[RESEND PATCH v2 0/4] x86/boot: a few cleanups

[RESEND PATCH v2 0/4] x86/boot: a few cleanups

[Mike Rapoport]

From: "Mike Rapoport (Microsoft)" <rppt@kernel.org>

Hi,

These patches cleanup x86::setup_arch() and e820.c.
No functional changes intended.

This is a resend of v2 rebased on v6.14-rc2

v2: update comments as suggested by Ingo
v1: https://lore.kernel.org/all/20241203112525.591496-1-rppt@kernel.org

Mike Rapoport (Microsoft) (4):
  x86/boot: move setting of memblock parameters to e820__memblock_setup()
  x86/boot: split kernel resources setup into a helper function
  x86/boot: split parsing of boot_params into a helper function
  x86/e820: drop E820_TYPE_RESERVED_KERN and related code

 arch/x86/include/asm/e820/api.h   |   1 -
 arch/x86/include/asm/e820/types.h |   9 --
 arch/x86/kernel/e820.c            |  95 ++++++++-------------
 arch/x86/kernel/setup.c           | 134 ++++++++++++++----------------
 arch/x86/kernel/tboot.c           |   3 +-
 arch/x86/mm/init_64.c             |   8 --
 6 files changed, 97 insertions(+), 153 deletions(-)


base-commit: a64dcfb451e254085a7daee5fe51bf22959d52d3
-- 
2.47.2

[RESEND PATCH v2 1/4] x86/boot: move setting of memblock parameters to e820__memblock_setup()

[Mike Rapoport]

From: "Mike Rapoport (Microsoft)" <rppt@kernel.org>

Changing memblock parameters, namely bottom_up and allocation upper
limit does not have any effect before memblock initialization in
e820__memblock_setup().

Move the calls to memblock_set_bottom_up() and memblock_set_current_limit()
to e820__memblock_setup() to group all the memblock initial setup and make
setup_arch() more readable.

No functional changes.

Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
---
 arch/x86/kernel/e820.c  | 30 ++++++++++++++++++++++++++++++
 arch/x86/kernel/setup.c | 25 -------------------------
 2 files changed, 30 insertions(+), 25 deletions(-)

diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 82b96ed9890a..8d8bd039d2ac 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -1302,6 +1302,36 @@ void __init e820__memblock_setup(void)
 	int i;
 	u64 end;
 
+#ifdef CONFIG_MEMORY_HOTPLUG
+	/*
+	 * Memory used by the kernel cannot be hot-removed because Linux
+	 * cannot migrate the kernel pages. When memory hotplug is
+	 * enabled, we should prevent memblock from allocating memory
+	 * for the kernel.
+	 *
+	 * ACPI SRAT records all hotpluggable memory ranges. But before
+	 * SRAT is parsed, we don't know about it.
+	 *
+	 * The kernel image is loaded into memory at very early time. We
+	 * cannot prevent this anyway. So on NUMA system, we set any
+	 * node the kernel resides in as un-hotpluggable.
+	 *
+	 * Since on modern servers, one node could have double-digit
+	 * gigabytes memory, we can assume the memory around the kernel
+	 * image is also un-hotpluggable. So before SRAT is parsed, just
+	 * allocate memory near the kernel image to try the best to keep
+	 * the kernel away from hotpluggable memory.
+	 */
+	if (movable_node_is_enabled())
+		memblock_set_bottom_up(true);
+#endif
+
+	/*
+	 * At this point only the first megabyte is mapped for sure, the
+	 * rest of the memory cannot be used for memblock resizing
+	 */
+	memblock_set_current_limit(ISA_END_ADDRESS);
+
 	/*
 	 * The bootstrap memblock region count maximum is 128 entries
 	 * (INIT_MEMBLOCK_REGIONS), but EFI might pass us more E820 entries
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index cebee310e200..4baadeab8a2e 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -866,30 +866,6 @@ void __init setup_arch(char **cmdline_p)
 	if (efi_enabled(EFI_BOOT))
 		efi_memblock_x86_reserve_range();
 
