From 3e909599215456928e6b42a04f11c2517881570b Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Wed, 15 Jan 2014 13:21:22 +0000 Subject: efi: Move facility flags to struct efi As we grow support for more EFI architectures they're going to want the ability to query which EFI features are available on the running system. Instead of storing this information in an architecture-specific place, stick it in the global 'struct efi', which is already the central location for EFI state. While we're at it, let's change the return value of efi_enabled() to be bool and replace all references to 'facility' with 'feature', which is the usual word used to describe the attributes of the running system. Signed-off-by: Matt Fleming --- arch/x86/include/asm/efi.h | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index 3d6b9f81cc68..a7a8bd89024e 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -119,7 +119,6 @@ extern void __iomem *efi_ioremap(unsigned long addr, unsigned long size, #endif /* CONFIG_X86_32 */ extern int add_efi_memmap; -extern unsigned long x86_efi_facility; extern struct efi_scratch efi_scratch; extern void efi_set_executable(efi_memory_desc_t *md, bool executable); extern int efi_memblock_x86_reserve_range(void); -- cgit v1.2.3 From ef6bea6ddf0e76077d0798e57b374015b23a837e Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Sat, 18 Jan 2014 12:48:14 +0100 Subject: x86, ptdump: Add the functionality to dump an arbitrary pagetable With reusing the ->trampoline_pgd page table for mapping EFI regions in order to use them after having switched to EFI virtual mode, it is very useful to be able to dump aforementioned page table in dmesg. This adds that functionality through the walk_pgd_level() interface which can be called from somewhere else. The original functionality of dumping to debugfs remains untouched. Cc: Arjan van de Ven Signed-off-by: Borislav Petkov Tested-by: Toshi Kani Signed-off-by: Matt Fleming --- arch/x86/include/asm/pgtable.h | 3 +- arch/x86/mm/dump_pagetables.c | 84 +++++++++++++++++++++++++++--------------- 2 files changed, 56 insertions(+), 31 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 5ad38ad07890..938ef1d0458e 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -15,9 +15,10 @@ : (prot)) #ifndef __ASSEMBLY__ - #include +void ptdump_walk_pgd_level(struct seq_file *m, pgd_t *pgd); + /* * ZERO_PAGE is a global shared page that is always zero: used * for zero-mapped memory areas etc.. diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c index 0002a3a33081..20621d753d5f 100644 --- a/arch/x86/mm/dump_pagetables.c +++ b/arch/x86/mm/dump_pagetables.c @@ -30,6 +30,7 @@ struct pg_state { unsigned long start_address; unsigned long current_address; const struct addr_marker *marker; + bool to_dmesg; }; struct addr_marker { @@ -88,10 +89,28 @@ static struct addr_marker address_markers[] = { #define PUD_LEVEL_MULT (PTRS_PER_PMD * PMD_LEVEL_MULT) #define PGD_LEVEL_MULT (PTRS_PER_PUD * PUD_LEVEL_MULT) +#define pt_dump_seq_printf(m, to_dmesg, fmt, args...) \ +({ \ + if (to_dmesg) \ + printk(KERN_INFO fmt, ##args); \ + else \ + if (m) \ + seq_printf(m, fmt, ##args); \ +}) + +#define pt_dump_cont_printf(m, to_dmesg, fmt, args...) \ +({ \ + if (to_dmesg) \ + printk(KERN_CONT fmt, ##args); \ + else \ + if (m) \ + seq_printf(m, fmt, ##args); \ +}) + /* * Print a readable form of a pgprot_t to the seq_file */ -static void printk_prot(struct seq_file *m, pgprot_t prot, int level) +static void printk_prot(struct seq_file *m, pgprot_t prot, int level, bool dmsg) { pgprotval_t pr = pgprot_val(prot); static const char * const level_name[] = @@ -99,47 +118,47 @@ static void printk_prot(struct seq_file *m, pgprot_t prot, int level) if (!pgprot_val(prot)) { /* Not present */ - seq_printf(m, " "); + pt_dump_cont_printf(m, dmsg, " "); } else { if (pr & _PAGE_USER) - seq_printf(m, "USR "); + pt_dump_cont_printf(m, dmsg, "USR "); else - seq_printf(m, " "); + pt_dump_cont_printf(m, dmsg, " "); if (pr & _PAGE_RW) - seq_printf(m, "RW "); + pt_dump_cont_printf(m, dmsg, "RW "); else - seq_printf(m, "ro "); + pt_dump_cont_printf(m, dmsg, "ro "); if (pr & _PAGE_PWT) - seq_printf(m, "PWT "); + pt_dump_cont_printf(m, dmsg, "PWT "); else - seq_printf(m, " "); + pt_dump_cont_printf(m, dmsg, " "); if (pr & _PAGE_PCD) - seq_printf(m, "PCD "); + pt_dump_cont_printf(m, dmsg, "PCD "); else - seq_printf(m, " "); + pt_dump_cont_printf(m, dmsg, " "); /* Bit 9 has a different meaning on level 3 vs 4 */ if (level <= 3) { if (pr & _PAGE_PSE) - seq_printf(m, "PSE "); + pt_dump_cont_printf(m, dmsg, "PSE "); else - seq_printf(m, " "); + pt_dump_cont_printf(m, dmsg, " "); } else { if (pr & _PAGE_PAT) - seq_printf(m, "pat "); + pt_dump_cont_printf(m, dmsg, "pat "); else - seq_printf(m, " "); + pt_dump_cont_printf(m, dmsg, " "); } if (pr & _PAGE_GLOBAL) - seq_printf(m, "GLB "); + pt_dump_cont_printf(m, dmsg, "GLB "); else - seq_printf(m, " "); + pt_dump_cont_printf(m, dmsg, " "); if (pr & _PAGE_NX) - seq_printf(m, "NX "); + pt_dump_cont_printf(m, dmsg, "NX "); else - seq_printf(m, "x "); + pt_dump_cont_printf(m, dmsg, "x "); } - seq_printf(m, "%s\n", level_name[level]); + pt_dump_cont_printf(m, dmsg, "%s\n", level_name[level]); } /* @@ -178,7 +197,8 @@ static void note_page(struct seq_file *m, struct pg_state *st, st->current_prot = new_prot; st->level = level; st->marker = address_markers; - seq_printf(m, "---[ %s ]---\n", st->marker->name); + pt_dump_seq_printf(m, st->to_dmesg, "---[ %s ]---\n", + st->marker->name); } else if (prot != cur || level != st->level || st->current_address >= st->marker[1].start_address) { const char *unit = units; @@ -188,17 +208,17 @@ static void note_page(struct seq_file *m, struct pg_state *st, /* * Now print the actual finished series */ - seq_printf(m, "0x%0*lx-0x%0*lx ", - width, st->start_address, - width, st->current_address); + pt_dump_seq_printf(m, st->to_dmesg, "0x%0*lx-0x%0*lx ", + width, st->start_address, + width, st->current_address); delta = (st->current_address - st->start_address) >> 10; while (!