/*
*
* Trampoline.S Derived from Setup.S by Linus Torvalds
*
* 4 Jan 1997 Michael Chastain: changed to gnu as.
* 15 Sept 2005 Eric Biederman: 64bit PIC support
*
* Entry: CS:IP point to the start of our code, we are
* in real mode with no stack, but the rest of the
* trampoline page to make our stack and everything else
* is a mystery.
*
* In fact we don't actually need a stack so we don't
* set one up.
*
* On entry to trampoline_data, the processor is in real mode
* with 16-bit addressing and 16-bit data. CS has some value
* and IP is zero. Thus, data addresses need to be absolute
* (no relocation) and are taken with regard to r_base.
*
* With the addition of trampoline_level4_pgt this code can
* now enter a 64bit kernel that lives at arbitrary 64bit
* physical addresses.
*
* If you work on this file, check the object module with objdump
* --full-contents --reloc to make sure there are no relocation
* entries.
*/
#include <linux/linkage.h>
#include <asm/pgtable.h>
#include <asm/page.h>
#include <asm/msr.h>
#include <asm/segment.h>
.data
.code16
ENTRY(trampoline_data)
r_base = .
cli # We should be safe anyway
wbinvd
mov %cs, %ax # Code and data in the same place
mov %ax, %ds
mov %ax, %es
mov %ax, %ss
movl $0xA5A5A5A5, trampoline_data - r_base
# write marker for master knows we're running
# Setup stack
movw $(trampoline_stack_end - r_base), %sp
call verify_cpu # Verify the cpu supports long mode
mov %cs, %ax
movzx %ax, %esi # Find the 32bit trampoline location
shll $4, %esi
# Fixup the vectors
addl %esi, startup_32_vector - r_base
addl %esi, startup_64_vector - r_base
addl %esi, tgdt + 2 - r_base # Fixup the gdt pointer
/*
* GDT tables in non default location kernel can be beyond 16MB and
* lgdt will not be able to load the address as in real mode default
* operand size is 16bit. Use lgdtl instead to force operand size
* to 32 bit.
*/
lidtl tidt - r_base # load idt with 0, 0
lgdtl tgdt - r_base # load gdt with whatever is appropriate
xor %ax, %ax
inc %ax # protected mode (PE) bit
lmsw %ax # into protected mode
# flush prefetch and jump to startup_32
ljmpl *(startup_32_vector - r_base)
.code32
.balign 4
startup_32:
movl $__KERNEL_DS, %eax # Initialize the %ds segment register
movl %eax, %ds
xorl %eax, %eax
btsl $5, %eax # Enable PAE mode
movl %eax, %cr4
# Setup trampoline 4 level pagetables
leal (trampoline_level4_pgt - r_base)(%esi), %eax
movl %eax, %cr3
movl $MSR_EFER, %ecx
movl $(1 << _EFER_LME), %eax # Enable Long Mode
xorl %edx, %edx
wrmsr
xorl %eax, %eax
btsl $31, %eax # Enable paging and in turn activate Long Mode
btsl $0, %eax # Enable protected mode
movl %eax, %cr0
/*
* At this point we're in long mode but in 32bit compatibility mode
* with EFER.LME = 1, CS.L = 0, CS.D = 1 (and in turn
* EFER.LMA = 1). Now we want to jump in 64bit mode, to do that we use
* the new gdt/idt that has __KERNEL_CS with CS.L = 1.
*/
ljmp *(startup_64_vector - r_base)(%esi)
.code64
.balign 4
startup_64:
# Now jump into the kernel using virtual addresses
movq $secondary_startup_64, %rax
jmp *%rax
.code16
verify_cpu:
pushl $0 # Kill any dangerous flags
popfl
/* minimum CPUID flags for x86-64 */
/* see http://www.x86-64.org/lists/discuss/msg02971.html */
#define REQUIRED_MASK1 ((1<<0)|(1<<3)|(1<<4)|(1<<5)|(1<<6)|(1<<8)|\
(1<<13)|(1<<15)|(1<<24)|(1<<25)|(1<<26))
#define REQUIRED_MASK2 (1<<29)
pushfl # check for cpuid
popl %eax
movl %eax, %ebx
xorl $0x200000,%eax
pushl %eax
popfl
pushfl
popl %eax
pushl %ebx
popfl
cmpl %eax, %ebx
jz no_longmode
xorl %eax, %eax # See if cpuid 1 is implemented
cpuid
cmpl $0x1, %eax
jb no_longmode
movl $0x01, %eax # Does the cpu have what it takes?
cpuid
andl $REQUIRED_MASK1, %edx
xorl $REQUIRED_MASK1, %edx
jnz no_longmode
movl $0x80000000, %eax # See if extended cpuid is implemented
cpuid
cmpl $0x80000001, %eax
jb no_longmode
movl $0x80000001, %eax # Does the cpu have what it takes?
cpuid
andl $REQUIRED_MASK2, %edx
xorl $REQUIRED_MASK2, %edx
jnz no_longmode
ret # The cpu supports long mode
no_longmode:
hlt
jmp no_longmode
# Careful these need to be in the same 64K segment as the above;
tidt:
.word 0 # idt limit = 0
.word 0, 0 # idt base = 0L
# Duplicate the global descriptor table
# so the kernel can live anywhere
.balign 4
tgdt:
.short tgdt_end - tgdt # gdt limit
.long tgdt - r_base
.short 0
.quad 0x00cf9b000000ffff # __KERNEL32_CS
.quad 0x00af9b000000ffff # __KERNEL_CS
.quad 0x00cf93000000ffff # __KERNEL_DS
tgdt_end:
.balign 4
startup_32_vector:
.long startup_32 - r_base
.word __KERNEL32_CS, 0
.balign 4
startup_64_vector:
.long startup_64 - r_base
.word __KERNEL_CS, 0
trampoline_stack:
.org 0x1000
trampoline_stack_end:
ENTRY(trampoline_level4_pgt)
.quad level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE
.fill 510,8,0
.quad level3_kernel_pgt - __START_KERNEL_map + _KERNPG_TABLE
ENTRY(trampoline_end)