-#ifdef CONFIG_MEMORY_HOTPLUG
-	/*
-	 * Memory used by the kernel cannot be hot-removed because Linux
-	 * cannot migrate the kernel pages. When memory hotplug is
-	 * enabled, we should prevent memblock from allocating memory
-	 * for the kernel.
-	 *
-	 * ACPI SRAT records all hotpluggable memory ranges. But before
-	 * SRAT is parsed, we don't know about it.
-	 *
-	 * The kernel image is loaded into memory at very early time. We
-	 * cannot prevent this anyway. So on NUMA system, we set any
-	 * node the kernel resides in as un-hotpluggable.
-	 *
-	 * Since on modern servers, one node could have double-digit
-	 * gigabytes memory, we can assume the memory around the kernel
-	 * image is also un-hotpluggable. So before SRAT is parsed, just
-	 * allocate memory near the kernel image to try the best to keep
-	 * the kernel away from hotpluggable memory.
-	 */
-	if (movable_node_is_enabled())
-		memblock_set_bottom_up(true);
-#endif
-
 	x86_report_nx();
 
 	apic_setup_apic_calls();
@@ -990,7 +966,6 @@ void __init setup_arch(char **cmdline_p)
 
 	cleanup_highmap();
 
-	memblock_set_current_limit(ISA_END_ADDRESS);
 	e820__memblock_setup();
 
 	/*
-- 
2.47.2

[RESEND PATCH v2 2/4] x86/boot: split kernel resources setup into a helper function

[Mike Rapoport]

From: "Mike Rapoport (Microsoft)" <rppt@kernel.org>

Makes setup_arch a bit easier to comprehend.

No functional changes.

Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
---
 arch/x86/kernel/setup.c | 36 ++++++++++++++++++++++--------------
 1 file changed, 22 insertions(+), 14 deletions(-)

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 4baadeab8a2e..3d95946ab749 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -527,6 +527,23 @@ void __init reserve_standard_io_resources(void)
 
 }
 
+static void __init setup_kernel_resources(void)
+{
+	code_resource.start = __pa_symbol(_text);
+	code_resource.end = __pa_symbol(_etext)-1;
+	rodata_resource.start = __pa_symbol(__start_rodata);
+	rodata_resource.end = __pa_symbol(__end_rodata)-1;
+	data_resource.start = __pa_symbol(_sdata);
+	data_resource.end = __pa_symbol(_edata)-1;
+	bss_resource.start = __pa_symbol(__bss_start);
+	bss_resource.end = __pa_symbol(__bss_stop)-1;
+
+	insert_resource(&iomem_resource, &code_resource);
+	insert_resource(&iomem_resource, &rodata_resource);
+	insert_resource(&iomem_resource, &data_resource);
+	insert_resource(&iomem_resource, &bss_resource);
+}
+
 static bool __init snb_gfx_workaround_needed(void)
 {
 #ifdef CONFIG_PCI
@@ -845,15 +862,6 @@ void __init setup_arch(char **cmdline_p)
 		root_mountflags &= ~MS_RDONLY;
 	setup_initial_init_mm(_text, _etext, _edata, (void *)_brk_end);
 
-	code_resource.start = __pa_symbol(_text);
-	code_resource.end = __pa_symbol(_etext)-1;
-	rodata_resource.start = __pa_symbol(__start_rodata);
-	rodata_resource.end = __pa_symbol(__end_rodata)-1;
-	data_resource.start = __pa_symbol(_sdata);
-	data_resource.end = __pa_symbol(_edata)-1;
-	bss_resource.start = __pa_symbol(__bss_start);
-	bss_resource.end = __pa_symbol(__bss_stop)-1;
-
 	/*
 	 * x86_configure_nx() is called before parse_early_param() to detect
 	 * whether hardware doesn't support NX (so that the early EHCI debug
@@ -897,11 +905,11 @@ void __init setup_arch(char **cmdline_p)
 	tsc_early_init();
 	x86_init.resources.probe_roms();
 
