diff options
author | dfr <dfr@FreeBSD.org> | 2009-03-11 15:30:12 +0000 |
---|---|---|
committer | dfr <dfr@FreeBSD.org> | 2009-03-11 15:30:12 +0000 |
commit | 598fb4217fe304b2d9dae488867c35e7213e3ba5 (patch) | |
tree | 633dc40712503605071c68975d2c972638554000 /sys/amd64/include/xen | |
parent | 2ee25eb1a489668e88e623440b968811e9f6d5b3 (diff) | |
parent | 0b3e97d804d73af27fba381fd51fe0dd48a9cb06 (diff) | |
download | FreeBSD-src-598fb4217fe304b2d9dae488867c35e7213e3ba5.zip FreeBSD-src-598fb4217fe304b2d9dae488867c35e7213e3ba5.tar.gz |
Merge in support for Xen HVM on amd64 architecture.
Diffstat (limited to 'sys/amd64/include/xen')
-rw-r--r-- | sys/amd64/include/xen/hypercall.h | 415 | ||||
-rw-r--r-- | sys/amd64/include/xen/synch_bitops.h | 129 | ||||
-rw-r--r-- | sys/amd64/include/xen/xen-os.h | 296 | ||||
-rw-r--r-- | sys/amd64/include/xen/xenfunc.h | 83 | ||||
-rw-r--r-- | sys/amd64/include/xen/xenpmap.h | 227 | ||||
-rw-r--r-- | sys/amd64/include/xen/xenvar.h | 122 |
6 files changed, 1272 insertions, 0 deletions
diff --git a/sys/amd64/include/xen/hypercall.h b/sys/amd64/include/xen/hypercall.h new file mode 100644 index 0000000..50fa376 --- /dev/null +++ b/sys/amd64/include/xen/hypercall.h @@ -0,0 +1,415 @@ +/****************************************************************************** + * hypercall.h + * + * Linux-specific hypervisor handling. + * + * Copyright (c) 2002-2004, K A Fraser + * + * 64-bit updates: + * Benjamin Liu <benjamin.liu@intel.com> + * Jun Nakajima <jun.nakajima@intel.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#ifndef __MACHINE_XEN_HYPERCALL_H__ +#define __MACHINE_XEN_HYPERCALL_H__ + +#include <sys/systm.h> + +#ifndef __XEN_HYPERVISOR_H__ +# error "please don't include this file directly" +#endif + +#define __STR(x) #x +#define STR(x) __STR(x) +#define ENOXENSYS 38 +#define CONFIG_XEN_COMPAT 0x030002 +#define __must_check + +#ifdef XEN +#define HYPERCALL_STR(name) \ + "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)" +#else +#define HYPERCALL_STR(name) \ + "mov $("STR(__HYPERVISOR_##name)" * 32),%%eax; "\ + "add hypercall_stubs(%%rip),%%rax; " \ + "call *%%rax" +#endif + +#define _hypercall0(type, name) \ +({ \ + type __res; \ + __asm__ volatile ( \ + HYPERCALL_STR(name) \ + : "=a" (__res) \ + : \ + : "memory" ); \ + __res; \ +}) + +#define _hypercall1(type, name, a1) \ +({ \ + type __res; \ + long __ign1; \ + __asm__ volatile ( \ + HYPERCALL_STR(name) \ + : "=a" (__res), "=D" (__ign1) \ + : "1" ((long)(a1)) \ + : "memory" ); \ + __res; \ +}) + +#define _hypercall2(type, name, a1, a2) \ +({ \ + type __res; \ + long __ign1, __ign2; \ + __asm__ volatile ( \ + HYPERCALL_STR(name) \ + : "=a" (__res), "=D" (__ign1), "=S" (__ign2) \ + : "1" ((long)(a1)), "2" ((long)(a2)) \ + : "memory" ); \ + __res; \ +}) + +#define _hypercall3(type, name, a1, a2, a3) \ +({ \ + type __res; \ + long __ign1, __ign2, __ign3; \ + __asm__ volatile ( \ + HYPERCALL_STR(name) \ + : "=a" (__res), "=D" (__ign1), "=S" (__ign2), \ + "=d" (__ign3) \ + : "1" ((long)(a1)), "2" ((long)(a2)), \ + "3" ((long)(a3)) \ + : "memory" ); \ + __res; \ +}) + +#define _hypercall4(type, name, a1, a2, a3, a4) \ +({ \ + type __res; \ + long __ign1, __ign2, __ign3; \ + register long __arg4 __asm__("r10") = (long)(a4); \ + __asm__ volatile ( \ + HYPERCALL_STR(name) \ + : "=a" (__res), "=D" (__ign1), "=S" (__ign2), \ + "=d" (__ign3), "+r" (__arg4) \ + : "1" ((long)(a1)), "2" ((long)(a2)), \ + "3" ((long)(a3)) \ + : "memory" ); \ + __res; \ +}) + +#define _hypercall5(type, name, a1, a2, a3, a4, a5) \ +({ \ + type __res; \ + long __ign1, __ign2, __ign3; \ + register long __arg4 __asm__("r10") = (long)(a4); \ + register long __arg5 __asm__("r8") = (long)(a5); \ + __asm__ volatile ( \ + HYPERCALL_STR(name) \ + : "=a" (__res), "=D" (__ign1), "=S" (__ign2), \ + "=d" (__ign3), "+r" (__arg4), "+r" (__arg5) \ + : "1" ((long)(a1)), "2" ((long)(a2)), \ + "3" ((long)(a3)) \ + : "memory" ); \ + __res; \ +}) + +static inline int __must_check +HYPERVISOR_set_trap_table( + const