x86/efi: Add efi_fake_mem support for EFI_MEMORY_SP

Given that EFI_MEMORY_SP is platform BIOS policy decision for marking
memory ranges as "reserved for a specific purpose" there will inevitably
be scenarios where the BIOS omits the attribute in situations where it
is desired. Unlike other attributes if the OS wants to reserve this
memory from the kernel the reservation needs to happen early in init. So
early, in fact, that it needs to happen before e820__memblock_setup()
which is a pre-requisite for efi_fake_memmap() that wants to allocate
memory for the updated table.

Introduce an x86 specific efi_fake_memmap_early() that can search for
attempts to set EFI_MEMORY_SP via efi_fake_mem and update the e820 table
accordingly.

The KASLR code that scans the command line looking for user-directed
memory reservations also needs to be updated to consider
"efi_fake_mem=nn@ss:0x40000" requests.

Acked-by: Ard Biesheuvel <[e-mail]>
Reviewed-by: Dave Hansen <[e-mail]>
Signed-off-by: Dan Williams <[e-mail]>
Acked-by: Thomas Gleixner <[e-mail]>
Signed-off-by: Rafael J. Wysocki <[e-mail]>

Documentation/admin-guide/kernel-parameters.txt

···
 			updating original EFI memory map.
 			Region of memory which aa attribute is added to is
 			from ss to ss+nn.
+
 			If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
 			is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
 			attribute is added to range 0x100000000-0x180000000 and
 			0x10a0000000-0x1120000000.
 
+			If efi_fake_mem=8G@9G:0x40000 is specified, the
+			EFI_MEMORY_SP(0x40000) attribute is added to
+			range 0x240000000-0x43fffffff.
+
 			Using this parameter you can do debugging of EFI memmap
-			related feature. For example, you can do debugging of
+			related features. For example, you can do debugging of
 			Address Range Mirroring feature even if your box
-			doesn't support it.
+			doesn't support it, or mark specific memory as
+			"soft reserved".
 
 	efivar_ssdt=	[EFI; X86] Name of an EFI variable that contains an SSDT
 			that is to be dynamically loaded by Linux. If there are

arch/x86/boot/compressed/kaslr.c

···
 #include "../../../../lib/ctype.c"
 #include "../../../../lib/cmdline.c"
 
+enum parse_mode {
+	PARSE_MEMMAP,
+	PARSE_EFI,
+};
+
 static int
-parse_memmap(char *p, unsigned long long *start, unsigned long long *size)
+parse_memmap(char *p, unsigned long long *start, unsigned long long *size,
+		enum parse_mode mode)
 {
 	char *oldp;
 
···
 		*start = memparse(p + 1, &p);
 		return 0;
 	case '@':
-		/* memmap=nn@ss specifies usable region, should be skipped */
-		*size = 0;
+		if (mode == PARSE_MEMMAP) {
+			/*
+			 * memmap=nn@ss specifies usable region, should
+			 * be skipped
+			 */
+			*size = 0;
+		} else {
+			unsigned long long flags;
+
+			/*
+			 * efi_fake_mem=nn@ss:attr the attr specifies
+			 * flags that might imply a soft-reservation.
+			 */
+			*start = memparse(p + 1, &p);
+			if (p && *p == ':') {
+				p++;
+				if (kstrtoull(p, 0, &flags) < 0)
+					*size = 0;
+				else if (flags & EFI_MEMORY_SP)
+					return 0;
+			}
+			*size = 0;
+		}
 		/* Fall through */
 	default:
 		/*
···
 	return -EINVAL;
 }
 
-static void mem_avoid_memmap(char *str)
+static void mem_avoid_memmap(enum parse_mode mode, char *str)
 {
 	static int i;
 
···
 		if (k)
 			*k++ = 0;
 
-		rc = parse_memmap(str, &start, &size);
+		rc = parse_memmap(str, &start, &size, mode);
 		if (rc < 0)
 			break;
 		str = k;
···
 	}
 }
 