(delta & 1023) && unit[1]) { delta >>= 10; unit++; } - seq_printf(m, "%9lu%c ", delta, *unit); - printk_prot(m, st->current_prot, st->level); + pt_dump_cont_printf(m, st->to_dmesg, "%9lu%c ", delta, *unit); + printk_prot(m, st->current_prot, st->level, st->to_dmesg); /* * We print markers for special areas of address space, @@ -207,7 +227,8 @@ static void note_page(struct seq_file *m, struct pg_state *st, */ if (st->current_address >= st->marker[1].start_address) { st->marker++; - seq_printf(m, "---[ %s ]---\n", st->marker->name); + pt_dump_seq_printf(m, st->to_dmesg, "---[ %s ]---\n", + st->marker->name); } st->start_address = st->current_address; @@ -296,7 +317,7 @@ static void walk_pud_level(struct seq_file *m, struct pg_state *st, pgd_t addr, #define pgd_none(a) pud_none(__pud(pgd_val(a))) #endif -static void walk_pgd_level(struct seq_file *m) +void ptdump_walk_pgd_level(struct seq_file *m, pgd_t *pgd) { #ifdef CONFIG_X86_64 pgd_t *start = (pgd_t *) &init_level4_pgt; @@ -304,9 +325,12 @@ static void walk_pgd_level(struct seq_file *m) pgd_t *start = swapper_pg_dir; #endif int i; - struct pg_state st; + struct pg_state st = {}; - memset(&st, 0, sizeof(st)); + if (pgd) { + start = pgd; + st.to_dmesg = true; + } for (i = 0; i < PTRS_PER_PGD; i++) { st.current_address = normalize_addr(i * PGD_LEVEL_MULT); @@ -331,7 +355,7 @@ static void walk_pgd_level(struct seq_file *m) static int ptdump_show(struct seq_file *m, void *v) { - walk_pgd_level(m); + ptdump_walk_pgd_level(m, NULL); return 0; } -- cgit v1.2.3 From 11cc851254b4bc3bd4430be8db2a41469303a427 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Sat, 18 Jan 2014 12:48:15 +0100 Subject: x86/efi: Dump the EFI page table This is very useful for debugging issues with the recently added pagetable switching code for EFI virtual mode. Signed-off-by: Borislav Petkov Tested-by: Toshi Kani Signed-off-by: Matt Fleming --- arch/x86/Kconfig.debug | 9 +++++++++ arch/x86/include/asm/efi.h | 1 + arch/x86/platform/efi/efi.c | 1 + arch/x86/platform/efi/efi_32.c | 1 + arch/x86/platform/efi/efi_64.c | 9 +++++++++ 5 files changed, 21 insertions(+) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 321a52ccf63a..61bd2ad94281 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -81,6 +81,15 @@ config X86_PTDUMP kernel. If in doubt, say "N" +config EFI_PGT_DUMP + bool "Dump the EFI pagetable" + depends on EFI && X86_PTDUMP + ---help--- + Enable this if you want to dump the EFI page table before + enabling virtual mode. This can be used to debug miscellaneous + issues with the mapping of the EFI runtime regions into that + table. + config DEBUG_RODATA bool "Write protect kernel read-only data structures" default y diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index a7a8bd89024e..4afd3b3b658f 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -133,6 +133,7 @@ extern void efi_setup_page_tables(void); extern void __init old_map_region(efi_memory_desc_t *md); extern void __init runtime_code_page_mkexec(void); extern void __init efi_runtime_mkexec(void); +extern void __init efi_dump_pagetable(void); struct efi_setup_data { u64 fw_vendor; diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index 876dc5b9e4a3..37f20d71ec4b 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -1021,6 +1021,7 @@ void __init efi_enter_virtual_mode(void) efi_setup_page_tables(); efi_sync_low_kernel_mappings(); + efi_dump_pagetable(); if (!efi_setup) { status = phys_efi_set_virtual_address_map( diff --git a/arch/x86/platform/efi/efi_32.c b/arch/x86/platform/efi/efi_32.c index 0b74cdf7f816..39496ae3928a 100644 --- a/arch/x86/platform/efi/efi_32.c +++ b/arch/x86/platform/efi/efi_32.c @@ -41,6 +41,7 @@ static unsigned long efi_rt_eflags; void efi_sync_low_kernel_mappings(void) {} void efi_setup_page_tables(void) {} +void __init efi_dump_pagetable(void) {} void __init efi_map_region(efi_memory_desc_t *md) { diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index 0c2a234fef1e..e05c69b46f05 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -242,3 +242,12 @@ void __init efi_runtime_mkexec(void) if (__supported_pte_mask & _PAGE_NX) runtime_code_page_mkexec(); } + +void __init efi_dump_pagetable(void) +{ +#ifdef CONFIG_EFI_PGT_DUMP + pgd_t *pgd = (pgd_t *)__va(real_mode_header->trampoline_pgd); + + ptdump_walk_pgd_level(NULL, pgd); +#endif +} -- cgit v1.2.3 From 42a5477251f0e0f33ad5f6a95c48d685ec03191e Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Sat, 18 Jan 2014 12:48:16 +0100 Subject: x86, pageattr: Export page unmapping interface We will use it in efi so expose it. Signed-off-by: Borislav Petkov Tested-by: Toshi Kani Signed-off-by: Matt Fleming --- arch/x86/include/asm/pgtable_types.h | 2 ++ arch/x86/mm/pageattr.c | 44 +++++++++++++++++++++++++----------- 2 files changed, 33 insertions(+), 13 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h index 1aa9ccd43223..94e40f1efdfd 100644 --- a/arch/x86/include/asm/pgtable_types.h +++ b/arch/x86/include/asm/pgtable_types.h @@ -385,6 +385,8 @@ extern pte_t *lookup_address(unsigned long address, unsigned int *level); extern phys_addr_t slow_virt_to_phys(void *__address); extern int kernel_map_pages_in_pgd(pgd_t *pgd, u64 pfn, unsigned long address, unsigned numpages, unsigned long page_flags); +void kernel_unmap_pages_in_pgd(pgd_t *root, unsigned long address, + unsigned numpages); #endif /* !__ASSEMBLY__ */ #endif /* _ASM_X86_PGTABLE_DEFS_H */ diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index b3b19f46c016..a3488689e301 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -692,6 +692,18 @@ static bool try_to_free_pmd_page(pmd_t *pmd) return true; } +static bool try_to_free_pud_page(pud_t *pud) +{ + int i; + + for (i = 0; i < PTRS_PER_PUD; i++) + if (!pud_none(pud[i])) + return false; + + free_page((unsigned long)pud); + return true; +} + static bool unmap_pte_range(pmd_t *pmd, unsigned long start, unsigned long end) { pte_t *pte = pte_offset_kernel(pmd, start); @@ -805,6 +817,16 @@ static void unmap_pud_range(pgd_t *pgd, unsigned long start, unsigned long end) */ } +static void unmap_pgd_range(pgd_t *root, unsigned long addr, unsigned long end) +{ + pgd_t *pgd_entry = root + pgd_index(addr); + + unmap_pud_range(pgd_entry, addr, end); + + if (try_to_free_pud_page((pud_t *)pgd_page_vaddr(*pgd_entry))) + pgd_clear(pgd_entry); +} + static int alloc_pte_page(pmd_t *pmd) { pte_t *pte = (pte_t *)get_zeroed_page(GFP_KERNEL | __GFP_NOTRACK); @@ -999,9 +1021,8 @@ static int populate_pud(struct cpa_data *cpa, unsigned long start, pgd_t *pgd, static int populate_pgd(struct cpa_data *cpa, unsigned long addr) { pgprot_t pgprot = __pgprot(_KERNPG_TABLE); - bool allocd_pgd = false; - pgd_t *pgd_entry; pud_t *pud = NULL; /* shut up gcc */ + pgd_t *pgd_entry; int ret; pgd_entry = cpa->pgd + pgd_index(addr); @@ -1015,7 +1036,6 @@ static int populate_pgd(struct cpa_data *cpa, unsigned long addr) return -1; set_pgd(pgd_entry, __pgd(__pa(pud) | _KERNPG_TABLE)); - allocd_pgd = true; } pgprot_val(pgprot) &= ~pgprot_val(cpa->mask_clr); @@ -1023,19 +1043,11 @@ static int populate_pgd(struct cpa_data *cpa, unsigned long addr) ret = populate_pud(cpa, addr, pgd_entry, pgprot); if (ret < 0) { - unmap_pud_range(pgd_entry, addr, + unmap_pgd_range(cpa->pgd, addr, addr + (cpa->numpages << PAGE_SHIFT)); - - if (allocd_pgd) { - /* - * If I allocated this PUD page, I can just as well - * free it in this error path. - */ - pgd_clear(pgd_entry); - free_page((unsigned long)pud); - } return ret; } + cpa->numpages = ret; return 0; } @@ -1861,6 +1873,12 @@ out: return retval; } +void kernel_unmap_pages_in_pgd(pgd_t *root, unsigned long address, + unsigned numpages) +{ + unmap_pgd_range(root, address, address + (numpages << PAGE_SHIFT)); +} + /* * The testcases use internal knowledge of the implementation that shouldn't * be exposed to the rest of the kernel. Include these directly here. -- cgit v1.2.3 From b7b898ae0c0a82489511a1ce1b35f26215e6beb5 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Sat, 18 Jan 2014 12:48:17 +0100 Subject: x86/efi: Make efi virtual runtime map passing more robust Currently, running SetVirtualAddressMap() and