMBOL_GPL(__supported_pte_mask); #ifdef CONFIG_X86_PAE diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 365eaea..747f6c9 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -88,7 +88,7 @@ early_param("gbpages", parse_direct_gbpages_on); int after_bootmem; -unsigned long __supported_pte_mask __read_mostly = ~0UL; +pteval_t __supported_pte_mask __read_mostly = ~_PAGE_IOMAP; EXPORT_SYMBOL_GPL(__supported_pte_mask); static int do_not_nx __cpuinitdata; @@ -195,9 +195,6 @@ set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte) } pte = pte_offset_kernel(pmd, vaddr); - if (!pte_none(*pte) && pte_val(new_pte) && - pte_val(*pte) != (pte_val(new_pte) & __supported_pte_mask)) - pte_ERROR(*pte); set_pte(pte, new_pte); /* @@ -312,7 +309,7 @@ static __ref void *alloc_low_page(unsigned long *phys) if (pfn >= table_top) panic("alloc_low_page: ran out of memory"); - adr = early_ioremap(pfn * PAGE_SIZE, PAGE_SIZE); + adr = early_memremap(pfn * PAGE_SIZE, PAGE_SIZE); memset(adr, 0, PAGE_SIZE); *phys = pfn * PAGE_SIZE; return adr; @@ -775,7 +772,7 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, old_start = mr[i].start; memmove(&mr[i], &mr[i+1], (nr_range - 1 - i) * sizeof (struct map_range)); - mr[i].start = old_start; + mr[i--].start = old_start; nr_range--; } diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index 6ab3196..ddb3751 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -24,18 +24,47 @@ #ifdef CONFIG_X86_64 +static inline int phys_addr_valid(unsigned long addr) +{ + return addr < (1UL << boot_cpu_data.x86_phys_bits); +} + unsigned long __phys_addr(unsigned long x) { - if (x >= __START_KERNEL_map) - return x - __START_KERNEL_map + phys_base; - return x - PAGE_OFFSET; + if (x >= __START_KERNEL_map) { + x -= __START_KERNEL_map; + VIRTUAL_BUG_ON(x >= KERNEL_IMAGE_SIZE); + x += phys_base; + } else { + VIRTUAL_BUG_ON(x < PAGE_OFFSET); + x -= PAGE_OFFSET; + VIRTUAL_BUG_ON(system_state == SYSTEM_BOOTING ? x > MAXMEM : + !phys_addr_valid(x)); + } + return x; } EXPORT_SYMBOL(__phys_addr); -static inline int phys_addr_valid(unsigned long addr) +bool __virt_addr_valid(unsigned long x) { - return addr < (1UL << boot_cpu_data.x86_phys_bits); + if (x >= __START_KERNEL_map) { + x -= __START_KERNEL_map; + if (x >= KERNEL_IMAGE_SIZE) + return false; + x += phys_base; + } else { + if (x < PAGE_OFFSET) + return false; + x -= PAGE_OFFSET; + if (system_state == SYSTEM_BOOTING ? + x > MAXMEM : !phys_addr_valid(x)) { + return false; + } + } + + return pfn_valid(x >> PAGE_SHIFT); } +EXPORT_SYMBOL(__virt_addr_valid); #else @@ -44,6 +73,28 @@ static inline int phys_addr_valid(unsigned long addr) return 1; } +#ifdef CONFIG_DEBUG_VIRTUAL +unsigned long __phys_addr(unsigned long x) +{ + /* VMALLOC_* aren't constants; not available at the boot time */ + VIRTUAL_BUG_ON(x < PAGE_OFFSET); + VIRTUAL_BUG_ON(system_state != SYSTEM_BOOTING && + is_vmalloc_addr((void *) x)); + return x - PAGE_OFFSET; +} +EXPORT_SYMBOL(__phys_addr); +#endif + +bool __virt_addr_valid(unsigned long x) +{ + if (x < PAGE_OFFSET) + return false; + if (system_state != SYSTEM_BOOTING && is_vmalloc_addr((void *) x)) + return false; + return pfn_valid((x - PAGE_OFFSET) >> PAGE_SHIFT); +} +EXPORT_SYMBOL(__virt_addr_valid); + #endif int page_is_ram(unsigned long pagenr) @@ -223,16 +274,16 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr, switch (prot_val) { case _PAGE_CACHE_UC: default: - prot = PAGE_KERNEL_NOCACHE; + prot = PAGE_KERNEL_IO_NOCACHE; break; case _PAGE_CACHE_UC_MINUS: - prot = PAGE_KERNEL_UC_MINUS; + prot = PAGE_KERNEL_IO_UC_MINUS; break; case _PAGE_CACHE_WC: - prot = PAGE_KERNEL_WC; + prot = PAGE_KERNEL_IO_WC; break; case _PAGE_CACHE_WB: - prot = PAGE_KERNEL; + prot = PAGE_KERNEL_IO; break; } @@ -549,12 +600,12 @@ static void __init __early_set_fixmap(enum fixed_addresses idx, } static inline void __init early_set_fixmap(enum fixed_addresses idx, - unsigned long phys) + unsigned long phys, pgprot_t prot) { if (after_paging_init) - set_fixmap(idx, phys); + __set_fixmap(idx, phys, prot); else - __early_set_fixmap(idx, phys, PAGE_KERNEL); + __early_set_fixmap(idx, phys, prot); } static inline void __init early_clear_fixmap(enum fixed_addresses idx) @@ -565,16 +616,22 @@ static inline void __init early_clear_fixmap(enum fixed_addresses idx) __early_set_fixmap(idx, 0, __pgprot(0)); } - -static int __initdata early_ioremap_nested; - +static void *prev_map[FIX_BTMAPS_SLOTS] __initdata; +static unsigned long prev_size[FIX_BTMAPS_SLOTS] __initdata; static int __init check_early_ioremap_leak(void) { - if (!early_ioremap_nested) + int count = 0; + int i; + + for (i = 0; i < FIX_BTMAPS_SLOTS; i++) + if (prev_map[i]) + count++; + + if (!count) return 0; WARN(1, KERN_WARNING "Debug warning: early ioremap leak of %d areas detected.\n", - early_ioremap_nested); + count); printk(KERN_WARNING "please boot with early_ioremap_debug and report the dmesg.\n"); @@ -582,18 +639,33 @@ static int __init check_early_ioremap_leak(void) } late_initcall(check_early_ioremap_leak); -void __init *early_ioremap(unsigned long phys_addr, unsigned long size) +static void __init *__early_ioremap(unsigned long phys_addr, unsigned long size, pgprot_t prot) { unsigned long offset, last_addr; - unsigned int nrpages, nesting; + unsigned int nrpages; enum fixed_addresses idx0, idx; + int i, slot; WARN_ON(system_state != SYSTEM_BOOTING); - nesting = early_ioremap_nested; + slot = -1; + for (i = 0; i < FIX_BTMAPS_SLOTS; i++) { + if (!prev_map[i]) { + slot = i; + break; + } + } + + if (slot < 0) { + printk(KERN_INFO "early_iomap(%08lx, %08lx) not found slot\n", + phys_addr, size); + WARN_ON(1); + return NULL; + } + if (early_ioremap_debug) { printk(KERN_INFO "early_ioremap(%08lx, %08lx) [%d] => ", - phys_addr, size, nesting); + phys_addr, size, slot); dump_stack(); } @@ -604,11 +676,7 @@ void __init *early_ioremap(unsigned long phys_addr, unsigned long size) return NULL; } - if (nesting >= FIX_BTMAPS_NESTING) { - WARN_ON(1); - return NULL; - } - early_ioremap_nested++; + prev_size[slot] = size; /* * Mappings have to be page-aligned */ @@ -628,10 +696,10 @@ void __init *early_ioremap(unsigned long phys_addr, unsigned long size) /* * Ok, go for it.. */ - idx0 = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*nesting; + idx0 = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*slot; idx = idx0; while (nrpages > 0) { - early_set_fixmap(idx, phys_addr); + early_set_fixmap(idx, phys_addr, prot); phys_addr += PAGE_SIZE; --idx; --nrpages; @@ -639,7 +707,20 @@ void __init *early_ioremap(unsigned long phys_addr, unsigned long size) if (early_ioremap_debug) printk(KERN_CONT "%08lx + %08lx\n", offset, fix_to_virt(idx0)); - return (void *) (offset + fix_to_virt(idx0)); + prev_map[slot] = (void *) (offset + fix_to_virt(idx0)); + return prev_map[slot]; +} + +/* Remap an IO device */ +void __init *early_ioremap(unsigned long phys_addr, unsigned long size) +{ + return __early_ioremap(phys_addr, size, PAGE_KERNEL_IO); +} + +/* Remap memory */ +void __init *early_memremap(unsigned long phys_addr, unsigned long size) +{ + return __early_ioremap(phys_addr, size, PAGE_KERNEL); } void __init early_iounmap(void *addr, unsigned long size) @@ -648,15 +729,33 @@ void __init early_iounmap(void *addr, unsigned long size) unsigned long offset; unsigned int nrpages; enum fixed_addresses idx; - int nesting; + int i, slot; + + slot = -1; + for (i = 0; i < FIX_BTMAPS_SLOTS; i++) { + if (prev_map[i] == addr) { + slot = i; + break; + } + } - nesting = --early_ioremap_nested; - if (WARN_ON(nesting < 0)) + if (slot < 0) { + printk(KERN_INFO "early_iounmap(%p, %08lx) not found slot\n", + addr, size); + WARN_ON(1); return; + } + + if (prev_size[slot] != size) { + printk(KERN_INFO "early_iounmap(%p, %08lx) [%d] size not consistent %08lx\n", + addr, size, slot, prev_size[slot]); + WARN_ON(1); + return; + } if (early_ioremap_debug) { printk(KERN_INFO "early_iounmap(%p, %08lx) [%d]\n", addr, - size, nesting); + size, slot); dump_stack(); } @@ -668,12 +767,13 @@ void __init early_iounmap(void *addr, unsigned long size) offset = virt_addr & ~PAGE_MASK; nrpages = PAGE_ALIGN(offset + size - 1) >> PAGE_SHIFT; - idx = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*nesting; + idx = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*slot; while (nrpages > 0) { early_clear_fixmap(idx); --idx; --nrpages; } + prev_map[slot] = 0; } void __this_fixmap_does_not_exist(void) diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/numa_32.c similarity index 100% rename from arch/x86/mm/discontig_32.c rename to arch/x86/mm/numa_32.c diff --git a/arch/x86/xen/Kconfig b/arch/x86/xen/Kconfig index 3815e42..d3e6846 100644 --- a/arch/x86/xen/Kconfig +++ b/arch/x86/xen/Kconfig @@ -27,4 +27,12 @@ config XEN_MAX_DOMAIN_MEMORY config XEN_SAVE_RESTORE bool depends on PM - default y \ No newline at end of file + default y + +config XEN_DEBUG_FS + bool "Enable Xen debug and tuning parameters in debugfs" + depends on XEN && DEBUG_FS + default n + help + Enable statistics output and various tuning options in debugfs. + Enabling this option may incur a significant performance overhead. \ No newline at end of file diff --git a/arch/x86/xen/Makefile b/arch/x86/xen/Makefile index 59c1e53..3139479 100644 --- a/arch/x86/xen/Makefile +++ b/arch/x86/xen/Makefile @@ -1,4 +1,12 @@ -obj-y := enlighten.o setup.o multicalls.o mmu.o \ +ifdef CONFIG_FTRACE +# Do not profile debug and lowlevel utilities +CFLAGS_REMOVE_spinlock.o = -pg +CFLAGS_REMOVE_time.o = -pg +CFLAGS_REMOVE_irq.o = -pg +endif + +obj-y := enlighten.o setup.o multicalls.o mmu.o irq.o \ time.o xen-asm_$(BITS).o grant-table.o suspend.o -obj-$(CONFIG_SMP) += smp.o +obj-$(CONFIG_SMP) += smp.o spinlock.o +obj-$(CONFIG_XEN_DEBUG_FS) += debugfs.o \ No newline at end of file diff --git a/arch/x86/xen/debugfs.c b/arch/x86/xen/debugfs.c new file mode 100644 index 0000000..b53225d --- /dev/null +++ b/arch/x86/xen/debugfs.c @@ -0,0 +1,123 @@ +#include <linux/init.h> +#include <linux/debugfs.h> +#include <linux/module.h> + +#include "debugfs.h" + +static struct dentry *d_xen_debug; + +struct dentry * __init xen_init_debugfs(void) +{ + if (!d_xen_debug) { + d_xen_debug = debugfs_create_dir("xen", NULL); + + if (!d_xen_debug) + pr_warning("Could not create 'xen' debugfs directory\n"); + } + + return d_xen_debug; +} + +struct array_data +{ + void *array; + unsigned elements; +}; + +static int u32_array_open(struct inode *inode, struct file *file) +{ + file->private_data = NULL; + return nonseekable_open(inode, file); +} + +static size_t format_array(char *buf, size_t bufsize, const char *fmt, + u32 *array, unsigned array_size) +{ + size_t ret = 0; + unsigned i; + + for(i = 0; i < array_size; i++) { + size_t len; + + len = snprintf(buf, bufsize, fmt, array[i]); + len++; /* ' ' or '\n' */ + ret += len; + + if (buf) { + buf += len; + bufsize -= len; + buf[-1] = (i == array_size-1) ? '\n' : ' '; + } + } + + ret++; /* \0 */ + if (buf) + *buf = '\0'; + + return ret; +} + +static char *format_array_alloc(const char *fmt, u32 *array, unsigned array_size) +{ + size_t len = format_array(NULL, 0, fmt, array, array_size); + char *ret; + + ret = kmalloc(len, GFP_KERNEL); + if (ret == NULL) + return NULL; + + format_array(ret, len, fmt, array, array_size); + return ret; +} + +static ssize_t u32_array_read(struct file *file, char __user *buf, size_t len, + loff_t *ppos) +{ + struct inode *inode = file->f_path.dentry->d_inode; + struct array_data *data = inode->i_private; + size_t size; + + if (*ppos == 0) { + if (file->private_data) { + kfree(file->private_data); + file->private_data = NULL; + } + + file->private_data = format_array_alloc("%u", data->array, data->elements); + } + + size = 0; + if (file->private_data) + size = strlen(file->private_data); + + return simple_read_from_buffer(buf, len, ppos, file->private_data, size); +} + +static int xen_array_release(struct inode *inode, struct file *file) +{ + kfree(file->private_data); + + return 0; +} + +static struct file_operations u32_array_fops = { + .owner = THIS_MODULE, + .open = u32_array_open, + .release= xen_array_release, + .read = u32_array_read, +}; + +struct dentry *xen_debugfs_create_u32_array(const char *name, mode_t mode, + struct dentry *parent, + u32 *array, unsigned elements) +{ + struct array_data *data = kmalloc(sizeof(*data), GFP_KERNEL); + + if (data == NULL) + return NULL; + + data->array = array; + data->elements = elements; + + return debugfs_create_file(name, mode, parent, data, &u32_array_fops); +} diff --git a/arch/x86/xen/debugfs.h b/arch/x86/xen/debugfs.h new file mode 100644 index 0000000..e281320 --- /dev/null +++ b/arch/x86/xen/debugfs.h @@ -0,0 +1,10 @@ +#ifndef _XEN_DEBUGFS_H +#define _XEN_DEBUGFS_H + +struct dentry * __init xen_init_debugfs(void); + +struct dentry *xen_debugfs_create_u32_array(const char *name, mode_t mode, + struct dentry *parent, + u32 *array, unsigned elements); + +#endif /* _XEN_DEBUGFS_H */ diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index a27d562..708b8ea 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -30,7 +30,6 @@ #include <xen/interface/xen.h> #include <xen/interface/physdev.h> #include <xen/interface/vcpu.h> -#include <xen/interface/sched.h> #include <xen/features.h> #include <xen/page.h> #include <xen/hvc-console.h> @@ -58,6 +57,9 @@ EXPORT_SYMBOL_GPL(hypercall_page); DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu); DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info); +enum xen_domain_type xen_domain_type = XEN_NATIVE; +EXPORT_SYMBOL_GPL(xen_domain_type); + /* * Identity map, in addition to plain kernel map. This needs to be * large enough to allocate page table pages to allocate the rest. @@ -227,103 +229,68 @@ static unsigned long xen_get_debugreg(int reg) return HYPERVISOR_get_debugreg(reg); } -static unsigned long xen_save_fl(void) +static void xen_leave_lazy(void) { - struct vcpu_info *vcpu; - unsigned long flags; - - vcpu = x86_read_percpu(xen_vcpu); - - /* flag has opposite sense of mask */ - flags = !vcpu->evtchn_upcall_mask; - - /* convert to IF type flag - -0 -> 0x00000000 - -1 -> 0xffffffff - */ - return (-flags) & X86_EFLAGS_IF; + paravirt_leave_lazy(paravirt_get_lazy_mode()); + xen_mc_flush(); } -static void xen_restore_fl(unsigned long flags) +static unsigned long xen_store_tr(void) { - struct vcpu_info *vcpu; - - /* convert from IF type flag */ - flags = !