"nosoftreserve", "disable_early_pci_dma",
"no_disable_early_pci_dma" }
old_map [X86-64]: switch to the old ioremap-based EFI
- runtime services mapping. 32-bit still uses this one by
- default.
+ runtime services mapping. [Needs CONFIG_X86_UV=y]
nochunk: disable reading files in "chunks" in the EFI
boot stub, as chunking can cause problems with some
firmware implementations.
* This is the main reason why we're doing stable VA mappings for RT
* services.
*
- * This flag is used in conjunction with a chicken bit called
- * "efi=old_map" which can be used as a fallback to the old runtime
- * services mapping method in case there's some b0rkage with a
- * particular EFI implementation (haha, it is hard to hold up the
- * sarcasm here...).
+ * SGI UV1 machines are known to be incompatible with this scheme, so we
+ * provide an opt-out for these machines via a DMI quirk that sets the
+ * attribute below.
*/
-#define EFI_OLD_MEMMAP EFI_ARCH_1
+#define EFI_UV1_MEMMAP EFI_ARCH_1
+
+static inline bool efi_have_uv1_memmap(void)
+{
+ return IS_ENABLED(CONFIG_X86_UV) && efi_enabled(EFI_UV1_MEMMAP);
+}
#define EFI32_LOADER_SIGNATURE "EL32"
#define EFI64_LOADER_SIGNATURE "EL64"
kernel_fpu_begin(); \
firmware_restrict_branch_speculation_start(); \
\
- if (!efi_enabled(EFI_OLD_MEMMAP)) \
+ if (!efi_have_uv1_memmap()) \
efi_switch_mm(&efi_mm); \
})
#define arch_efi_call_virt_teardown() \
({ \
- if (!efi_enabled(EFI_OLD_MEMMAP)) \
+ if (!efi_have_uv1_memmap()) \
efi_switch_mm(efi_scratch.prev_mm); \
\
firmware_restrict_branch_speculation_end(); \
extern void efi_switch_mm(struct mm_struct *mm);
extern void efi_recover_from_page_fault(unsigned long phys_addr);
extern void efi_free_boot_services(void);
+extern pgd_t * __init efi_uv1_memmap_phys_prolog(void);
+extern void __init efi_uv1_memmap_phys_epilog(pgd_t *save_pgd);
struct efi_setup_data {
u64 fw_vendor;
if (IS_ENABLED(CONFIG_X86_64) == efi_enabled(EFI_64BIT))
return true;
- if (IS_ENABLED(CONFIG_EFI_MIXED) && !efi_enabled(EFI_OLD_MEMMAP))
- return true;
-
- return false;
+ return IS_ENABLED(CONFIG_EFI_MIXED);
}
extern void parse_efi_setup(u64 phys_addr, u32 data_len);
* acpi_rsdp=<addr> on kernel command line to make second kernel boot
* without efi.
*/
- if (efi_enabled(EFI_OLD_MEMMAP))
+ if (efi_have_uv1_memmap())
return 0;
params->secure_boot = boot_params.secure_boot;
efi_print_memmap();
}
+#if defined(CONFIG_X86_32) || defined(CONFIG_X86_UV)
+
void __init efi_set_executable(efi_memory_desc_t *md, bool executable)
{
u64 addr, npages;
(unsigned long long)md->phys_addr);
}
+#endif
+
/* Merge contiguous regions of the same type and attribute */
static void __init efi_merge_regions(void)
{
*/
static void *efi_map_next_entry(void *entry)
{
- if (!efi_enabled(EFI_OLD_MEMMAP) && efi_enabled(EFI_64BIT)) {
+ if (!efi_have_uv1_memmap() && efi_enabled(EFI_64BIT)) {
/*
* Starting in UEFI v2.5 the EFI_PROPERTIES_TABLE
* config table feature requires us to map all entries
/*
* We don't do virtual mode, since we don't do runtime services, on
- * non-native EFI. With efi=old_map, we don't do runtime services in
+ * non-native EFI. With the UV1 memmap, we don't do runtime services in
* kexec kernel because in the initial boot something else might
* have been mapped at these virtual addresses.
*/
- if (efi_is_mixed() || efi_enabled(EFI_OLD_MEMMAP)) {
+ if (efi_is_mixed() || efi_have_uv1_memmap()) {
efi_memmap_unmap();
clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
return;
*
* The old method which used to update that memory descriptor with the
* virtual address obtained from ioremap() is still supported when the
- * kernel is booted with efi=old_map on its command line. Same old
- * method enabled the runtime services to be called without having to
- * thunk back into physical mode for every invocation.