-
 static void handle_mem_options(void)
 {
 	char *args = (char *)get_cmd_line_ptr();
···
 		}
 
 		if (!strcmp(param, "memmap")) {
-			mem_avoid_memmap(val);
+			mem_avoid_memmap(PARSE_MEMMAP, val);
 		} else if (strstr(param, "hugepages")) {
 			parse_gb_huge_pages(param, val);
 		} else if (!strcmp(param, "mem")) {
···
 				goto out;
 
 			mem_limit = mem_size;
+		} else if (!strcmp(param, "efi_fake_mem")) {
+			mem_avoid_memmap(PARSE_EFI, val);
 		}
 	}
 

arch/x86/include/asm/efi.h

···
 }
 #endif /* CONFIG_EFI */
 
+#ifdef CONFIG_EFI_FAKE_MEMMAP
+extern void __init efi_fake_memmap_early(void);
+#else
+static inline void efi_fake_memmap_early(void)
+{
+}
+#endif
+
 #endif /* _ASM_X86_EFI_H */

arch/x86/platform/efi/efi.c

···
 	if (add_efi_memmap || do_efi_soft_reserve())
 		do_add_efi_memmap();
 
+	efi_fake_memmap_early();
+
 	WARN(efi.memmap.desc_version != 1,
 	     "Unexpected EFI_MEMORY_DESCRIPTOR version %ld",
 	     efi.memmap.desc_version);

drivers/firmware/efi/Makefile

···
 obj-$(CONFIG_EFI_RUNTIME_MAP)		+= runtime-map.o
 obj-$(CONFIG_EFI_RUNTIME_WRAPPERS)	+= runtime-wrappers.o
 obj-$(CONFIG_EFI_STUB)			+= libstub/
-obj-$(CONFIG_EFI_FAKE_MEMMAP)		+= fake_mem.o
+obj-$(CONFIG_EFI_FAKE_MEMMAP)		+= fake_map.o
 obj-$(CONFIG_EFI_BOOTLOADER_CONTROL)	+= efibc.o
 obj-$(CONFIG_EFI_TEST)			+= test/
 obj-$(CONFIG_EFI_DEV_PATH_PARSER)	+= dev-path-parser.o
 obj-$(CONFIG_APPLE_PROPERTIES)		+= apple-properties.o
 obj-$(CONFIG_EFI_RCI2_TABLE)		+= rci2-table.o
+
+fake_map-y				+= fake_mem.o
+fake_map-$(CONFIG_X86)			+= x86_fake_mem.o
 
 arm-obj-$(CONFIG_EFI)			:= arm-init.o arm-runtime.o
 obj-$(CONFIG_ARM)			+= $(arm-obj-y)

drivers/firmware/efi/fake_mem.c

···
 #include <linux/memblock.h>
 #include <linux/types.h>
 #include <linux/sort.h>
-#include <asm/efi.h>
+#include "fake_mem.h"
 
-#define EFI_MAX_FAKEMEM CONFIG_EFI_MAX_FAKE_MEM
-
-static struct efi_mem_range fake_mems[EFI_MAX_FAKEMEM];
-static int nr_fake_mem;
+struct efi_mem_range efi_fake_mems[EFI_MAX_FAKEMEM];
+int nr_fake_mem;
 
 static int __init cmp_fake_mem(const void *x1, const void *x2)
 {
···
 	/* count up the number of EFI memory descriptor */
 	for (i = 0; i < nr_fake_mem; i++) {
 		for_each_efi_memory_desc(md) {
-			struct range *r = &fake_mems[i].range;
+			struct range *r = &efi_fake_mems[i].range;
 
 			new_nr_map += efi_memmap_split_count(md, r);
 		}
···
 	}
 
 	for (i = 0; i < nr_fake_mem; i++)
-		efi_memmap_insert(&efi.memmap, new_memmap, &fake_mems[i]);
+		efi_memmap_insert(&efi.memmap, new_memmap, &efi_fake_mems[i]);
 
 	/* swap into new EFI memmap */
 	early_memunmap(new_memmap, efi.memmap.desc_size * new_nr_map);
···
 		if (nr_fake_mem >= EFI_MAX_FAKEMEM)
 			break;
 