(flags & X86_EFLAGS_IF); - - /* There's a one instruction preempt window here. We need to - make sure we're don't switch CPUs between getting the vcpu - pointer and updating the mask. */ - preempt_disable(); - vcpu = x86_read_percpu(xen_vcpu); - vcpu->evtchn_upcall_mask = flags; - preempt_enable_no_resched(); - - /* Doesn't matter if we get preempted here, because any - pending event will get dealt with anyway. */ - - if (flags == 0) { - preempt_check_resched(); - barrier(); /* unmask then check (avoid races) */ - if (unlikely(vcpu->evtchn_upcall_pending)) - force_evtchn_callback(); - } + return 0; } -static void xen_irq_disable(void) +/* + * Set the page permissions for a particular virtual address. If the + * address is a vmalloc mapping (or other non-linear mapping), then + * find the linear mapping of the page and also set its protections to + * match. + */ +static void set_aliased_prot(void *v, pgprot_t prot) { - /* There's a one instruction preempt window here. We need to - make sure we're don't switch CPUs between getting the vcpu - pointer and updating the mask. */ - preempt_disable(); - x86_read_percpu(xen_vcpu)->evtchn_upcall_mask = 1; - preempt_enable_no_resched(); -} + int level; + pte_t *ptep; + pte_t pte; + unsigned long pfn; + struct page *page; -static void xen_irq_enable(void) -{ - struct vcpu_info *vcpu; + ptep = lookup_address((unsigned long)v, &level); + BUG_ON(ptep == NULL); - /* We don't need to worry about being preempted here, since - either a) interrupts are disabled, so no preemption, or b) - the caller is confused and is trying to re-enable interrupts - on an indeterminate processor. */ + pfn = pte_pfn(*ptep); + page = pfn_to_page(pfn); - vcpu = x86_read_percpu(xen_vcpu); - vcpu->evtchn_upcall_mask = 0; + pte = pfn_pte(pfn, prot); - /* Doesn't matter if we get preempted here, because any - pending event will get dealt with anyway. */ + if (HYPERVISOR_update_va_mapping((unsigned long)v, pte, 0)) + BUG(); - barrier(); /* unmask then check (avoid races) */ - if (unlikely(vcpu->evtchn_upcall_pending)) - force_evtchn_callback(); -} + if (!PageHighMem(page)) { + void *av = __va(PFN_PHYS(pfn)); -static void xen_safe_halt(void) -{ - /* Blocking includes an implicit local_irq_enable(). */ - if (HYPERVISOR_sched_op(SCHEDOP_block, NULL) != 0) - BUG(); + if (av != v) + if (HYPERVISOR_update_va_mapping((unsigned long)av, pte, 0)) + BUG(); + } else + kmap_flush_unused(); } -static void xen_halt(void) +static void xen_alloc_ldt(struct desc_struct *ldt, unsigned entries) { - if (irqs_disabled()) - HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL); - else - xen_safe_halt(); -} + const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE; + int i; -static void xen_leave_lazy(void) -{ - paravirt_leave_lazy(paravirt_get_lazy_mode()); - xen_mc_flush(); + for(i = 0; i < entries; i += entries_per_page) + set_aliased_prot(ldt + i, PAGE_KERNEL_RO); } -static unsigned long xen_store_tr(void) +static void xen_free_ldt(struct desc_struct *ldt, unsigned entries) { - return 0; + const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE; + int i; + + for(i = 0; i < entries; i += entries_per_page) + set_aliased_prot(ldt + i, PAGE_KERNEL); } static void xen_set_ldt(const void *addr, unsigned entries) @@ -426,8 +393,7 @@ static void xen_load_gs_index(unsigned int idx) static void xen_write_ldt_entry(struct desc_struct *dt, int entrynum, const void *ptr) { - unsigned long lp = (unsigned long)&dt[entrynum]; - xmaddr_t mach_lp = virt_to_machine(lp); + xmaddr_t mach_lp = arbitrary_virt_to_machine(&dt[entrynum]); u64 entry = *(u64 *)ptr; preempt_disable(); @@ -560,7 +526,7 @@ static void xen_write_gdt_entry(struct desc_struct *dt, int entry, } static void xen_load_sp0(struct tss_struct *tss, - struct thread_struct *thread) + struct thread_struct *thread) { struct multicall_space mcs = xen_mc_entry(0); MULTI_stack_switch(mcs.mc, __KERNEL_DS, thread->sp0); @@ -835,6 +801,19 @@ static int xen_write_msr_safe(unsigned int msr, unsigned low, unsigned high) ret = -EFAULT; break; #endif + + case MSR_STAR: + case MSR_CSTAR: + case MSR_LSTAR: + case MSR_SYSCALL_MASK: + case MSR_IA32_SYSENTER_CS: + case MSR_IA32_SYSENTER_ESP: + case MSR_IA32_SYSENTER_EIP: + /* Fast syscall setup is all done in hypercalls, so + these are all ignored. Stub them out here to stop + Xen console noise. */ + break; + default: ret = native_write_msr_safe(msr, low, high); } @@ -878,8 +857,8 @@ static void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn, unsigned l SetPagePinned(page); if (!PageHighMem(page)) { - make_lowmem_page_readonly(__va(PFN_PHYS(pfn))); - if (level == PT_PTE) + make_lowmem_page_readonly(__va(PFN_PHYS((unsigned long)pfn))); + if (level == PT_PTE && USE_SPLIT_PTLOCKS) pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn); } else /* make sure there are no stray mappings of @@ -947,7 +926,7 @@ static void xen_release_ptpage(unsigned long pfn, unsigned level) if (PagePinned(page)) { if (!PageHighMem(page)) { - if (level == PT_PTE) + if (level == PT_PTE && USE_SPLIT_PTLOCKS) pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn); make_lowmem_page_readwrite(__va(PFN_PHYS(pfn))); } @@ -1252,6 +1231,9 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = { .load_gs_index = xen_load_gs_index, #endif + .alloc_ldt = xen_alloc_ldt, + .free_ldt = xen_free_ldt, + .store_gdt = native_store_gdt, .store_idt = native_store_idt, .store_tr = xen_store_tr, @@ -1273,36 +1255,6 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = { }, }; -static void __init __xen_init_IRQ(void) -{ -#ifdef CONFIG_X86_64 - int i; - - /* Create identity vector->irq map */ - for(i = 0; i < NR_VECTORS; i++) { - int cpu; - - for_each_possible_cpu(cpu) - per_cpu(vector_irq, cpu)[i] = i; - } -#endif /* CONFIG_X86_64 */ - - xen_init_IRQ(); -} - -static const struct pv_irq_ops xen_irq_ops __initdata = { - .init_IRQ = __xen_init_IRQ, - .save_fl = xen_save_fl, - .restore_fl = xen_restore_fl, - .irq_disable = xen_irq_disable, - .irq_enable = xen_irq_enable, - .safe_halt = xen_safe_halt, - .halt = xen_halt, -#ifdef CONFIG_X86_64 - .adjust_exception_frame = xen_adjust_exception_frame, -#endif -}; - static const struct pv_apic_ops xen_apic_ops __initdata = { #ifdef CONFIG_X86_LOCAL_APIC .setup_boot_clock = paravirt_nop, @@ -1694,6 +1646,8 @@ asmlinkage void __init xen_start_kernel(void) if (!xen_start_info) return; + xen_domain_type = XEN_PV_DOMAIN; + BUG_ON(memcmp(xen_start_info->magic, "xen-3", 5) != 0); xen_setup_features(); @@ -1703,10 +1657,11 @@ asmlinkage void __init xen_start_kernel(void) pv_init_ops = xen_init_ops; pv_time_ops = xen_time_ops; pv_cpu_ops = xen_cpu_ops; - pv_irq_ops = xen_irq_ops; pv_apic_ops = xen_apic_ops; pv_mmu_ops = xen_mmu_ops; + xen_init_irq_ops(); + #ifdef CONFIG_X86_LOCAL_APIC /* * set up the basic apic ops. @@ -1737,7 +1692,7 @@ asmlinkage void __init xen_start_kernel(void) /* Prevent unwanted bits from being set in PTEs. */ __supported_pte_mask &= ~_PAGE_GLOBAL; - if (!is_initial_xendomain()) + if (!xen_initial_domain()) __supported_pte_mask &= ~(_PAGE_PWT | _PAGE_PCD); /* Don't do the full vcpu_info placement stuff until we have a @@ -1772,7 +1727,7 @@ asmlinkage void __init xen_start_kernel(void) boot_params.hdr.ramdisk_size = xen_start_info->mod_len; boot_params.hdr.cmd_line_ptr = __pa(xen_start_info->cmd_line); - if (!is_initial_xendomain()) { + if (!xen_initial_domain()) { add_preferred_console("xenboot", 0, NULL); add_preferred_console("tty", 0, NULL); add_preferred_console("hvc", 0, NULL); diff --git a/arch/x86/xen/irq.c b/arch/x86/xen/irq.c new file mode 100644 index 0000000..28b85ab --- /dev/null +++ b/arch/x86/xen/irq.c @@ -0,0 +1,143 @@ +#include <linux/hardirq.h> + +#include <xen/interface/xen.h> +#include <xen/interface/sched.h> +#include <xen/interface/vcpu.h> + +#include <asm/xen/hypercall.h> +#include <asm/xen/hypervisor.h> + +#include "xen-ops.h" + +/* + * Force a proper event-channel callback from Xen after clearing the + * callback mask. We do this in a very simple manner, by making a call + * down into Xen. The pending flag will be checked by Xen on return. + */ +void xen_force_evtchn_callback(void) +{ + (void)HYPERVISOR_xen_version(0, NULL); +} + +static void __init __xen_init_IRQ(void) +{ +#ifdef CONFIG_X86_64 + int i; + + /* Create identity vector->irq map */ + for(i = 0; i < NR_VECTORS; i++) { + int cpu; + + for_each_possible_cpu(cpu) + per_cpu(vector_irq, cpu)[i] = i; + } +#endif /* CONFIG_X86_64 */ + + xen_init_IRQ(); +} + +static unsigned long xen_save_fl(void) +{ + struct vcpu_info *vcpu; + unsigned long flags; + + vcpu = x86_read_percpu(xen_vcpu); + + /* flag has opposite sense of mask */ + flags = !vcpu->evtchn_upcall_mask; + + /* convert to IF type flag + -0 -> 0x00000000 + -1 -> 0xffffffff + */ + return (-flags) & X86_EFLAGS_IF; +} + +static void xen_restore_fl(unsigned long flags) +{ + struct vcpu_info *vcpu; + + /* convert from IF type flag */ + flags = !(flags & X86_EFLAGS_IF); + + /* There's a one instruction preempt window here. We need to + make sure we're don't switch CPUs between getting the vcpu + pointer and updating the mask. */ + preempt_disable(); + vcpu = x86_read_percpu(xen_vcpu); + vcpu->evtchn_upcall_mask = flags; + preempt_enable_no_resched(); + + /* Doesn't matter if we get preempted here, because any + pending event will get dealt with anyway. */ + + if (flags == 0) { + preempt_check_resched(); + barrier(); /* unmask then check (avoid races) */ + if (unlikely(vcpu->evtchn_upcall_pending)) + xen_force_evtchn_callback(); + } +} + +static void xen_irq_disable(void) +{ + /* There's a one instruction preempt window here. We need to + make sure we're don't switch CPUs between getting the vcpu + pointer and updating the mask. */ + preempt_disable(); + x86_read_percpu(xen_vcpu)->evtchn_upcall_mask = 1; + preempt_enable_no_resched(); +} + +static void xen_irq_enable(void) +{ + struct vcpu_info *vcpu; + + /* We don't need to worry about being preempted here, since + either a) interrupts are disabled, so no preemption, or b) + the caller is confused and is trying to re-enable interrupts + on an indeterminate processor. */ + + vcpu = x86_read_percpu(xen_vcpu); + vcpu->evtchn_upcall_mask = 0; + + /* Doesn't matter if we get preempted here, because any + pending event will get dealt with anyway. */ + + barrier(); /* unmask then check (avoid races) */ + if (unlikely(vcpu->evtchn_upcall_pending)) + xen_force_evtchn_callback(); +} + +static void xen_safe_halt(void) +{ + /* Blocking includes an implicit local_irq_enable(). */ + if (HYPERVISOR_sched_op(SCHEDOP_block, NULL) != 0) + BUG(); +} + +static void xen_halt(void) +{ + if (irqs_disabled()) + HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL); + else + xen_safe_halt(); +} + +static const struct pv_irq_ops xen_irq_ops __initdata = { + .init_IRQ = __xen_init_IRQ, + .save_fl = xen_save_fl, + .restore_fl = xen_restore_fl, + .irq_disable = xen_irq_disable, + .irq_enable = xen_irq_enable, + .safe_halt = xen_safe_halt, + .halt = xen_halt, +#ifdef CONFIG_X86_64 + .adjust_exception_frame = xen_adjust_exception_frame, +#endif +}; + +void __init xen_init_irq_ops() +{ + pv_irq_ops = xen_irq_ops; +} diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index aa37469..64e5868 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -40,6 +40,7 @@ */ #include <linux/sched.h> #include <linux/highmem.h> +#include <linux/debugfs.h> #include <linux/bug.h> #include <asm/pgtable.h> @@ -57,6 +58,61 @@ #include "multicalls.h" #include "mmu.h" +#include "debugfs.h" + +#define MMU_UPDATE_HISTO 30 + +#ifdef CONFIG_XEN_DEBUG_FS + +static struct { + u32 pgd_update; + u32 pgd_update_pinned; + u32 pgd_update_batched; + + u32 pud_update; + u32 pud_update_pinned; + u32 pud_update_batched; + + u32 pmd_update; + u32 pmd_update_pinned; + u32 pmd_update_batched; + + u32 pte_update; + u32 pte_update_pinned; + u32 pte_update_batched; + + u32 mmu_update; + u32 mmu_update_extended; + u32 mmu_update_histo[MMU_UPDATE_HISTO]; + + u32 prot_commit; + u32 prot_commit_batched; + + u32 set_pte_at; + u32 set_pte_at_batched; + u32 set_pte_at_pinned; + u32 set_pte_at_current; + u32 set_pte_at_kernel; +} mmu_stats; + +static u8 zero_stats; + +static inline void check_zero(void) +{ + if (unlikely(zero_stats)) { + memset(&mmu_stats, 0, sizeof(mmu_stats)); + zero_stats = 0; + } +} + +#define ADD_STATS(elem, val) \ + do { check_zero(); mmu_stats.elem += (val); } while(0) + +#else /* !CONFIG_XEN_DEBUG_FS */ + +#define ADD_STATS(elem, val) do { (void)(val); } while(0) + +#endif /* CONFIG_XEN_DEBUG_FS */ /* * Just beyond the highest usermode address. STACK_TOP_MAX has a @@ -229,25 +285,35 @@ void make_lowmem_page_readwrite(void *vaddr) } -static bool page_pinned(void *ptr) +static bool xen_page_pinned(void *ptr) { struct page *page = virt_to_page(ptr); return PagePinned(page); } -static void extend_mmu_update(const struct mmu_update *update) +static void xen_extend_mmu_update(const struct mmu_update *update) { struct multicall_space mcs; struct mmu_update *u; mcs = xen_mc_extend_args(__HYPERVISOR_mmu_update, sizeof(*u)); - if (mcs.mc != NULL) + if (mcs.mc != NULL) { + ADD_STATS(mmu_update_extended, 1); + ADD_STATS(mmu_update_histo[mcs.mc->args[1]], -1); + mcs.mc->args[1]++; - else { + + if (mcs.mc->args[1] < MMU_UPDATE_HISTO) + ADD_STATS(mmu_update_histo[mcs.mc->args[1]], 1); + else + ADD_STATS(mmu_update_histo[0], 1); + } else { + ADD_STATS(mmu_update, 1); mcs = __xen_mc_entry(sizeof(*u)); MULTI_mmu_update(mcs.mc, mcs.args, 1, NULL, DOMID_SELF); + ADD_STATS(mmu_update_histo[1], 1); } u = mcs.args; @@ -265,7 +331,9 @@ void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val) /* ptr may be ioremapped for 64-bit pagetable setup */ u.ptr = arbitrary_virt_to_machine(ptr).maddr; u.val = pmd_val_ma(val); - extend_mmu_update(&u); + xen_extend_mmu_update(&u); + + ADD_STATS(pmd_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU); xen_mc_issue(PARAVIRT_LAZY_MMU); @@ -274,13 +342,17 @@ void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val) void xen_set_pmd(pmd_t *ptr, pmd_t val) { + ADD_STATS(pmd_update, 1); + /* If page is not pinned, we can just update the entry directly */ - if (!page_pinned(ptr)) { + if (!xen_page_pinned(ptr)) { *ptr = val; return; } + ADD_STATS(pmd_update_pinned, 1); + xen_set_pmd_hyper(ptr, val); } @@ -300,12 +372,18 @@ void xen_set_pte_at(struct mm_struct *mm, unsigned long addr, if (mm == &init_mm) preempt_disable(); + ADD_STATS(set_pte_at, 1); +// ADD_STATS(set_pte_at_pinned, xen_page_pinned(ptep)); + ADD_STATS(set_pte_at_current, mm == current->mm); + ADD_STATS(set_pte_at_kernel, mm == &init_mm); + if (mm == current->mm || mm == &init_mm) { if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) { struct multicall_space mcs; mcs = xen_mc_entry(0); MULTI_update_va_mapping(mcs.mc, addr, pteval, 0); + ADD_STATS(set_pte_at_batched, 1); xen_mc_issue(PARAVIRT_LAZY_MMU); goto out; } else @@ -334,7 +412,10 @@ void xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr, u.ptr = virt_to_machine(ptep).maddr | MMU_PT_UPDATE_PRESERVE_AD; u.val = pte_val_ma(pte); - extend_mmu_update(&u); + xen_extend_mmu_update(&u); + + ADD_STATS(prot_commit, 1); + ADD_STATS(prot_commit_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU); xen_mc_issue(PARAVIRT_LAZY_MMU); } @@ -400,7 +481,9 @@ void xen_set_pud_hyper(pud_t *ptr, pud_t val) /* ptr may be ioremapped for 64-bit pagetable setup */ u.ptr = arbitrary_virt_to_machine(ptr).maddr; u.val = pud_val_ma(val); - extend_mmu_update(&u); + xen_extend_mmu_update(&u); + + ADD_STATS(pud_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU); xen_mc_issue(PARAVIRT_LAZY_MMU); @@ -409,18 +492,26 @@ void xen_set_pud_hyper(pud_t *ptr, pud_t val) void xen_set_pud(pud_t *ptr, pud_t val) { + ADD_STATS(pud_update, 1); + /* If page is not pinned, we can just update the entry directly */ - if (!page_pinned(ptr)) { + if (!xen_page_pinned(ptr)) { *ptr = val; return; } + ADD_STATS(pud_update_pinned, 1); + xen_set_pud_hyper(ptr, val); } void xen_set_pte(pte_t *ptep, pte_t pte) { + ADD_STATS(pte_update, 1); +// ADD_STATS(pte_update_pinned, xen_page_pinned(ptep)); + ADD_STATS(pte_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU); + #ifdef CONFIG_X86_PAE ptep->pte_high = pte.pte_high; smp_wmb(); @@ -490,7 +581,7 @@ static void __xen_set_pgd_hyper(pgd_t *ptr, pgd_t val) u.ptr = virt_to_machine(ptr).maddr; u.val = pgd_val_ma(val); - extend_mmu_update(&u); + xen_extend_mmu_update(&u); } /* @@ -517,17 +608,22 @@ void xen_set_pgd(pgd_t *ptr, pgd_t val) { pgd_t *user_ptr = xen_get_user_pgd(ptr); + ADD_STATS(pgd_update, 1); + /* If page is not pinned, we can just update the entry directly */ - if (!page_pinned(ptr)) { + if (!xen_page_pinned(ptr)) { *ptr = val; if (user_ptr) { - WARN_ON(page_pinned(user_ptr)); + WARN_ON(xen_page_pinned(user_ptr)); *user_ptr = val; } return; } + ADD_STATS(pgd_update_pinned, 1); + ADD_STATS(pgd_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU); + /* If it's pinned, then we can at least batch the kernel and user updates together. */ xen_mc_batch(); @@ -555,8 +651,8 @@ void xen_set_pgd(pgd_t *ptr, pgd_t val) * For 64-bit, we must skip the Xen hole in the middle of the address * space, just after the big x86-64 virtual hole. */ -static int pgd_walk(pgd_t *pgd, int (*func)(struct page *, enum pt_level), - unsigned long limit) +static int xen_pgd_walk(pgd_t *pgd, int (*func)(struct page *, enum pt_level), + unsigned long limit) { int flush = 0; unsigned hole_low, hole_high; @@ -590,8 +686,6 @@ static int pgd_walk(pgd_t *pgd, int (*func)(struct page *, enum pt_level), pmdidx_limit = 0; #endif - flush |= (*func)(virt_to_page(pgd), PT_PGD); - for (pgdidx = 0; pgdidx <= pgdidx_limit; pgdidx++) { pud_t *pud; @@ -637,16 +731,22 @@ static int pgd_walk(pgd_t *pgd, int (*func)(struct page *, enum pt_level), } } } + out: + /* Do the top level last, so that the callbacks can use it as + a cue to do final things like tlb flushes. */ + flush |= (*func)(virt_to_page(pgd), PT_PGD); return flush; } -static spinlock_t *lock_pte(struct page *page) +/* If we're using split pte locks, then take the page's lock and + return a pointer to it. Otherwise return NULL. */ +static spinlock_t *xen_pte_lock(struct page *page) { spinlock_t *ptl = NULL; -#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS +#if USE_SPLIT_PTLOCKS ptl = __pte_lockptr(page); spin_lock(ptl); #endif @@ -654,7 +754,7 @@ static spinlock_t *lock_pte(struct page *page) return ptl; } -static void do_unlock(void *v) +static void xen_pte_unlock(void *v) { spinlock_t *ptl = v; spin_unlock(ptl); @@ -672,7 +772,7 @@ static void xen_do_pin(unsigned level, unsigned long pfn) MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF); } -static int pin_page(struct page *page, enum pt_level level) +static int xen_pin_page(struct page *page, enum pt_level level) { unsigned pgfl = TestSetPagePinned(page); int flush; @@ -691,21 +791,40 @@ static int pin_page(struct page *page, enum pt_level level) flush = 0; + /* + * We need to hold the pagetable lock between the time + * we make the pagetable RO and when we actually pin + * it. If we don't, then other users may come in and + * attempt to update the pagetable by writing it, + * which will fail because the memory is RO but not + * pinned, so Xen won't do the trap'n'emulate. + * + * If we're using split pte locks, we can't hold the + * entire pagetable's worth of locks during the + * traverse, because we may wrap the preempt count (8 + * bits). The solution is to mark RO and pin each PTE + * page while holding the lock. This means the number + * of locks we end up holding is never more than a + * batch size (~32 entries, at present). + * + * If we're not using split pte locks, we needn't pin + * the PTE pages independently, because we're + * protected by the overall pagetable lock. + */ ptl = NULL; if (level == PT_PTE) - ptl = lock_pte(page); + ptl = xen_pte_lock(page); MULTI_update_va_mapping(mcs.mc, (unsigned long)pt, pfn_pte(pfn, PAGE_KERNEL_RO), level == PT_PGD ? UVMF_TLB_FLUSH : 0); - if (level == PT_PTE) + if (ptl) { xen_do_pin(MMUEXT_PIN_L1_TABLE, pfn); - if (ptl) { /* Queue a deferred unlock for when this batch is completed. */ - xen_mc_callback(do_unlock, ptl); + xen_mc_callback(xen_pte_unlock, ptl); } } @@ -719,7 +838,7 @@ void xen_pgd_pin(pgd_t *pgd) { xen_mc_batch(); - if (pgd_walk(pgd, pin_page, USER_LIMIT)) { + if (xen_pgd_walk(pgd, xen_pin_page, USER_LIMIT)) { /* re-enable interrupts for kmap_flush_unused */ xen_mc_issue(0); kmap_flush_unused(); @@ -733,14 +852,14 @@ void xen_pgd_pin(pgd_t *pgd) xen_do_pin(MMUEXT_PIN_L4_TABLE, PFN_DOWN(__pa(pgd))); if (user_pgd) { - pin_page(virt_to_page(user_pgd), PT_PGD); + xen_pin_page(virt_to_page(user_pgd), PT_PGD); xen_do_pin(MMUEXT_PIN_L4_TABLE, PFN_DOWN(__pa(user_pgd))); } } #else /* CONFIG_X86_32 */ #ifdef CONFIG_X86_PAE /* Need to make sure unshared kernel PMD is pinnable */ - pin_page(virt_to_page(pgd_page(pgd[pgd_index(TASK_SIZE)])), PT_PMD); + xen_pin_page(virt_to_page(pgd_page(pgd[pgd_index(TASK_SIZE)])), PT_PMD); #endif xen_do_pin(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(pgd))); #endif /* CONFIG_X86_64 */ @@ -775,7 +894,7 @@ void xen_mm_pin_all(void) * that's before we have page structures to store the bits. So do all * the book-keeping now. */ -static __init int mark_pinned(struct page *page, enum pt_level level) +static __init int xen_mark_pinned(struct page *page, enum pt_level level) { SetPagePinned(page); return 0; @@ -783,10 +902,10 @@ static __init int mark_pinned(struct page *page, enum pt_level level) void __init xen_mark_init_mm_pinned(void) { - pgd_walk(init_mm.pgd, mark_pinned, FIXADDR_TOP); + xen_pgd_walk(init_mm.pgd, xen_mark_pinned, FIXADDR_TOP); } -static int unpin_page(struct page *page, enum pt_level level) +static int xen_unpin_page(struct page *page, enum pt_level level) { unsigned pgfl = TestClearPagePinned(page); @@ -796,10 +915,18 @@ static int unpin_page(struct page *page, enum pt_level level) spinlock_t *ptl = NULL; struct multicall_space mcs; + /* + * Do the converse to pin_page. If we're using split + * pte locks, we must be holding the lock for while + * the pte page is unpinned but still RO to prevent + * concurrent updates from seeing it in this + * partially-pinned state. + */ if (level == PT_PTE) { - ptl = lock_pte(page); + ptl = xen_pte_lock(page); - xen_do_pin(MMUEXT_UNPIN_TABLE, pfn); + if (ptl) + xen_do_pin(MMUEXT_UNPIN_TABLE, pfn); } mcs = __xen_mc_entry(0); @@ -810,7 +937,7 @@ static int unpin_page(struct page *page, enum pt_level level) if (ptl) { /* unlock when batch completed */ - xen_mc_callback(do_unlock, ptl); + xen_mc_callback(xen_pte_unlock, ptl); } } @@ -830,17 +957,17 @@ static void xen_pgd_unpin(pgd_t *pgd) if (user_pgd) { xen_do_pin(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(user_pgd))); - unpin_page(virt_to_page(user_pgd), PT_PGD); + xen_unpin_page(virt_to_page(user_pgd), PT_PGD); } } #endif #ifdef CONFIG_X86_PAE /* Need to make sure unshared kernel PMD is unpinned */ - pin_page(virt_to_page(pgd_page(pgd[pgd_index(TASK_SIZE)])), PT_PMD); + xen_unpin_page(virt_to_page(pgd_page(pgd[pgd_index(TASK_SIZE)])), PT_PMD); #endif - pgd_walk(pgd, unpin_page, USER_LIMIT); + xen_pgd_walk(pgd, xen_unpin_page, USER_LIMIT); xen_mc_issue(0); } @@ -907,7 +1034,7 @@ static void drop_other_mm_ref(void *info) } } -static void drop_mm_ref(struct mm_struct *mm) +static void xen_drop_mm_ref(struct mm_struct *mm) { cpumask_t mask; unsigned cpu; @@ -937,7 +1064,7 @@ static void drop_mm_ref(struct mm_struct *mm) smp_call_function_mask(mask, drop_other_mm_ref, mm, 1); } #else -static void drop_mm_ref(struct mm_struct *mm) +static void xen_drop_mm_ref(struct mm_struct *mm) { if (current->active_mm == mm) load_cr3(swapper_pg_dir); @@ -961,14 +1088,77 @@ static void drop_mm_ref(struct mm_struct *mm) void xen_exit_mmap(struct mm_struct *mm) { get_cpu(); /* make sure we don't move around */ - drop_mm_ref(mm); + xen_drop_mm_ref(mm); put_cpu(); spin_lock(&mm->page_table_lock); /* pgd may not be pinned in the error exit path of execve */ - if (page_pinned(mm->pgd)) + if (xen_page_pinned(mm->pgd)) xen_pgd_unpin(mm->pgd); spin_unlock(&mm->page_table_lock); } + +#ifdef CONFIG_XEN_DEBUG_FS + +static struct dentry *d_mmu_debug; + +static int __init xen_mmu_debugfs(void) +{ + struct dentry *d_xen = xen_init_debugfs(); + + if (d_xen == NULL) + return -ENOMEM; + + d_mmu_debug = debugfs_create_dir("mmu", d_xen); + + debugfs_create_u8("zero_stats", 0644, d_mmu_debug, &zero_stats); + + debugfs_create_u32("pgd_update", 0444, d_mmu_debug, &mmu_stats.pgd_update); + debugfs_create_u32("pgd_update_pinned", 0444, d_mmu_debug, + &mmu_stats.pgd_update_pinned); + debugfs_create_u32("pgd_update_batched", 0444, d_mmu_debug, + &mmu_stats.pgd_update_pinned); + + debugfs_create_u32("pud_update", 0444, d_mmu_debug, &mmu_stats.pud_update); + debugfs_create_u32("pud_update_pinned", 0444, d_mmu_debug, + &mmu_stats.pud_update_pinned); + debugfs_create_u32("pud_update_batched", 0444, d_mmu_debug, + &mmu_stats.pud_update_pinned); + + debugfs_create_u32("pmd_update", 0444, d_mmu_debug, &mmu_stats.pmd_update); + debugfs_create_u32("pmd_update_pinned", 0444, d_mmu_debug, + &mmu_stats.pmd_update_pinned); + debugfs_create_u32("pmd_update_batched", 0444, d_mmu_debug, + &mmu_stats.pmd_update_pinned); + + debugfs_create_u32("pte_update", 0444, d_mmu_debug, &mmu_stats.pte_update); +// debugfs_create_u32("pte_update_pinned", 0444, d_mmu_debug, +// &mmu_stats.pte_update_pinned); + debugfs_create_u32("pte_update_batched", 0444, d_mmu_debug, + &mmu_stats.pte_update_pinned); + + debugfs_create_u32("mmu_update", 0444, d_mmu_debug, &mmu_stats.mmu_update); + debugfs_create_u32("mmu_update_extended", 0444, d_mmu_debug, + &mmu_stats.mmu_update_extended); + xen_debugfs_create_u32_array("mmu_update_histo", 0444, d_mmu_debug, + mmu_stats.mmu_update_histo, 20); + + debugfs_create_u32("set_pte_at", 0444, d_mmu_debug, &mmu_stats.set_pte_at); + debugfs_create_u32("set_pte_at_batched", 0444, d_mmu_debug, + &mmu_stats.set_pte_at_batched); + debugfs_create_u32("set_pte_at_current", 0444, d_mmu_debug, + &mmu_stats.set_pte_at_current); + debugfs_create_u32("set_pte_at_kernel", 0444, d_mmu_debug, + &mmu_stats.set_pte_at_kernel); + + debugfs_create_u32("prot_commit", 0444, d_mmu_debug, &mmu_stats.prot_commit); + debugfs_create_u32("prot_commit_batched", 0444, d_mmu_debug, + &mmu_stats.prot_commit_batched); + + return 0; +} +fs_initcall(xen_mmu_debugfs); + +#endif /* CONFIG_XEN_DEBUG_FS */ diff --git a/arch/x86/xen/multicalls.c b/arch/x86/xen/multicalls.c index 9efd1c6..8ea8a0d 100644 --- a/arch/x86/xen/multicalls.c +++ b/arch/x86/xen/multicalls.c @@ -21,16 +21,20 @@ */ #include <linux/percpu.h> #include <linux/hardirq.h> +#include <linux/debugfs.h> #include <asm/xen/hypercall.h> #include "multicalls.h" +#include "debugfs.h" + +#define MC_BATCH 32 #define MC_DEBUG 1 -#define MC_BATCH 32 #define MC_ARGS (MC_BATCH * 16) + struct mc_buffer { struct multicall_entry entries[MC_BATCH]; #if MC_DEBUG @@ -47,6 +51,76 @@ struct mc_buffer { static DEFINE_PER_CPU(struct mc_buffer, mc_buffer); DEFINE_PER_CPU(unsigned long, xen_mc_irq_flags); +/* flush reasons 0- slots, 1- args, 2- callbacks */ +enum flush_reasons +{ + FL_SLOTS, + FL_ARGS, + FL_CALLBACKS, + + FL_N_REASONS +}; + +#ifdef CONFIG_XEN_DEBUG_FS +#define NHYPERCALLS 40 /* not really */ + +static struct { + unsigned histo[MC_BATCH+1]; + + unsigned issued; + unsigned arg_total; + unsigned hypercalls; + unsigned histo_hypercalls[NHYPERCALLS]; + + unsigned flush[FL_N_REASONS]; +} mc_stats; + +static u8 zero_stats; + +static inline void check_zero(void) +{ + if (unlikely(zero_stats)) { + memset(&mc_stats, 0, sizeof(mc_stats)); + zero_stats = 0; + } +} + +static void mc_add_stats(const struct mc_buffer *mc) +{ + int i; + + check_zero(); + + mc_stats.issued++; + mc_stats.hypercalls += mc->mcidx; + mc_stats.arg_total += mc->argidx; + + mc_stats.histo[mc->mcidx]++; + for(i = 0; i < mc->mcidx; i++) { + unsigned op = mc->entries[i].op; + if (op < NHYPERCALLS) + mc_stats.histo_hypercalls[op]++; + } +} + +static void mc_stats_flush(enum flush_reasons