-	/* after parse_early_param, so could debug it */
-	insert_resource(&iomem_resource, &code_resource);
-	insert_resource(&iomem_resource, &rodata_resource);
-	insert_resource(&iomem_resource, &data_resource);
-	insert_resource(&iomem_resource, &bss_resource);
+	/*
+	 * Add resources for kernel text and data to the iomem_resource.
+	 * Do it after parse_early_param, so it can be debugged.
+	 */
+	setup_kernel_resources();
 
 	e820_add_kernel_range();
 	trim_bios_range();
-- 
2.47.2

[RESEND PATCH v2 3/4] x86/boot: split parsing of boot_params into a helper function

[Mike Rapoport]

From: "Mike Rapoport (Microsoft)" <rppt@kernel.org>

Makes setup_arch a bit easier to comprehend.

No functional changes.

Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
---
 arch/x86/kernel/setup.c | 72 +++++++++++++++++++++++------------------
 1 file changed, 41 insertions(+), 31 deletions(-)

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 3d95946ab749..6fb9a851bc19 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -429,6 +429,46 @@ static void __init parse_setup_data(void)
 	}
 }
 
+/*
+ * Translate the fields of 'struct boot_param' into global variables
+ * representing these parameters.
+ */
+static void __init parse_boot_params(void)
+{
+	ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
+	screen_info = boot_params.screen_info;
+	edid_info = boot_params.edid_info;
+#ifdef CONFIG_X86_32
+	apm_info.bios = boot_params.apm_bios_info;
+	ist_info = boot_params.ist_info;
+#endif
+	saved_video_mode = boot_params.hdr.vid_mode;
+	bootloader_type = boot_params.hdr.type_of_loader;
+	if ((bootloader_type >> 4) == 0xe) {
+		bootloader_type &= 0xf;
+		bootloader_type |= (boot_params.hdr.ext_loader_type+0x10) << 4;
+	}
+	bootloader_version  = bootloader_type & 0xf;
+	bootloader_version |= boot_params.hdr.ext_loader_ver << 4;
+
+#ifdef CONFIG_BLK_DEV_RAM
+	rd_image_start = boot_params.hdr.ram_size & RAMDISK_IMAGE_START_MASK;
+#endif
+#ifdef CONFIG_EFI
+	if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
+		     EFI32_LOADER_SIGNATURE, 4)) {
+		set_bit(EFI_BOOT, &efi.flags);
+	} else if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
+		     EFI64_LOADER_SIGNATURE, 4)) {
+		set_bit(EFI_BOOT, &efi.flags);
+		set_bit(EFI_64BIT, &efi.flags);
+	}
+#endif
+
+	if (!boot_params.hdr.root_flags)
+		root_mountflags &= ~MS_RDONLY;
+}
+
 static void __init memblock_x86_reserve_range_setup_data(void)
 {
 	struct setup_indirect *indirect;
@@ -806,35 +846,7 @@ void __init setup_arch(char **cmdline_p)
 
 	setup_olpc_ofw_pgd();
 
-	ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
-	screen_info = boot_params.screen_info;
-	edid_info = boot_params.edid_info;
-#ifdef CONFIG_X86_32
-	apm_info.bios = boot_params.apm_bios_info;
-	ist_info = boot_params.ist_info;
-#endif
-	saved_video_mode = boot_params.hdr.vid_mode;
-	bootloader_type = boot_params.hdr.type_of_loader;
-	if ((bootloader_type >> 4) == 0xe) {
-		bootloader_type &= 0xf;
-		bootloader_type |= (boot_params.hdr.ext_loader_type+0x10) << 4;
-	}
-	bootloader_version  = bootloader_type & 0xf;
-	bootloader_version |= boot_params.hdr.ext_loader_ver << 4;
-
-#ifdef CONFIG_BLK_DEV_RAM
-	rd_image_start = boot_params.hdr.ram_size & RAMDISK_IMAGE_START_MASK;
-#endif
-#ifdef CONFIG_EFI
-	if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
-		     EFI32_LOADER_SIGNATURE, 4)) {
-		set_bit(EFI_BOOT, &efi.flags);
-	} else if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
-		     EFI64_LOADER_SIGNATURE, 4)) {
-		set_bit(EFI_BOOT, &efi.flags);
-		set_bit(EFI_64BIT, &efi.flags);
-	}
-#endif
+	parse_boot_params();
 