passing the physical address of the virtual map array was working only by a lucky coincidence because the memory was present in the EFI page table too. Until Toshi went and booted this on a big HP box - the krealloc() manner of resizing the memmap we're doing did allocate from such physical addresses which were not mapped anymore and boom: http://lkml.kernel.org/r/1386806463.1791.295.camel@misato.fc.hp.com One way to take care of that issue is to reimplement the krealloc thing but with pages. We start with contiguous pages of order 1, i.e. 2 pages, and when we deplete that memory (shouldn't happen all that often but you know firmware) we realloc the next power-of-two pages. Having the pages, it is much more handy and easy to map them into the EFI page table with the already existing mapping code which we're using for building the virtual mappings. Thanks to Toshi Kani and Matt for the great debugging help. Reported-by: Toshi Kani Signed-off-by: Borislav Petkov Tested-by: Toshi Kani Signed-off-by: Matt Fleming --- arch/x86/include/asm/efi.h | 3 +- arch/x86/platform/efi/efi.c | 99 +++++++++++++++++++++++++++++++++--------- arch/x86/platform/efi/efi_32.c | 6 ++- arch/x86/platform/efi/efi_64.c | 32 ++++++++++++-- 4 files changed, 114 insertions(+), 26 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index 4afd3b3b658f..e985d6bf7d3a 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -129,7 +129,8 @@ extern void efi_memory_uc(u64 addr, unsigned long size); extern void __init efi_map_region(efi_memory_desc_t *md); extern void __init efi_map_region_fixed(efi_memory_desc_t *md); extern void efi_sync_low_kernel_mappings(void); -extern void efi_setup_page_tables(void); +extern int efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages); +extern void efi_cleanup_page_tables(unsigned long pa_memmap, unsigned num_pages); extern void __init old_map_region(efi_memory_desc_t *md); extern void __init runtime_code_page_mkexec(void); extern void __init efi_runtime_mkexec(void); diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index 37f20d71ec4b..576bb126593a 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -926,14 +926,36 @@ static void __init efi_map_regions_fixed(void) } +static void *realloc_pages(void *old_memmap, int old_shift) +{ + void *ret; + + ret = (void *)__get_free_pages(GFP_KERNEL, old_shift + 1); + if (!ret) + goto out; + + /* + * A first-time allocation doesn't have anything to copy. + */ + if (!old_memmap) + return ret; + + memcpy(ret, old_memmap, PAGE_SIZE << old_shift); + +out: + free_pages((unsigned long)old_memmap, old_shift); + return ret; +} + /* - * Map efi memory ranges for runtime serivce and update new_memmap with virtual - * addresses. + * Map the efi memory ranges of the runtime services and update new_mmap with + * virtual addresses. */ -static void * __init efi_map_regions(int *count) +static void * __init efi_map_regions(int *count, int *pg_shift) { + void *p, *new_memmap = NULL; + unsigned long left = 0; efi_memory_desc_t *md; - void *p, *tmp, *new_memmap = NULL; for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { md = p; @@ -948,20 +970,23 @@ static void * __init efi_map_regions(int *count) efi_map_region(md); get_systab_virt_addr(md); - tmp = krealloc(new_memmap, (*count + 1) * memmap.desc_size, - GFP_KERNEL); - if (!tmp) - goto out; - new_memmap = tmp; + if (left < memmap.desc_size) { + new_memmap = realloc_pages(new_memmap, *pg_shift); + if (!new_memmap) + return NULL; + + left += PAGE_SIZE << *pg_shift; + (*pg_shift)++; + } + memcpy(new_memmap + (*count * memmap.desc_size), md, memmap.desc_size); + + left -= memmap.desc_size; (*count)++; } return new_memmap; -out: - kfree(new_memmap); - return NULL; } /* @@ -987,9 +1012,9 @@ out: */ void __init efi_enter_virtual_mode(void) { - efi_status_t status; + int err, count = 0, pg_shift = 0; void *new_memmap = NULL; - int err, count = 0; + efi_status_t status; efi.systab = NULL; @@ -1006,20 +1031,24 @@ void __init efi_enter_virtual_mode(void) efi_map_regions_fixed(); } else { efi_merge_regions(); - new_memmap = efi_map_regions(&count); + new_memmap = efi_map_regions(&count, &pg_shift); if (!new_memmap) { pr_err("Error reallocating memory, EFI runtime non-functional!\n"); return; } - } - err = save_runtime_map(); - if (err) - pr_err("Error saving runtime map, efi runtime on kexec non-functional!!\n"); + err = save_runtime_map(); + if (err) + pr_err("Error saving runtime map, efi runtime on kexec non-functional!!\n"); + } BUG_ON(!efi.systab); - efi_setup_page_tables(); + if (!efi_setup) { + if (efi_setup_page_tables(__pa(new_memmap), 1 << pg_shift)) + return; + } + efi_sync_low_kernel_mappings(); efi_dump_pagetable(); @@ -1060,7 +1089,35 @@ void __init efi_enter_virtual_mode(void) efi_runtime_mkexec(); - kfree(new_memmap); + + /* + * We mapped the descriptor array into the EFI pagetable above but we're + * not unmapping it here. Here's why: + * + * We're copying select PGDs from the kernel page table to the EFI page + * table and when we do so and make changes to those PGDs like unmapping + * stuff from them, those changes appear in the kernel page table and we + * go boom. + * + * From setup_real_mode(): + * + * ... + * trampoline_pgd[0] = init_level4_pgt[pgd_index(__PAGE_OFFSET)].pgd; + * + * In this particular case, our allocation is in PGD 0 of the EFI page + * table but we've copied that PGD from PGD[272] of the EFI page table: + * + * pgd_index(__PAGE_OFFSET = 0xffff880000000000) = 272 + * + * where the direct memory mapping in kernel space is. + * + * new_memmap's VA comes from that direct mapping and thus clearing it, + * it would get cleared in the kernel page table too. + * + * efi_cleanup_page_tables(__pa(new_memmap), 1 << pg_shift); + */ + if (!efi_setup) + free_pages((unsigned long)new_memmap, pg_shift); /* clean DUMMY object */ efi.set_variable(efi_dummy_name, &EFI_DUMMY_GUID, diff --git a/arch/x86/platform/efi/efi_32.c b/arch/x86/platform/efi/efi_32.c index 39496ae3928a..9ee3491e31fb 100644 --- a/arch/x86/platform/efi/efi_32.c +++ b/arch/x86/platform/efi/efi_32.c @@ -40,8 +40,12 @@ static unsigned long efi_rt_eflags; void efi_sync_low_kernel_mappings(void) {} -void efi_setup_page_tables(void) {} void __init efi_dump_pagetable(void) {} +int efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages) +{ + return 0; +} +void efi_cleanup_page_tables(unsigned long pa_memmap, unsigned num_pages) {} void __init efi_map_region(efi_memory_desc_t *md) { diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index e05c69b46f05..19280900ec25 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -137,12 +137,38 @@ void efi_sync_low_kernel_mappings(void) sizeof(pgd_t) * num_pgds); } -void efi_setup_page_tables(void) +int efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages) { + pgd_t *pgd; + + if (efi_enabled(EFI_OLD_MEMMAP)) + return 0; + efi_scratch.efi_pgt = (pgd_t *)(unsigned long)real_mode_header->trampoline_pgd; + pgd = __va(efi_scratch.efi_pgt); - if (!efi_enabled(EFI_OLD_MEMMAP)) - efi_scratch.use_pgd = true; + /* + * It can happen that the physical address of new_memmap lands in memory + * which is not mapped in the EFI page table. Therefore we need to go + * and ident-map those pages containing the map before calling + * phys_efi_set_virtual_address_map(). + */ + if (kernel_map_pages_in_pgd(pgd, pa_memmap, pa_memmap, num_pages, _PAGE_NX)) { + pr_err("Error ident-mapping new memmap (0x%lx)!