idx) +{ + check_zero(); + + mc_stats.flush[idx]++; +} + +#else /* !CONFIG_XEN_DEBUG_FS */ + +static inline void mc_add_stats(const struct mc_buffer *mc) +{ +} + +static inline void mc_stats_flush(enum flush_reasons idx) +{ +} +#endif /* CONFIG_XEN_DEBUG_FS */ + void xen_mc_flush(void) { struct mc_buffer *b = &__get_cpu_var(mc_buffer); @@ -60,6 +134,8 @@ void xen_mc_flush(void) something in the middle */ local_irq_save(flags); + mc_add_stats(b); + if (b->mcidx) { #if MC_DEBUG memcpy(b->debug, b->entries, @@ -115,6 +191,7 @@ struct multicall_space __xen_mc_entry(size_t args) if (b->mcidx == MC_BATCH || (argidx + args) > MC_ARGS) { + mc_stats_flush(b->mcidx == MC_BATCH ? FL_SLOTS : FL_ARGS); xen_mc_flush(); argidx = roundup(b->argidx, sizeof(u64)); } @@ -158,10 +235,44 @@ void xen_mc_callback(void (*fn)(void *), void *data) struct mc_buffer *b = &__get_cpu_var(mc_buffer); struct callback *cb; - if (b->cbidx == MC_BATCH) + if (b->cbidx == MC_BATCH) { + mc_stats_flush(FL_CALLBACKS); xen_mc_flush(); + } cb = &b->callbacks[b->cbidx++]; cb->fn = fn; cb->data = data; } + +#ifdef CONFIG_XEN_DEBUG_FS + +static struct dentry *d_mc_debug; + +static int __init xen_mc_debugfs(void) +{ + struct dentry *d_xen = xen_init_debugfs(); + + if (d_xen == NULL) + return -ENOMEM; + + d_mc_debug = debugfs_create_dir("multicalls", d_xen); + + debugfs_create_u8("zero_stats", 0644, d_mc_debug, &zero_stats); + + debugfs_create_u32("batches", 0444, d_mc_debug, &mc_stats.issued); + debugfs_create_u32("hypercalls", 0444, d_mc_debug, &mc_stats.hypercalls); + debugfs_create_u32("arg_total", 0444, d_mc_debug, &mc_stats.arg_total); + + xen_debugfs_create_u32_array("batch_histo", 0444, d_mc_debug, + mc_stats.histo, MC_BATCH); + xen_debugfs_create_u32_array("hypercall_histo", 0444, d_mc_debug, + mc_stats.histo_hypercalls, NHYPERCALLS); + xen_debugfs_create_u32_array("flush_reasons", 0444, d_mc_debug, + mc_stats.flush, FL_N_REASONS); + + return 0; +} +fs_initcall(xen_mc_debugfs); + +#endif /* CONFIG_XEN_DEBUG_FS */ diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c index d8faf79..d77da61 100644 --- a/arch/x86/xen/smp.c +++ b/arch/x86/xen/smp.c @@ -11,11 +11,8 @@ * useful topology information for the kernel to make use of. As a * result, all CPUs are treated as if they're single-core and * single-threaded. - * - * This does not handle HOTPLUG_CPU yet. */ #include <linux/sched.h> -#include <linux/kernel_stat.h> #include <linux/err.h> #include <linux/smp.h> @@ -36,8 +33,6 @@ #include "xen-ops.h" #include "mmu.h" -static void __cpuinit xen_init_lock_cpu(int cpu); - cpumask_t xen_cpu_initialized_map; static DEFINE_PER_CPU(int, resched_irq); @@ -64,11 +59,12 @@ static irqreturn_t xen_reschedule_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -static __cpuinit void cpu_bringup_and_idle(void) +static __cpuinit void cpu_bringup(void) { int cpu = smp_processor_id(); cpu_init(); + touch_softlockup_watchdog(); preempt_disable(); xen_enable_sysenter(); @@ -89,6 +85,11 @@ static __cpuinit void cpu_bringup_and_idle(void) local_irq_enable(); wmb(); /* make sure everything is out */ +} + +static __cpuinit void cpu_bringup_and_idle(void) +{ + cpu_bringup(); cpu_idle(); } @@ -212,8 +213,6 @@ static void __init xen_smp_prepare_cpus(unsigned int max_cpus) cpu_set(cpu, cpu_present_map); } - - //init_xenbus_allowed_cpumask(); } static __cpuinit int @@ -281,12 +280,6 @@ static int __cpuinit xen_cpu_up(unsigned int cpu) struct task_struct *idle = idle_task(cpu); int rc; -#if 0 - rc = cpu_up_check(cpu); - if (rc) - return rc; -#endif - #ifdef CONFIG_X86_64 /* Allocate node local memory for AP pdas */ WARN_ON(cpu == 0); @@ -339,6 +332,60 @@ static void xen_smp_cpus_done(unsigned int max_cpus) { } +#ifdef CONFIG_HOTPLUG_CPU +static int xen_cpu_disable(void) +{ + unsigned int cpu = smp_processor_id(); + if (cpu == 0) + return -EBUSY; + + cpu_disable_common(); + + load_cr3(swapper_pg_dir); + return 0; +} + +static void xen_cpu_die(unsigned int cpu) +{ + while (HYPERVISOR_vcpu_op(VCPUOP_is_up, cpu, NULL)) { + current->state = TASK_UNINTERRUPTIBLE; + schedule_timeout(HZ/10); + } + unbind_from_irqhandler(per_cpu(resched_irq, cpu), NULL); + unbind_from_irqhandler(per_cpu(callfunc_irq, cpu), NULL); + unbind_from_irqhandler(per_cpu(debug_irq, cpu), NULL); + unbind_from_irqhandler(per_cpu(callfuncsingle_irq, cpu), NULL); + xen_uninit_lock_cpu(cpu); + xen_teardown_timer(cpu); + + if (num_online_cpus() == 1) + alternatives_smp_switch(0); +} + +static void xen_play_dead(void) +{ + play_dead_common(); + HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL); + cpu_bringup(); +} + +#else /* !CONFIG_HOTPLUG_CPU */ +static int xen_cpu_disable(void) +{ + return -ENOSYS; +} + +static void xen_cpu_die(unsigned int cpu) +{ + BUG(); +} + +static void xen_play_dead(void) +{ + BUG(); +} + +#endif static void stop_self(void *v) { int cpu = smp_processor_id(); @@ -419,176 +466,16 @@ static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -struct xen_spinlock { - unsigned char lock; /* 0 -> free; 1 -> locked */ - unsigned short spinners; /* count of waiting cpus */ -}; - -static int xen_spin_is_locked(struct raw_spinlock *lock) -{ - struct xen_spinlock *xl = (struct xen_spinlock *)lock; - - return xl->lock != 0; -} - -static int xen_spin_is_contended(struct raw_spinlock *lock) -{ - struct xen_spinlock *xl = (struct xen_spinlock *)lock; - - /* Not strictly true; this is only the count of contended - lock-takers entering the slow path. */ - return xl->spinners != 0; -} - -static int xen_spin_trylock(struct raw_spinlock *lock) -{ - struct xen_spinlock *xl = (struct xen_spinlock *)lock; - u8 old = 1; - - asm("xchgb %b0,%1" - : "+q" (old), "+m" (xl->lock) : : "memory"); - - return old == 0; -} - -static DEFINE_PER_CPU(int, lock_kicker_irq) = -1; -static DEFINE_PER_CPU(struct xen_spinlock *, lock_spinners); - -static inline void spinning_lock(struct xen_spinlock *xl) -{ - __get_cpu_var(lock_spinners) = xl; - wmb(); /* set lock of interest before count */ - asm(LOCK_PREFIX " incw %0" - : "+m" (xl->spinners) : : "memory"); -} - -static inline void unspinning_lock(struct xen_spinlock *xl) -{ - asm(LOCK_PREFIX " decw %0" - : "+m" (xl->spinners) : : "memory"); - wmb(); /* decrement count before clearing lock */ - __get_cpu_var(lock_spinners) = NULL; -} - -static noinline int xen_spin_lock_slow(struct raw_spinlock *lock) -{ - struct xen_spinlock *xl = (struct xen_spinlock *)lock; - int irq = __get_cpu_var(lock_kicker_irq); - int ret; - - /* If kicker interrupts not initialized yet, just spin */ - if (irq == -1) - return 0; - - /* announce we're spinning */ - spinning_lock(xl); - - /* clear pending */ - xen_clear_irq_pending(irq); - - /* check again make sure it didn't become free while - we weren't looking */ - ret = xen_spin_trylock(lock); - if (ret) - goto out; - - /* block until irq becomes pending */ - xen_poll_irq(irq); - kstat_this_cpu.irqs[irq]++; - -out: - unspinning_lock(xl); - return ret; -} - -static void xen_spin_lock(struct raw_spinlock *lock) -{ - struct xen_spinlock *xl = (struct xen_spinlock *)lock; - int timeout; - u8 oldval; - - do { - timeout = 1 << 10; - - asm("1: xchgb %1,%0\n" - " testb %1,%1\n" - " jz 3f\n" - "2: rep;nop\n" - " cmpb $0,%0\n" - " je 1b\n" - " dec %2\n" - " jnz 2b\n" - "3:\n" - : "+m" (xl->lock), "=q" (oldval), "+r" (timeout) - : "1" (1) - : "memory"); - - } while (unlikely(oldval != 0 && !xen_spin_lock_slow(lock))); -} - -static noinline void xen_spin_unlock_slow(struct xen_spinlock *xl) -{ - int cpu; - - for_each_online_cpu(cpu) { - /* XXX should mix up next cpu selection */ - if (per_cpu(lock_spinners, cpu) == xl) { - xen_send_IPI_one(cpu, XEN_SPIN_UNLOCK_VECTOR); - break; - } - } -} - -static void xen_spin_unlock(struct raw_spinlock *lock) -{ - struct xen_spinlock *xl = (struct xen_spinlock *)lock; - - smp_wmb(); /* make sure no writes get moved after unlock */ - xl->lock = 0; /* release lock */ - - /* make sure unlock happens before kick */ - barrier(); - - if (unlikely(xl->spinners)) - xen_spin_unlock_slow(xl); -} - -static __cpuinit void xen_init_lock_cpu(int cpu) -{ - int irq; - const char *name; - - name = kasprintf(GFP_KERNEL, "spinlock%d", cpu); - irq = bind_ipi_to_irqhandler(XEN_SPIN_UNLOCK_VECTOR, - cpu, - xen_reschedule_interrupt, - IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING, - name, - NULL); - - if (irq >= 0) { - disable_irq(irq); /* make sure it's never delivered */ - per_cpu(lock_kicker_irq, cpu) = irq; - } - - printk("cpu %d spinlock event irq %d\n", cpu, irq); -} - -static void __init xen_init_spinlocks(void) -{ - pv_lock_ops.spin_is_locked = xen_spin_is_locked; - pv_lock_ops.spin_is_contended = xen_spin_is_contended; - pv_lock_ops.spin_lock = xen_spin_lock; - pv_lock_ops.spin_trylock = xen_spin_trylock; - pv_lock_ops.spin_unlock = xen_spin_unlock; -} - static const struct smp_ops xen_smp_ops __initdata = { .smp_prepare_boot_cpu = xen_smp_prepare_boot_cpu, .smp_prepare_cpus = xen_smp_prepare_cpus, - .cpu_up = xen_cpu_up, .smp_cpus_done = xen_smp_cpus_done, + .cpu_up = xen_cpu_up, + .cpu_die = xen_cpu_die, + .cpu_disable = xen_cpu_disable, + .play_dead = xen_play_dead, + .smp_send_stop = xen_smp_send_stop, .smp_send_reschedule = xen_smp_send_reschedule, diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c new file mode 100644 index 0000000..dd71e3a --- /dev/null +++ b/arch/x86/xen/spinlock.c @@ -0,0 +1,428 @@ +/* + * Split spinlock implementation out into its own file, so it can be + * compiled in a FTRACE-compatible way. + */ +#include <linux/kernel_stat.h> +#include <linux/spinlock.h> +#include <linux/debugfs.h> +#include <linux/log2.h> + +#include <asm/paravirt.h> + +#include <xen/interface/xen.h> +#include <xen/events.h> + +#include "xen-ops.h" +#include "debugfs.h" + +#ifdef CONFIG_XEN_DEBUG_FS +static struct xen_spinlock_stats +{ + u64 taken; + u32 taken_slow; + u32 taken_slow_nested; + u32 taken_slow_pickup; + u32 taken_slow_spurious; + u32 taken_slow_irqenable; + + u64 released; + u32 released_slow; + u32 released_slow_kicked; + +#define HISTO_BUCKETS 30 + u32 histo_spin_total[HISTO_BUCKETS+1]; + u32 histo_spin_spinning[HISTO_BUCKETS+1]; + u32 histo_spin_blocked[HISTO_BUCKETS+1]; + + u64 time_total; + u64 time_spinning; + u64 time_blocked; +} spinlock_stats; + +static u8 zero_stats; + +static unsigned lock_timeout = 1 << 10; +#define TIMEOUT lock_timeout + +static inline void check_zero(void) +{ + if (unlikely(zero_stats)) { + memset(&spinlock_stats, 0, sizeof(spinlock_stats)); + zero_stats = 0; + } +} + +#define ADD_STATS(elem, val) \ + do { check_zero(); spinlock_stats.elem += (val); } while(0) + +static inline u64 spin_time_start(void) +{ + return xen_clocksource_read(); +} + +static void __spin_time_accum(u64 delta, u32 *array) +{ + unsigned index = ilog2(delta); + + check_zero(); + + if (index < HISTO_BUCKETS) + array[index]++; + else + array[HISTO_BUCKETS]++; +} + +static inline void spin_time_accum_spinning(u64 start) +{ + u32 delta = xen_clocksource_read() - start; + + __spin_time_accum(delta, spinlock_stats.histo_spin_spinning); + spinlock_stats.time_spinning += delta; +} + +static inline void spin_time_accum_total(u64 start) +{ + u32 delta = xen_clocksource_read() - start; + + __spin_time_accum(delta, spinlock_stats.histo_spin_total); + spinlock_stats.time_total += delta; +} + +static inline void spin_time_accum_blocked(u64 start) +{ + u32 delta = xen_clocksource_read() - start; + + __spin_time_accum(delta, spinlock_stats.histo_spin_blocked); + spinlock_stats.time_blocked += delta; +} +#else /* !CONFIG_XEN_DEBUG_FS */ +#define TIMEOUT (1 << 10) +#define ADD_STATS(elem, val) do { (void)(val); } while(0) + +static inline u64 spin_time_start(void) +{ + return 0; +} + +static inline void spin_time_accum_total(u64 start) +{ +} +static inline void spin_time_accum_spinning(u64 start) +{ +} +static inline void spin_time_accum_blocked(u64 start) +{ +} +#endif /* CONFIG_XEN_DEBUG_FS */ + +struct xen_spinlock { + unsigned char lock; /* 0 -> free; 1 -> locked */ + unsigned short spinners; /* count of waiting cpus */ +}; + +static int xen_spin_is_locked(struct raw_spinlock *lock) +{ + struct xen_spinlock *xl = (struct xen_spinlock *)lock; + + return xl->lock != 0; +} + +static int xen_spin_is_contended(struct raw_spinlock *lock) +{ + struct xen_spinlock *xl = (struct xen_spinlock *)lock; + + /* Not strictly true; this is only the count of contended + lock-takers entering the slow path. */ + return xl->spinners != 0; +} + +static int xen_spin_trylock(struct raw_spinlock *lock) +{ + struct xen_spinlock *xl = (struct xen_spinlock *)lock; + u8 old = 1; + + asm("xchgb %b0,%1" + : "+q" (old), "+m" (xl->lock) : : "memory"); + + return old == 0; +} + +static DEFINE_PER_CPU(int, lock_kicker_irq) = -1; +static DEFINE_PER_CPU(struct xen_spinlock *, lock_spinners); + +/* + * Mark a cpu as interested in a lock. Returns the CPU's previous + * lock of interest, in case we got preempted by an interrupt. + */ +static inline struct xen_spinlock *spinning_lock(struct xen_spinlock *xl) +{ + struct xen_spinlock *prev; + + prev = __get_cpu_var(lock_spinners); + __get_cpu_var(lock_spinners) = xl; + + wmb(); /* set lock of interest before count */ + + asm(LOCK_PREFIX " incw %0" + : "+m" (xl->spinners) : : "memory"); + + return prev; +} + +/* + * Mark a cpu as no longer interested in a lock. Restores previous + * lock of interest (NULL for none). + */ +static inline void unspinning_lock(struct xen_spinlock *xl, struct xen_spinlock *prev) +{ + asm(LOCK_PREFIX " decw %0" + : "+m" (xl->spinners) : : "memory"); + wmb(); /* decrement count before restoring lock */ + __get_cpu_var(lock_spinners) = prev; +} + +static noinline int xen_spin_lock_slow(struct raw_spinlock *lock, bool irq_enable) +{ + struct xen_spinlock *xl = (struct xen_spinlock *)lock; + struct xen_spinlock *prev; + int irq = __get_cpu_var(lock_kicker_irq); + int ret; + unsigned long flags; + u64 start; + + /* If kicker interrupts not initialized yet, just spin */ + if (irq == -1) + return 0; + + start = spin_time_start(); + + /* announce we're spinning */ + prev = spinning_lock(xl); + + flags = __raw_local_save_flags(); + if (irq_enable) { + ADD_STATS(taken_slow_irqenable, 