+ * kernel is booted on SG1 UV1 hardware. Same old method enabled the
+ * runtime services to be called without having to thunk back into
+ * physical mode for every invocation.
*
* The new method does a pagetable switch in a preemption-safe manner
* so that we're in a different address space when calling a runtime
efi_dump_pagetable();
}
-static int __init arch_parse_efi_cmdline(char *str)
-{
- if (!str) {
- pr_warn("need at least one option\n");
- return -EINVAL;
- }
-
- if (parse_option_str(str, "old_map"))
- set_bit(EFI_OLD_MEMMAP, &efi.flags);
-
- return 0;
-}
-early_param("efi", arch_parse_efi_cmdline);
-
bool efi_is_table_address(unsigned long phys_addr)
{
unsigned int i;
struct efi_scratch efi_scratch;
-static void __init early_code_mapping_set_exec(int executable)
-{
- efi_memory_desc_t *md;
-
- if (!(__supported_pte_mask & _PAGE_NX))
- return;
-
- /* Make EFI service code area executable */
- for_each_efi_memory_desc(md) {
- if (md->type == EFI_RUNTIME_SERVICES_CODE ||
- md->type == EFI_BOOT_SERVICES_CODE)
- efi_set_executable(md, executable);
- }
-}
-
-static void __init efi_old_memmap_phys_epilog(pgd_t *save_pgd);
-
-static pgd_t * __init efi_old_memmap_phys_prolog(void)
-{
- unsigned long vaddr, addr_pgd, addr_p4d, addr_pud;
- pgd_t *save_pgd, *pgd_k, *pgd_efi;
- p4d_t *p4d, *p4d_k, *p4d_efi;
- pud_t *pud;
-
- int pgd;
- int n_pgds, i, j;
-
- early_code_mapping_set_exec(1);
-
- n_pgds = DIV_ROUND_UP((max_pfn << PAGE_SHIFT), PGDIR_SIZE);
- save_pgd = kmalloc_array(n_pgds, sizeof(*save_pgd), GFP_KERNEL);
- if (!save_pgd)
- return NULL;
-
- /*
- * Build 1:1 identity mapping for efi=old_map usage. Note that
- * PAGE_OFFSET is PGDIR_SIZE aligned when KASLR is disabled, while
- * it is PUD_SIZE ALIGNED with KASLR enabled. So for a given physical
- * address X, the pud_index(X) != pud_index(__va(X)), we can only copy
- * PUD entry of __va(X) to fill in pud entry of X to build 1:1 mapping.
- * This means here we can only reuse the PMD tables of the direct mapping.
- */
- for (pgd = 0; pgd < n_pgds; pgd++) {
- addr_pgd = (unsigned long)(pgd * PGDIR_SIZE);
- vaddr = (unsigned long)__va(pgd * PGDIR_SIZE);
- pgd_efi = pgd_offset_k(addr_pgd);
- save_pgd[pgd] = *pgd_efi;
-
- p4d = p4d_alloc(&init_mm, pgd_efi, addr_pgd);
- if (!p4d) {
- pr_err("Failed to allocate p4d table!\n");
- goto out;
- }
-
- for (i = 0; i < PTRS_PER_P4D; i++) {
- addr_p4d = addr_pgd + i * P4D_SIZE;
- p4d_efi = p4d + p4d_index(addr_p4d);
-
- pud = pud_alloc(&init_mm, p4d_efi, addr_p4d);
- if (!pud) {
- pr_err("Failed to allocate pud table!\n");
- goto out;
- }
-
- for (j = 0; j < PTRS_PER_PUD; j++) {
- addr_pud = addr_p4d + j * PUD_SIZE;
-
- if (addr_pud > (max_pfn << PAGE_SHIFT))
- break;
-
- vaddr = (unsigned long)__va(addr_pud);
-
- pgd_k = pgd_offset_k(vaddr);
- p4d_k = p4d_offset(pgd_k, vaddr);
- pud[j] = *pud_offset(p4d_k, vaddr);
- }
- }
- pgd_offset_k(pgd * PGDIR_SIZE)->pgd &= ~_PAGE_NX;
- }
-
- __flush_tlb_all();
- return save_pgd;
-out:
- efi_old_memmap_phys_epilog(save_pgd);
- return NULL;
-}
-
-static void __init efi_old_memmap_phys_epilog(pgd_t *save_pgd)
-{
- /*
- * After the lock is released, the original page table is restored.