 	x86_init.oem.arch_setup();
 
@@ -858,8 +870,6 @@ void __init setup_arch(char **cmdline_p)
 
 	copy_edd();
 
-	if (!boot_params.hdr.root_flags)
-		root_mountflags &= ~MS_RDONLY;
 	setup_initial_init_mm(_text, _etext, _edata, (void *)_brk_end);
 
 	/*
-- 
2.47.2

[RESEND PATCH v2 4/4] x86/e820: drop E820_TYPE_RESERVED_KERN and related code

[Mike Rapoport]

From: "Mike Rapoport (Microsoft)" <rppt@kernel.org>

E820_TYPE_RESERVED_KERN is a relict from the ancient history that was used
to early reserve setup_data (see commit 28bb22379513 ("x86: move
reserve_setup_data to setup.c")).

Nowadays setup_data is anyway reserved in memblock and there is no point in
carrying E820_TYPE_RESERVED_KERN that behaves exactly like E820_TYPE_RAM
but only complicates the code.

A bonus for removing E820_TYPE_RESERVED_KERN is a small but measurable
speedup of 20 microseconds in init_mem_mappings() on a VM with 32GB or RAM.

Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
---
 arch/x86/include/asm/e820/api.h   |  1 -
 arch/x86/include/asm/e820/types.h |  9 -----
 arch/x86/kernel/e820.c            | 65 ++-----------------------------
 arch/x86/kernel/setup.c           |  1 -
 arch/x86/kernel/tboot.c           |  3 +-
 arch/x86/mm/init_64.c             |  8 ----
 6 files changed, 4 insertions(+), 83 deletions(-)

diff --git a/arch/x86/include/asm/e820/api.h b/arch/x86/include/asm/e820/api.h
index 2e74a7f0e935..c83645d5b2a8 100644
--- a/arch/x86/include/asm/e820/api.h
+++ b/arch/x86/include/asm/e820/api.h
@@ -29,7 +29,6 @@ extern unsigned long e820__end_of_low_ram_pfn(void);
 extern u64  e820__memblock_alloc_reserved(u64 size, u64 align);
 extern void e820__memblock_setup(void);
 
-extern void e820__reserve_setup_data(void);
 extern void e820__finish_early_params(void);
 extern void e820__reserve_resources(void);
 extern void e820__reserve_resources_late(void);
diff --git a/arch/x86/include/asm/e820/types.h b/arch/x86/include/asm/e820/types.h
index 314f75d886d0..80c4a7266629 100644
--- a/arch/x86/include/asm/e820/types.h
+++ b/arch/x86/include/asm/e820/types.h
@@ -35,15 +35,6 @@ enum e820_type {
 	 * marking it with the IORES_DESC_SOFT_RESERVED designation.
 	 */
 	E820_TYPE_SOFT_RESERVED	= 0xefffffff,
-
-	/*
-	 * Reserved RAM used by the kernel itself if
-	 * CONFIG_INTEL_TXT=y is enabled, memory of this type
-	 * will be included in the S3 integrity calculation
-	 * and so should not include any memory that the BIOS
-	 * might alter over the S3 transition:
-	 */
-	E820_TYPE_RESERVED_KERN	= 128,
 };
 
