From: Matt Tolentino Attached is another patch that enables EFI boot-up support in ia32 kernels. This one has numerous changes based on the various comments and suggestions made on the last one several weeks ago. The patch is against 2.6.0-test6-mm2. This patch departs from the last in that I am no longer adding separate boot parameters at the end of the the empty_zero_page; rather, based on Eric's suggestion, I've added EFI parameters via offset grouped with the rest of the existing boot parameters. This allows the reuse of several parameters which were redundant (command-line, initrd, etc.). This also simplifies some of the code in setup_arch(). In order to continue to determine whether the kernel should initialize using EFI tables, I've temporarily added a check on the LOADER_TYPE boot parameter. Although I haven't requested that elilo be assigned an id for this yet, I've used this to determine whether the kernel should use the EFI initialization path as well as a check to see if the EFI_SYSTAB boot parameter contains anything. If someone has a better suggestion for determining this, I'm open... This patch also uses the existing ioremapping functions to map the efi tables into kernel virtual address space. I've added an option such that I could use Dave Hansen's boot_ioremap() before paging_init(). After paging_init, I then remap the efi memmap using bt_ioremap for use later. This has eliminated the need for several functions...thanks for the suggestions and thanks for your help Dave. Still this could use a look-see. Something else that I'd like to solicit suggestions on is how to deal with CONFIG_ACPI_EFI. One of the goals of this patch is to enable the kernel to "just know" at boot time when to use EFI information to initialize. That way, the same kernel image would theoretically still boot on existing systems with a different BIOS. I think this patch accomplishes that goal to some degree, but how best to deal with when/when not to look at the EFI config table for the RSDP is still dealt with at compile time. I'd like to dynamically determine this as well, but am not sure of the best way...suggestions? Len? Andrew? Note, I'm still working on verifying this works properly on a system with >4GB system memory, as well as with the 4G/4G split (hence the reason this is against the -mm kernel)...hopefully, I'll be able to steal some time on one of those monsters next week. arch/i386/Kconfig | 3 arch/i386/kernel/Makefile | 2 arch/i386/kernel/acpi/boot.c | 5 arch/i386/kernel/efi.c | 606 +++++++++++++++++++++++++++++++++++++++++++ arch/i386/kernel/efi_stub.S | 125 ++++++++ arch/i386/kernel/reboot.c | 12 arch/i386/kernel/setup.c | 133 ++++++++- arch/i386/kernel/time.c | 66 ++++ arch/i386/mm/init.c | 21 + drivers/acpi/Kconfig | 12 drivers/acpi/osl.c | 1 include/asm-i386/setup.h | 5 include/linux/efi.h | 33 ++ init/main.c | 3 14 files changed, 999 insertions(+), 28 deletions(-) diff -puN arch/i386/Kconfig~ia32-efi-support arch/i386/Kconfig --- 25/arch/i386/Kconfig~ia32-efi-support 2003-10-15 10:15:56.000000000 -0700 +++ 25-akpm/arch/i386/Kconfig 2003-10-15 10:15:56.000000000 -0700 @@ -793,7 +793,8 @@ config HAVE_DEC_LOCK # Summit needs it only when NUMA is on config BOOT_IOREMAP bool - depends on ((X86_SUMMIT || X86_GENERICARCH) && NUMA) + depends on X86_PC +# depends on (((X86_SUMMIT || X86_GENERICARCH) && NUMA)) || X86_GENERICARCH default y endmenu diff -puN arch/i386/kernel/acpi/boot.c~ia32-efi-support arch/i386/kernel/acpi/boot.c --- 25/arch/i386/kernel/acpi/boot.c~ia32-efi-support 2003-10-15 10:15:56.000000000 -0700 +++ 25-akpm/arch/i386/kernel/acpi/boot.c 2003-10-15 10:15:56.000000000 -0700 @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include @@ -296,6 +297,10 @@ acpi_find_rsdp (void) { unsigned long rsdp_phys = 0; + if (efi.acpi20) + return __pa(efi.acpi20); + else if (efi.acpi) + return __pa(efi.acpi); /* * Scan memory looking for the RSDP signature. First search EBDA (low * memory) paragraphs and then search upper memory (E0000-FFFFF). diff -puN /dev/null arch/i386/kernel/efi.c --- /dev/null 2002-08-30 16:31:37.000000000 -0700 +++ 25-akpm/arch/i386/kernel/efi.c 2003-10-15 10:15:56.000000000 -0700 @@ -0,0 +1,606 @@ +/* + * Extensible Firmware Interface + * + * Based on Extensible Firmware Interface Specification version 1.0 + * + * Copyright (C) 1999 VA Linux Systems + * Copyright (C) 1999 Walt Drummond + * Copyright (C) 1999-2002 Hewlett-Packard Co. + * David Mosberger-Tang + * Stephane Eranian + * + * All EFI Runtime Services are not implemented yet as EFI only + * supports physical mode addressing on SoftSDV. This is to be fixed + * in a future version. --drummond 1999-07-20 + * + * Implemented EFI runtime services and virtual mode calls. --davidm + * + * Goutham Rao: + * Skip non-WB memory and ignore empty memory ranges. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + +#define EFI_DEBUG 0 +#define PFX "EFI: " + +extern efi_status_t asmlinkage efi_call_phys(void *, ...); + +struct efi efi; +struct efi efi_phys __initdata; +struct efi_memory_map memmap __initdata; + +/* + * We require an early boot_ioremap mapping mechanism initially + */ +extern void * boot_ioremap(unsigned long, unsigned long); + +/* + * efi_dir is allocated here, but the directory isn't created + * here, as proc_mkdir() doesn't work this early in the bootup + * process. Therefore, each module, like efivars, must test for + * if (!efi_dir) efi_dir = proc_mkdir("efi", NULL); + * prior to creating their own entries under /proc/efi. + */ +#ifdef CONFIG_PROC_FS +struct proc_dir_entry *efi_dir; +#endif + + +/* + * To make EFI call EFI runtime service in physical addressing mode we need + * prelog/epilog before/after the invocation to disable interrupt, to + * claim EFI runtime service handler exclusively and to duplicate a memory in + * low memory space say 0 - 3G. + */ + +static unsigned long efi_rt_eflags; +static spinlock_t efi_rt_lock = SPIN_LOCK_UNLOCKED; +static pgd_t efi_bak_pg_dir_pointer[2]; + +static void efi_call_phys_prelog(void) +{ + unsigned long cr4; + unsigned long temp; + + spin_lock(&efi_rt_lock); + local_irq_save(efi_rt_eflags); + + /* + * If I don't have PSE, I should just duplicate two entries in page + * directory. If I have PSE, I just need to duplicate one entry in + * page directory. + */ + __asm__ __volatile__("movl %%cr4, %0":"=r"(cr4)); + + if (cr4 & X86_CR4_PSE) { + efi_bak_pg_dir_pointer[0].pgd = + swapper_pg_dir[pgd_index(0)].pgd; + swapper_pg_dir[0].pgd = + swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd; + } else { + efi_bak_pg_dir_pointer[0].pgd = + swapper_pg_dir[pgd_index(0)].pgd; + efi_bak_pg_dir_pointer[1].pgd = + swapper_pg_dir[pgd_index(0x400000)].pgd; + swapper_pg_dir[pgd_index(0)].pgd = + swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd; + temp = PAGE_OFFSET + 0x400000; + swapper_pg_dir[pgd_index(0x400000)].pgd = + swapper_pg_dir[pgd_index(temp)].pgd; + } + + /* + * After the lock is released, the original page table is restored. + */ + local_flush_tlb(); + + cpu_gdt_descr[0].address = __pa(cpu_gdt_descr[0].address); + __asm__ __volatile__("lgdt %0":"=m" + (*(struct Xgt_desc_struct *) __pa(&cpu_gdt_descr[0]))); +} + +static void efi_call_phys_epilog(void) +{ + unsigned long cr4; + + cpu_gdt_descr[0].address = + (unsigned long) __va(cpu_gdt_descr[0].address); + __asm__ __volatile__("lgdt %0":"=m"(cpu_gdt_descr)); + __asm__ __volatile__("movl %%cr4, %0":"=r"(cr4)); + + if (cr4 & X86_CR4_PSE) { + swapper_pg_dir[pgd_index(0)].pgd = + efi_bak_pg_dir_pointer[0].pgd; + } else { + swapper_pg_dir[pgd_index(0)].pgd = + efi_bak_pg_dir_pointer[0].pgd; + swapper_pg_dir[pgd_index(0x400000)].pgd = + efi_bak_pg_dir_pointer[1].pgd; + } + + /* + * After the lock is released, the original page table is restored. + */ + local_flush_tlb(); + + local_irq_restore(efi_rt_eflags); + spin_unlock(&efi_rt_lock); +} + +static efi_status_t +phys_efi_set_virtual_address_map(unsigned long memory_map_size, + unsigned long descriptor_size, + u32 descriptor_version, + efi_memory_desc_t *virtual_map) +{ + efi_status_t status; + + efi_call_phys_prelog(); + status = efi_call_phys(efi_phys.set_virtual_address_map, + memory_map_size, descriptor_size, + descriptor_version, virtual_map); + efi_call_phys_epilog(); + return status; +} + +efi_status_t +phys_efi_get_time(efi_time_t *tm, efi_time_cap_t *tc) +{ + efi_status_t status; + + efi_call_phys_prelog(); + status = efi_call_phys(efi_phys.get_time, tm, tc); + efi_call_phys_epilog(); + return status; +} + +void efi_gettimeofday(struct timespec *tv) +{ + efi_time_t tm; + + memset(tv, 0, sizeof(*tv)); + if ((*efi.get_time) (&tm, 0) != EFI_SUCCESS) + return; + + tv->tv_sec = mktime(tm.year, tm.month, tm.day, tm.hour, tm.minute, + tm.second); + tv->tv_nsec = tm.nanosecond; +} + +int is_available_memory(efi_memory_desc_t * md) +{ + if (!(md->attribute & EFI_MEMORY_WB)) + return 0; + + switch (md->type) { + case EFI_LOADER_CODE: + case EFI_LOADER_DATA: + case EFI_BOOT_SERVICES_CODE: + case EFI_BOOT_SERVICES_DATA: + case EFI_CONVENTIONAL_MEMORY: + return 1; + } + return 0; +} + +/* + * We need to map the EFI memory map again after paging_init(). + */ +void __init efi_map_memmap(void) +{ + memmap.map = NULL; + + memmap.map = (efi_memory_desc_t *) + bt_ioremap((unsigned long) memmap.phys_map, + (memmap.nr_map * sizeof(efi_memory_desc_t))); + + if (memmap.map == NULL) + printk(KERN_ERR PFX "Could not remap the EFI memmap!\n"); +} + +void __init print_efi_memmap(void) +{ + efi_memory_desc_t *md; + int i; + + for (i = 0; i < memmap.nr_map; i++) { + md = &memmap.map[i]; + printk(KERN_INFO "mem%02u: type=%u, attr=0x%llx, " + "range=[0x%016llx-0x%016llx) (%lluMB)\n", + i, md->type, md->attribute, md->phys_addr, + md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT), + (md->num_pages >> (20 - EFI_PAGE_SHIFT))); + } +} + +/* + * Walks the EFI memory map and calls CALLBACK once for each EFI + * memory descriptor that has memory that is available for kernel use. + */ +void efi_memmap_walk(efi_freemem_callback_t callback, void *arg) +{ + int prev_valid = 0; + struct range { + unsigned long start; + unsigned long end; + } prev, curr; + efi_memory_desc_t *md; + unsigned long start, end; + int i; + + for (i = 0; i < memmap.nr_map; i++) { + md = &memmap.map[i]; + + if ((md->num_pages == 0) || (!is_available_memory(md))) + continue; + + curr.start = md->phys_addr; + curr.end = curr.start + (md->num_pages << EFI_PAGE_SHIFT); + + if (!prev_valid) { + prev = curr; + prev_valid = 1; + } else { + if (curr.start < prev.start) + printk(KERN_INFO PFX "Unordered memory map\n"); + if (prev.end == curr.start) + prev.end = curr.end; + else { + start = + (unsigned long) (PAGE_ALIGN(prev.start)); + end = (unsigned long) (prev.end & PAGE_MASK); + if ((end > start) + && (*callback) (start, end, arg) < 0) + return; + prev = curr; + } + } + } + if (prev_valid) { + start = (unsigned long) PAGE_ALIGN(prev.start); + end = (unsigned long) (prev.end & PAGE_MASK); + if (end > start) + (*callback) (start, end, arg); + } +} + +void __init efi_init(void) +{ + efi_config_table_t *config_tables; + efi_runtime_services_t *runtime; + efi_char16_t *c16; + char vendor[100] = "unknown"; + unsigned long num_config_tables; + int i = 0; + + memset(&efi, 0, sizeof(efi) ); + memset(&efi_phys, 0, sizeof(efi_phys)); + + efi_phys.systab = EFI_SYSTAB; + memmap.phys_map = EFI_MEMMAP; + memmap.nr_map = EFI_MEMMAP_SIZE/EFI_MEMDESC_SIZE; + memmap.desc_version = EFI_MEMDESC_VERSION; + + efi.systab = (efi_system_table_t *) + boot_ioremap((unsigned long) efi_phys.systab, + sizeof(efi_system_table_t)); + /* + * Verify the EFI Table + */ + if (efi.systab == NULL) + printk(KERN_ERR PFX "Woah! Couldn't map the EFI system table.\n"); + if (efi.systab->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE) + printk(KERN_ERR PFX "Woah! EFI system table signature incorrect\n"); + if ((efi.systab->hdr.revision ^ EFI_SYSTEM_TABLE_REVISION) >> 16 != 0) + printk(KERN_ERR PFX + "Warning: EFI system table major version mismatch: " + "got %d.%02d, expected %d.%02d\n", + efi.systab->hdr.revision >> 16, + efi.systab->hdr.revision & 0xffff, + EFI_SYSTEM_TABLE_REVISION >> 16, + EFI_SYSTEM_TABLE_REVISION & 0xffff); + /* + * Grab some details from the system table + */ + num_config_tables = efi.systab->nr_tables; + config_tables = (efi_config_table_t *)efi.systab->tables; + runtime = efi.systab->runtime; + + /* + * Show what we know for posterity + */ + c16 = (efi_char16_t *) boot_ioremap(efi.systab->fw_vendor, 2); + if (c16) { + for (i = 0; i < sizeof(vendor) && *c16; ++i) + vendor[i] = *c16++; + vendor[i] = '\0'; + } else + printk(KERN_ERR PFX "Could not map the firmware vendor!\n"); + + printk(KERN_INFO PFX "EFI v%u.%.02u by %s \n", + efi.systab->hdr.revision >> 16, + efi.systab->hdr.revision & 0xffff, vendor); + + /* + * Let's see what config tables the firmware passed to us. + */ + config_tables = (efi_config_table_t *) + boot_ioremap((unsigned long) config_tables, + num_config_tables * sizeof(efi_config_table_t)); + + if (config_tables == NULL) + printk(KERN_ERR PFX "Could not map EFI Configuration Table!\n"); + + for (i = 0; i < num_config_tables; i++) { + if (efi_guidcmp(config_tables[i].guid, MPS_TABLE_GUID) == 0) { + efi.mps = (void *)config_tables[i].table; + printk(KERN_INFO " MPS=0x%lx ", config_tables[i].table); + } else + if (efi_guidcmp(config_tables[i].guid, ACPI_20_TABLE_GUID) == 0) { + efi.acpi20 = __va(config_tables[i].table); + printk(KERN_INFO " ACPI 2.0=0x%lx ", config_tables[i].table); + } else + if (efi_guidcmp(config_tables[i].guid, ACPI_TABLE_GUID) == 0) { + efi.acpi = __va(config_tables[i].table); + printk(KERN_INFO " ACPI=0x%lx ", config_tables[i].table); + } else + if (efi_guidcmp(config_tables[i].guid, SMBIOS_TABLE_GUID) == 0) { + efi.smbios = (void *) config_tables[i].table; + printk(KERN_INFO " SMBIOS=0x%lx ", config_tables[i].table); + } else + if (efi_guidcmp(config_tables[i].guid, HCDP_TABLE_GUID) == 0) { + efi.hcdp = (void *)config_tables[i].table; + printk(KERN_INFO " HCDP=0x%lx ", config_tables[i].table); + } else + if (efi_guidcmp(config_tables[i].guid, UGA_IO_PROTOCOL_GUID) == 0) { + efi.uga = (void *)config_tables[i].table; + printk(KERN_INFO " UGA=0x%lx ", config_tables[i].table); + } + } + printk("\n"); + + /* + * Check out the runtime services table. We need to map + * the runtime services table so that we can grab the physical + * address of several of the EFI runtime functions, needed to + * set the firmware into virtual mode. + */ + + runtime = (efi_runtime_services_t *) boot_ioremap((unsigned long) + runtime, + sizeof(efi_runtime_services_t)); + if (runtime != NULL) { + /* + * We will only need *early* access to the following + * two EFI runtime services before set_virtual_address_map + * is invoked. + */ + efi_phys.get_time = (efi_get_time_t *) runtime->get_time; + efi_phys.set_virtual_address_map = + (efi_set_virtual_address_map_t *) + runtime->set_virtual_address_map; + } else + printk(KERN_ERR PFX "Could not map the runtime service table!