1); + raw_local_irq_enable(); + } + + ADD_STATS(taken_slow, 1); + ADD_STATS(taken_slow_nested, prev != NULL); + + do { + /* clear pending */ + xen_clear_irq_pending(irq); + + /* check again make sure it didn't become free while + we weren't looking */ + ret = xen_spin_trylock(lock); + if (ret) { + ADD_STATS(taken_slow_pickup, 1); + + /* + * If we interrupted another spinlock while it + * was blocking, make sure it doesn't block + * without rechecking the lock. + */ + if (prev != NULL) + xen_set_irq_pending(irq); + goto out; + } + + /* + * Block until irq becomes pending. If we're + * interrupted at this point (after the trylock but + * before entering the block), then the nested lock + * handler guarantees that the irq will be left + * pending if there's any chance the lock became free; + * xen_poll_irq() returns immediately if the irq is + * pending. + */ + xen_poll_irq(irq); + ADD_STATS(taken_slow_spurious, !xen_test_irq_pending(irq)); + } while (!xen_test_irq_pending(irq)); /* check for spurious wakeups */ + + kstat_this_cpu.irqs[irq]++; + +out: + raw_local_irq_restore(flags); + unspinning_lock(xl, prev); + spin_time_accum_blocked(start); + + return ret; +} + +static inline void __xen_spin_lock(struct raw_spinlock *lock, bool irq_enable) +{ + struct xen_spinlock *xl = (struct xen_spinlock *)lock; + unsigned timeout; + u8 oldval; + u64 start_spin; + + ADD_STATS(taken, 1); + + start_spin = spin_time_start(); + + do { + u64 start_spin_fast = spin_time_start(); + + timeout = TIMEOUT; + + asm("1: xchgb %1,%0\n" + " testb %1,%1\n" + " jz 3f\n" + "2: rep;nop\n" + " cmpb $0,%0\n" + " je 1b\n" + " dec %2\n" + " jnz 2b\n" + "3:\n" + : "+m" (xl->lock), "=q" (oldval), "+r" (timeout) + : "1" (1) + : "memory"); + + spin_time_accum_spinning(start_spin_fast); + + } while (unlikely(oldval != 0 && + (TIMEOUT == ~0 || !xen_spin_lock_slow(lock, irq_enable)))); + + spin_time_accum_total(start_spin); +} + +static void xen_spin_lock(struct raw_spinlock *lock) +{ + __xen_spin_lock(lock, false); +} + +static void xen_spin_lock_flags(struct raw_spinlock *lock, unsigned long flags) +{ + __xen_spin_lock(lock, !raw_irqs_disabled_flags(flags)); +} + +static noinline void xen_spin_unlock_slow(struct xen_spinlock *xl) +{ + int cpu; + + ADD_STATS(released_slow, 1); + + for_each_online_cpu(cpu) { + /* XXX should mix up next cpu selection */ + if (per_cpu(lock_spinners, cpu) == xl) { + ADD_STATS(released_slow_kicked, 1); + xen_send_IPI_one(cpu, XEN_SPIN_UNLOCK_VECTOR); + break; + } + } +} + +static void xen_spin_unlock(struct raw_spinlock *lock) +{ + struct xen_spinlock *xl = (struct xen_spinlock *)lock; + + ADD_STATS(released, 1); + + smp_wmb(); /* make sure no writes get moved after unlock */ + xl->lock = 0; /* release lock */ + + /* make sure unlock happens before kick */ + barrier(); + + if (unlikely(xl->spinners)) + xen_spin_unlock_slow(xl); +} + +static irqreturn_t dummy_handler(int irq, void *dev_id) +{ + BUG(); + return IRQ_HANDLED; +} + +void __cpuinit xen_init_lock_cpu(int cpu) +{ + int irq; + const char *name; + + name = kasprintf(GFP_KERNEL, "spinlock%d", cpu); + irq = bind_ipi_to_irqhandler(XEN_SPIN_UNLOCK_VECTOR, + cpu, + dummy_handler, + IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING, + name, + NULL); + + if (irq >= 0) { + disable_irq(irq); /* make sure it's never delivered */ + per_cpu(lock_kicker_irq, cpu) = irq; + } + + printk("cpu %d spinlock event irq %d\n", cpu, irq); +} + +void xen_uninit_lock_cpu(int cpu) +{ + unbind_from_irqhandler(per_cpu(lock_kicker_irq, cpu), NULL); +} + +void __init xen_init_spinlocks(void) +{ + pv_lock_ops.spin_is_locked = xen_spin_is_locked; + pv_lock_ops.spin_is_contended = xen_spin_is_contended; + pv_lock_ops.spin_lock = xen_spin_lock; + pv_lock_ops.spin_lock_flags = xen_spin_lock_flags; + pv_lock_ops.spin_trylock = xen_spin_trylock; + pv_lock_ops.spin_unlock = xen_spin_unlock; +} + +#ifdef CONFIG_XEN_DEBUG_FS + +static struct dentry *d_spin_debug; + +static int __init xen_spinlock_debugfs(void) +{ + struct dentry *d_xen = xen_init_debugfs(); + + if (d_xen == NULL) + return -ENOMEM; + + d_spin_debug = debugfs_create_dir("spinlocks", d_xen); + + debugfs_create_u8("zero_stats", 0644, d_spin_debug, &zero_stats); + + debugfs_create_u32("timeout", 0644, d_spin_debug, &lock_timeout); + + debugfs_create_u64("taken", 0444, d_spin_debug, &spinlock_stats.taken); + debugfs_create_u32("taken_slow", 0444, d_spin_debug, + &spinlock_stats.taken_slow); + debugfs_create_u32("taken_slow_nested", 0444, d_spin_debug, + &spinlock_stats.taken_slow_nested); + debugfs_create_u32("taken_slow_pickup", 0444, d_spin_debug, + &spinlock_stats.taken_slow_pickup); + debugfs_create_u32("taken_slow_spurious", 0444, d_spin_debug, + &spinlock_stats.taken_slow_spurious); + debugfs_create_u32("taken_slow_irqenable", 0444, d_spin_debug, + &spinlock_stats.taken_slow_irqenable); + + debugfs_create_u64("released", 0444, d_spin_debug, &spinlock_stats.released); + debugfs_create_u32("released_slow", 0444, d_spin_debug, + &spinlock_stats.released_slow); + debugfs_create_u32("released_slow_kicked", 0444, d_spin_debug, + &spinlock_stats.released_slow_kicked); + + debugfs_create_u64("time_spinning", 0444, d_spin_debug, + &spinlock_stats.time_spinning); + debugfs_create_u64("time_blocked", 0444, d_spin_debug, + &spinlock_stats.time_blocked); + debugfs_create_u64("time_total", 0444, d_spin_debug, + &spinlock_stats.time_total); + + xen_debugfs_create_u32_array("histo_total", 0444, d_spin_debug, + spinlock_stats.histo_spin_total, HISTO_BUCKETS + 1); + xen_debugfs_create_u32_array("histo_spinning", 0444, d_spin_debug, + spinlock_stats.histo_spin_spinning, HISTO_BUCKETS + 1); + xen_debugfs_create_u32_array("histo_blocked", 0444, d_spin_debug, + spinlock_stats.histo_spin_blocked, HISTO_BUCKETS + 1); + + return 0; +} +fs_initcall(xen_spinlock_debugfs); + +#endif /* CONFIG_XEN_DEBUG_FS */ diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index 685b774..004ba86 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -30,8 +30,6 @@ #define TIMER_SLOP 100000 #define NS_PER_TICK (1000000000LL / HZ) -static cycle_t xen_clocksource_read(void); - /* runstate info updated by Xen */ static DEFINE_PER_CPU(struct vcpu_runstate_info, runstate); @@ -213,7 +211,7 @@ unsigned long xen_tsc_khz(void) return xen_khz; } -static cycle_t xen_clocksource_read(void) +cycle_t xen_clocksource_read(void) { struct pvclock_vcpu_time_info *src; cycle_t ret; @@ -452,6 +450,14 @@ void xen_setup_timer(int cpu) setup_runstate_info(cpu); } +void xen_teardown_timer(int cpu) +{ + struct clock_event_device *evt; + BUG_ON(cpu == 0); + evt = &per_cpu(xen_clock_events, cpu); + unbind_from_irqhandler(evt->irq, NULL); +} + void xen_setup_cpu_clockevents(void) { BUG_ON(preemptible()); diff --git a/arch/x86/xen/xen-asm_32.S b/arch/x86/xen/xen-asm_32.S index 2497a30..42786f5 100644 --- a/arch/x86/xen/xen-asm_32.S +++ b/arch/x86/xen/xen-asm_32.S @@ -298,7 +298,7 @@ check_events: push %eax push %ecx push %edx - call force_evtchn_callback + call xen_force_evtchn_callback pop %edx pop %ecx pop %eax diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S index 7f58304..3b9bda4 100644 --- a/arch/x86/xen/xen-asm_64.S +++ b/arch/x86/xen/xen-asm_64.S @@ -122,7 +122,7 @@ check_events: push %r9 push %r10 push %r11 - call force_evtchn_callback + call xen_force_evtchn_callback pop %r11 pop %r10 pop %r9 diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h index dd3c231..d7422dc 100644 --- a/arch/x86/xen/xen-ops.h +++ b/arch/x86/xen/xen-ops.h @@ -2,6 +2,7 @@ #define XEN_OPS_H #include <linux/init.h> +#include <linux/clocksource.h> #include <linux/irqreturn.h> #include <xen/xen-ops.h> @@ -31,7 +32,10 @@ void xen_vcpu_restore(void); void __init xen_build_dynamic_phys_to_machine(void); +void xen_init_irq_ops(void); void xen_setup_timer(int cpu); +void xen_teardown_timer(int cpu); +cycle_t xen_clocksource_read(void); void xen_setup_cpu_clockevents(void); unsigned long xen_tsc_khz(void); void __init xen_time_init(void); @@ -50,6 +54,10 @@ void __init xen_setup_vcpu_info_placement(void); #ifdef CONFIG_SMP void xen_smp_init(void); +void __init xen_init_spinlocks(void); +__cpuinit void xen_init_lock_cpu(int cpu); +void xen_uninit_lock_cpu(int cpu); + extern cpumask_t xen_cpu_initialized_map; #else static inline void xen_smp_init(void) {} diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 3ca643c..d5e7532 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -1032,7 +1032,7 @@ static struct xenbus_driver blkfront = { static int __init xlblk_init(void) { - if (!is_running_on_xen()) + if (!xen_domain()) return -ENODEV; if (register_blkdev(XENVBD_MAJOR, DEV_NAME)) { diff --git a/drivers/char/hvc_xen.c b/drivers/char/hvc_xen.c index 6b70aa6..538ceea 100644 --- a/drivers/char/hvc_xen.c +++ b/drivers/char/hvc_xen.c @@ -108,8 +108,8 @@ static int __init xen_init(void) { struct hvc_struct *hp; - if (!is_running_on_xen() || - is_initial_xendomain() || + if (!xen_pv_domain() || + xen_initial_domain() || !xen_start_info->console.domU.evtchn) return -ENODEV; @@ -142,7 +142,7 @@ static void __exit xen_fini(void) static int xen_cons_init(void) { - if (!is_running_on_xen()) + if (!xen_pv_domain()) return 0; hvc_instantiate(HVC_COOKIE, 0, &hvc_ops); diff --git a/drivers/input/xen-kbdfront.c b/drivers/input/xen-kbdfront.c index 9ce3b3b..3ab6362 100644 --- a/drivers/input/xen-kbdfront.c +++ b/drivers/input/xen-kbdfront.c @@ -335,11 +335,11 @@ static struct xenbus_driver xenkbd = { static int __init xenkbd_init(void) { - if (!is_running_on_xen()) + if (!xen_domain()) return -ENODEV; /* Nothing to do if running in dom0. */ - if (is_initial_xendomain()) + if (xen_initial_domain()) return -ENODEV; return xenbus_register_frontend(&xenkbd); diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index c749bdb..3c3dd40 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -1794,10 +1794,10 @@ static struct xenbus_driver netfront = { static int __init netif_init(void) { - if (!is_running_on_xen()) + if (!xen_domain()) return -ENODEV; - if (is_initial_xendomain()) + if (xen_initial_domain()) return 0; printk(KERN_INFO "Initialising Xen virtual ethernet driver.\n"); @@ -1809,7 +1809,7 @@ module_init(netif_init); static void __exit netif_exit(void) { - if (is_initial_xendomain()) + if (xen_initial_domain()) return; xenbus_unregister_driver(&netfront); diff --git a/drivers/video/xen-fbfront.c b/drivers/video/xen-fbfront.c index 47ed39b..a463b3d 100644 --- a/drivers/video/xen-fbfront.c +++ b/drivers/video/xen-fbfront.c @@ -680,11 +680,11 @@ static struct xenbus_driver xenfb = { static int __init xenfb_init(void) { - if (!is_running_on_xen()) + if (!xen_domain()) return -ENODEV; /* Nothing to do if running in dom0. */ - if (is_initial_xendomain()) + if (xen_initial_domain()) return -ENODEV; return xenbus_register_frontend(&xenfb); diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile index 363286c..d2a8fdf 100644 --- a/drivers/xen/Makefile +++ b/drivers/xen/Makefile @@ -1,4 +1,5 @@ obj-y += grant-table.o features.o events.o manage.o obj-y += xenbus/ +obj-$(CONFIG_HOTPLUG_CPU) += cpu_hotplug.o obj-$(CONFIG_XEN_XENCOMM) += xencomm.o obj-$(CONFIG_XEN_BALLOON) += balloon.o diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c index 2e15da5..a51f3e1 100644 --- a/drivers/xen/balloon.c +++ b/drivers/xen/balloon.c @@ -226,9 +226,8 @@ static int increase_reservation(unsigned long nr_pages) } set_xen_guest_handle(reservation.extent_start, frame_list); - reservation.nr_extents = nr_pages; - rc = HYPERVISOR_memory_op( - XENMEM_populate_physmap, &reservation); + reservation.nr_extents = nr_pages; + rc = HYPERVISOR_memory_op(XENMEM_populate_physmap, &reservation); if (rc < nr_pages) { if (rc > 0) { int ret; @@ -236,7 +235,7 @@ static int increase_reservation(unsigned long nr_pages) /* We hit the Xen hard limit: reprobe. */ reservation.nr_extents = rc; ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, - &reservation); + &reservation); BUG_ON(ret != rc); } if (rc >= 0) @@ -420,7 +419,7 @@ static int __init balloon_init(void) unsigned long pfn; struct page *page; - if (!is_running_on_xen()) + if (!xen_pv_domain()) return -ENODEV; pr_info("xen_balloon: Initialising balloon driver.\n"); @@ -464,136 +463,13 @@ static void balloon_exit(void) module_exit(balloon_exit); -static void balloon_update_driver_allowance(long delta) -{ - unsigned long flags; - - spin_lock_irqsave(&balloon_lock, flags); - balloon_stats.driver_pages += delta; - spin_unlock_irqrestore(&balloon_lock, flags); -} - -static int dealloc_pte_fn( - pte_t *pte, struct page *pmd_page, unsigned long addr, void *data) -{ - unsigned long mfn = pte_mfn(*pte); - int ret; - struct xen_memory_reservation reservation = { - .nr_extents = 1, - .extent_order = 0, - .domid = DOMID_SELF - }; - set_xen_guest_handle(reservation.extent_start, &mfn); - set_pte_at(&init_mm, addr, pte, __pte_ma(0ull)); - set_phys_to_machine(__pa(addr) >> PAGE_SHIFT, INVALID_P2M_ENTRY); - ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation); - BUG_ON(ret != 1); - return 0; -} - -static struct page **alloc_empty_pages_and_pagevec(int nr_pages) -{ - unsigned long vaddr, flags; - struct page *page, **pagevec; - int i, ret; - - pagevec = kmalloc(sizeof(page) * nr_pages, GFP_KERNEL); - if (pagevec == NULL) - return NULL; - - for (i = 0; i < nr_pages; i++) { - page = pagevec[i] = alloc_page(GFP_KERNEL); - if (page == NULL) - goto err; - - vaddr = (unsigned long)page_address(page); - - scrub_page(page); - - spin_lock_irqsave(&balloon_lock, flags); - - if (xen_feature(XENFEAT_auto_translated_physmap)) { - unsigned long gmfn = page_to_pfn(page); - struct xen_memory_reservation reservation = { - .nr_extents = 1, - .extent_order = 0, - .domid = DOMID_SELF - }; - set_xen_guest_handle(reservation.extent_start, &gmfn); - ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, - &reservation); - if (ret == 1) - ret = 0; /* success */ - } else { - ret = apply_to_page_range(&init_mm, vaddr, PAGE_SIZE, - dealloc_pte_fn, NULL); - } - - if (ret != 0) { - spin_unlock_irqrestore(&balloon_lock, flags); - __free_page(page); - goto err; - } - - totalram_pages = --balloon_stats.current_pages; - - spin_unlock_irqrestore(&balloon_lock, flags); - } - - out: - schedule_work(&balloon_worker); - flush_tlb_all(); - return pagevec; - - err: - spin_lock_irqsave(&balloon_lock, flags); - while (--i >= 0) - balloon_append(pagevec[i]); - spin_unlock_irqrestore(&balloon_lock, flags); - kfree(pagevec); - pagevec = NULL; - goto out; -} - -static void free_empty_pages_and_pagevec(struct page **pagevec, int nr_pages) -{ - unsigned long flags; - int i; - - if (pagevec == NULL) - return; - - spin_lock_irqsave(&balloon_lock, flags); - for (i = 0; i < nr_pages; i++) { - BUG_ON(page_count(pagevec[i]) != 1); - balloon_append(pagevec[i]); - } - spin_unlock_irqrestore(&balloon_lock, flags); - - kfree(pagevec); - - schedule_work(&balloon_worker); -} - -static void balloon_release_driver_page(struct page *page) -{ - unsigned long flags; - - spin_lock_irqsave(&balloon_lock, flags); - balloon_append(page); - balloon_stats.driver_pages--; - spin_unlock_irqrestore(&balloon_lock, flags); - - schedule_work(&balloon_worker); -} - - -#define BALLOON_SHOW(name, format, args...) \ - static ssize_t show_##name(struct sys_device *dev, \ - char *buf) \ - { \ - return sprintf(buf, format, ##args); \ - } \ +#define BALLOON_SHOW(name, format, args...) \ + static ssize_t show_##name(struct sys_device *dev, \ + struct sysdev_attribute *attr, \ + char *buf) \ + { \ + return sprintf(buf, format, ##args); \ + } \ static SYSDEV_ATTR(name, S_IRUGO, show_##name, NULL) BALLOON_SHOW(current_kb, "%lu\n", PAGES2KB(balloon_stats.current_pages)); @@ -604,7 +480,8 @@ BALLOON_SHOW(hard_limit_kb, (balloon_stats.hard_limit!