 /*
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 8d8bd039d2ac..9fb67abc32cc 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -187,8 +187,7 @@ void __init e820__range_add(u64 start, u64 size, enum e820_type type)
 static void __init e820_print_type(enum e820_type type)
 {
 	switch (type) {
-	case E820_TYPE_RAM:		/* Fall through: */
-	case E820_TYPE_RESERVED_KERN:	pr_cont("usable");			break;
+	case E820_TYPE_RAM:		pr_cont("usable");			break;
 	case E820_TYPE_RESERVED:	pr_cont("reserved");			break;
 	case E820_TYPE_SOFT_RESERVED:	pr_cont("soft reserved");		break;
 	case E820_TYPE_ACPI:		pr_cont("ACPI data");			break;
@@ -764,7 +763,7 @@ void __init e820__register_nosave_regions(unsigned long limit_pfn)
 
 		pfn = PFN_DOWN(entry->addr + entry->size);
 
-		if (entry->type != E820_TYPE_RAM && entry->type != E820_TYPE_RESERVED_KERN)
+		if (entry->type != E820_TYPE_RAM)
 			register_nosave_region(PFN_UP(entry->addr), pfn);
 
 		if (pfn >= limit_pfn)
@@ -990,60 +989,6 @@ static int __init parse_memmap_opt(char *str)
 }
 early_param("memmap", parse_memmap_opt);
 
-/*
- * Reserve all entries from the bootloader's extensible data nodes list,
- * because if present we are going to use it later on to fetch e820
- * entries from it:
- */
-void __init e820__reserve_setup_data(void)
-{
-	struct setup_indirect *indirect;
-	struct setup_data *data;
-	u64 pa_data, pa_next;
-	u32 len;
-
-	pa_data = boot_params.hdr.setup_data;
-	if (!pa_data)
-		return;
-
-	while (pa_data) {
-		data = early_memremap(pa_data, sizeof(*data));
-		if (!data) {
-			pr_warn("e820: failed to memremap setup_data entry\n");
-			return;
-		}
-
-		len = sizeof(*data);
-		pa_next = data->next;
-
-		e820__range_update(pa_data, sizeof(*data)+data->len, E820_TYPE_RAM, E820_TYPE_RESERVED_KERN);
-
-		if (data->type == SETUP_INDIRECT) {
-			len += data->len;
-			early_memunmap(data, sizeof(*data));
-			data = early_memremap(pa_data, len);
-			if (!data) {
-				pr_warn("e820: failed to memremap indirect setup_data\n");
-				return;
-			}
-
-			indirect = (struct setup_indirect *)data->data;
-
-			if (indirect->type != SETUP_INDIRECT)
-				e820__range_update(indirect->addr, indirect->len,
-						   E820_TYPE_RAM, E820_TYPE_RESERVED_KERN);
-		}
-
-		pa_data = pa_next;
-		early_memunmap(data, len);
-	}
-
-	e820__update_table(e820_table);
-
-	pr_info("extended physical RAM map:\n");
-	e820__print_table("reserve setup_data");
-}
-
 /*
  * Called after parse_early_param(), after early parameters (such as mem=)
  * have been processed, in which case we already have an E820 table filled in
@@ -1063,7 +1008,6 @@ void __init e820__finish_early_params(void)
 static const char *__init e820_type_to_string(struct e820_entry *entry)
 {
 	switch (entry->type) {
-	case E820_TYPE_RESERVED_KERN:	/* Fall-through: */
 	case E820_TYPE_RAM:		return "System RAM";
 	case E820_TYPE_ACPI:		return "ACPI Tables";
 	case E820_TYPE_NVS:		return "ACPI Non-volatile Storage";
@@ -1079,7 +1023,6 @@ static const char *__init e820_type_to_string(struct e820_entry *entry)
 static unsigned long __init e820_type_to_iomem_type(struct e820_entry *entry)
 {
 	switch (entry->type) {
-	case E820_TYPE_RESERVED_KERN:	/* Fall-through: */
 	case E820_TYPE_RAM:		return IORESOURCE_SYSTEM_RAM;
 	case E820_TYPE_ACPI:		/* Fall-through: */
 	case E820_TYPE_NVS:		/* Fall-through: */
@@ -1101,7 +1044,6 @@ static unsigned long __init e820_type_to_iores_desc(struct e820_entry *entry)
 	case E820_TYPE_PRAM:		return IORES_DESC_PERSISTENT_MEMORY_LEGACY;
 	case E820_TYPE_RESERVED:	return IORES_DESC_RESERVED;
 	case E820_TYPE_SOFT_RESERVED:	return IORES_DESC_SOFT_RESERVED;
-	case E820_TYPE_RESERVED_KERN:	/* Fall-through: */
 	case E820_TYPE_RAM:		/* Fall-through: */
 	case E820_TYPE_UNUSABLE:	/* Fall-through: */
 	default:			return IORES_DESC_NONE;
@@ -1124,7 +1066,6 @@ static bool __init do_mark_busy(enum e820_type type, struct resource *res)
 	case E820_TYPE_PRAM:
 	case E820_TYPE_PMEM:
 		return false;
-	case E820_TYPE_RESERVED_KERN:
 	case E820_TYPE_RAM:
 	case E820_TYPE_ACPI:
 	case E820_TYPE_NVS:
@@ -1353,7 +1294,7 @@ void __init e820__memblock_setup(void)
 		if (entry->type == E820_TYPE_SOFT_RESERVED)
 			memblock_reserve(entry->addr, entry->size);
 