\n"); + + /* Map the EFI memory map for use until paging_init() */ + + memmap.map = (efi_memory_desc_t *) + boot_ioremap((unsigned long) EFI_MEMMAP, EFI_MEMMAP_SIZE); + + if (memmap.map == NULL) + printk(KERN_ERR PFX "Could not map the EFI memory map!\n"); + + if (EFI_MEMDESC_SIZE != sizeof(efi_memory_desc_t)) { + printk(KERN_WARNING PFX "Warning! Kernel-defined memdesc doesn't " + "match the one from EFI!\n"); + } +#if EFI_DEBUG + print_efi_memmap(); +#endif +} + +/* + * This function will switch the EFI runtime services to virtual mode. + * Essentially, look through the EFI memmap and map every region that + * has the runtime attribute bit set in its memory descriptor and update + * that memory descriptor with the virtual address obtained from ioremap(). + * This enables the runtime services to be called without having to + * thunk back into physical mode for every invocation. + */ + +void __init efi_enter_virtual_mode(void) +{ + efi_memory_desc_t *md; + efi_status_t status; + int i; + + efi.systab = NULL; + + for (i = 0; i < memmap.nr_map; i++) { + md = &memmap.map[i]; + + if (md->attribute & EFI_MEMORY_RUNTIME) { + md->virt_addr = + (u64) ioremap((unsigned long) md->phys_addr, + (unsigned long) (md->num_pages + << EFI_PAGE_SHIFT)); + if (!(unsigned long) md->virt_addr) { + printk(KERN_ERR PFX "ioremap of md: 0x%lX failed \n", + (unsigned long) md->phys_addr); + } + + if (((unsigned long)md->phys_addr <= (unsigned long)efi_phys.systab) && ((unsigned long)efi_phys.systab < md->phys_addr + ((unsigned long) md->num_pages << EFI_PAGE_SHIFT))) { + efi.systab = (efi_system_table_t *) + ((md->virt_addr - md->phys_addr) + + (u64)efi_phys.systab); + } + } + } + + if (!efi.systab) + BUG(); + + status = phys_efi_set_virtual_address_map( + sizeof(efi_memory_desc_t) * memmap.nr_map, + sizeof(efi_memory_desc_t), + memmap.desc_version, + memmap.phys_map); + + if (status != EFI_SUCCESS) { + printk (KERN_ALERT "You are screwed! " + "Unable to switch EFI into virtual mode " + "(status=%lx)\n", status); + panic("EFI call to SetVirtualAddressMap() failed!"); + } + + /* + * Now that EFI is in virtual mode, update the function + * pointers in the runtime service table to the new virtual addresses. + */ + + efi.get_time = (efi_get_time_t *) efi.systab->runtime->get_time; + efi.set_time = (efi_set_time_t *) efi.systab->runtime->set_time; + efi.get_wakeup_time = (efi_get_wakeup_time_t *) + efi.systab->runtime->get_wakeup_time; + efi.set_wakeup_time = (efi_set_wakeup_time_t *) + efi.systab->runtime->set_wakeup_time; + efi.get_variable = (efi_get_variable_t *) + efi.systab->runtime->get_variable; + efi.get_next_variable = (efi_get_next_variable_t *) + efi.systab->runtime->get_next_variable; + efi.set_variable = (efi_set_variable_t *) + efi.systab->runtime->set_variable; + efi.get_next_high_mono_count = (efi_get_next_high_mono_count_t *) + efi.systab->runtime->get_next_high_mono_count; + efi.reset_system = (efi_reset_system_t *) + efi.systab->runtime->reset_system; +} + +void __init +efi_initialize_iomem_resources(struct resource *code_resource, + struct resource *data_resource) +{ + struct resource *res; + efi_memory_desc_t *md; + int i; + + for (i = 0; i < memmap.nr_map; i++) { + md = &memmap.map[i]; + + if ((md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT)) > + 0x100000000ULL) + continue; + res = alloc_bootmem_low(sizeof(struct resource)); + switch (md->type) { + case EFI_RESERVED_TYPE: + res->name = "Reserved Memory"; + break; + case EFI_LOADER_CODE: + res->name = "Loader Code"; + break; + case EFI_LOADER_DATA: + res->name = "Loader Data"; + break; + case EFI_BOOT_SERVICES_DATA: + res->name = "BootServices Data"; + break; + case EFI_BOOT_SERVICES_CODE: + res->name = "BootServices Code"; + break; + case EFI_RUNTIME_SERVICES_CODE: + res->name = "Runtime Service Code"; + break; + case EFI_RUNTIME_SERVICES_DATA: + res->name = "Runtime Service Data"; + break; + case EFI_CONVENTIONAL_MEMORY: + res->name = "Conventional Memory"; + break; + case EFI_UNUSABLE_MEMORY: + res->name = "Unusable Memory"; + break; + case EFI_ACPI_RECLAIM_MEMORY: + res->name = "ACPI Reclaim"; + break; + case EFI_ACPI_MEMORY_NVS: + res->name = "ACPI NVS"; + break; + case EFI_MEMORY_MAPPED_IO: + res->name = "Memory Mapped IO"; + break; + case EFI_MEMORY_MAPPED_IO_PORT_SPACE: + res->name = "Memory Mapped IO Port Space"; + break; + default: + res->name = "Reserved"; + break; + } + res->start = md->phys_addr; + res->end = res->start + ((md->num_pages << EFI_PAGE_SHIFT) - 1); + res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; + if (request_resource(&iomem_resource, res) < 0) + printk(KERN_ERR PFX "Failed to allocate res %s : 0x%lx-0x%lx\n", + res->name, res->start, res->end); + /* + * We don't know which region contains kernel data so we try + * it repeatedly and let the resource manager test it. + */ + if (md->type == EFI_CONVENTIONAL_MEMORY) { + request_resource(res, code_resource); + request_resource(res, data_resource); + } + } +} + +/* + * Convenience functions to obtain memory types and attributes + */ + +u32 efi_mem_type(unsigned long phys_addr) +{ + efi_memory_desc_t *md; + int i; + + for (i = 0; i < memmap.nr_map; i++) { + md = &memmap.map[i]; + if ((md->phys_addr <= phys_addr) && (phys_addr < + (md->phys_addr + (md-> num_pages << EFI_PAGE_SHIFT)) )) + return md->type; + } + return 0; +} + +u64 efi_mem_attributes(unsigned long phys_addr) +{ + efi_memory_desc_t *md; + int i; + + for (i = 0; i < memmap.nr_map; i++) { + md = &memmap.map[i]; + if ((md->phys_addr <= phys_addr) && (phys_addr < + (md->phys_addr + (md-> num_pages << EFI_PAGE_SHIFT)) )) + return md->attribute; + } + return 0; +} + diff -puN /dev/null arch/i386/kernel/efi_stub.S --- /dev/null 2002-08-30 16:31:37.000000000 -0700 +++ 25-akpm/arch/i386/kernel/efi_stub.S 2003-10-15 10:15:56.000000000 -0700 @@ -0,0 +1,125 @@ +/* + * EFI call stub for IA32. + * + * This stub allows us to make EFI calls in physical mode with interrupts + * turned off. + */ + +#include +#include +#include +#include + +/* + * efi_call_phys(void *, ...) is a function with variable parameters. + * All the callers of this function assure that all the parameters are 4-bytes. + */ + +/* + * In gcc calling convention, EBX, ESP, EBP, ESI and EDI are all callee save. + * So we'd better save all of them at the beginning of this function and restore + * at the end no matter how many we use, because we can not assure EFI runtime + * service functions will comply with gcc calling convention, too. + */ + +.text +.section .text, "a" +ENTRY(efi_call_phys) + /* + * 0. The function can only be called in Linux kernel. So CS has been + * set to 0x0010, DS and SS have been set to 0x0018. In EFI, I found + * the values of these registers are the same. And, the corresponding + * GDT entries are identical. So I will do nothing about segment reg + * and GDT, but change GDT base register in prelog and epilog. + */ + + /* + * 1. Now I am running with EIP = + PAGE_OFFSET. + * But to make it smoothly switch from virtual mode to