=~0UL) ? PAGES2KB(balloon_stats.hard_limit) : 0); BALLOON_SHOW(driver_kb, "%lu\n", PAGES2KB(balloon_stats.driver_pages)); -static ssize_t show_target_kb(struct sys_device *dev, char *buf) +static ssize_t show_target_kb(struct sys_device *dev, struct sysdev_attribute *attr, + char *buf) { return sprintf(buf, "%lu\n", PAGES2KB(balloon_stats.target_pages)); } @@ -614,19 +491,14 @@ static ssize_t store_target_kb(struct sys_device *dev, const char *buf, size_t count) { - char memstring[64], *endchar; + char *endchar; unsigned long long target_bytes; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (count <= 1) - return -EBADMSG; /* runt */ - if (count > sizeof(memstring)) - return -EFBIG; /* too long */ - strcpy(memstring, buf); + target_bytes = memparse(buf, &endchar); - target_bytes = memparse(memstring, &endchar); balloon_set_new_target(target_bytes >> PAGE_SHIFT); return count; @@ -694,20 +566,4 @@ static int register_balloon(struct sys_device *sysdev) return error; } -static void unregister_balloon(struct sys_device *sysdev) -{ - int i; - - sysfs_remove_group(&sysdev->kobj, &balloon_info_group); - for (i = 0; i < ARRAY_SIZE(balloon_attrs); i++) - sysdev_remove_file(sysdev, balloon_attrs[i]); - sysdev_unregister(sysdev); - sysdev_class_unregister(&balloon_sysdev_class); -} - -static void balloon_sysfs_exit(void) -{ - unregister_balloon(&balloon_sysdev); -} - MODULE_LICENSE("GPL"); diff --git a/drivers/xen/cpu_hotplug.c b/drivers/xen/cpu_hotplug.c new file mode 100644 index 0000000..565280e --- /dev/null +++ b/drivers/xen/cpu_hotplug.c @@ -0,0 +1,90 @@ +#include <linux/notifier.h> + +#include <xen/xenbus.h> + +#include <asm-x86/xen/hypervisor.h> +#include <asm/cpu.h> + +static void enable_hotplug_cpu(int cpu) +{ + if (!cpu_present(cpu)) + arch_register_cpu(cpu); + + cpu_set(cpu, cpu_present_map); +} + +static void disable_hotplug_cpu(int cpu) +{ + if (cpu_present(cpu)) + arch_unregister_cpu(cpu); + + cpu_clear(cpu, cpu_present_map); +} + +static void vcpu_hotplug(unsigned int cpu) +{ + int err; + char dir[32], state[32]; + + if (!cpu_possible(cpu)) + return; + + sprintf(dir, "cpu/%u", cpu); + err = xenbus_scanf(XBT_NIL, dir, "availability", "%s", state); + if (err != 1) { + printk(KERN_ERR "XENBUS: Unable to read cpu state\n"); + return; + } + + if (strcmp(state, "online") == 0) { + enable_hotplug_cpu(cpu); + } else if (strcmp(state, "offline") == 0) { + (void)cpu_down(cpu); + disable_hotplug_cpu(cpu); + } else { + printk(KERN_ERR "XENBUS: unknown state(%s) on CPU%d\n", + state, cpu); + } +} + +static void handle_vcpu_hotplug_event(struct xenbus_watch *watch, + const char **vec, unsigned int len) +{ + unsigned int cpu; + char *cpustr; + const char *node = vec[XS_WATCH_PATH]; + + cpustr = strstr(node, "cpu/"); + if (cpustr != NULL) { + sscanf(cpustr, "cpu/%u", &cpu); + vcpu_hotplug(cpu); + } +} + +static int setup_cpu_watcher(struct notifier_block *notifier, + unsigned long event, void *data) +{ + static struct xenbus_watch cpu_watch = { + .node = "cpu", + .callback = handle_vcpu_hotplug_event}; + + (void)register_xenbus_watch(&cpu_watch); + + return NOTIFY_DONE; +} + +static int __init setup_vcpu_hotplug_event(void) +{ + static struct notifier_block xsn_cpu = { + .notifier_call = setup_cpu_watcher }; + + if (!xen_pv_domain()) + return -ENODEV; + + register_xenstore_notifier(&xsn_cpu); + + return 0; +} + +arch_initcall(setup_vcpu_hotplug_event); + diff --git a/drivers/xen/events.c b/drivers/xen/events.c index 0e0c285..c3290bc 100644 --- a/drivers/xen/events.c +++ b/drivers/xen/events.c @@ -84,17 +84,6 @@ static int irq_bindcount[NR_IRQS]; /* Xen will never allocate port zero for any purpose. */ #define VALID_EVTCHN(chn) ((chn) != 0) -/* - * Force a proper event-channel callback from Xen after clearing the - * callback mask. We do this in a very simple manner, by making a call - * down into Xen. The pending flag will be checked by Xen on return. - */ -void force_evtchn_callback(void) -{ - (void)HYPERVISOR_xen_version(0, NULL); -} -EXPORT_SYMBOL_GPL(force_evtchn_callback); - static struct irq_chip xen_dynamic_chip; /* Constructor for packed IRQ information. */ @@ -175,6 +164,12 @@ static inline void set_evtchn(int port) sync_set_bit(port, &s->evtchn_pending[0]); } +static inline int test_evtchn(int port) +{ + struct shared_info *s = HYPERVISOR_shared_info; + return sync_test_bit(port, &s->evtchn_pending[0]); +} + /** * notify_remote_via_irq - send event to remote end of event channel via irq @@ -365,6 +360,10 @@ static void unbind_from_irq(unsigned int irq) per_cpu(virq_to_irq, cpu_from_evtchn(evtchn)) [index_from_irq(irq)] = -1; break; + case IRQT_IPI: + per_cpu(ipi_to_irq, cpu_from_evtchn(evtchn)) + [index_from_irq(irq)] = -1; + break; default: break; } @@ -743,6 +742,25 @@ void xen_clear_irq_pending(int irq) clear_evtchn(evtchn); } +void xen_set_irq_pending(int irq) +{ + int evtchn = evtchn_from_irq(irq); + + if (VALID_EVTCHN(evtchn)) + set_evtchn(evtchn); +} + +bool xen_test_irq_pending(int irq) +{ + int evtchn = evtchn_from_irq(irq); + bool ret = false; + + if (VALID_EVTCHN(evtchn)) + ret = test_evtchn(evtchn); + + return ret; +} + /* Poll waiting for an irq to become pending. In the usual case, the irq will be disabled so it won't deliver an interrupt. */ void xen_poll_irq(int irq) diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c index e9e1116..06592b9 100644 --- a/drivers/xen/grant-table.c +++ b/drivers/xen/grant-table.c @@ -508,7 +508,7 @@ static int __devinit gnttab_init(void) unsigned int max_nr_glist_frames, nr_glist_frames; unsigned int nr_init_grefs; - if (!is_running_on_xen()) + if (!xen_domain()) return -ENODEV; nr_grant_frames = 1; diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c index 57ceb53..7f24a98 100644 --- a/drivers/xen/xenbus/xenbus_probe.c +++ b/drivers/xen/xenbus/xenbus_probe.c @@ -814,7 +814,7 @@ static int __init xenbus_probe_init(void) DPRINTK(""); err = -ENODEV; - if (!is_running_on_xen()) + if (!xen_domain()) goto out_error; /* Register ourselves with the kernel bus subsystem */ @@ -829,7 +829,7 @@ static int __init xenbus_probe_init(void) /* * Domain0 doesn't have a store_evtchn or store_mfn yet. */ - if (is_initial_xendomain()) { + if (xen_initial_domain()) { /* dom0 not yet supported */ } else { xenstored_ready = 1; @@ -846,7 +846,7 @@ static int __init xenbus_probe_init(void) goto out_unreg_back; } - if (!is_initial_xendomain()) + if (!xen_initial_domain()) xenbus_probe(NULL); return 0; @@ -937,7 +937,7 @@ static void wait_for_devices(struct xenbus_driver *xendrv) unsigned long timeout = jiffies + 10*HZ; struct device_driver *drv = xendrv ? &xendrv->driver : NULL; - if (!ready_to_wait_for_devices || !is_running_on_xen()) + if (!ready_to_wait_for_devices || !xen_domain()) return; while (exists_disconnected_device(drv)) { diff --git a/include/asm-x86/acpi.h b/include/asm-x86/acpi.h index 392e173..672e1e1 100644 --- a/include/asm-x86/acpi.h +++ b/include/asm-x86/acpi.h @@ -102,9 +102,6 @@ static inline void disable_acpi(void) acpi_noirq = 1; } -/* Fixmap pages to reserve for ACPI boot-time tables (see fixmap.h) */ -#define FIX_ACPI_PAGES 4 - extern int acpi_gsi_to_irq(u32 gsi, unsigned int *irq); static inline void acpi_noirq_set(void) { acpi_noirq = 1; } diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h index b73fea5..f06adac 100644 --- a/include/asm-x86/desc.h +++ b/include/asm-x86/desc.h @@ -24,6 +24,11 @@ static inline void fill_ldt(struct desc_struct *desc, desc->d = info->seg_32bit; desc->g = info->limit_in_pages; desc->base2 = (info->base_addr & 0xff000000) >> 24; + /* + * Don't allow setting of the lm bit. It is useless anyway + * because 64bit system calls require __USER_CS: + */ + desc->l = 0; } extern struct desc_ptr idt_descr; @@ -97,7 +102,15 @@ static inline int desc_empty(const void *ptr) native_write_gdt_entry(dt, entry, desc, type) #define write_idt_entry(dt, entry, g) \ native_write_idt_entry(dt, entry, g) -#endif + +static inline void paravirt_alloc_ldt(struct desc_struct *ldt, unsigned entries) +{ +} + +static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries) +{ +} +#endif /* CONFIG_PARAVIRT */ static inline void native_write_idt_entry(gate_desc *idt, int entry, const gate_desc *gate) @@ -338,20 +351,16 @@ static inline void set_system_intr_gate(unsigned int n, void *addr) _set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS); } -static inline void set_trap_gate(unsigned int n, void *addr) +static inline void set_system_trap_gate(unsigned int n, void *addr) { BUG_ON((unsigned)n > 0xFF); - _set_gate(n, GATE_TRAP, addr, 0, 0, __KERNEL_CS); + _set_gate(n, GATE_TRAP, addr, 0x3, 0, __KERNEL_CS); } -static inline void set_system_gate(unsigned int n, void *addr) +static inline void set_trap_gate(unsigned int n, void *addr) { BUG_ON((unsigned)n > 0xFF); -#ifdef CONFIG_X86_32 - _set_gate(n, GATE_TRAP, addr, 0x3, 0, __KERNEL_CS); -#else - _set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS); -#endif + _set_gate(n, GATE_TRAP, addr, 0, 0, __KERNEL_CS); } static inline void set_task_gate(unsigned int n, unsigned int gdt_entry) @@ -366,7 +375,7 @@ static inline void set_intr_gate_ist(int n, void *addr, unsigned ist) _set_gate(n, GATE_INTERRUPT, addr, 0, ist, __KERNEL_CS); } -static inline void set_system_gate_ist(int n, void *addr, unsigned ist) +static inline void set_system_intr_gate_ist(int n, void *addr, unsigned ist) { BUG_ON((unsigned)n > 0xFF); _set_gate(n, GATE_INTERRUPT, addr, 0x3, ist, __KERNEL_CS); diff --git a/include/asm-x86/es7000/mpparse.h b/include/asm-x86/es7000/mpparse.h index 7b5c889..ed5a3ca 100644 --- a/include/asm-x86/es7000/mpparse.h +++ b/include/asm-x86/es7000/mpparse.h @@ -5,6 +5,7 @@ extern int parse_unisys_oem (char *oemptr); extern int find_unisys_acpi_oem_table(unsigned long *oem_addr); +extern void unmap_unisys_acpi_oem_table(unsigned long oem_addr); extern void setup_unisys(void); #ifndef CONFIG_X86_GENERICARCH diff --git a/include/asm-x86/fixmap_32.h b/include/asm-x86/fixmap_32.h index 784e3e7..c6fde10 100644 --- a/include/asm-x86/fixmap_32.h +++ b/include/asm-x86/fixmap_32.h @@ -94,15 +94,11 @@ enum fixed_addresses { * can have a single pgd entry and a single pte table: */ #define NR_FIX_BTMAPS 64 -#define FIX_BTMAPS_NESTING 4 +#define FIX_BTMAPS_SLOTS 4 FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 - (__end_of_permanent_fixed_addresses & 255), - FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1, + FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1, FIX_WP_TEST, -#ifdef CONFIG_ACPI - FIX_ACPI_BEGIN, - FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1, -#endif #ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT FIX_OHCI1394_BASE, #endif diff --git a/include/asm-x86/fixmap_64.h b/include/asm-x86/fixmap_64.h index dafb24b..a6ff01d 100644 --- a/include/asm-x86/fixmap_64.h +++ b/include/asm-x86/fixmap_64.h @@ -49,26 +49,22 @@ enum fixed_addresses { #ifdef CONFIG_PARAVIRT FIX_PARAVIRT_BOOTMAP, #endif -#ifdef CONFIG_ACPI - FIX_ACPI_BEGIN, - FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1, -#endif + __end_of_permanent_fixed_addresses, #ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT FIX_OHCI1394_BASE, #endif - __end_of_permanent_fixed_addresses, /* * 256 temporary boot-time mappings, used by early_ioremap(), * before ioremap() is functional. * - * We round it up to the next 512 pages boundary so that we + * We round it up to the next 256 pages boundary so that we * can have a single pgd entry and a single pte table: */ #define NR_FIX_BTMAPS 64 -#define FIX_BTMAPS_NESTING 4 - FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 512 - - (__end_of_permanent_fixed_addresses & 511), - FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1, +#define FIX_BTMAPS_SLOTS 4 + FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 - + (__end_of_permanent_fixed_addresses & 255), + FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1, __end_of_fixed_addresses }; diff --git a/include/asm-x86/io.h b/include/asm-x86/io.h index 72b7719..a233f83 100644 --- a/include/asm-x86/io.h +++ b/include/asm-x86/io.h @@ -5,20 +5,6 @@ #include <linux/compiler.h> -/* - * early_ioremap() and early_iounmap() are for temporary early boot-time - * mappings, before the real ioremap() is functional. - * A boot-time mapping is currently limited to at most 16 pages. - */ -#ifndef __ASSEMBLY__ -extern void early_ioremap_init(void); -extern void early_ioremap_clear(void); -extern void early_ioremap_reset(void); -extern void *early_ioremap(unsigned long offset, unsigned long size); -extern void early_iounmap(void *addr, unsigned long