- */
- int pgd_idx, i;
- int nr_pgds;
- pgd_t *pgd;
- p4d_t *p4d;
- pud_t *pud;
-
- nr_pgds = DIV_ROUND_UP((max_pfn << PAGE_SHIFT) , PGDIR_SIZE);
-
- for (pgd_idx = 0; pgd_idx < nr_pgds; pgd_idx++) {
- pgd = pgd_offset_k(pgd_idx * PGDIR_SIZE);
- set_pgd(pgd_offset_k(pgd_idx * PGDIR_SIZE), save_pgd[pgd_idx]);
-
- if (!pgd_present(*pgd))
- continue;
-
- for (i = 0; i < PTRS_PER_P4D; i++) {
- p4d = p4d_offset(pgd,
- pgd_idx * PGDIR_SIZE + i * P4D_SIZE);
-
- if (!p4d_present(*p4d))
- continue;
-
- pud = (pud_t *)p4d_page_vaddr(*p4d);
- pud_free(&init_mm, pud);
- }
-
- p4d = (p4d_t *)pgd_page_vaddr(*pgd);
- p4d_free(&init_mm, p4d);
- }
-
- kfree(save_pgd);
-
- __flush_tlb_all();
- early_code_mapping_set_exec(0);
-}
-
EXPORT_SYMBOL_GPL(efi_mm);
/*
pud_t *pud;
gfp_t gfp_mask;
- if (efi_enabled(EFI_OLD_MEMMAP))
+ if (efi_have_uv1_memmap())
return 0;
gfp_mask = GFP_KERNEL | __GFP_ZERO;
pud_t *pud_k, *pud_efi;
pgd_t *efi_pgd = efi_mm.pgd;
- if (efi_enabled(EFI_OLD_MEMMAP))
+ if (efi_have_uv1_memmap())
return;
/*
unsigned npages;
pgd_t *pgd = efi_mm.pgd;
- if (efi_enabled(EFI_OLD_MEMMAP))
+ if (efi_have_uv1_memmap())
return 0;
/*
unsigned long size = md->num_pages << PAGE_SHIFT;
u64 pa = md->phys_addr;
- if (efi_enabled(EFI_OLD_MEMMAP))
+ if (efi_have_uv1_memmap())
return old_map_region(md);
/*
__map_region(md, md->virt_addr);
}
-void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size,
- u32 type, u64 attribute)
-{
- unsigned long last_map_pfn;
-
- if (type == EFI_MEMORY_MAPPED_IO)
- return ioremap(phys_addr, size);
-
- last_map_pfn = init_memory_mapping(phys_addr, phys_addr + size);
- if ((last_map_pfn << PAGE_SHIFT) < phys_addr + size) {
- unsigned long top = last_map_pfn << PAGE_SHIFT;
- efi_ioremap(top, size - (top - phys_addr), type, attribute);
- }
-
- if (!(attribute & EFI_MEMORY_WB))
- efi_memory_uc((u64)(unsigned long)__va(phys_addr), size);
-
- return (void __iomem *)__va(phys_addr);
-}
-
void __init parse_efi_setup(u64 phys_addr, u32 data_len)
{
efi_setup = phys_addr + sizeof(struct setup_data);
{
efi_memory_desc_t *md;
- if (efi_enabled(EFI_OLD_MEMMAP)) {
+ if (efi_have_uv1_memmap()) {
if (__supported_pte_mask & _PAGE_NX)
runtime_code_page_mkexec();
return;
void __init efi_dump_pagetable(void)
{
#ifdef CONFIG_EFI_PGT_DUMP
- if (efi_enabled(EFI_OLD_MEMMAP))
+ if (efi_have_uv1_memmap())
ptdump_walk_pgd_level(NULL, swapper_pg_dir);
else
ptdump_walk_pgd_level(NULL, efi_mm.pgd);
descriptor_version,
virtual_map);
- if (efi_enabled(EFI_OLD_MEMMAP)) {
- save_pgd = efi_old_memmap_phys_prolog();
+ if (efi_have_uv1_memmap()) {
+ save_pgd = efi_uv1_memmap_phys_prolog();
if (!save_pgd)
return EFI_ABORTED;
} else {
kernel_fpu_end();
if (save_pgd)
- efi_old_memmap_phys_epilog(save_pgd);
+ efi_uv1_memmap_phys_epilog(save_pgd);
else
efi_switch_mm(efi_scratch.prev_mm);
/*
* To Do: Remove this check after adding functionality to unmap EFI boot
- * services code/data regions from direct mapping area because
- * "efi=old_map" maps EFI regions in swapper_pg_dir.