flat mode. + * The mapping of lower virtual memory has been created in prelog and + * epilog. + */ + movl $1f, %edx + subl $__PAGE_OFFSET, %edx + jmp *%edx +1: + + /* + * 2. Now on the top of stack is the return + * address in the caller of efi_call_phys(), then parameter 1, + * parameter 2, ..., param n. To make things easy, we save the return + * address of efi_call_phys in a global variable. + */ + popl %edx + movl %edx, saved_return_addr + /* get the function pointer into ECX*/ + popl %ecx + movl %ecx, efi_rt_function_ptr + movl $2f, %edx + subl $__PAGE_OFFSET, %edx + pushl %edx + + /* + * 3. Clear PG bit in %CR0. + */ + movl %cr0, %edx + andl $0x7fffffff, %edx + movl %edx, %cr0 + jmp 1f +1: + + /* + * 4. Adjust stack pointer. + */ + subl $__PAGE_OFFSET, %esp + + /* + * 5. Call the physical function. + */ + jmp *%ecx + +2: + /* + * 6. After EFI runtime service returns, control will return to + * following instruction. We'd better readjust stack pointer first. + */ + addl $__PAGE_OFFSET, %esp + + /* + * 7. Restore PG bit + */ + movl %cr0, %edx + orl $0x80000000, %edx + movl %edx, %cr0 + jmp 1f +1: + /* + * 8. Now restore the virtual mode from flat mode by + * adding EIP with PAGE_OFFSET. + */ + movl $1f, %edx + jmp *%edx +1: + + /* + * 9. Balance the stack. And because EAX contain the return value, + * we'd better not clobber it. + */ + leal efi_rt_function_ptr, %edx + movl (%edx), %ecx + pushl %ecx + + /* + * 10. Push the saved return address onto the stack and return. + */ + leal saved_return_addr, %edx + movl (%edx), %ecx + pushl %ecx + ret +.previous + +.data +saved_return_addr: + .long 0 +efi_rt_function_ptr: + .long 0 diff -puN arch/i386/kernel/Makefile~ia32-efi-support arch/i386/kernel/Makefile --- 25/arch/i386/kernel/Makefile~ia32-efi-support 2003-10-15 10:15:56.000000000 -0700 +++ 25-akpm/arch/i386/kernel/Makefile 2003-10-15 10:15:56.000000000 -0700 @@ -7,7 +7,7 @@ extra-y := head.o init_task.o vmlinux.ld obj-y := process.o semaphore.o signal.o entry.o traps.o irq.o vm86.o \ ptrace.o i8259.o ioport.o ldt.o setup.o time.o sys_i386.o \ pci-dma.o i386_ksyms.o i387.o dmi_scan.o bootflag.o \ - doublefault.o + doublefault.o efi.o efi_stub.o obj-y += cpu/ obj-y += timers/ diff -puN arch/i386/kernel/reboot.c~ia32-efi-support arch/i386/kernel/reboot.c --- 25/arch/i386/kernel/reboot.c~ia32-efi-support 2003-10-15 10:15:56.000000000 -0700 +++ 25-akpm/arch/i386/kernel/reboot.c 2003-10-15 10:15:56.000000000 -0700 @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include "mach_reboot.h" @@ -263,7 +264,12 @@ void machine_restart(char * __unused) disable_IO_APIC(); #endif - if(!reboot_thru_bios) { + if (!reboot_thru_bios) { + if (efi_enabled) { + efi.reset_system(EFI_RESET_COLD, EFI_SUCCESS, 0, 0); + __asm__ __volatile__("lidt %0": :"m" (no_idt)); + __asm__ __volatile__("int3"); + } /* rebooting needs to touch the page at absolute addr 0 */ *((unsigned short *)__va(0x472)) = reboot_mode; for (;;) { @@ -273,6 +279,8 @@ void machine_restart(char * __unused) __asm__ __volatile__("int3"); } } + if (efi_enabled) + efi.reset_system(EFI_RESET_WARM, EFI_SUCCESS, 0, 0); machine_real_restart(jump_to_bios, sizeof(jump_to_bios)); } @@ -287,6 +295,8 @@ EXPORT_SYMBOL(machine_halt); void machine_power_off(void) { + if (efi_enabled) + efi.reset_system(EFI_RESET_SHUTDOWN, EFI_SUCCESS, 0, 0); if (pm_power_off) pm_power_off(); } diff -puN arch/i386/kernel/setup.c~ia32-efi-support arch/i386/kernel/setup.c --- 25/arch/i386/kernel/setup.c~ia32-efi-support 2003-10-15 10:15:56.000000000 -0700 +++ 25-akpm/arch/i386/kernel/setup.c 2003-10-15 10:15:56.000000000 -0700 @@ -36,6 +36,8 @@ #include #include #include +#include +#include #include