\n", pa_memmap); + return 1; + } + + efi_scratch.use_pgd = true; + + + return 0; +} + +void efi_cleanup_page_tables(unsigned long pa_memmap, unsigned num_pages) +{ + pgd_t *pgd = (pgd_t *)__va(real_mode_header->trampoline_pgd); + + kernel_unmap_pages_in_pgd(pgd, pa_memmap, num_pages); } static void __init __map_region(efi_memory_desc_t *md, u64 va) -- cgit v1.2.3 From 426e34cc4f6094cefe4f3175764cdf583128e7cd Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Fri, 6 Dec 2013 21:13:04 +0000 Subject: x86/mm/pageattr: Always dump the right page table in an oops Now that we have EFI-specific page tables we need to lookup the pgd when dumping those page tables, rather than assuming that swapper_pgdir is the current pgdir. Remove the double underscore prefix, which is usually reserved for static functions. Acked-by: Borislav Petkov Signed-off-by: Matt Fleming --- arch/x86/include/asm/pgtable_types.h | 2 ++ arch/x86/mm/fault.c | 7 ++++++- arch/x86/mm/pageattr.c | 12 ++++++++---- 3 files changed, 16 insertions(+), 5 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h index 1aa9ccd43223..24ae4211ba34 100644 --- a/arch/x86/include/asm/pgtable_types.h +++ b/arch/x86/include/asm/pgtable_types.h @@ -382,6 +382,8 @@ static inline void update_page_count(int level, unsigned long pages) { } * as a pte too. */ extern pte_t *lookup_address(unsigned long address, unsigned int *level); +extern pte_t *lookup_address_in_pgd(pgd_t *pgd, unsigned long address, + unsigned int *level); extern phys_addr_t slow_virt_to_phys(void *__address); extern int kernel_map_pages_in_pgd(pgd_t *pgd, u64 pfn, unsigned long address, unsigned numpages, unsigned long page_flags); diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 6dea040cc3a1..6055a2001b27 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -584,8 +584,13 @@ show_fault_oops(struct pt_regs *regs, unsigned long error_code, if (error_code & PF_INSTR) { unsigned int level; + pgd_t *pgd; + pte_t *pte; - pte_t *pte = lookup_address(address, &level); + pgd = __va(read_cr3() & PHYSICAL_PAGE_MASK); + pgd += pgd_index(address); + + pte = lookup_address_in_pgd(pgd, address, &level); if (pte && pte_present(*pte) && !pte_exec(*pte)) printk(nx_warning, from_kuid(&init_user_ns, current_uid())); diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index b3b19f46c016..14c656f0483b 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -323,8 +323,12 @@ static inline pgprot_t static_protections(pgprot_t prot, unsigned long address, return prot; } -static pte_t *__lookup_address_in_pgd(pgd_t *pgd, unsigned long address, - unsigned int *level) +/* + * Lookup the page table entry for a virtual address in a specific pgd. + * Return a pointer to the entry and the level of the mapping. + */ +pte_t *lookup_address_in_pgd(pgd_t *pgd, unsigned long address, + unsigned int *level) { pud_t *pud; pmd_t *pmd; @@ -365,7 +369,7 @@ static pte_t *__lookup_address_in_pgd(pgd_t *pgd, unsigned long address, */ pte_t *lookup_address(unsigned long address, unsigned int *level) { - return __lookup_address_in_pgd(pgd_offset_k(address), address, level); + return lookup_address_in_pgd(pgd_offset_k(address), address, level); } EXPORT_SYMBOL_GPL(lookup_address); @@ -373,7 +377,7 @@ static pte_t *_lookup_address_cpa(struct cpa_data *cpa, unsigned long address, unsigned int *level) { if (cpa->pgd) - return __lookup_address_in_pgd(cpa->pgd + pgd_index(address), + return lookup_address_in_pgd(cpa->pgd + pgd_index(address), address, level); return lookup_address(address, level); -- cgit v1.2.3 From b8ff87a6158886771677e6dc8139bac6e3cba717 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Fri, 10 Jan 2014 15:54:31 +0000 Subject: x86/efi: Firmware agnostic handover entry points The EFI handover code only works if the "bitness" of the firmware and the kernel match, i.e. 64-bit firmware and 64-bit kernel - it is not possible to mix the two. This goes against the tradition that a 32-bit kernel can be loaded on a 64-bit BIOS platform without having to do anything special in the boot loader. Linux distributions, for one thing, regularly run only 32-bit kernels on their live media. Despite having only one 'handover_offset' field in the kernel header, EFI boot loaders use two separate entry points to enter the kernel based on the architecture the boot loader was compiled for, (1) 32-bit loader: handover_offset (2) 64-bit loader: handover_offset + 512 Since we already have two entry points, we can leverage them to infer the bitness of the firmware we're running on, without requiring any boot loader modifications, by making (1) and (2) valid entry points for both CONFIG_X86_32 and CONFIG_X86_64 kernels. To be clear, a 32-bit boot loader will always use (1) and a 64-bit boot loader will always use (2). It's just that, if a single kernel image supports (1) and (2) that image can be used with both 32-bit and 64-bit boot loaders, and hence both 32-bit and 64-bit EFI. (1) and (2) must be 512 bytes apart at all times, but that is already part of the boot ABI and we could never change that delta without breaking existing boot loaders anyhow. Signed-off-by: Matt Fleming --- arch/x86/boot/Makefile | 2 +- arch/x86/boot/compressed/eboot.c | 9 ++++-- arch/x86/boot/compressed/head_32.S | 2 +- arch/x86/boot/compressed/head_64.S | 62 +++++++++++++++++++++++++++++++++----- arch/x86/boot/tools/build.c | 22 +++++++++----- arch/x86/include/asm/efi.h | 6 ++-- 6 files changed, 80 insertions(+), 23 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile index 878df7e88cd4..abb9eba61b50 100644 --- a/arch/x86/boot/Makefile +++ b/arch/x86/boot/Makefile @@ -80,7 +80,7 @@ targets += voffset.h $(obj)/voffset.h: vmlinux FORCE $(call if_changed,voffset) -sed-zoffset := -e 's/^\([0-9a-fA-F]*\) . \(startup_32\|startup_64\|efi_pe_entry\|efi_stub_entry\|input_data\|_end\|z_.*\)$$/\#define ZO_\2 0x\1/p' +sed-zoffset := -e 's/^\([0-9a-fA-F]*\) . \(startup_32\|startup_64\|efi32_stub_entry\|efi64_stub_entry\|efi_pe_entry\|input_data\|_end\|z_.