+ * services code/data regions from direct mapping area because the UV1
+ * memory map maps EFI regions in swapper_pg_dir.
*/
- if (efi_enabled(EFI_OLD_MEMMAP))
+ if (efi_have_uv1_memmap())
return;
/*
return ret;
}
-static const struct dmi_system_id sgi_uv1_dmi[] = {
+static const struct dmi_system_id sgi_uv1_dmi[] __initconst = {
{ NULL, "SGI UV1",
{ DMI_MATCH(DMI_PRODUCT_NAME, "Stoutland Platform"),
DMI_MATCH(DMI_PRODUCT_VERSION, "1.0"),
}
/* UV2+ BIOS has a fix for this issue. UV1 still needs the quirk. */
- if (dmi_check_system(sgi_uv1_dmi))
- set_bit(EFI_OLD_MEMMAP, &efi.flags);
+ if (dmi_check_system(sgi_uv1_dmi)) {
+ if (IS_ENABLED(CONFIG_X86_UV)) {
+ set_bit(EFI_UV1_MEMMAP, &efi.flags);
+ } else {
+ pr_warn("EFI runtime disabled, needs CONFIG_X86_UV=y on UV1\n");
+ clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+ efi_memmap_unmap();
+ }
+ }
}
/*
/*
* Make sure that an efi runtime service caused the page fault.
* "efi_mm" cannot be used to check if the page fault had occurred
- * in the firmware context because efi=old_map doesn't use efi_pgd.
+ * in the firmware context because the UV1 memmap doesn't use efi_pgd.
*/
if (efi_rts_work.efi_rts_id == EFI_NONE)
return;
return BIOS_STATUS_UNIMPLEMENTED;
/*
- * If EFI_OLD_MEMMAP is set, we need to fall back to using our old EFI
+ * If EFI_UV1_MEMMAP is set, we need to fall back to using our old EFI
* callback method, which uses efi_call() directly, with the kernel page tables:
*/
- if (unlikely(efi_enabled(EFI_OLD_MEMMAP))) {
+ if (unlikely(efi_enabled(EFI_UV1_MEMMAP))) {
kernel_fpu_begin();
ret = efi_call((void *)__va(tab->function), (u64)which, a1, a2, a3, a4, a5);
kernel_fpu_end();
pr_info("UV: UVsystab: Revision:%x\n", uv_systab->revision);
return 0;
}
+
+static void __init early_code_mapping_set_exec(int executable)
+{
+ efi_memory_desc_t *md;
+
+ if (!(__supported_pte_mask & _PAGE_NX))
+ return;
+
+ /* Make EFI service code area executable */
+ for_each_efi_memory_desc(md) {
+ if (md->type == EFI_RUNTIME_SERVICES_CODE ||
+ md->type == EFI_BOOT_SERVICES_CODE)
+ efi_set_executable(md, executable);
+ }
+}
+
+void __init efi_uv1_memmap_phys_epilog(pgd_t *save_pgd)
+{
+ /*
+ * After the lock is released, the original page table is restored.
