251 lines
6.2 KiB
C
251 lines
6.2 KiB
C
/*
|
|
* Copyright (C) 2018 Intel Corporation. All rights reserved.
|
|
*
|
|
* SPDX-License-Identifier: BSD-3-Clause
|
|
*/
|
|
|
|
#include <hypervisor.h>
|
|
|
|
int dm_emulate_pio_post(struct vcpu *vcpu)
|
|
{
|
|
uint16_t cur = vcpu->vcpu_id;
|
|
int cur_context = vcpu->arch_vcpu.cur_context;
|
|
union vhm_request_buffer *req_buf = NULL;
|
|
uint32_t mask =
|
|
0xFFFFFFFFUL >> (32 - 8 * vcpu->req.reqs.pio_request.size);
|
|
uint64_t *rax;
|
|
|
|
req_buf = (union vhm_request_buffer *)(vcpu->vm->sw.io_shared_page);
|
|
|
|
rax = &vcpu->arch_vcpu.contexts[cur_context].guest_cpu_regs.regs.rax;
|
|
vcpu->req.reqs.pio_request.value =
|
|
req_buf->req_queue[cur].reqs.pio_request.value;
|
|
|
|
/* VHM emulation data already copy to req, mark to free slot now */
|
|
req_buf->req_queue[cur].valid = false;
|
|
|
|
if (req_buf->req_queue[cur].processed != REQ_STATE_SUCCESS)
|
|
return -1;
|
|
|
|
if (vcpu->req.reqs.pio_request.direction == REQUEST_READ)
|
|
*rax = ((*rax) & ~mask) |
|
|
(vcpu->req.reqs.pio_request.value & mask);
|
|
|
|
return 0;
|
|
}
|
|
|
|
static void dm_emulate_pio_pre(struct vcpu *vcpu, uint64_t exit_qual,
|
|
uint32_t sz, uint64_t req_value)
|
|
{
|
|
vcpu->req.type = REQ_PORTIO;
|
|
if (VM_EXIT_IO_INSTRUCTION_ACCESS_DIRECTION(exit_qual) != 0U)
|
|
vcpu->req.reqs.pio_request.direction = REQUEST_READ;
|
|
else
|
|
vcpu->req.reqs.pio_request.direction = REQUEST_WRITE;
|
|
|
|
vcpu->req.reqs.pio_request.address =
|
|
VM_EXIT_IO_INSTRUCTION_PORT_NUMBER(exit_qual);
|
|
vcpu->req.reqs.pio_request.size = sz;
|
|
vcpu->req.reqs.pio_request.value = req_value;
|
|
}
|
|
|
|
int io_instr_vmexit_handler(struct vcpu *vcpu)
|
|
{
|
|
uint32_t sz;
|
|
uint32_t mask;
|
|
uint32_t port;
|
|
int8_t direction;
|
|
struct vm_io_handler *handler;
|
|
uint64_t exit_qual;
|
|
struct vm *vm = vcpu->vm;
|
|
int cur_context_idx = vcpu->arch_vcpu.cur_context;
|
|
struct run_context *cur_context;
|
|
int status = -EINVAL;
|
|
|
|
cur_context = &vcpu->arch_vcpu.contexts[cur_context_idx];
|
|
exit_qual = vcpu->arch_vcpu.exit_qualification;
|
|
|
|
sz = VM_EXIT_IO_INSTRUCTION_SIZE(exit_qual) + 1;
|
|
port = VM_EXIT_IO_INSTRUCTION_PORT_NUMBER(exit_qual);
|
|
direction = VM_EXIT_IO_INSTRUCTION_ACCESS_DIRECTION(exit_qual);
|
|
mask = 0xfffffffful >> (32 - 8 * sz);
|
|
|
|
TRACE_4I(TRACE_VMEXIT_IO_INSTRUCTION, port, direction, sz,
|
|
cur_context_idx);
|
|
|
|
for (handler = vm->arch_vm.io_handler;
|
|
handler; handler = handler->next) {
|
|
|
|
if ((port >= handler->desc.addr + handler->desc.len) ||
|
|
(port + sz <= handler->desc.addr))
|
|
continue;
|
|
else if (!((port >= handler->desc.addr) && ((port + sz)
|
|
<= (handler->desc.addr + handler->desc.len)))) {
|
|
pr_fatal("Err:IO, port 0x%04x, size=%u spans devices",
|
|
port, sz);
|
|
return -EIO;
|
|
}
|
|
|
|
if (direction == 0) {
|
|
handler->desc.io_write(handler, vm, port, sz,
|
|
cur_context->guest_cpu_regs.regs.rax);
|
|
|
|
pr_dbg("IO write on port %04x, data %08x", port,
|
|
cur_context->guest_cpu_regs.regs.rax & mask);
|
|
|
|
status = 0;
|
|
break;
|
|
} else {
|
|
uint32_t data = handler->desc.io_read(handler, vm,
|
|
port, sz);
|
|
|
|
cur_context->guest_cpu_regs.regs.rax &= ~mask;
|
|
cur_context->guest_cpu_regs.regs.rax |= data & mask;
|
|
|
|
pr_dbg("IO read on port %04x, data %08x", port, data);
|
|
|
|
status = 0;
|
|
break;
|
|
}
|
|
}
|
|
|
|
/* Go for VHM */
|
|
if (status != 0) {
|
|
uint64_t *rax = &cur_context->guest_cpu_regs.regs.rax;
|
|
|
|