*\)$$/\#define ZO_\2 0x\1/p' quiet_cmd_zoffset = ZOFFSET $@ cmd_zoffset = $(NM) $< | sed -n $(sed-zoffset) > $@ diff --git a/arch/x86/boot/compressed/eboot.c b/arch/x86/boot/compressed/eboot.c index ab1f3a2f1e1e..5e1ba4fa3f79 100644 --- a/arch/x86/boot/compressed/eboot.c +++ b/arch/x86/boot/compressed/eboot.c @@ -1256,12 +1256,13 @@ static efi_status_t alloc_e820ext(u32 nr_desc, struct setup_data **e820ext, } static efi_status_t exit_boot(struct boot_params *boot_params, - void *handle) + void *handle, bool is64) { struct efi_info *efi = &boot_params->efi_info; unsigned long map_sz, key, desc_size; efi_memory_desc_t *mem_map; struct setup_data *e820ext; + const char *signature; __u32 e820ext_size; __u32 nr_desc, prev_nr_desc; efi_status_t status; @@ -1295,7 +1296,9 @@ get_map: goto get_map; /* Allocated memory, get map again */ } - memcpy(&efi->efi_loader_signature, EFI_LOADER_SIGNATURE, sizeof(__u32)); + signature = is64 ? EFI64_LOADER_SIGNATURE : EFI32_LOADER_SIGNATURE; + memcpy(&efi->efi_loader_signature, signature, sizeof(__u32)); + efi->efi_systab = (unsigned long)sys_table; efi->efi_memdesc_size = desc_size; efi->efi_memdesc_version = desc_version; @@ -1408,7 +1411,7 @@ struct boot_params *efi_main(struct efi_config *c, hdr->code32_start = bzimage_addr; } - status = exit_boot(boot_params, handle); + status = exit_boot(boot_params, handle, is64); if (status != EFI_SUCCESS) goto fail; diff --git a/arch/x86/boot/compressed/head_32.S b/arch/x86/boot/compressed/head_32.S index eed23c087d6c..cccc05f0681c 100644 --- a/arch/x86/boot/compressed/head_32.S +++ b/arch/x86/boot/compressed/head_32.S @@ -64,7 +64,7 @@ ENTRY(efi_pe_entry) pushl %ecx jmp 2f /* Skip efi_config initialization */ -ENTRY(efi_stub_entry) +ENTRY(efi32_stub_entry) add $0x4, %esp popl %ecx popl %edx diff --git a/arch/x86/boot/compressed/head_64.S b/arch/x86/boot/compressed/head_64.S index 1bc206fa4bd0..37c741b0d2ac 100644 --- a/arch/x86/boot/compressed/head_64.S +++ b/arch/x86/boot/compressed/head_64.S @@ -178,6 +178,13 @@ ENTRY(startup_32) */ pushl $__KERNEL_CS leal startup_64(%ebp), %eax +#ifdef CONFIG_EFI_MIXED + movl efi32_config(%ebp), %ebx + cmp $0, %ebx + jz 1f + leal handover_entry(%ebp), %eax +1: +#endif pushl %eax /* Enter paged protected Mode, activating Long Mode */ @@ -188,6 +195,30 @@ ENTRY(startup_32) lret ENDPROC(startup_32) +#ifdef CONFIG_EFI_MIXED + .org 0x190 +ENTRY(efi32_stub_entry) + add $0x4, %esp /* Discard return address */ + popl %ecx + popl %edx + popl %esi + + leal (BP_scratch+4)(%esi), %esp + call 1f +1: pop %ebp + subl $1b, %ebp + + movl %ecx, efi32_config(%ebp) + movl %edx, efi32_config+8(%ebp) + sgdtl efi32_boot_gdt(%ebp) + + leal efi32_config(%ebp), %eax + movl %eax, efi_config(%ebp) + + jmp startup_32 +ENDPROC(efi32_stub_entry) +#endif + .code64 .org 0x200 ENTRY(startup_64) @@ -231,13 +262,7 @@ ENTRY(efi_pe_entry) mov %rax, %rsi jmp 2f /* Skip the relocation */ -ENTRY(efi_stub_entry) - movq %rdi, efi64_config(%rip) /* Handle */ - movq %rsi, efi64_config+8(%rip) /* EFI System table pointer */ - - leaq efi64_config(%rip), %rax - movq %rax, efi_config(%rip) - +handover_entry: call 1f 1: popq %rbp subq $1b, %rbp @@ -247,7 +272,6 @@ ENTRY(efi_stub_entry) */ movq efi_config(%rip), %rax addq %rbp, 88(%rax) - movq %rdx, %rsi 2: movq efi_config(%rip), %rdi call efi_main @@ -336,6 +360,20 @@ preferred_addr: leaq relocated(%rbx), %rax jmp *%rax +#ifdef CONFIG_EFI_STUB + .org 0x390 +ENTRY(efi64_stub_entry) + movq %rdi, efi64_config(%rip) /* Handle */ + movq %rsi, efi64_config+8(%rip) /* EFI System table pointer */ + + leaq efi64_config(%rip), %rax + movq %rax, efi_config(%rip) + + movq %rdx, %rsi + jmp handover_entry +ENDPROC(efi64_stub_entry) +#endif + .text relocated: @@ -404,6 +442,14 @@ gdt_end: efi_config: .quad 0 +#ifdef CONFIG_EFI_MIXED + .global efi32_config +efi32_config: + .fill 11,8,0 + .quad efi64_thunk + .byte 0 +#endif + .global efi64_config efi64_config: .fill 11,8,0 diff --git a/arch/x86/boot/tools/build.c b/arch/x86/boot/tools/build.c index bf262077ec92..4f07df5ac5d9 100644 --- a/arch/x86/boot/tools/build.c +++ b/arch/x86/boot/tools/build.c @@ -53,7 +53,8 @@ int is_big_kernel; #define PECOFF_RELOC_RESERVE 0x20 -unsigned long efi_stub_entry; +unsigned long efi32_stub_entry; +unsigned long efi64_stub_entry; unsigned long efi_pe_entry; unsigned long startup_64; @@ -231,20 +232,26 @@ static void efi_stub_defaults(void) /* Defaults for old kernel */ #ifdef CONFIG_X86_32 efi_pe_entry = 0x10; - efi_stub_entry = 0x30; #else efi_pe_entry = 0x210; - efi_stub_entry = 0x230; startup_64 = 0x200; #endif } static void efi_stub_entry_update(void) { -#ifdef CONFIG_X86_64 /* Yes, this is really how we defined it :( */ - efi_stub_entry -= 0x200; + unsigned long addr = efi32_stub_entry; + +#ifdef CONFIG_X86_64 + /* Yes, this is really how we defined it :( */ + addr = efi64_stub_entry - 0x200; +#endif + +#ifdef CONFIG_EFI_MIXED + if (efi32_stub_entry != addr) + die("32-bit and 64-bit EFI entry points do not match\n"); #endif - put_unaligned_le32(efi_stub_entry, &buf[0x264]); + put_unaligned_le32(addr, &buf[0x264]); } #else @@ -289,7 +296,8 @@ static void parse_zoffset(char *fname) p = (char *)buf; while (p && *p) { - PARSE_ZOFS(p, efi_stub_entry); + PARSE_ZOFS(p, efi32_stub_entry); + PARSE_ZOFS(p, efi64_stub_entry); PARSE_ZOFS(p, efi_pe_entry); PARSE_ZOFS(p, startup_64); diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index 3d6b9f81cc68..85856868507b 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -19,9 +19,11 @@ */ #define EFI_OLD_MEMMAP EFI_ARCH_1 +#define EFI32_LOADER_SIGNATURE "EL32" +#define EFI64_LOADER_SIGNATURE "EL64" + #ifdef CONFIG_X86_32 -#define EFI_LOADER_SIGNATURE "EL32" extern unsigned long asmlinkage efi_call_phys(void *, ...); @@ -57,8 +59,6 @@ extern unsigned long asmlinkage efi_call_phys(void *, ...); #else /* !CONFIG_X86_32 */ -#define EFI_LOADER_SIGNATURE "EL64" - extern u64 efi_call0(void *fp); extern u64 efi_call1(void *fp, u64 arg1); extern u64 efi_call2(void *fp, u64 arg1, u64 arg2); -- cgit v1.2.3 From 4f9dbcfc40299ddaa780fe8c1cd74998c1be3af5 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Fri, 10 Jan 2014 18:48:30 +0000 Subject: x86/efi: Add mixed runtime services support Setup the runtime services based on whether we're booting in EFI native mode or not. For non-native mode we need to thunk from 64-bit into 32-bit mode before invoking the EFI runtime services. Using the runtime services after SetVirtualAddressMap() is slightly more complicated because we need to ensure that all the addresses we pass to the firmware are below the 4GB boundary so that they can be addressed with 32-bit pointers, see efi_setup_page_tables(). Signed-off-by: Matt Fleming --- arch/x86/include/asm/efi.h | 21 +++ arch/x86/platform/efi/Makefile | 1 + arch/x86/platform/efi/efi.c | 127 +++++++++---- arch/x86/platform/efi/efi_64.c | 335 ++++++++++++++++++++++++++++++++++- arch/x86/platform/efi/efi_stub_64.S | 1 + arch/x86/platform/efi/efi_thunk_64.S | 65 +++++++ 6 files changed, 517 insertions(+), 33 deletions(-) create mode 100644 arch/x86/platform/efi/efi_thunk_64.S (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index 85856868507b..85935e6f69f3 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -154,6 +154,27 @@ static inline bool efi_is_native(void) extern struct console early_efi_console; extern void parse_efi_setup(u64 phys_addr, u32 data_len); + +#ifdef CONFIG_EFI_MIXED +extern void efi_thunk_runtime_setup(void); +extern efi_status_t efi_thunk_set_virtual_address_map( + void *phys_set_virtual_address_map, + unsigned long memory_map_size, + unsigned long descriptor_size, + u32 descriptor_version, + efi_memory_desc_t *virtual_map); +#else +static inline void efi_thunk_runtime_setup(void) {} +static inline efi_status_t efi_thunk_set_virtual_address_map( + void *phys_set_virtual_address_map, + unsigned long memory_map_size, + unsigned long descriptor_size, + u32 descriptor_version, + efi_memory_desc_t *virtual_map) +{ + return EFI_SUCCESS; +} +#endif /* CONFIG_EFI_MIXED */ #else /* * IF EFI is not configured, have the EFI calls return -ENOSYS. diff --git a/arch/x86/platform/efi/Makefile b/arch/x86/platform/efi/Makefile index b7b0b35c1981..d51045afcaaf 100644 --- a/arch/x86/platform/efi/Makefile +++ b/arch/x86/platform/efi/Makefile @@ -1,3 +1,4 @@ obj-$(CONFIG_EFI) += efi.o efi_$(BITS).o efi_stub_$(BITS).o obj-$(CONFIG_ACPI_BGRT) += efi-bgrt.o obj-$(CONFIG_EARLY_PRINTK_EFI) += early_printk.o +obj-$(CONFIG_EFI_MIXED) += efi_thunk_$(BITS).o diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index b96ae7918a16..39f5b7bba695 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -588,37 +588,85 @@ static int __init efi_systab_init(void *phys) return 0; } -static int __init efi_runtime_init(void) +static int __init efi_runtime_init32(void) { - efi_runtime_services_t *runtime; + efi_runtime_services_32_t *runtime; + + runtime = early_ioremap((unsigned long)efi.systab->runtime, + sizeof(efi_runtime_services_32_t)); + if (!runtime) { + pr_err("Could not map the runtime service table!\n"); + return -ENOMEM; + } /* - * 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. + * 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 *) + (unsigned long)runtime->get_time; + efi_phys.set_virtual_address_map = + (efi_set_virtual_address_map_t *) + (unsigned long)runtime->set_virtual_address_map; + /* + * Make efi_get_time can be called before entering + * virtual mode. + */ + efi.get_time = phys_efi_get_time; + early_iounmap(runtime, sizeof(efi_runtime_services_32_t)); + + return 0; +} + +static int __init efi_runtime_init64(void) +{ + efi_runtime_services_64_t *runtime; + runtime = early_ioremap((unsigned long)efi.systab->runtime, - sizeof(efi_runtime_services_t)); + sizeof(efi_runtime_services_64_t)); if (!runtime) { pr_err("Could not map the runtime service table!\n"); return -ENOMEM; } + /* - * We will only need *early* access to the following - * two EFI runtime services before set_virtual_address_map + * 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.get_time = (efi_get_time_t *) + (unsigned long)runtime->get_time; efi_phys.set_virtual_address_map = - (efi_set_virtual_address_map_t *) - runtime->set_virtual_address_map; + (efi_set_virtual_address_map_t *) + (unsigned long)runtime->set_virtual_address_map; /* * Make efi_get_time can be called before entering * virtual mode. */ efi.get_time = phys_efi_get_time; - early_iounmap(runtime, sizeof(efi_runtime_services_t)); + early_iounmap(runtime, sizeof(efi_runtime_services_64_t)); + + return 0; +} + +static int __init efi_runtime_init(void) +{ + int rv; + + /* + * 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. + */ + if (efi_enabled(EFI_64BIT)) + rv = efi_runtime_init64(); + else + rv = efi_runtime_init32(); + + if (rv) + return rv; return 0; } @@ -841,6 +889,22 @@ void __init old_map_region(efi_memory_desc_t *md) (unsigned long long)md->phys_addr); } +static void native_runtime_setup(void) +{ + efi.get_time = virt_efi_get_time; + efi.set_time = virt_efi_set_time; + efi.get_wakeup_time = virt_efi_get_wakeup_time; + efi.set_wakeup_time = virt_efi_set_wakeup_time; + efi.get_variable = virt_efi_get_variable; + efi.get_next_variable = virt_efi_get_next_variable; + efi.set_variable = virt_efi_set_variable; + efi.get_next_high_mono_count = virt_efi_get_next_high_mono_count; + efi.reset_system = virt_efi_reset_system; + efi.query_variable_info = virt_efi_query_variable_info; + efi.update_capsule = virt_efi_update_capsule; + efi.query_capsule_caps = virt_efi_query_capsule_caps; +} + /* Merge contiguous regions of the same type and attribute */ static void __init efi_merge_regions(void) { @@ -1023,11 +1087,20 @@ void __init efi_enter_virtual_mode(void) efi_sync_low_kernel_mappings(); if (!efi_setup) { - status = phys_efi_set_virtual_address_map( - memmap.desc_size * count, - memmap.desc_size, - memmap.desc_version, - (efi_memory_desc_t *)__pa(new_memmap)); + if (efi_is_native()) { + status = phys_efi_set_virtual_address_map( + memmap.desc_size * count, + memmap.desc_size, + memmap.desc_version, + (efi_memory_desc_t *)__pa(new_memmap)); + } else { + status = efi_thunk_set_virtual_address_map( + efi_phys.set_virtual_address_map, + memmap.desc_size * count, + memmap.desc_size, + memmap.desc_version, + (efi_memory_desc_t *)__pa(new_memmap)); + } if (status != EFI_SUCCESS) { pr_alert("Unable to switch EFI into virtual mode (status=%lx)!\n", @@ -1043,19 +1116,13 @@ void __init efi_enter_virtual_mode(void) * Call EFI services through wrapper functions. */ efi.runtime_version = efi_systab.hdr.revision; - efi.get_time = virt_efi_get_time; - efi.set_time = virt_efi_set_time; - efi.get_wakeup_time = virt_efi_get_wakeup_time; - efi.set_wakeup_time = virt_efi_set_wakeup_time; - efi.get_variable = virt_efi_get_variable; - efi.get_next_variable = virt_efi_get_next_variable; - efi.set_variable = virt_efi_set_variable; - efi.get_next_high_mono_count = virt_efi_get_next_high_mono_count; - efi.reset_system = virt_efi_reset_system; + + if (efi_is_native()) + native_runtime_setup(); + else + efi_thunk_runtime_setup(); + efi.set_virtual_address_map = NULL; - efi.query_variable_info = virt_efi_query_variable_info; - efi.update_capsule = virt_efi_update_capsule; - efi.query_capsule_caps = virt_efi_query_capsule_caps; efi_runtime_mkexec(); diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index 0c2a234fef1e..12112ab4fd40 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -39,6 +39,7 @@ #include #include #include +#include static pgd_t *save_pgd __initdata; static unsigned long efi_flags __initdata; @@ -58,7 +59,8 @@ struct efi_scratch { u64 prev_cr3; pgd_t *efi_pgt; bool use_pgd; -}; + u64 phys_stack; +} __packed; static void __init early_code_mapping_set_exec(int executable) { @@ -139,10 +141,40 @@ void efi_sync_low_kernel_mappings(void) void efi_setup_page_tables(void) { + unsigned long text; + unsigned npages; + struct page *page; + efi_scratch.efi_pgt = (pgd_t *)(unsigned long)real_mode_header->trampoline_pgd; - if (!efi_enabled(EFI_OLD_MEMMAP)) - efi_scratch.use_pgd = true; + if (efi_enabled(EFI_OLD_MEMMAP)) + return; + + efi_scratch.use_pgd = true; + + /* + * When making calls to the firmware everything needs to be 1:1 + * mapped and addressable with 32-bit pointers. Map the kernel + * text and allocate a new stack because we can't rely on the + * stack pointer being < 4GB. + */ + if (!IS_ENABLED(CONFIG_EFI_MIXED)) + return; + + page = alloc_page(GFP_KERNEL|__GFP_DMA32); + if (!page) + panic("Unable to allocate EFI runtime stack < 4GB\n"); + + efi_scratch.phys_stack = virt_to_phys(page_address(page)); + efi_scratch.phys_stack += PAGE_SIZE; /* stack grows down */ + + npages = (_end - _text) >> PAGE_SHIFT; + text = __pa(_text); + + if (kernel_map_pages_in_pgd(__va(efi_scratch.efi_pgt), + text >> PAGE_SHIFT, text, npages, 0)) { + pr_err("Failed to map kernel text 1:1\n"); + } } static