size); -extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys); -#endif - #define build_mmio_read(name, size, type, reg, barrier) \ static inline type name(const volatile void __iomem *addr) \ { type ret; asm volatile("mov" size " %1,%0":reg (ret) \ @@ -97,6 +83,7 @@ extern void early_ioremap_init(void); extern void early_ioremap_clear(void); extern void early_ioremap_reset(void); extern void *early_ioremap(unsigned long offset, unsigned long size); +extern void *early_memremap(unsigned long offset, unsigned long size); extern void early_iounmap(void *addr, unsigned long size); extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys); diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h index 64429e9..ee6e086 100644 --- a/include/asm-x86/io_64.h +++ b/include/asm-x86/io_64.h @@ -165,9 +165,6 @@ static inline void *phys_to_virt(unsigned long address) #include <asm-generic/iomap.h> -extern void *early_ioremap(unsigned long addr, unsigned long size); -extern void early_iounmap(void *addr, unsigned long size); - /* * This one maps high address device memory and turns off caching for that area. * it's useful if some control registers are in such an area and write combining diff --git a/include/asm-x86/irqflags.h b/include/asm-x86/irqflags.h index 424acb4..2bdab21 100644 --- a/include/asm-x86/irqflags.h +++ b/include/asm-x86/irqflags.h @@ -166,27 +166,6 @@ static inline int raw_irqs_disabled(void) return raw_irqs_disabled_flags(flags); } -/* - * makes the traced hardirq state match with the machine state - * - * should be a rarely used function, only in places where its - * otherwise impossible to know the irq state, like in traps. - */ -static inline void trace_hardirqs_fixup_flags(unsigned long flags) -{ - if (raw_irqs_disabled_flags(flags)) - trace_hardirqs_off(); - else - trace_hardirqs_on(); -} - -static inline void trace_hardirqs_fixup(void) -{ - unsigned long flags = __raw_local_save_flags(); - - trace_hardirqs_fixup_flags(flags); -} - #else #ifdef CONFIG_X86_64 diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h index 5ec3ad3..fbbab66 100644 --- a/include/asm-x86/kdebug.h +++ b/include/asm-x86/kdebug.h @@ -27,10 +27,9 @@ extern void printk_address(unsigned long address, int reliable); extern void die(const char *, struct pt_regs *,long); extern int __must_check __die(const char *, struct pt_regs *, long); extern void show_registers(struct pt_regs *regs); -extern void __show_registers(struct pt_regs *, int all); extern void show_trace(struct task_struct *t, struct pt_regs *regs, unsigned long *sp, unsigned long bp); -extern void __show_regs(struct pt_regs *regs); +extern void __show_regs(struct pt_regs *regs, int all); extern void show_regs(struct pt_regs *regs); extern unsigned long oops_begin(void); extern void oops_end(unsigned long, struct pt_regs *, int signr); diff --git a/include/asm-x86/kprobes.h b/include/asm-x86/kprobes.h index bd84078..8a0748d 100644 --- a/include/asm-x86/kprobes.h +++ b/include/asm-x86/kprobes.h @@ -82,15 +82,6 @@ struct kprobe_ctlblk { struct prev_kprobe prev_kprobe; }; -/* trap3/1 are intr gates for kprobes. So, restore the status of IF, - * if necessary, before executing the original int3/1 (trap) handler. - */ -static inline void restore_interrupts(struct pt_regs *regs) -{ - if (regs->flags & X86_EFLAGS_IF) - local_irq_enable(); -} - extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr); extern int kprobe_exceptions_notify(struct notifier_block *self, unsigned long val, void *data); diff --git a/include/asm-x86/mach-default/mach_traps.h b/include/asm-x86/mach-default/mach_traps.h index de9ac3f..ff8778f 100644 --- a/include/asm-x86/mach-default/mach_traps.h +++ b/include/asm-x86/mach-default/mach_traps.h @@ -7,12 +7,6 @@ #include <asm/mc146818rtc.h> -static inline void clear_mem_error(unsigned char reason) -{ - reason = (reason & 0xf) | 4; - outb(reason, 0x61); -} - static inline unsigned char get_nmi_reason(void) { return inb(0x61); diff --git a/include/asm-x86/mmzone_64.h b/include/asm-x86/mmzone_64.h index 626b03a..6480f33 100644 --- a/include/asm-x86/mmzone_64.h +++ b/include/asm-x86/mmzone_64.h @@ -7,7 +7,7 @@ #ifdef CONFIG_NUMA -#define VIRTUAL_BUG_ON(x) +#include <linux/mmdebug.h> #include <asm/smp.h> @@ -29,7 +29,6 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr) { unsigned nid; VIRTUAL_BUG_ON(!memnodemap); - VIRTUAL_BUG_ON((addr >> memnode_shift) >= memnodemapsize); nid = memnodemap[addr >> memnode_shift]; VIRTUAL_BUG_ON(nid >= MAX_NUMNODES || !node_data[nid]); return nid; diff --git a/include/asm-x86/module.h b/include/asm-x86/module.h index 48dc3e0..864f200 100644 --- a/include/asm-x86/module.h +++ b/include/asm-x86/module.h @@ -52,8 +52,6 @@ struct mod_arch_specific {}; #define MODULE_PROC_FAMILY "EFFICEON " #elif defined CONFIG_MWINCHIPC6 #define MODULE_PROC_FAMILY "WINCHIPC6 " -#elif defined CONFIG_MWINCHIP2 -#define MODULE_PROC_FAMILY "WINCHIP2 " #elif defined CONFIG_MWINCHIP3D #define MODULE_PROC_FAMILY "WINCHIP3D " #elif defined CONFIG_MCYRIXIII diff --git a/include/asm-x86/nmi.h b/include/asm-x86/nmi.h index d5e715f..a53f829 100644 --- a/include/asm-x86/nmi.h +++ b/include/asm-x86/nmi.h @@ -15,10 +15,6 @@ */ int do_nmi_callback(struct pt_regs *regs, int cpu); -#ifdef CONFIG_X86_64 -extern void default_do_nmi(struct pt_regs *); -#endif - extern void die_nmi(char *str, struct pt_regs *regs, int do_panic); extern int check_nmi_watchdog(void); extern int nmi_watchdog_enabled; diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h index c915747..d4f1d57 100644 --- a/include/asm-x86/page.h +++ b/include/asm-x86/page.h @@ -179,6 +179,7 @@ static inline pteval_t native_pte_flags(pte_t pte) #endif /* CONFIG_PARAVIRT */ #define __pa(x) __phys_addr((unsigned long)(x)) +#define __pa_nodebug(x) __phys_addr_nodebug((unsigned long)(x)) /* __pa_symbol should be used for C visible symbols. This seems to be the official gcc blessed way to do such arithmetic. */ #define __pa_symbol(x) __pa(__phys_reloc_hide((unsigned long)(x))) @@ -188,9 +189,14 @@ static inline pteval_t native_pte_flags(pte_t pte) #define __boot_va(x) __va(x) #define __boot_pa(x) __pa(x) +/* + * virt_to_page(kaddr) returns a valid pointer if and only if + * virt_addr_valid(kaddr) returns true. + */ #define virt_to_page(kaddr) pfn_to_page(__pa(kaddr) >> PAGE_SHIFT) #define pfn_to_kaddr(pfn) __va((pfn) << PAGE_SHIFT) -#define virt_addr_valid(kaddr) pfn_valid(__pa(kaddr) >> PAGE_SHIFT) +extern bool __virt_addr_valid(unsigned long kaddr); +#define virt_addr_valid(kaddr) __virt_addr_valid((unsigned long) (kaddr)) #endif /* __ASSEMBLY__ */ diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h index 72f7305..e8d80d1 100644 --- a/include/asm-x86/page_32.h +++ b/include/asm-x86/page_32.h @@ -20,6 +20,12 @@ #endif #define THREAD_SIZE (PAGE_SIZE << THREAD_ORDER) +#define STACKFAULT_STACK 0 +#define DOUBLEFAULT_STACK 1 +#define NMI_STACK 0 +#define DEBUG_STACK 0 +#define MCE_STACK 0 +#define N_EXCEPTION_STACKS 1 #ifdef CONFIG_X86_PAE /* 44=32+12, the limit we can fit into an unsigned long pfn */ @@ -73,7 +79,12 @@ typedef struct page *pgtable_t; #endif #ifndef __ASSEMBLY__ -#define __phys_addr(x) ((x) - PAGE_OFFSET) +#define __phys_addr_nodebug(x) ((x) - PAGE_OFFSET) +#ifdef CONFIG_DEBUG_VIRTUAL +extern unsigned long __phys_addr(unsigned long); +#else +#define __phys_addr(x) __phys_addr_nodebug(x) +#endif #define __phys_reloc_hide(x) RELOC_HIDE((x), 0) #ifdef CONFIG_FLATMEM diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h index d7d358a..8d6ae2f 100644 --- a/include/asm-x86/paravirt.h +++ b/include/asm-x86/paravirt.h @@ -124,6 +124,9 @@ struct pv_cpu_ops { int entrynum, const void *desc, int size); void (*write_idt_entry)(gate_desc *, int entrynum, const gate_desc *gate); + void (*alloc_ldt)(struct desc_struct *ldt, unsigned entries); + void (*free_ldt)(struct desc_struct *ldt, unsigned entries); + void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t); void (*set_iopl_mask)(unsigned mask); @@ -325,6 +328,7 @@ struct pv_lock_ops { int (*spin_is_locked)(struct raw_spinlock *lock); int (*spin_is_contended)(struct raw_spinlock *lock); void (*spin_lock)(struct raw_spinlock *lock); + void (*spin_lock_flags)(struct raw_spinlock *lock, unsigned long flags); int (*spin_trylock)(struct raw_spinlock *lock); void (*spin_unlock)(struct raw_spinlock *lock); }; @@ -830,6 +834,16 @@ do { \ (aux) = __aux; \ } while (0) +static inline void paravirt_alloc_ldt(struct desc_struct *ldt, unsigned entries) +{ + PVOP_VCALL2(pv_cpu_ops.alloc_ldt, ldt, entries); +} + +static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries) +{ + PVOP_VCALL2(pv_cpu_ops.free_ldt, ldt, entries); +} + static inline void load_TR_desc(void) { PVOP_VCALL0(pv_cpu_ops.load_tr_desc); @@ -1394,6 +1408,12 @@ static __always_inline void __raw_spin_lock(struct raw_spinlock *lock) PVOP_VCALL1(pv_lock_ops.spin_lock, lock); } +static __always_inline void __raw_spin_lock_flags(struct raw_spinlock *lock, + unsigned long flags) +{ + PVOP_VCALL2(pv_lock_ops.spin_lock_flags, lock, flags); +} + static __always_inline int __raw_spin_trylock(struct raw_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_trylock, lock); diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index ed93245..182f9d4 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -15,7 +15,7 @@ #define _PAGE_BIT_PAT 7 /* on 4KB pages */ #define _PAGE_BIT_GLOBAL 8 /* Global TLB entry PPro+ */ #define _PAGE_BIT_UNUSED1 9 /* available for programmer */ -#define _PAGE_BIT_UNUSED2 10 +#define _PAGE_BIT_IOMAP 10 /* flag used to indicate IO mapping */ #define _PAGE_BIT_UNUSED3 11 #define _PAGE_BIT_PAT_LARGE 12 /* On 2MB or 1GB pages */ #define _PAGE_BIT_SPECIAL _PAGE_BIT_UNUSED1 @@ -32,7 +32,7 @@ #define _PAGE_PSE (_AT(pteval_t, 1) << _PAGE_BIT_PSE) #define _PAGE_GLOBAL (_AT(pteval_t, 1) << _PAGE_BIT_GLOBAL) #define _PAGE_UNUSED1 (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED1) -#define _PAGE_UNUSED2 (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED2) +#define _PAGE_IOMAP (_AT(pteval_t, 1) << _PAGE_BIT_IOMAP) #define _PAGE_UNUSED3 (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED3) #define _PAGE_PAT (_AT(pteval_t, 1) << _PAGE_BIT_PAT) #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE) @@ -99,6 +99,11 @@ #define __PAGE_KERNEL_LARGE_NOCACHE (__PAGE_KERNEL | _PAGE_CACHE_UC | _PAGE_PSE) #define __PAGE_KERNEL_LARGE_EXEC (__PAGE_KERNEL_EXEC | _PAGE_PSE) +#define __PAGE_KERNEL_IO (__PAGE_KERNEL | _PAGE_IOMAP) +#define __PAGE_KERNEL_IO_NOCACHE (__PAGE_KERNEL_NOCACHE | _PAGE_IOMAP) +#define __PAGE_KERNEL_IO_UC_MINUS (__PAGE_KERNEL_UC_MINUS | _PAGE_IOMAP) +#define __PAGE_KERNEL_IO_WC (__PAGE_KERNEL_WC | _PAGE_IOMAP) + #define PAGE_KERNEL __pgprot(__PAGE_KERNEL) #define PAGE_KERNEL_RO __pgprot(__PAGE_KERNEL_RO) #define PAGE_KERNEL_EXEC __pgprot(__PAGE_KERNEL_EXEC) @@ -113,6 +118,11 @@ #define PAGE_KERNEL_VSYSCALL __pgprot(__PAGE_KERNEL_VSYSCALL) #define PAGE_KERNEL_VSYSCALL_NOCACHE __pgprot(__PAGE_KERNEL_VSYSCALL_NOCACHE) +#define PAGE_KERNEL_IO __pgprot(__PAGE_KERNEL_IO) +#define PAGE_KERNEL_IO_NOCACHE __pgprot(__PAGE_KERNEL_IO_NOCACHE) +#define PAGE_KERNEL_IO_UC_MINUS __pgprot(__PAGE_KERNEL_IO_UC_MINUS) +#define PAGE_KERNEL_IO_WC __pgprot(__PAGE_KERNEL_IO_WC) + /* xwr */ #define __P000 PAGE_NONE #define __P001 PAGE_READONLY @@ -196,7 +206,7 @@ static inline int pte_exec(pte_t pte) static inline int pte_special(pte_t pte) { - return pte_val(pte) & _PAGE_SPECIAL; + return pte_flags(pte) & _PAGE_SPECIAL; } static inline unsigned long pte_pfn(pte_t pte) diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h index ac578f1..a202552 100644 --- a/include/asm-x86/ptrace.h +++ b/include/asm-x86/ptrace.h @@ -174,12 +174,8 @@ extern unsigned long profile_pc(struct pt_regs *regs); extern unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs); - -#ifdef CONFIG_X86_32 extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code, int si_code); -#endif - void signal_fault(struct pt_regs *regs, void __user *frame, char *where); extern long syscall_trace_enter(struct pt_regs *); diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 29324c1..a6afc29 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -50,12 +50,16 @@ extern struct { struct smp_ops { void (*smp_prepare_boot_cpu)(void); void (*smp_prepare_cpus)(unsigned max_cpus); - int (*cpu_up)(unsigned cpu); void (*smp_cpus_done)(unsigned max_cpus); void (*smp_send_stop)(void); void (*smp_send_reschedule)(int cpu); + int (*cpu_up)(unsigned cpu); + int (*cpu_disable)(void); + void (*cpu_die)(unsigned int cpu); + void (*play_dead)(void); + void (*send_call_func_ipi)(cpumask_t mask); void (*send_call_func_single_ipi)(int cpu); }; @@ -94,6 +98,21 @@ static inline int __cpu_up(unsigned int cpu) return smp_ops.cpu_up(cpu); } +static inline int __cpu_disable(void) +{ + return smp_ops.cpu_disable(); +} + +static inline void __cpu_die(unsigned int cpu) +{ + smp_ops.cpu_die(cpu); +} + +static inline void play_dead(void) +{ + smp_ops.play_dead(); +} + static inline void smp_send_reschedule(int cpu) { smp_ops.smp_send_reschedule(cpu); @@ -109,15 +128,20 @@ static inline void arch_send_call_function_ipi(cpumask_t mask) smp_ops.send_call_func_ipi(mask); } +void cpu_disable_common(void); void native_smp_prepare_boot_cpu(void); void native_smp_prepare_cpus(unsigned int max_cpus); void native_smp_cpus_done(unsigned int max_cpus); int native_cpu_up(unsigned int cpunum); +int native_cpu_disable(void); +void native_cpu_die(unsigned int cpu); +void native_play_dead(void); +void play_dead_common(void); + void native_send_call_func_ipi(cpumask_t mask); void native_send_call_func_single_ipi(int cpu); -extern int __cpu_disable(void); -extern void __cpu_die(unsigned int cpu); +extern void prefill_possible_map(void); void smp_store_cpu_info(int id); #define cpu_physical_id(cpu) per_cpu(x86_cpu_to_apicid, cpu) @@ -127,15 +151,11 @@ static inline int num_booting_cpus(void) { return cpus_weight(cpu_callout_map); } -#endif /* CONFIG_SMP */ - -#if defined(CONFIG_SMP) && defined(CONFIG_HOTPLUG_CPU) -extern void prefill_possible_map(void); #else static inline void prefill_possible_map(void) { } -#endif +#endif /* CONFIG_SMP */ extern