(void)memset(&vcpu->req, 0, sizeof(struct vhm_request));
|
|
dm_emulate_pio_pre(vcpu, exit_qual, sz, *rax);
|
|
status = acrn_insert_request_wait(vcpu, &vcpu->req);
|
|
}
|
|
|
|
if (status != 0) {
|
|
pr_fatal("Err:IO %s access to port 0x%04x, size=%u",
|
|
(direction != 0) ? "read" : "write", port, sz);
|
|
|
|
}
|
|
|
|
return status;
|
|
}
|
|
|
|
static void register_io_handler(struct vm *vm, struct vm_io_handler *hdlr)
|
|
{
|
|
if (vm->arch_vm.io_handler != NULL)
|
|
hdlr->next = vm->arch_vm.io_handler;
|
|
|
|
vm->arch_vm.io_handler = hdlr;
|
|
}
|
|
|
|
static void empty_io_handler_list(struct vm *vm)
|
|
{
|
|
struct vm_io_handler *handler = vm->arch_vm.io_handler;
|
|
struct vm_io_handler *tmp;
|
|
|
|
while (handler != NULL) {
|
|
tmp = handler;
|
|
handler = tmp->next;
|
|
free(tmp);
|
|
}
|
|
vm->arch_vm.io_handler = NULL;
|
|
}
|
|
|
|
void free_io_emulation_resource(struct vm *vm)
|
|
{
|
|
empty_io_handler_list(vm);
|
|
|
|
/* Free I/O emulation bitmaps */
|
|
free(vm->arch_vm.iobitmap[0]);
|
|
free(vm->arch_vm.iobitmap[1]);
|
|
}
|
|
|
|
void allow_guest_io_access(struct vm *vm, uint32_t address, uint32_t nbytes)
|
|
{
|
|
uint32_t *b;
|
|
uint32_t i;
|
|
uint32_t a;
|
|
|
|
b = vm->arch_vm.iobitmap[0];
|
|
for (i = 0U; i < nbytes; i++) {
|
|
if ((address & 0x8000U) != 0U)
|
|
b = vm->arch_vm.iobitmap[1];
|
|
a = address & 0x7fffU;
|
|
b[a >> 5] &= ~(1 << (a & 0x1fU));
|
|
address++;
|
|
}
|
|
}
|
|
|
|
static void deny_guest_io_access(struct vm *vm, uint32_t address, uint32_t nbytes)
|
|
{
|
|
uint32_t *b;
|
|
uint32_t i;
|
|
uint32_t a;
|
|
|
|
b = vm->arch_vm.iobitmap[0];
|
|
for (i = 0U; i < nbytes; i++) {
|
|
if ((address & 0x8000U) != 0U)
|
|
b = vm->arch_vm.iobitmap[1];
|
|
a = address & 0x7fffU;
|
|
b[a >> 5U] |= (1U << (a & 0x1fU));
|
|
address++;
|
|
}
|
|
}
|
|
|
|
static struct vm_io_handler *create_io_handler(uint32_t port, uint32_t len,
|
|
io_read_fn_t io_read_fn_ptr,
|
|
io_write_fn_t io_write_fn_ptr)
|
|
{
|
|
|
|
struct vm_io_handler *handler;
|
|
|
|
handler = calloc(1, sizeof(struct vm_io_handler));
|
|
|
|
if (handler != NULL) {
|
|
handler->desc.addr = port;
|
|
handler->desc.len = len;
|
|
handler->desc.io_read = io_read_fn_ptr;
|
|
handler->desc.io_write = io_write_fn_ptr;
|
|
} else {
|
|
pr_err("Error: out of memory");
|
|
}
|
|
|
|
return handler;
|
|
}
|
|
|
|
void setup_io_bitmap(struct vm *vm)
|
|
{
|
|
/* Allocate VM architecture state and IO bitmaps A and B */
|
|
vm->arch_vm.iobitmap[0] = alloc_page();
|
|
vm->arch_vm.iobitmap[1] = alloc_page();
|
|
|
|
ASSERT((vm->arch_vm.iobitmap[0] != NULL) &&
|
|
(vm->arch_vm.iobitmap[1] != NULL), "");
|
|
|
|
if (is_vm0(vm)) {
|
|
(void)memset(vm->arch_vm.iobitmap[0], 0x00, CPU_PAGE_SIZE);
|
|
(void)memset(vm->arch_vm.iobitmap[1], 0x00, CPU_PAGE_SIZE);
|
|
} else {
|
|
/* block all IO port access from Guest */
|
|
(void)memset(vm->arch_vm.iobitmap[0], 0xFF, CPU_PAGE_SIZE);
|
|
(void)memset(vm->arch_vm.iobitmap[1], 0xFF, CPU_PAGE_SIZE);
|
|
}
|
|
}
|
|
|
|
void register_io_emulation_handler(struct vm *vm, struct vm_io_range *range,
|
|
io_read_fn_t io_read_fn_ptr,
|
|
io_write_fn_t io_write_fn_ptr)
|
|
{
|
|
struct vm_io_handler *handler = NULL;
|
|
|
|
if (io_read_fn_ptr == NULL || io_write_fn_ptr == NULL) {
|
|
pr_err("Invalid IO handler.");
|
|
return;
|
|
}
|
|
|
|
if (is_vm0(vm))
|
|
deny_guest_io_access(vm, range->base, range->len);
|
|
|
|
handler = create_io_handler(range->base,
|
|
range->len, io_read_fn_ptr, io_write_fn_ptr);
|
|
|
|
register_io_handler(vm, handler);
|
|
}
|