zephyr/arch/x86_64/include/kernel_arch_func.h

117 lines
2.5 KiB
C
Raw Normal View History

arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
/*
* Copyright (c) 2018 Intel Corporation
*
* SPDX-License-Identifier: Apache-2.0
*/
#ifndef _KERNEL_ARCH_FUNC_H
#define _KERNEL_ARCH_FUNC_H
#include <irq.h>
#include <xuk-switch.h>
static inline void kernel_arch_init(void)
{
/* This is a noop, we already took care of things before
* z_cstart() is entered
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
*/
}
static inline struct _cpu *z_arch_curr_cpu(void)
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
{
long long ret, off = 0;
/* The struct _cpu pointer for the current CPU lives at the
* start of the the FS segment
*/
__asm__("movq %%fs:(%1), %0" : "=r"(ret) : "r"(off));
return (struct _cpu *)(long)ret;
}
static inline unsigned int z_arch_irq_lock(void)
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
{
unsigned long long key;
__asm__ volatile("pushfq; cli; popq %0" : "=r"(key));
return (int)key;
}
static inline void z_arch_irq_unlock(unsigned int key)
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
{
if (key & 0x200) {
__asm__ volatile("sti");
}
}
/**
* Returns true if interrupts were unlocked prior to the
* z_arch_irq_lock() call that produced the key argument.
*/
static inline bool z_arch_irq_unlocked(unsigned int key)
{
return (key & 0x200) != 0;
}
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
static inline void arch_nop(void)
{
__asm__ volatile("nop");
}
void z_arch_irq_disable(unsigned int irq);
void z_arch_irq_enable(unsigned int irq);
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
/* Not a standard Zephyr function, but probably will be */
static inline unsigned long long z_arch_k_cycle_get_64(void)
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
{
unsigned int hi, lo;
__asm__ volatile("rdtsc" : "=d"(hi), "=a"(lo));
return (((unsigned long long)hi) << 32) | lo;
}
static inline unsigned int z_arch_k_cycle_get_32(void)
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
{
#ifdef CONFIG_HPET_TIMER
extern u32_t z_timer_cycle_get_32(void);
return z_timer_cycle_get_32();
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
#else
return (u32_t)z_arch_k_cycle_get_64();
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
#endif
}
#define z_arch_is_in_isr() (z_arch_curr_cpu()->nested != 0)
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
static inline void z_arch_switch(void *switch_to, void **switched_from)
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
{
xuk_switch(switch_to, switched_from);
}
static inline u32_t x86_apic_scaled_tsc(void)
{
u32_t lo, hi;
u64_t tsc;
__asm__ volatile("rdtsc" : "=a"(lo), "=d"(hi));
tsc = (((u64_t)hi) << 32) | lo;
return (u32_t)(tsc >> CONFIG_XUK_APIC_TSC_SHIFT);
}
void x86_apic_set_timeout(u32_t cyc_from_now);
#define Z_ARCH_IRQ_CONNECT(irq, pri, isr, arg, flags) \
z_arch_irq_connect_dynamic(irq, pri, isr, arg, flags)
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
extern int x86_64_except_reason;
/* Vector 5 is the "bounds" exception which is otherwise vestigial
* (BOUND is an illegal instruction in long mode)
*/
#define Z_ARCH_EXCEPT(reason) do { \
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
x86_64_except_reason = reason; \
__asm__ volatile("int $5"); \
} while (false)
void z_arch_sched_ipi(void);
arch/x86_64: New architecture added This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr. Only the basic architecture support needed to run 64 bit code is added; no drivers are added, though a low-level console exists and is wired to printk(). The support is built on top of a "X86 underkernel" layer, which can be built in isolation as a unit test on a Linux host. Limitations: + Right now the SDK lacks an x86_64 toolchain. The build will fall back to a host toolchain if it finds no cross compiler defined, which is tested to work on gcc 8.2.1 right now. + No x87/SSE/AVX usage is allowed. This is a stronger limitation than other architectures where the instructions work from one thread even if the context switch code doesn't support it. We are passing -no-sse to prevent gcc from automatically generating SSE instructions for non-floating-point purposes, which has the side effect of changing the ABI. Future work to handle the FPU registers will need to be combined with an "application" ABI distinct from the kernel one (or just to require USERSPACE). + Paging is enabled (it has to be in long mode), but is a 1:1 mapping of all memory. No MMU/USERSPACE support yet. + We are building with -mno-red-zone for stack size reasons, but this is a valuable optimization. Enabling it requires automatic stack switching, which requires a TSS, which means it has to happen after MMU support. + The OS runs in 64 bit mode, but for compatibility reasons is compiled to the 32 bit "X32" ABI. So while the full 64 bit registers and instruction set are available, C pointers are 32 bits long and Zephyr is constrained to run in the bottom 4G of memory. Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2018-08-19 12:24:48 -07:00
#endif /* _KERNEL_ARCH_FUNC_H */