unsigned disabled_cpus __cpuinitdata; @@ -205,9 +225,5 @@ static inline int hard_smp_processor_id(void) #endif /* CONFIG_X86_LOCAL_APIC */ -#ifdef CONFIG_HOTPLUG_CPU -extern void cpu_uninit(void); -#endif - #endif /* __ASSEMBLY__ */ #endif /* ASM_X86__SMP_H */ diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h index 93adae3..8badab0 100644 --- a/include/asm-x86/spinlock.h +++ b/include/asm-x86/spinlock.h @@ -182,8 +182,6 @@ static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock) } #endif -#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) - #ifdef CONFIG_PARAVIRT /* * Define virtualization-friendly old-style lock byte lock, for use in @@ -272,6 +270,13 @@ static __always_inline void __raw_spin_unlock(raw_spinlock_t *lock) { __ticket_spin_unlock(lock); } + +static __always_inline void __raw_spin_lock_flags(raw_spinlock_t *lock, + unsigned long flags) +{ + __raw_spin_lock(lock); +} + #endif /* CONFIG_PARAVIRT */ static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock) diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h index 34505dd..b20c894 100644 --- a/include/asm-x86/system.h +++ b/include/asm-x86/system.h @@ -64,7 +64,10 @@ do { \ \ /* regparm parameters for __switch_to(): */ \ [prev] "a" (prev), \ - [next] "d" (next)); \ + [next] "d" (next) \ + \ + : /* reloaded segment registers */ \ + "memory"); \ } while (0) /* diff --git a/include/asm-x86/tlbflush.h b/include/asm-x86/tlbflush.h index ef68b76..3cdd08b 100644 --- a/include/asm-x86/tlbflush.h +++ b/include/asm-x86/tlbflush.h @@ -119,6 +119,10 @@ static inline void native_flush_tlb_others(const cpumask_t *cpumask, { } +static inline void reset_lazy_tlbstate(void) +{ +} + #else /* SMP */ #include <asm/smp.h> @@ -151,6 +155,12 @@ struct tlb_state { char __cacheline_padding[L1_CACHE_BYTES-8]; }; DECLARE_PER_CPU(struct tlb_state, cpu_tlbstate); + +void reset_lazy_tlbstate(void); +#else +static inline void reset_lazy_tlbstate(void) +{ +} #endif #endif /* SMP */ diff --git a/include/asm-x86/traps.h b/include/asm-x86/traps.h index 7a692ba..6c3dc2c 100644 --- a/include/asm-x86/traps.h +++ b/include/asm-x86/traps.h @@ -3,7 +3,12 @@ #include <asm/debugreg.h> -/* Common in X86_32 and X86_64 */ +#ifdef CONFIG_X86_32 +#define dotraplinkage +#else +#define dotraplinkage asmlinkage +#endif + asmlinkage void divide_error(void); asmlinkage void debug(void); asmlinkage void nmi(void); @@ -12,31 +17,47 @@ asmlinkage void overflow(void); asmlinkage void bounds(void); asmlinkage void invalid_op(void); asmlinkage void device_not_available(void); +#ifdef CONFIG_X86_64 +asmlinkage void double_fault(void); +#endif asmlinkage void coprocessor_segment_overrun(void); asmlinkage void invalid_TSS(void); asmlinkage void segment_not_present(void); asmlinkage void stack_segment(void); asmlinkage void general_protection(void); asmlinkage void page_fault(void); +asmlinkage void spurious_interrupt_bug(void); asmlinkage void coprocessor_error(void); -asmlinkage void simd_coprocessor_error(void); asmlinkage void alignment_check(void); -asmlinkage void spurious_interrupt_bug(void); #ifdef CONFIG_X86_MCE asmlinkage void machine_check(void); #endif /* CONFIG_X86_MCE */ +asmlinkage void simd_coprocessor_error(void); -void do_divide_error(struct pt_regs *, long); -void do_overflow(struct pt_regs *, long); -void do_bounds(struct pt_regs *, long); -void do_coprocessor_segment_overrun(struct pt_regs *, long); -void do_invalid_TSS(struct pt_regs *, long); -void do_segment_not_present(struct pt_regs *, long); -void do_stack_segment(struct pt_regs *, long); -void do_alignment_check(struct pt_regs *, long); -void do_invalid_op(struct pt_regs *, long); -void do_general_protection(struct pt_regs *, long); -void do_nmi(struct pt_regs *, long); +dotraplinkage void do_divide_error(struct pt_regs *, long); +dotraplinkage void do_debug(struct pt_regs *, long); +dotraplinkage void do_nmi(struct pt_regs *, long); +dotraplinkage void do_int3(struct pt_regs *, long); +dotraplinkage void do_overflow(struct pt_regs *, long); +dotraplinkage void do_bounds(struct pt_regs *, long); +dotraplinkage void do_invalid_op(struct pt_regs *, long); +dotraplinkage void do_device_not_available(struct pt_regs *, long); +dotraplinkage void do_coprocessor_segment_overrun(struct pt_regs *, long); +dotraplinkage void do_invalid_TSS(struct pt_regs *, long); +dotraplinkage void do_segment_not_present(struct pt_regs *, long); +dotraplinkage void do_stack_segment(struct pt_regs *, long); +dotraplinkage void do_general_protection(struct pt_regs *, long); +dotraplinkage void do_page_fault(struct pt_regs *, unsigned long); +dotraplinkage void do_spurious_interrupt_bug(struct pt_regs *, long); +dotraplinkage void do_coprocessor_error(struct pt_regs *, long); +dotraplinkage void do_alignment_check(struct pt_regs *, long); +#ifdef CONFIG_X86_MCE +dotraplinkage void do_machine_check(struct pt_regs *, long); +#endif +dotraplinkage void do_simd_coprocessor_error(struct pt_regs *, long); +#ifdef CONFIG_X86_32 +dotraplinkage void do_iret_error(struct pt_regs *, long); +#endif static inline int get_si_code(unsigned long condition) { @@ -52,31 +73,9 @@ extern int panic_on_unrecovered_nmi; extern int kstack_depth_to_print; #ifdef CONFIG_X86_32 - -void do_iret_error(struct pt_regs *, long); -void do_int3(struct pt_regs *, long); -void do_debug(struct pt_regs *, long); void math_error(void __user *); -void do_coprocessor_error(struct pt_regs *, long); -void do_simd_coprocessor_error(struct pt_regs *, long); -void do_spurious_interrupt_bug(struct pt_regs *, long); unsigned long patch_espfix_desc(unsigned long, unsigned long); asmlinkage void math_emulate(long); +#endif -void do_page_fault(struct pt_regs *regs, unsigned long error_code); - -#else /* CONFIG_X86_32 */ - -asmlinkage void double_fault(void); - -asmlinkage void do_int3(struct pt_regs *, long); -asmlinkage void do_stack_segment(struct pt_regs *, long); -asmlinkage void do_debug(struct pt_regs *, unsigned long); -asmlinkage void do_coprocessor_error(struct pt_regs *); -asmlinkage void do_simd_coprocessor_error(struct pt_regs *); -asmlinkage void do_spurious_interrupt_bug(struct pt_regs *); - -asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code); - -#endif /* CONFIG_X86_32 */ #endif /* ASM_X86__TRAPS_H */ diff --git a/include/asm-x86/xen/hypervisor.h b/include/asm-x86/xen/hypervisor.h index 0ef3a88..445a247 100644 --- a/include/asm-x86/xen/hypervisor.h +++ b/include/asm-x86/xen/hypervisor.h @@ -54,7 +54,6 @@ /* arch/i386/kernel/setup.c */ extern struct shared_info *HYPERVISOR_shared_info; extern struct start_info *xen_start_info; -#define is_initial_xendomain() (xen_start_info->flags & SIF_INITDOMAIN) /* arch/i386/mach-xen/evtchn.c */ /* Force a proper event-channel callback from Xen. */ @@ -67,6 +66,17 @@ u64 jiffies_to_st(unsigned long jiffies); #define MULTI_UVMFLAGS_INDEX 3 #define MULTI_UVMDOMID_INDEX 4 -#define is_running_on_xen() (xen_start_info ? 1 : 0) +enum xen_domain_type { + XEN_NATIVE, + XEN_PV_DOMAIN, + XEN_HVM_DOMAIN, +}; + +extern enum xen_domain_type xen_domain_type; + +#define xen_domain() (xen_domain_type != XEN_NATIVE) +#define xen_pv_domain() (xen_domain_type == XEN_PV_DOMAIN) +#define xen_initial_domain() (xen_pv_domain() && xen_start_info->flags & SIF_INITDOMAIN) +#define xen_hvm_domain() (xen_domain_type == XEN_HVM_DOMAIN) #endif /* ASM_X86__XEN__HYPERVISOR_H */ diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 2651f80..75d81f1 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -182,7 +182,7 @@ extern int vsscanf(const char *, const char *, va_list) extern int get_option(char **str, int *pint); extern char *get_options(const char *str, int nints, int *ints); -extern unsigned long long memparse(char *ptr, char **retptr); +extern unsigned long long memparse(const char *ptr, char **retptr); extern int core_kernel_text(unsigned long addr); extern int __kernel_text_address(unsigned long addr); diff --git a/include/linux/mm.h b/include/linux/mm.h index 72a15dc..c61ba10 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -7,6 +7,7 @@ #include <linux/gfp.h> #include <linux/list.h> +#include <linux/mmdebug.h> #include <linux/mmzone.h> #include <linux/rbtree.h> #include <linux/prio_tree.h> @@ -219,12 +220,6 @@ struct inode; */ #include <linux/page-flags.h> -#ifdef CONFIG_DEBUG_VM -#define VM_BUG_ON(cond) BUG_ON(cond) -#else -#define VM_BUG_ON(condition) do { } while(0) -#endif - /* * Methods to modify the page usage count. * @@ -919,7 +914,7 @@ static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long a } #endif /* CONFIG_MMU && !__ARCH_HAS_4LEVEL_HACK */ -#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS +#if USE_SPLIT_PTLOCKS /* * We tuck a spinlock to guard each pagetable page into its struct page, * at page->private, with BUILD_BUG_ON to make sure that this will not @@ -932,14 +927,14 @@ static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long a } while (0) #define pte_lock_deinit(page) ((page)->mapping = NULL) #define pte_lockptr(mm, pmd) ({(void)(mm); __pte_lockptr(pmd_page(*(pmd)));}) -#else +#else /* !USE_SPLIT_PTLOCKS */ /* * We use mm->page_table_lock to guard all pagetable pages of the mm. */ #define pte_lock_init(page) do {} while (0) #define pte_lock_deinit(page) do {} while (0) #define pte_lockptr(mm, pmd) ({(void)(pmd); &(mm)->page_table_lock;}) -#endif /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */ +#endif /* USE_SPLIT_PTLOCKS */ static inline void pgtable_page_ctor(struct page *page) { diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index bf33413..9d49fa3 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -21,11 +21,13 @@ struct address_space; -#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS +#define USE_SPLIT_PTLOCKS (NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS) + +#if USE_SPLIT_PTLOCKS typedef atomic_long_t mm_counter_t; -#else /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */ +#else /* !USE_SPLIT_PTLOCKS */ typedef unsigned long mm_counter_t; -#endif /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */ +#endif /* !USE_SPLIT_PTLOCKS */ /* * Each physical page in the system has a struct page associated with @@ -65,7 +67,7 @@ struct page { * see PAGE_MAPPING_ANON below. */ }; -#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS +#if USE_SPLIT_PTLOCKS spinlock_t ptl; #endif struct kmem_cache *slab; /* SLUB: Pointer to slab */ diff --git a/include/linux/mmdebug.h b/include/linux/mmdebug.h new file mode 100644 index 0000000..8a55098 --- /dev/null +++ b/include/linux/mmdebug.h @@ -0,0 +1,18 @@ +#ifndef LINUX_MM_DEBUG_H +#define LINUX_MM_DEBUG_H 1 + +#include <linux/autoconf.h> + +#ifdef CONFIG_DEBUG_VM +#define VM_BUG_ON(cond) BUG_ON(cond) +#else +#define VM_BUG_ON(cond) do { } while (0) +#endif + +#ifdef CONFIG_DEBUG_VIRTUAL +#define VIRTUAL_BUG_ON(cond) BUG_ON(cond) +#else +#define VIRTUAL_BUG_ON(cond) do { } while (0) +#endif + +#endif diff --git a/include/linux/sched.h b/include/linux/sched.h index 3d9120c..272c353 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -352,7 +352,7 @@ arch_get_unmapped_area_topdown(struct file *filp, unsigned long addr, extern void arch_unmap_area(struct mm_struct *, unsigned long); extern void arch_unmap_area_topdown(struct mm_struct *, unsigned long); -#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS +#if USE_SPLIT_PTLOCKS /* * The mm counters are not protected by its page_table_lock, * so must be incremented atomically. @@ -363,7 +363,7 @@ extern void arch_unmap_area_topdown(struct mm_struct *, unsigned long); #define inc_mm_counter(mm, member) atomic_long_inc(&(mm)->_##member) #define dec_mm_counter(mm, member) atomic_long_dec(&(mm)->_##member) -#else /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */ +#else /* !USE_SPLIT_PTLOCKS */ /* * The mm counters are protected by its page_table_lock, * so can be incremented directly. @@ -374,7 +374,7 @@ extern void arch_unmap_area_topdown(struct mm_struct *, unsigned long); #define inc_mm_counter(mm, member) (mm)->_##member++ #define dec_mm_counter(mm, member) (mm)->_##member-- -#endif /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */ +#endif /* !USE_SPLIT_PTLOCKS */ #define get_mm_rss(mm) \ (get_mm_counter(mm, file_rss) + get_mm_counter(mm, anon_rss)) diff --git a/include/xen/events.h b/include/xen/events.h index 4680ff3..0d5f1ad 100644 --- a/include/xen/events.h +++ b/include/xen/events.h @@ -46,6 +46,8 @@ extern void xen_irq_resume(void); /* Clear an irq's pending state, in preparation for polling on it */ void xen_clear_irq_pending(int irq); +void xen_set_irq_pending(int irq); +bool xen_test_irq_pending(int irq); /* Poll waiting for an irq to become pending. In the usual case, the irq will be disabled so it won't deliver an interrupt. */ diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 0b50481..082235b 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -495,6 +495,15 @@ config DEBUG_VM If unsure, say N. +config DEBUG_VIRTUAL + bool "Debug VM translations" + depends on DEBUG_KERNEL && X86 + help + Enable some costly sanity checks in virtual to page code. This can + catch mistakes with virt_to_page() and friends. + + If unsure, say N. + config DEBUG_WRITECOUNT bool "Debug filesystem writers count" depends on DEBUG_KERNEL diff --git a/lib/cmdline.c b/lib/cmdline.c index 5ba8a94..f5f3ad8 100644 --- a/lib/cmdline.c +++ b/lib/cmdline.c @@ -126,7 +126,7 @@ char *get_options(const char *str, int nints, int *ints) * megabyte, or one gigabyte, respectively. */ -unsigned long long memparse(char *ptr, char **retptr) +unsigned long long memparse(const char *ptr, char **retptr) { char *endptr; /* local pointer to end of parsed string */ diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 85b9a0d..bba06c4 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -180,6 +180,13 @@ struct page *vmalloc_to_page(const void *vmalloc_addr) pmd_t *pmd; pte_t *ptep, pte; + /* + * XXX we might need to change this if we add VIRTUAL_BUG_ON for + * architectures that do not vmalloc module space + */ + VIRTUAL_BUG_ON(!is_vmalloc_addr(vmalloc_addr) && + !is_module_address(addr)); + if (!pgd_none(*pgd)) { pud = pud_offset(pgd, addr); if (!pud_none(*pud)) {
| |