trap_info_t *table) +{ + return _hypercall1(int, set_trap_table, table); +} + +static inline int __must_check +HYPERVISOR_mmu_update( + mmu_update_t *req, unsigned int count, unsigned int *success_count, + domid_t domid) +{ + return _hypercall4(int, mmu_update, req, count, success_count, domid); +} + +static inline int __must_check +HYPERVISOR_mmuext_op( + struct mmuext_op *op, unsigned int count, unsigned int *success_count, + domid_t domid) +{ + return _hypercall4(int, mmuext_op, op, count, success_count, domid); +} + +static inline int __must_check +HYPERVISOR_set_gdt( + unsigned long *frame_list, unsigned int entries) +{ + return _hypercall2(int, set_gdt, frame_list, entries); +} + +static inline int __must_check +HYPERVISOR_stack_switch( + unsigned long ss, unsigned long esp) +{ + return _hypercall2(int, stack_switch, ss, esp); +} + +static inline int __must_check +HYPERVISOR_set_callbacks( + unsigned long event_address, unsigned long failsafe_address, + unsigned long syscall_address) +{ + return _hypercall3(int, set_callbacks, + event_address, failsafe_address, syscall_address); +} + +static inline int +HYPERVISOR_fpu_taskswitch( + int set) +{ + return _hypercall1(int, fpu_taskswitch, set); +} + +static inline int __must_check +HYPERVISOR_sched_op_compat( + int cmd, unsigned long arg) +{ + return _hypercall2(int, sched_op_compat, cmd, arg); +} + +static inline int __must_check +HYPERVISOR_sched_op( + int cmd, void *arg) +{ + return _hypercall2(int, sched_op, cmd, arg); +} + +static inline long __must_check +HYPERVISOR_set_timer_op( + uint64_t timeout) +{ + return _hypercall1(long, set_timer_op, timeout); +} + +static inline int __must_check +HYPERVISOR_platform_op( + struct xen_platform_op *platform_op) +{ + platform_op->interface_version = XENPF_INTERFACE_VERSION; + return _hypercall1(int, platform_op, platform_op); +} + +static inline int __must_check +HYPERVISOR_set_debugreg( + unsigned int reg, unsigned long value) +{ + return _hypercall2(int, set_debugreg, reg, value); +} + +static inline unsigned long __must_check +HYPERVISOR_get_debugreg( + unsigned int reg) +{ + return _hypercall1(unsigned long, get_debugreg, reg); +} + +static inline int __must_check +HYPERVISOR_update_descriptor( + unsigned long ma, unsigned long word) +{ + return _hypercall2(int, update_descriptor, ma, word); +} + +static inline int __must_check +HYPERVISOR_memory_op( + unsigned int cmd, void *arg) +{ + return _hypercall2(int, memory_op, cmd, arg); +} + +static inline int __must_check +HYPERVISOR_multicall( + multicall_entry_t *call_list, unsigned int nr_calls) +{ + return _hypercall2(int, multicall, call_list, nr_calls); +} + +static inline int __must_check +HYPERVISOR_update_va_mapping( + unsigned long va, uint64_t new_val, unsigned long flags) +{ + return _hypercall3(int, update_va_mapping, va, new_val, flags); +} + +static inline int __must_check +HYPERVISOR_event_channel_op( + int cmd, void *arg) +{ + int rc = _hypercall2(int, event_channel_op, cmd, arg); + +#if CONFIG_XEN_COMPAT <= 0x030002 + if (unlikely(rc == -ENOXENSYS)) { + struct evtchn_op op; + op.cmd = cmd; + memcpy(&op.u, arg, sizeof(op.u)); + rc = _hypercall1(int, event_channel_op_compat, &op); + memcpy(arg, &op.u, sizeof(op.u)); + } +#endif + + return rc; +} + +static inline int __must_check +HYPERVISOR_xen_version( + int cmd, void *arg) +{ + return _hypercall2(int, xen_version, cmd, arg); +} + +static inline int __must_check +HYPERVISOR_console_io( + int cmd, unsigned int count, char *str) +{ + return _hypercall3(int, console_io, cmd, count, str); +} + +static inline int __must_check +HYPERVISOR_physdev_op( + int cmd, void *arg) +{ + int rc = _hypercall2(int, physdev_op, cmd, arg); + +#if CONFIG_XEN_COMPAT <= 0x030002 + if (unlikely(rc == -ENOXENSYS)) { + struct physdev_op op; + op.cmd = cmd; + memcpy(&op.u, arg, sizeof(op.u)); + rc = _hypercall1(int, physdev_op_compat, &op); + memcpy(arg, &op.u, sizeof(op.u)); + } +#endif + + return rc; +} + +static inline int __must_check +HYPERVISOR_grant_table_op( + unsigned int cmd, void *uop, unsigned int count) +{ + return _hypercall3(int, grant_table_op, cmd, uop, count); +} + +static inline int __must_check +HYPERVISOR_update_va_mapping_otherdomain( + unsigned long va, uint64_t new_val, unsigned long flags, domid_t domid) +{ + return _hypercall4(int, update_va_mapping_otherdomain, va, + new_val, flags, domid); +} + +static inline int __must_check +HYPERVISOR_vm_assist( + unsigned int cmd, unsigned int type) +{ + return _hypercall2(int, vm_assist, cmd, type); +} + +static inline int __must_check +HYPERVISOR_vcpu_op( + int cmd, unsigned int vcpuid, void *extra_args) +{ + return _hypercall3(int, vcpu_op, cmd, vcpuid, extra_args); +} + +static inline int __must_check +HYPERVISOR_set_segment_base( + int reg, unsigned long value) +{ + return _hypercall2(int, set_segment_base, reg, value); +} + +static inline int __must_check +HYPERVISOR_suspend( + unsigned long srec) +{ + struct sched_shutdown sched_shutdown = { + .reason = SHUTDOWN_suspend + }; + + int rc = _hypercall3(int, sched_op, SCHEDOP_shutdown, + &sched_shutdown, srec); + +#if CONFIG_XEN_COMPAT <= 0x030002 + if (rc == -ENOXENSYS) + rc = _hypercall3(int, sched_op_compat, SCHEDOP_shutdown, + SHUTDOWN_suspend, srec); +#endif + + return rc; +} + +#if CONFIG_XEN_COMPAT <= 0x030002 +static inline int +HYPERVISOR_nmi_op( + unsigned long op, void *arg) +{ + return _hypercall2(int, nmi_op, op, arg); +} +#endif + +#ifndef CONFIG_XEN +static inline unsigned long __must_check +HYPERVISOR_hvm_op( + int op, void *arg) +{ + return _hypercall2(unsigned long, hvm_op, op, arg); +} +#endif + +static inline int __must_check +HYPERVISOR_callback_op( + int cmd, const void *arg) +{ + return _hypercall2(int, callback_op, cmd, arg); +} + +static inline int __must_check +HYPERVISOR_xenoprof_op( + int op, void *arg) +{ + return _hypercall2(int, xenoprof_op, op, arg); +} + +static inline int __must_check +HYPERVISOR_kexec_op( + unsigned long op, void *args) +{ + return _hypercall2(int, kexec_op, op, args); +} + +#undef __must_check + +#endif /* __MACHINE_XEN_HYPERCALL_H__ */ diff --git a/sys/amd64/include/xen/synch_bitops.h b/sys/amd64/include/xen/synch_bitops.h new file mode 100644 index 0000000..746687a --- /dev/null +++ b/sys/amd64/include/xen/synch_bitops.h @@ -0,0 +1,129 @@ +#ifndef __XEN_SYNCH_BITOPS_H__ +#define __XEN_SYNCH_BITOPS_H__ + +/* + * Copyright 1992, Linus Torvalds. + * Heavily modified to provide guaranteed strong synchronisation + * when communicating with Xen or other guest OSes running on other CPUs. + */ + + +#define ADDR (*(volatile long *) addr) + +static __inline__ void synch_set_bit(int nr, volatile void * addr) +{ + __asm__ __volatile__ ( + "lock btsl %1,%0" + : "=m" (ADDR) : "Ir" (nr) : "memory" ); +} + +static __inline__ void synch_clear_bit(int nr, volatile void * addr) +{ + __asm__ __volatile__ ( + "lock btrl %1,%0" + : "=m" (ADDR) : "Ir" (nr) : "memory" ); +} + +static __inline__ void synch_change_bit(int nr, volatile void * addr) +{ + __asm__ __volatile__ ( + "lock btcl %1,%0" + : "=m" (ADDR) : "Ir" (nr) : "memory" ); +} + +static __inline__ int synch_test_and_set_bit(int nr, volatile void * addr) +{ + int oldbit; + __asm__ __volatile__ ( + "lock btsl %2,%1\n\tsbbl %0,%0" + : "=r" (oldbit), "=m" (ADDR) : "Ir" (nr) : "memory"); + return oldbit; +} + +static __inline__ int synch_test_and_clear_bit(int nr, volatile void * addr) +{ + int oldbit; + __asm__ __volatile__ ( + "lock btrl %2,%1\n\tsbbl %0,%0" + : "=r" (oldbit), "=m" (ADDR) : "Ir" (nr) : "memory"); + return oldbit; +} + +static __inline__ int synch_test_and_change_bit(int nr, volatile void * addr) +{ + int oldbit; + + __asm__ __volatile__ ( + "lock btcl %2,%1\n\tsbbl %0,%0" + : "=r" (oldbit), "=m" (ADDR) : "Ir" (nr) : "memory"); + return oldbit; +} + +struct __synch_xchg_dummy { unsigned long a[100]; }; +#define __synch_xg(x) ((volatile struct __synch_xchg_dummy *)(x)) + +#define synch_cmpxchg(ptr, old, new) \ +((__typeof__(*(ptr)))__synch_cmpxchg((ptr),\ + (unsigned long)(old), \ + (unsigned long)(new), \ + sizeof(*(ptr)))) + +static inline unsigned long __synch_cmpxchg(volatile void *ptr, + unsigned long old, + unsigned long new, int size) +{ + unsigned long prev; + switch (size) { + case 1: + __asm__ __volatile__("lock; cmpxchgb %b1,%2" + : "=a"(prev) + : "q"(new), "m"(*__synch_xg(ptr)), + "0"(old) + : "memory"); + return prev; + case 2: + __asm__ __volatile__("lock; cmpxchgw %w1,%2" + : "=a"(prev) + : "q"(new), "m"(*__synch_xg(ptr)), + "0"(old) + : "memory"); + return prev; + case 4: + __asm__ __volatile__("lock; cmpxchgl %k1,%2" + : "=a"(prev) + : "q"(new), "m"(*__synch_xg(ptr)), + "0"(old) + : "memory"); + return prev; + case 8: + __asm__ __volatile__("lock; cmpxchgq %1,%2" + : "=a"(prev) + : "q"(new), "m"(*__synch_xg(ptr)), + "0"(old) + : "memory"); + return prev; + } + return old; +} + +static __inline__ int synch_const_test_bit(int nr, const volatile void * addr) +{ + return ((1UL << (nr & 31)) & + (((const volatile unsigned int *) addr)[nr >> 5])) != 0; +} + +static __inline__ int synch_var_test_bit(int nr, volatile void * addr) +{ + int oldbit; + __asm__ __volatile__ ( + "btl %2,%1\n\tsbbl %0,%0" + : "=r" (oldbit) : "m" (ADDR), "Ir" (nr) ); + return oldbit; +} + +#define synch_test_bit(nr,addr) \ +(__builtin_constant_p(nr) ? \ + synch_const_test_bit((nr),(addr)) : \ + synch_var_test_bit((nr),(addr))) + +#endif /* __XEN_SYNCH_BITOPS_H__ */ diff --git a/sys/amd64/include/xen/xen-os.h b/sys/amd64/include/xen/xen-os.h new file mode 100644 index 0000000..163e7f2 --- /dev/null +++ b/sys/amd64/include/xen/xen-os.h @@ -0,0 +1,296 @@ +/****************************************************************************** + * os.h + * + * random collection of macros and definition + */ + +#ifndef _XEN_OS_H_ +#define _XEN_OS_H_ + +#ifdef PAE +#define CONFIG_X86_PAE +#endif + +#if !defined(__XEN_INTERFACE_VERSION__) +/* + * Can update to a more recent version when we implement + * the hypercall page + */ +#define __XEN_INTERFACE_VERSION__ 0x00030204 +#endif + +#include <xen/interface/xen.h> + +/* Force a proper event-channel callback from Xen. */ +void force_evtchn_callback(void); + +extern int gdtset; + +extern shared_info_t *HYPERVISOR_shared_info; + +/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */ +static inline void rep_nop(void) +{ + __asm__ __volatile__ ( "rep;nop" : : : "memory" ); +} +#define cpu_relax() rep_nop() + +/* crude memory allocator for memory allocation early in + * boot + */ +void *bootmem_alloc(unsigned int size); +void bootmem_free(void *ptr, unsigned int size); + + +/* Everything below this point is not included by assembler (.S) files. */ +#ifndef __ASSEMBLY__ + +void printk(const char *fmt, ...); + +/* some function prototypes */ +void trap_init(void); + +#define likely(x) __builtin_expect((x),1) +#define unlikely(x) __builtin_expect((x),0) + +#ifndef XENHVM + +/* + * STI/CLI equivalents. These basically set and clear the virtual + * event_enable flag in teh shared_info structure. Note that when + * the enable bit is set, there may be pending events to be handled. + * We may therefore call into do_hypervisor_callback() directly. + */ + +#define __cli() \ +do { \ + vcpu_info_t *_vcpu; \ + _vcpu = &HYPERVISOR_shared_info->vcpu_info[PCPU_GET(cpuid)]; \ + _vcpu->evtchn_upcall_mask = 1; \ + barrier(); \ +} while (0) + +#define __sti() \ +do { \ + vcpu_info_t *_vcpu; \ + barrier(); \ + _vcpu = &HYPERVISOR_shared_info->vcpu_info[PCPU_GET(cpuid)]; \ + _vcpu->evtchn_upcall_mask = 0; \ + barrier(); /* unmask then check (avoid races) */ \ + if ( unlikely(_vcpu->evtchn_upcall_pending) ) \ + force_evtchn_callback(); \ +} while (0) + +#define __restore_flags(x) \ +do { \ + vcpu_info_t *_vcpu; \ + barrier(); \ + _vcpu = &HYPERVISOR_shared_info->vcpu_info[PCPU_GET(cpuid)]; \ + if ((_vcpu->evtchn_upcall_mask = (x)) == 0) { \ + barrier(); /* unmask then check (avoid races) */ \ + if ( unlikely(_vcpu->evtchn_upcall_pending) ) \ + force_evtchn_callback(); \ + } \ +} while (0) + +/* + * Add critical_{enter, exit}? + * + */ +#define __save_and_cli(x) \ +do { \ + vcpu_info_t *_vcpu; \ + _vcpu = &HYPERVISOR_shared_info->vcpu_info[PCPU_GET(cpuid)]; \ + (x) = _vcpu->evtchn_upcall_mask; \ + _vcpu->evtchn_upcall_mask = 1; \ + barrier(); \ +} while (0) + + +#define cli() __cli() +#define sti() __sti() +#define save_flags(x) __save_flags(x) +#define restore_flags(x) __restore_flags(x) +#define save_and_cli(x) __save_and_cli(x) + +#define local_irq_save(x) __save_and_cli(x) +#define local_irq_restore(x) __restore_flags(x) +#define local_irq_disable() __cli() +#define local_irq_enable() __sti() + +#define mtx_lock_irqsave(lock, x) {local_irq_save((x)); mtx_lock_spin((lock));} +#define mtx_unlock_irqrestore(lock, x) {mtx_unlock_spin((lock)); local_irq_restore((x)); } +#define spin_lock_irqsave mtx_lock_irqsave +#define spin_unlock_irqrestore mtx_unlock_irqrestore + +#else +#endif + +#ifndef mb +#define mb() __asm__ __volatile__("mfence":::"memory") +#endif +#ifndef rmb +#define rmb() __asm__ __volatile__("lfence":::"memory"); +#endif +#ifndef wmb +#define wmb() barrier() +#endif +#ifdef SMP +#define smp_mb() mb() +#define smp_rmb() rmb() +#define smp_wmb() wmb() +#define smp_read_barrier_depends() read_barrier_depends() +#define set_mb(var, value) do { xchg(&var, value); } while (0) +#else +#define smp_mb() barrier() +#define smp_rmb() barrier() +#define smp_wmb() barrier() +#define smp_read_barrier_depends() do { } while(0) +#define set_mb(var, value) do { var = value; barrier(); } while (0) +#endif + + +/* This is a barrier for the compiler only, NOT the processor! */ +#define barrier() __asm__ __volatile__("": : :"memory") + +#define LOCK_PREFIX "" +#define LOCK "" +#define ADDR (*(volatile long *) addr) +/* + * Make sure gcc doesn't try to be clever and move things around + * on us. We need to use _exactly_ the address the user gave us, + * not some alias that contains the same information. + */ +typedef struct { volatile int counter; } atomic_t; + + + +#define xen_xchg(ptr,v) \ + ((__typeof__(*(ptr)))__xchg((unsigned long)(v),(ptr),sizeof(*(ptr)))) +struct __xchg_dummy { unsigned long a[100]; }; +#define __xg(x) ((volatile struct __xchg_dummy *)(x)) +static __inline unsigned long __xchg(unsigned long x, volatile void * ptr, + int size) +{ + switch (size) { + case 1: + __asm__ __volatile__("xchgb %b0,%1" + :"=q" (x) + :"m" (*__xg(ptr)), "0" (x) + :"memory"); + break; + case 2: + __asm__ __volatile__("xchgw %w0,%1" + :"=r" (x) + :"m" (*__xg(ptr)), "0" (x) + :"memory"); + break; + case 4: + __asm__ __volatile__("xchgl %0,%1" + :"=r" (x) + :"m" (*__xg(ptr)), "0" (x) + :"memory"); + break; + } + return x; +} + +/** + * test_and_clear_bit - Clear a bit and return its old value + * @nr: Bit to set + * @addr: Address to count from + * + * This operation is atomic and cannot be reordered. + * It also implies a memory barrier. + */ +static __inline int test_and_clear_bit(int nr, volatile void * addr) +{ + int oldbit; + + __asm__ __volatile__( LOCK_PREFIX + "btrl %2,%1\n\tsbbl %0,%0" + :"=r" (oldbit),"=m" (ADDR) + :"Ir" (nr) : "memory"); + return oldbit; +} + +static __inline int constant_test_bit(int nr, const volatile void * addr) +{ + return ((1UL << (nr & 31)) & (((const volatile unsigned int *) addr)[nr >> 5])) != 0; +} + +static __inline int variable_test_bit(int nr, volatile void * addr) +{ + int oldbit; + + __asm__ __volatile__( + "btl %2,%1\n\tsbbl %0,%0" + :"=r" (oldbit) + :"m" (ADDR),"Ir" (nr)); + return oldbit; +} + +#define test_bit(nr,addr) \ +(__builtin_constant_p(nr) ? \ + constant_test_bit((nr),(addr)) : \ + variable_test_bit((nr),(addr))) + + +/** + * set_bit - Atomically set a bit in memory + * @nr: the bit to set + * @addr: the address to start counting from + * + * This function is atomic and may not be reordered. See __set_bit() + * if you do not require the atomic guarantees. + * Note that @nr may be almost arbitrarily large; this function is not + * restricted to acting on a single-word quantity. + */ +static __inline__ void set_bit(int nr, volatile void * addr) +{ + __asm__ __volatile__( LOCK_PREFIX + "btsl %1,%0" + :"=m" (ADDR) + :"Ir" (nr)); +} + +/** + * clear_bit - Clears a bit in memory + * @nr: Bit to clear + * @addr: Address to start counting from + * + * clear_bit() is atomic and may not be reordered. However, it does + * not contain a memory barrier, so if it is used for locking purposes, + * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit() + * in order to ensure changes are visible on other processors. + */ +static __inline__ void clear_bit(int nr, volatile void * addr) +{ + __asm__ __volatile__( LOCK_PREFIX + "btrl %1,%0" + :"=m" (ADDR) + :"Ir" (nr)); +} + +/** + * atomic_inc - increment atomic variable + * @v: pointer of type atomic_t + * + * Atomically increments @v by 1. Note that the guaranteed + * useful range of an atomic_t is only 24 bits. + */ +static __inline__ void atomic_inc(atomic_t *v) +{ + __asm__ __volatile__( + LOCK "incl %0" + :"=m" (v->counter) + :"m" (v->counter)); +} + + +#define rdtscll(val) \ + __asm__ __volatile__("rdtsc" : "=A" (val)) + +#endif /* !__ASSEMBLY__ */ + +#endif /* _OS_H_ */ diff --git a/sys/amd64/include/xen/xenfunc.h b/sys/amd64/include/xen/xenfunc.h new file mode 100644 index 0000000..b3a6672 --- /dev/null +++ b/sys/amd64/include/xen/xenfunc.h @@ -0,0 +1,83 @@ +/* + * + * Copyright (c) 2004,2005 Kip Macy + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. The name of the author may not be used to endorse or promote products + * derived from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES + * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. + * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, + * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF + * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + + +#ifndef _XEN_XENFUNC_H_ +#define _XEN_XENFUNC_H_ + +#ifdef XENHVM +#include <machine/xen/xenvar.h> +#else +#include <machine/xen/xenpmap.h> +#include <machine/segments.h> +#endif + +#define BKPT __asm__("int3"); +#define XPQ_CALL_DEPTH 5 +#define XPQ_CALL_COUNT 2 +#define PG_PRIV PG_AVAIL3 +typedef struct { + unsigned long pt_ref; + unsigned long pt_eip[XPQ_CALL_COUNT][XPQ_CALL_DEPTH]; +} pteinfo_t; + +extern pteinfo_t *pteinfo_list; +#ifdef XENDEBUG_LOW +#define __PRINTK(x) printk x +#else +#define __PRINTK(x) +#endif + +char *xen_setbootenv(char *cmd_line); + +int xen_boothowto(char *envp); + +void _xen_machphys_update(vm_paddr_t, vm_paddr_t, char *file, int line); + +#ifdef INVARIANTS +#define xen_machphys_update(a, b) _xen_machphys_update((a), (b), __FILE__, __LINE__) +#else +#define xen_machphys_update(a, b) _xen_machphys_update((a), (b), NULL, 0) +#endif + +#ifndef XENHVM +void xen_update_descriptor(union descriptor *, union descriptor *); +#endif + +extern struct mtx balloon_lock; +#if 0 +#define balloon_lock(__flags) mtx_lock_irqsave(&balloon_lock, __flags) +#define balloon_unlock(__flags) mtx_unlock_irqrestore(&balloon_lock, __flags) +#else +#define balloon_lock(__flags) __flags = 1 +#define balloon_unlock(__flags) __flags = 0 +#endif + + + +#endif /* _XEN_XENFUNC_H_ */ diff --git a/sys/amd64/include/xen/xenpmap.h b/sys/amd64/include/xen/xenpmap.h new file mode 100644 index 0000000..d768dad --- /dev/null +++ b/sys/amd64/include/xen/xenpmap.h @@ -0,0 +1,227 @@ +/* + * + * Copyright (c) 2004 Christian Limpach. + * Copyright (c) 2004,2005 Kip Macy + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. All advertising materials mentioning features or use of this software + * must display the following acknowledgement: + * This product includes software developed by Christian Limpach. + * 4. The name of the author may not be used to endorse or promote products + * derived from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES + * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. + * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, + * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF + * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + + +#ifndef _XEN_XENPMAP_H_ +#define _XEN_XENPMAP_H_ + +#include <machine/xen/features.h> + +void _xen_queue_pt_update(vm_paddr_t, vm_paddr_t, char *, int); +void xen_pt_switch(vm_paddr_t); +void xen_set_ldt(vm_paddr_t, unsigned long); +void xen_pgdpt_pin(vm_paddr_t); +void xen_pgd_pin(vm_paddr_t); +void xen_pgd_unpin(vm_paddr_t); +void xen_pt_pin(vm_paddr_t); +void xen_pt_unpin(vm_paddr_t); +void xen_flush_queue(void); +void xen_check_queue(void); +#if 0 +void pmap_ref(pt_entry_t *pte, vm_paddr_t ma); +#endif + +#ifdef INVARIANTS +#define xen_queue_pt_update(a, b) _xen_queue_pt_update((a), (b), __FILE__, __LINE__) +#else +#define xen_queue_pt_update(a, b) _xen_queue_pt_update((a), (b), NULL, 0) +#endif + +#ifdef PMAP_DEBUG +#define PMAP_REF pmap_ref +#define PMAP_DEC_REF_PAGE pmap_dec_ref_page +#define PMAP_MARK_PRIV pmap_mark_privileged +#define PMAP_MARK_UNPRIV pmap_mark_unprivileged +#else +#define PMAP_MARK_PRIV(a) +#define PMAP_MARK_UNPRIV(a) +#define PMAP_REF(a, b) +#define PMAP_DEC_REF_PAGE(a) +#endif + +#define ALWAYS_SYNC 0 + +#ifdef PT_DEBUG +#define PT_LOG() printk("WP PT_SET %s:%d\n", __FILE__, __LINE__) +#else +#define PT_LOG() +#endif + +#define INVALID_P2M_ENTRY (~0UL) + +#define pmap_valid_entry(E) ((E) & PG_V) /* is PDE or PTE valid? */ + +#define SH_PD_SET_VA 1 +#define SH_PD_SET_VA_MA 2 +#define SH_PD_SET_VA_CLEAR 3 + +struct pmap; +void pd_set(struct pmap *pmap, int ptepindex, vm_paddr_t val, int type); +#ifdef notyet +static vm_paddr_t +vptetomachpte(vm_paddr_t *pte) +{ + vm_offset_t offset, ppte; + vm_paddr_t pgoffset, retval, *pdir_shadow_ptr; + int pgindex; + + ppte = (vm_offset_t)pte; + pgoffset = (ppte & PAGE_MASK); + offset = ppte - (vm_offset_t)PTmap; + pgindex = ppte >> PDRSHIFT; + + pdir_shadow_ptr = (vm_paddr_t *)PCPU_GET(pdir_shadow); + retval = (pdir_shadow_ptr[pgindex] & ~PAGE_MASK) + pgoffset; + return (retval); +} +#endif +#define PT_GET(_ptp) \ + (pmap_valid_entry(*(_ptp)) ? xpmap_mtop(*(_ptp)) : (0)) + +#ifdef WRITABLE_PAGETABLES + +#define PT_SET_VA(_ptp,_npte,sync) do { \ + PMAP_REF((_ptp), xpmap_ptom(_npte)); \ + PT_LOG(); \ + *(_ptp) = xpmap_ptom((_npte)); \ +} while (/*CONSTCOND*/0) +#define PT_SET_VA_MA(_ptp,_npte,sync) do { \ + PMAP_REF((_ptp), (_npte)); \ + PT_LOG(); \ + *(_ptp) = (_npte); \ +} while (/*CONSTCOND*/0) +#define PT_CLEAR_VA(_ptp, sync) do { \ + PMAP_REF((pt_entry_t *)(_ptp), 0); \ + PT_LOG(); \ + *(_ptp) = 0; \ +} while (/*CONSTCOND*/0) + +#define PD_SET_VA(_pmap, _ptp, _npte, sync) do { \ + PMAP_REF((_ptp), xpmap_ptom(_npte)); \ + pd_set((_pmap),(_ptp),(_npte), SH_PD_SET_VA); \ + if (sync || ALWAYS_SYNC) xen_flush_queue(); \ +} while (/*CONSTCOND*/0) +#define PD_SET_VA_MA(_pmap, _ptp, _npte, sync) do { \ + PMAP_REF((_ptp), (_npte)); \ + pd_set((_pmap),(_ptp),(_npte), SH_PD_SET_VA_MA); \ + if (sync || ALWAYS_SYNC) xen_flush_queue(); \ +} while (/*CONSTCOND*/0) +#define PD_CLEAR_VA(_pmap, _ptp, sync) do { \ + PMAP_REF((pt_entry_t *)(_ptp), 0); \ + pd_set((_pmap),(_ptp), 0, SH_PD_SET_VA_CLEAR); \ + if (sync || ALWAYS_SYNC) xen_flush_queue(); \ +} while (/*CONSTCOND*/0) + +#else /* !WRITABLE_PAGETABLES */ + +#define PT_SET_VA(_ptp,_npte,sync) do { \ + PMAP_REF((_ptp), xpmap_ptom(_npte)); \ + xen_queue_pt_update(vtomach(_ptp), \ + xpmap_ptom(_npte)); \ + if (sync || ALWAYS_SYNC) xen_flush_queue(); \ +} while (/*CONSTCOND*/0) +#define PT_SET_VA_MA(_ptp,_npte,sync) do { \ + PMAP_REF((_ptp), (_npte)); \ + xen_queue_pt_update(vtomach(_ptp), _npte); \ + if (sync || ALWAYS_SYNC) xen_flush_queue(); \ +} while (/*CONSTCOND*/0) +#define PT_CLEAR_VA(_ptp, sync) do { \ + PMAP_REF((pt_entry_t *)(_ptp), 0); \ + xen_queue_pt_update(vtomach(_ptp), 0); \ + if (sync || ALWAYS_SYNC) \ + xen_flush_queue(); \ +} while (/*CONSTCOND*/0) + +#define PD_SET_VA(_pmap, _ptepindex,_npte,sync) do { \ + PMAP_REF((_ptp), xpmap_ptom(_npte)); \ + pd_set((_pmap),(_ptepindex),(_npte), SH_PD_SET_VA); \ + if (sync || ALWAYS_SYNC) xen_flush_queue(); \ +} while (/*CONSTCOND*/0) +#define PD_SET_VA_MA(_pmap, _ptepindex,_npte,sync) do { \ + PMAP_REF((_ptp), (_npte)); \ + pd_set((_pmap),(_ptepindex),(_npte), SH_PD_SET_VA_MA); \ + if (sync || ALWAYS_SYNC) xen_flush_queue(); \ +} while (/*CONSTCOND*/0) +#define PD_CLEAR_VA(_pmap, _ptepindex, sync) do { \ + PMAP_REF((pt_entry_t *)(_ptp), 0); \ + pd_set((_pmap),(_ptepindex), 0, SH_PD_SET_VA_CLEAR); \ + if (sync || ALWAYS_SYNC) xen_flush_queue(); \ +} while (/*CONSTCOND*/0) + +#endif + +#define PT_SET_MA(_va, _ma) \ +do { \ + PANIC_IF(HYPERVISOR_update_va_mapping(((unsigned long)(_va)),\ + (_ma), \ + UVMF_INVLPG| UVMF_ALL) < 0); \ +} while (/*CONSTCOND*/0) + +#define PT_UPDATES_FLUSH() do { \ + xen_flush_queue(); \ +} while (/*CONSTCOND*/0) + +static __inline vm_paddr_t +xpmap_mtop(vm_paddr_t mpa) +{ + vm_paddr_t tmp = (mpa & PG_FRAME); + + return machtophys(tmp) | (mpa & ~PG_FRAME); +} + +static __inline vm_paddr_t +xpmap_ptom(vm_paddr_t ppa) +{ + vm_paddr_t tmp = (ppa & PG_FRAME); + + return phystomach(tmp) | (ppa & ~PG_FRAME); +} + +static __inline void +set_phys_to_machine(unsigned long pfn, unsigned long mfn) +{ +#ifdef notyet + PANIC_IF(max_mapnr && pfn >= max_mapnr); +#endif + if (xen_feature(XENFEAT_auto_translated_physmap)) { +#ifdef notyet + PANIC_IF((pfn != mfn && mfn != INVALID_P2M_ENTRY)); +#endif + return; + } + xen_phys_machine[pfn] = mfn; +} + + + + +#endif /* _XEN_XENPMAP_H_ */ diff --git a/sys/amd64/include/xen/xenvar.h b/sys/amd64/include/xen/xenvar.h new file mode 100644 index 0000000..1433b76 --- /dev/null +++ b/sys/amd64/include/xen/xenvar.h @@ -0,0 +1,122 @@ +/* + * Copyright (c) 2008 Kip Macy + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES + * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. + * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, + * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF + * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * + * $FreeBSD$ + */ +#ifndef XENVAR_H_ +#define XENVAR_H_ +#define XBOOTUP 0x1 +#define XPMAP 0x2 +extern int xendebug_flags; +#ifndef NOXENDEBUG +#define XENPRINTF printk +#else +#define XENPRINTF printf +#endif +#include <xen/features.h> + +#if 0 +#define TRACE_ENTER XENPRINTF("(file=%s, line=%d) entered %s\n", __FILE__, __LINE__, __FUNCTION__) +#define TRACE_EXIT XENPRINTF("(file=%s, line=%d) exiting %s\n", __FILE__, __LINE__, __FUNCTION__) +#define TRACE_DEBUG(argflags, _f, _a...) \ +if (xendebug_flags & argflags) XENPRINTF("(file=%s, line=%d) " _f "\n", __FILE__, __LINE__, ## _a); +#else +#define TRACE_ENTER +#define TRACE_EXIT +#define TRACE_DEBUG(argflags, _f, _a...) +#endif + +#ifdef XENHVM + +static inline vm_paddr_t +phystomach(vm_paddr_t pa) +{ + + return (pa); +} + +static inline vm_paddr_t +machtophys(vm_paddr_t ma) +{ + + return (ma); +} + +#define vtomach(va) pmap_kextract((vm_offset_t) (va)) +#define PFNTOMFN(pa) (pa) +#define MFNTOPFN(ma) (ma) + +#define set_phys_to_machine(pfn, mfn) ((void)0) +#define phys_to_machine_mapping_valid(pfn) (TRUE) +#define PT_UPDATES_FLUSH() ((void)0) + +#else + +extern xen_pfn_t *xen_phys_machine; + + +extern xen_pfn_t *xen_machine_phys; +/* Xen starts physical pages after the 4MB ISA hole - + * FreeBSD doesn't + */ + + +#undef ADD_ISA_HOLE /* XXX */ + +#ifdef ADD_ISA_HOLE +#define ISA_INDEX_OFFSET 1024 +#define ISA_PDR_OFFSET 1 +#else +#define ISA_INDEX_OFFSET 0 +#define ISA_PDR_OFFSET 0 +#endif + + +#define PFNTOMFN(i) (xen_phys_machine[(i)]) +#define MFNTOPFN(i) ((vm_paddr_t)xen_machine_phys[(i)]) + +#define VTOP(x) ((((uintptr_t)(x))) - KERNBASE) +#define PTOV(x) (((uintptr_t)(x)) + KERNBASE) + +#define VTOPFN(x) (VTOP(x) >> PAGE_SHIFT) +#define PFNTOV(x) PTOV((vm_paddr_t)(x) << PAGE_SHIFT) + +#define VTOMFN(va) (vtomach(va) >> PAGE_SHIFT) +#define PFN_UP(x) (((x) + PAGE_SIZE-1) >> PAGE_SHIFT) + +#define phystomach(pa) (((vm_paddr_t)(PFNTOMFN((pa) >> PAGE_SHIFT))) << PAGE_SHIFT) +#define machtophys(ma) (((vm_paddr_t)(MFNTOPFN((ma) >> PAGE_SHIFT))) << PAGE_SHIFT) + +#endif + +void xpq_init(void); + +int xen_create_contiguous_region(vm_page_t pages, int npages); + +void xen_destroy_contiguous_region(void * addr, int npages); + +#endif |