void __init __map_region(efi_memory_desc_t *md, u64 va) @@ -173,6 +205,16 @@ void __init efi_map_region(efi_memory_desc_t *md) */ __map_region(md, md->phys_addr); + /* + * Enforce the 1:1 mapping as the default virtual address when + * booting in EFI mixed mode, because even though we may be + * running a 64-bit kernel, the firmware may only be 32-bit. + */ + if (!efi_is_native () && IS_ENABLED(CONFIG_EFI_MIXED)) { + md->virt_addr = md->phys_addr; + return; + } + efi_va -= size; /* Is PA 2M-aligned? */ @@ -242,3 +284,290 @@ void __init efi_runtime_mkexec(void) if (__supported_pte_mask & _PAGE_NX) runtime_code_page_mkexec(); } + +#ifdef CONFIG_EFI_MIXED +extern efi_status_t efi64_thunk(u32, ...); + +#define runtime_service32(func) \ +({ \ + u32 table = (u32)(unsigned long)efi.systab; \ + u32 *rt, *___f; \ + \ + rt = (u32 *)(table + offsetof(efi_system_table_32_t, runtime)); \ + ___f = (u32 *)(*rt + offsetof(efi_runtime_services_32_t, func)); \ + *___f; \ +}) + +/* + * Switch to the EFI page tables early so that we can access the 1:1 + * runtime services mappings which are not mapped in any other page + * tables. This function must be called before runtime_service32(). + * + * Also, disable interrupts because the IDT points to 64-bit handlers, + * which aren't going to function correctly when we switch to 32-bit. + */ +#define efi_thunk(f, ...) \ +({ \ + efi_status_t __s; \ + unsigned long flags; \ + u32 func; \ + \ + efi_sync_low_kernel_mappings(); \ + local_irq_save(flags); \ + \ + efi_scratch.prev_cr3 = read_cr3(); \ + write_cr3((unsigned long)efi_scratch.efi_pgt); \ + __flush_tlb_all(); \ + \ + func = runtime_service32(f); \ + __s = efi64_thunk(func, __VA_ARGS__); \ + \ + write_cr3(efi_scratch.prev_cr3); \ + __flush_tlb_all(); \ + local_irq_restore(flags); \ + \ + __s; \ +}) + +efi_status_t efi_thunk_set_virtual_address_map( + void *phys_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; + unsigned long flags; + u32 func; + + efi_sync_low_kernel_mappings(); + local_irq_save(flags); + + efi_scratch.prev_cr3 = read_cr3(); + write_cr3((unsigned long)efi_scratch.efi_pgt); + __flush_tlb_all(); + + func = (u32)(unsigned long)phys_set_virtual_address_map; + status = efi64_thunk(func, memory_map_size, descriptor_size, + descriptor_version, virtual_map); + + write_cr3(efi_scratch.prev_cr3); + __flush_tlb_all(); + local_irq_restore(flags); + + return status; +} + +static efi_status_t efi_thunk_get_time(efi_time_t *tm, efi_time_cap_t *tc) +{ + efi_status_t status; + u32 phys_tm, phys_tc; + + spin_lock(&rtc_lock); + + phys_tm = virt_to_phys(tm); + phys_tc = virt_to_phys(tc); + + status = efi_thunk(get_time, phys_tm, phys_tc); + + spin_unlock(&rtc_lock); + + return status; +} + +static efi_status_t efi_thunk_set_time(efi_time_t *tm) +{ + efi_status_t status; + u32 phys_tm; + + spin_lock(&rtc_lock); + + phys_tm = virt_to_phys(tm); + + status = efi_thunk(set_time, phys_tm); + + spin_unlock(&rtc_lock); + + return status; +} + +static efi_status_t +efi_thunk_get_wakeup_time(efi_bool_t *enabled, efi_bool_t *pending, + efi_time_t *tm) +{ + efi_status_t status; + u32 phys_enabled, phys_pending, phys_tm; + + spin_lock(&rtc_lock); + + phys_enabled = virt_to_phys(enabled); + phys_pending = virt_to_phys(pending); + phys_tm = virt_to_phys(tm); + + status = efi_thunk(get_wakeup_time, phys_enabled, + phys_pending, phys_tm); + + spin_unlock(&rtc_lock); + + return status; +} + +static efi_status_t +efi_thunk_set_wakeup_time(efi_bool_t enabled, efi_time_t *tm) +{ + efi_status_t status; + u32 phys_tm; + + spin_lock(&rtc_lock); + + phys_tm = virt_to_phys(tm); + + status = efi_thunk(set_wakeup_time, enabled, phys_tm); + + spin_unlock(&rtc_lock); + + return status; +} + + +static efi_status_t +efi_thunk_get_variable(efi_char16_t *name, efi_guid_t *vendor, + u32 *attr, unsigned long *data_size, void *data) +{ + efi_status_t status; + u32 phys_name, phys_vendor, phys_attr; + u32 phys_data_size, phys_data; + + phys_data_size = virt_to_phys(data_size); + phys_vendor = virt_to_phys(vendor); + phys_name = virt_to_phys(name); + phys_attr = virt_to_phys(attr); + phys_data = virt_to_phys(data); + + status = efi_thunk(get_variable, phys_name, phys_vendor, + phys_attr, phys_data_size, phys_data); + + return status; +} + +static efi_status_t +efi_thunk_set_variable(efi_char16_t *name, efi_guid_t *vendor, + u32 attr, unsigned long data_size, void *data) +{ + u32 phys_name, phys_vendor, phys_data; + efi_status_t status; + + phys_name = virt_to_phys(name); + phys_vendor = virt_to_phys(vendor); + phys_data = virt_to_phys(data); + + /* If data_size is > sizeof(u32) we've got problems */ + status = efi_thunk(set_variable, phys_name, phys_vendor, + attr, data_size, phys_data); + + return status; +} + +static efi_status_t +efi_thunk_get_next_variable(unsigned long *name_size, + efi_char16_t *name, + efi_guid_t *vendor) +{ + efi_status_t status; + u32 phys_name_size, phys_name, phys_vendor; + + phys_name_size = virt_to_phys(name_size); + phys_vendor = virt_to_phys(vendor); + phys_name = virt_to_phys(name); + + status = efi_thunk(get_next_variable, phys_name_size, + phys_name, phys_vendor); + + return status; +} + +static efi_status_t +efi_thunk_get_next_high_mono_count(u32 *count) +{ + efi_status_t status; + u32 phys_count; + + phys_count = virt_to_phys(count); + status = efi_thunk(get_next_high_mono_count, phys_count); + + return status; +} + +static void +efi_thunk_reset_system(int reset_type, efi_status_t status, + unsigned long data_size, efi_char16_t *data) +{ + u32 phys_data; + + phys_data = virt_to_phys(data); + + efi_thunk(reset_system, reset_type, status, data_size, phys_data); +} + +static efi_status_t +efi_thunk_update_capsule(efi_capsule_header_t **capsules, + unsigned long count, unsigned long sg_list) +{ + /* + * To properly support this function we would need to repackage + * 'capsules' because the firmware doesn't understand 64-bit + * pointers. + */ + return EFI_UNSUPPORTED; +} + +static efi_status_t +efi_thunk_query_variable_info(u32 attr, u64 *storage_space, + u64 *remaining_space, + u64 *max_variable_size) +{ + efi_status_t status; + u32 phys_storage, phys_remaining, phys_max; + + if (efi.runtime_version < EFI_2_00_SYSTEM_TABLE_REVISION) + return EFI_UNSUPPORTED; + + phys_storage = virt_to_phys(storage_space); + phys_remaining = virt_to_phys(remaining_space); + phys_max = virt_to_phys(max_variable_size); + + status = efi_thunk(query_variable_info, phys_storage, + phys_remaining, phys_max); + + return status; +} + +static efi_status_t +efi_thunk_query_capsule_caps(efi_capsule_header_t **capsules, + unsigned long count, u64 *max_size, + int *reset_type) +{ + /* + * To properly support this function we would need to repackage + * 'capsules' because the firmware doesn't understand 64-bit + * pointers. + */ + return EFI_UNSUPPORTED; +} + +void efi_thunk_runtime_setup(void) +{ + efi.get_time = efi_thunk_get_time; + efi.set_time = efi_thunk_set_time; + efi.get_wakeup_time = efi_thunk_get_wakeup_time; + efi.set_wakeup_time = efi_thunk_set_wakeup_time; + efi.get_variable = efi_thunk_get_variable; + efi.get_next_variable = efi_thunk_get_next_variable; + efi.set_variable = efi_thunk_set_variable; + efi.get_next_high_mono_count = efi_thunk_get_next_high_mono_count; + efi.reset_system = efi_thunk_reset_system; + efi.query_variable_info = efi_thunk_query_variable_info; + efi.update_capsule = efi_thunk_update_capsule; + efi.query_capsule_caps = efi_thunk_query_capsule_caps; +} +#endif /* CONFIG_EFI_MIXED */ diff --git a/arch/x86/platform/efi/efi_stub_64.S b/arch/x86/platform/efi/efi_stub_64.S index a790d69cc85e..e811514beeac 100644 --- a/arch/x86/platform/efi/efi_stub_64.S +++ b/arch/x86/platform/efi/efi_stub_64.S @@ -318,3 +318,4 @@ efi_gdt64_end: ENTRY(efi_scratch) .fill 3,8,0 .byte 0 + .quad 0 diff --git a/arch/x86/platform/efi/efi_thunk_64.S b/arch/x86/platform/efi/efi_thunk_64.S new file mode 100644 index 000000000000..8806fa73e6e6 --- /dev/null +++ b/arch/x86/platform/efi/efi_thunk_64.S @@ -0,0 +1,65 @@ +/* + * Copyright (C) 2014 Intel Corporation; author Matt Fleming + */ + +#include +#include + + .text + .code64 +ENTRY(efi64_thunk) + push %rbp + push %rbx + + /* + * Switch to 1:1 mapped 32-bit stack pointer. + */ + movq %rsp, efi_saved_sp(%rip) + movq efi_scratch+25(%rip), %rsp + + /* + * Calculate the physical address of the kernel text. + */ + movq $__START_KERNEL_map, %rax + subq phys_base(%rip), %rax + + /* + * Push some physical addresses onto the stack. This is easier + * to do now in a code64 section while the assembler can address + * 64-bit values. Note that all the addresses on the stack are + * 32-bit. + */ + subq $16, %rsp + leaq efi_exit32(%rip), %rbx + subq %rax, %rbx + movl %ebx, 8(%rsp) + leaq efi_gdt64(%rip), %rbx + subq %rax, %rbx + movl %ebx, 2(%ebx) + movl %ebx, 4(%rsp) + leaq efi_gdt32(%rip), %rbx + subq %rax, %rbx + movl %ebx, 2(%ebx) + movl %ebx, (%rsp) + + leaq __efi64_thunk(%rip), %rbx + subq %rax, %rbx + call *%rbx + + movq efi_saved_sp(%rip), %rsp + pop %rbx + pop %rbp + retq +ENDPROC(efi64_thunk) + + .data +efi_gdt32: + .word efi_gdt32_end - efi_gdt32 + .long 0 /* Filled out above */ + .word 0 + .quad 0x0000000000000000 /* NULL descriptor */ + .quad 0x00cf9a000000ffff /* __KERNEL_CS */ + .quad 0x00cf93000000ffff /* __KERNEL_DS */ +efi_gdt32_end: + +efi_saved_sp: .quad 0 -- cgit v1.2.3 From 7d453eee36ae4cf30fc2f2faae54f634c4f863b7 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Fri, 10 Jan 2014 18:52:06 +0000 Subject: x86/efi: Wire up CONFIG_EFI_MIXED Add the Kconfig option and bump the kernel header version so that boot loaders can check whether the handover code is available if they want. The xloadflags field in the bzImage header is also updated to reflect that the kernel supports both entry points by setting both of XLF_EFI_HANDOVER_32 and XLF_EFI_HANDOVER_64 when CONFIG_EFI_MIXED=y. XLF_CAN_BE_LOADED_ABOVE_4G is disabled so that the kernel text is guaranteed to be addressable with 32-bits. Note that no boot loaders should be using the bits set in xloadflags to decide which entry point to jump to. The entire scheme is based on the concept that 32-bit bootloaders always jump to ->handover_offset and 64-bit loaders always jump to ->handover_offset + 512. We set both bits merely to inform the boot loader that it's safe to use the native handover offset even if the machine type in the PE/COFF header claims otherwise. Signed-off-by: Matt Fleming --- arch/x86/Kconfig | 14 ++++++++++++++ arch/x86/boot/header.S | 15 ++++++++++----- arch/x86/include/asm/efi.h | 11 +++++++++++ arch/x86/kernel/setup.c | 2 +- arch/x86/platform/efi/efi.c | 2 +- 5 files changed, 37 insertions(+), 7 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 0af5250d914f..8453fe1342ea 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1585,6 +1585,20 @@ config EFI_STUB See Documentation/efi-stub.txt for more information. +config EFI_MIXED + bool "EFI mixed-mode support" + depends on EFI_STUB && X86_64 + ---help--- + Enabling this feature allows a 64-bit kernel to be booted + on a 32-bit firmware, provided that your CPU supports 64-bit + mode. + + Note that it is not possible to boot a mixed-mode enabled + kernel via the EFI boot stub - a bootloader that supports + the EFI handover protocol must be used. + + If unsure, say N. + config SECCOMP def_bool y prompt "Enable seccomp to safely compute untrusted bytecode" diff --git a/arch/x86/boot/header.S b/arch/x86/boot/header.S index d69d96653325..256388260c88 100644 --- a/arch/x86/boot/header.S +++ b/arch/x86/boot/header.S @@ -283,7 +283,7 @@ _start: # Part 2 of the header, from the old setup.S .ascii "HdrS" # header signature - .word 0x020c # header version number (>= 0x0105) + .word 0x020d # header version number (>= 0x0105) # or else old loadlin-1.5 will fail) .globl realmode_swtch realmode_swtch: .word 0, 0 # default_switch, SETUPSEG @@ -375,7 +375,8 @@ xloadflags: # define XLF0 0 #endif -#if defined(CONFIG_RELOCATABLE) && defined(CONFIG_X86_64) +#if defined(CONFIG_RELOCATABLE) && defined(CONFIG_X86_64) && \ + !defined(CONFIG_EFI_MIXED) /* kernel/boot_param/ramdisk could be loaded above 4g */ # define XLF1 XLF_CAN_BE_LOADED_ABOVE_4G #else @@ -383,10 +384,14 @@ xloadflags: #endif #ifdef CONFIG_EFI_STUB -# ifdef CONFIG_X86_64 -# define XLF23 XLF_EFI_HANDOVER_64 /* 64-bit EFI handover ok */ +# ifdef CONFIG_EFI_MIXED +# define XLF23 (XLF_EFI_HANDOVER_32|XLF_EFI_HANDOVER_64) # else -# define XLF23 XLF_EFI_HANDOVER_32 /* 32-bit EFI handover ok */ +# ifdef CONFIG_X86_64 +# define XLF23 XLF_EFI_HANDOVER_64 /* 64-bit EFI handover ok */ +# else +# define XLF23 XLF_EFI_HANDOVER_32 /* 32-bit EFI handover ok */ +# endif # endif #else # define XLF23 0 diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index 85935e6f69f3..a3837254e320 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -152,6 +152,17 @@ static inline bool efi_is_native(void) return IS_ENABLED(CONFIG_X86_64) == efi_enabled(EFI_64BIT); } +static inline bool efi_runtime_supported(void) +{ + if (efi_is_native()) + return true; + + if (IS_ENABLED(CONFIG_EFI_MIXED) && !efi_enabled(EFI_OLD_MEMMAP)) + return true; + + return false; +} + extern struct console early_efi_console; extern void parse_efi_setup(u64 phys_addr, u32 data_len); diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 06853e670354..ff9b3a62a05c 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -1243,7 +1243,7 @@ void __init setup_arch(char **cmdline_p) * mismatched firmware/kernel archtectures since there is no * support for runtime services. */ - if (efi_enabled(EFI_BOOT) && !efi_is_native()) { + if (efi_enabled(EFI_BOOT) && !efi_runtime_supported()) { pr_info("efi: Setup done, disabling due to 32/64-bit mismatch\n"); efi_unmap_memmap(); } diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index 39f5b7bba695..395decedfa2b 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -802,7 +802,7 @@ void __init efi_init(void) * that doesn't match the kernel 32/64-bit mode. */ - if (!efi_is_native()) + if (!efi_runtime_supported()) pr_info("No EFI runtime due to 32/64-bit mismatch with kernel\n"); else { if (disable_runtime || efi_runtime_init()) -- cgit v1.2.3