+ */
+ int pgd_idx, i;
+ int nr_pgds;
+ pgd_t *pgd;
+ p4d_t *p4d;
+ pud_t *pud;
+
+ nr_pgds = DIV_ROUND_UP((max_pfn << PAGE_SHIFT) , PGDIR_SIZE);
+
+ for (pgd_idx = 0; pgd_idx < nr_pgds; pgd_idx++) {
+ pgd = pgd_offset_k(pgd_idx * PGDIR_SIZE);
+ set_pgd(pgd_offset_k(pgd_idx * PGDIR_SIZE), save_pgd[pgd_idx]);
+
+ if (!pgd_present(*pgd))
+ continue;
+
+ for (i = 0; i < PTRS_PER_P4D; i++) {
+ p4d = p4d_offset(pgd,
+ pgd_idx * PGDIR_SIZE + i * P4D_SIZE);
+
+ if (!p4d_present(*p4d))
+ continue;
+
+ pud = (pud_t *)p4d_page_vaddr(*p4d);
+ pud_free(&init_mm, pud);
+ }
+
+ p4d = (p4d_t *)pgd_page_vaddr(*pgd);
+ p4d_free(&init_mm, p4d);
+ }
+
+ kfree(save_pgd);
+
+ __flush_tlb_all();
+ early_code_mapping_set_exec(0);
+}
+
+pgd_t * __init efi_uv1_memmap_phys_prolog(void)
+{
+ unsigned long vaddr, addr_pgd, addr_p4d, addr_pud;
+ pgd_t *save_pgd, *pgd_k, *pgd_efi;
+ p4d_t *p4d, *p4d_k, *p4d_efi;
+ pud_t *pud;
+
+ int pgd;
+ int n_pgds, i, j;
+
+ early_code_mapping_set_exec(1);
+
+ n_pgds = DIV_ROUND_UP((max_pfn << PAGE_SHIFT), PGDIR_SIZE);
+ save_pgd = kmalloc_array(n_pgds, sizeof(*save_pgd), GFP_KERNEL);
+ if (!save_pgd)
+ return NULL;
+
+ /*
+ * Build 1:1 identity mapping for UV1 memmap usage. Note that
+ * PAGE_OFFSET is PGDIR_SIZE aligned when KASLR is disabled, while
+ * it is PUD_SIZE ALIGNED with KASLR enabled. So for a given physical
+ * address X, the pud_index(X) != pud_index(__va(X)), we can only copy
+ * PUD entry of __va(X) to fill in pud entry of X to build 1:1 mapping.
+ * This means here we can only reuse the PMD tables of the direct mapping.
+ */
+ for (pgd = 0; pgd < n_pgds; pgd++) {
+ addr_pgd = (unsigned long)(pgd * PGDIR_SIZE);
+ vaddr = (unsigned long)__va(pgd * PGDIR_SIZE);
+ pgd_efi = pgd_offset_k(addr_pgd);
+ save_pgd[pgd] = *pgd_efi;
+
+ p4d = p4d_alloc(&init_mm, pgd_efi, addr_pgd);
+ if (!p4d) {
+ pr_err("Failed to allocate p4d table!\n");
+ goto out;
+ }
+
+ for (i = 0; i < PTRS_PER_P4D; i++) {
+ addr_p4d = addr_pgd + i * P4D_SIZE;
+ p4d_efi = p4d + p4d_index(addr_p4d);
+
+ pud = pud_alloc(&init_mm, p4d_efi, addr_p4d);
+ if (!pud) {
+ pr_err("Failed to allocate pud table!\n");
+ goto out;
+ }
+
+ for (j = 0; j < PTRS_PER_PUD; j++) {
+ addr_pud = addr_p4d + j * PUD_SIZE;
+
+ if (addr_pud > (max_pfn << PAGE_SHIFT))
+ break;
+
+ vaddr = (unsigned long)__va(addr_pud);
+
+ pgd_k = pgd_offset_k(vaddr);
+ p4d_k = p4d_offset(pgd_k, vaddr);
+ pud[j] = *pud_offset(p4d_k, vaddr);
+ }
+ }
+ pgd_offset_k(pgd * PGDIR_SIZE)->pgd &= ~_PAGE_NX;
+ }
+
+ __flush_tlb_all();
+ return save_pgd;
+out:
+ efi_uv1_memmap_phys_epilog(save_pgd);
+ return NULL;
+}
+
+void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size,
+ u32 type, u64 attribute)
+{
+ unsigned long last_map_pfn;
+
+ if (type == EFI_MEMORY_MAPPED_IO)
+ return ioremap(phys_addr, size);
+
+ last_map_pfn = init_memory_mapping(phys_addr, phys_addr + size);
+ if ((last_map_pfn << PAGE_SHIFT) < phys_addr + size) {
+ unsigned long top = last_map_pfn << PAGE_SHIFT;
+ efi_ioremap(top, size - (top - phys_addr), type, attribute);
+ }
+
+ if (!(attribute & EFI_MEMORY_WB))
+ efi_memory_uc((u64)(unsigned long)__va(phys_addr), size);
+
+ return (void __iomem *)__va(phys_addr);
+}
+
+static int __init arch_parse_efi_cmdline(char *str)
+{
+ if (!str) {
+ pr_warn("need at least one option\n");
+ return -EINVAL;
+ }
+
+ if (parse_option_str(str, "old_map"))
+ set_bit(EFI_UV1_MEMMAP, &efi.flags);
+
+ return 0;
+}
+early_param("efi", arch_parse_efi_cmdline);
projects / linux.git / commitdiff