-		fake_mems[nr_fake_mem].range.start = start;
-		fake_mems[nr_fake_mem].range.end = start + mem_size - 1;
-		fake_mems[nr_fake_mem].attribute = attribute;
+		efi_fake_mems[nr_fake_mem].range.start = start;
+		efi_fake_mems[nr_fake_mem].range.end = start + mem_size - 1;
+		efi_fake_mems[nr_fake_mem].attribute = attribute;
 		nr_fake_mem++;
 
 		if (*p == ',')
 			p++;
 	}
 
-	sort(fake_mems, nr_fake_mem, sizeof(struct efi_mem_range),
+	sort(efi_fake_mems, nr_fake_mem, sizeof(struct efi_mem_range),
 	     cmp_fake_mem, NULL);
 
 	for (i = 0; i < nr_fake_mem; i++)
 		pr_info("efi_fake_mem: add attr=0x%016llx to [mem 0x%016llx-0x%016llx]",
-			fake_mems[i].attribute, fake_mems[i].range.start,
-			fake_mems[i].range.end);
+			efi_fake_mems[i].attribute, efi_fake_mems[i].range.start,
+			efi_fake_mems[i].range.end);
 
 	return *p == '\0' ? 0 : -EINVAL;
 }

drivers/firmware/efi/fake_mem.h

···
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __EFI_FAKE_MEM_H__
+#define __EFI_FAKE_MEM_H__
+#include <asm/efi.h>
+
+#define EFI_MAX_FAKEMEM CONFIG_EFI_MAX_FAKE_MEM
+
+extern struct efi_mem_range efi_fake_mems[EFI_MAX_FAKEMEM];
+extern int nr_fake_mem;
+#endif /* __EFI_FAKE_MEM_H__ */

drivers/firmware/efi/x86_fake_mem.c

···
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2019 Intel Corporation. All rights reserved. */
+#include <linux/efi.h>
+#include <asm/e820/api.h>
+#include "fake_mem.h"
+
+void __init efi_fake_memmap_early(void)
+{
+	int i;
+
+	/*
+	 * The late efi_fake_mem() call can handle all requests if
+	 * EFI_MEMORY_SP support is disabled.
+	 */
+	if (!efi_soft_reserve_enabled())
+		return;
+
+	if (!efi_enabled(EFI_MEMMAP) || !nr_fake_mem)
+		return;
+
+	/*
+	 * Given that efi_fake_memmap() needs to perform memblock
+	 * allocations it needs to run after e820__memblock_setup().
+	 * However, if efi_fake_mem specifies EFI_MEMORY_SP for a given
+	 * address range that potentially needs to mark the memory as
+	 * reserved prior to e820__memblock_setup(). Update e820
+	 * directly if EFI_MEMORY_SP is specified for an
+	 * EFI_CONVENTIONAL_MEMORY descriptor.
+	 */
+	for (i = 0; i < nr_fake_mem; i++) {
+		struct efi_mem_range *mem = &efi_fake_mems[i];
+		efi_memory_desc_t *md;
+		u64 m_start, m_end;
+
+		if ((mem->attribute & EFI_MEMORY_SP) == 0)
+			continue;
+
+		m_start = mem->range.start;
+		m_end = mem->range.end;
+		for_each_efi_memory_desc(md) {
+			u64 start, end;
+
+			if (md->type != EFI_CONVENTIONAL_MEMORY)
+				continue;
+
+			start = md->phys_addr;
+			end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) - 1;
+
+			if (m_start <= end && m_end >= start)
+				/* fake range overlaps descriptor */;
+			else
+				continue;
+
+			/*
+			 * Trim the boundary of the e820 update to the
+			 * descriptor in case the fake range overlaps
+			 * !EFI_CONVENTIONAL_MEMORY
+			 */
+			start = max(start, m_start);
+			end = min(end, m_end);
+
+			if (end <= start)
+				continue;
+			e820__range_update(start, end - start + 1, E820_TYPE_RAM,
+					E820_TYPE_SOFT_RESERVED);
+			e820__update_table(e820_table);
+		}
+	}
+}