-		if (entry->type != E820_TYPE_RAM && entry->type != E820_TYPE_RESERVED_KERN)
+		if (entry->type != E820_TYPE_RAM)
 			continue;
 
 		memblock_add(entry->addr, entry->size);
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 6fb9a851bc19..46d92d04b6da 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -895,7 +895,6 @@ void __init setup_arch(char **cmdline_p)
 		setup_clear_cpu_cap(X86_FEATURE_APIC);
 	}
 
-	e820__reserve_setup_data();
 	e820__finish_early_params();
 
 	if (efi_enabled(EFI_BOOT))
diff --git a/arch/x86/kernel/tboot.c b/arch/x86/kernel/tboot.c
index 4c1bcb6053fc..46b8f1f16676 100644
--- a/arch/x86/kernel/tboot.c
+++ b/arch/x86/kernel/tboot.c
@@ -200,8 +200,7 @@ static int tboot_setup_sleep(void)
 	tboot->num_mac_regions = 0;
 
 	for (i = 0; i < e820_table->nr_entries; i++) {
-		if ((e820_table->entries[i].type != E820_TYPE_RAM)
-		 && (e820_table->entries[i].type != E820_TYPE_RESERVED_KERN))
+		if (e820_table->entries[i].type != E820_TYPE_RAM)
 			continue;
 
 		add_mac_region(e820_table->entries[i].addr, e820_table->entries[i].size);
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 01ea7c6df303..519aa53114fa 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -468,8 +468,6 @@ phys_pte_init(pte_t *pte_page, unsigned long paddr, unsigned long paddr_end,
 			if (!after_bootmem &&
 			    !e820__mapped_any(paddr & PAGE_MASK, paddr_next,
 					     E820_TYPE_RAM) &&
-			    !e820__mapped_any(paddr & PAGE_MASK, paddr_next,
-					     E820_TYPE_RESERVED_KERN) &&
 			    !e820__mapped_any(paddr & PAGE_MASK, paddr_next,
 					     E820_TYPE_ACPI))
 				set_pte_init(pte, __pte(0), init);
@@ -525,8 +523,6 @@ phys_pmd_init(pmd_t *pmd_page, unsigned long paddr, unsigned long paddr_end,
 			if (!after_bootmem &&
 			    !e820__mapped_any(paddr & PMD_MASK, paddr_next,
 					     E820_TYPE_RAM) &&
-			    !e820__mapped_any(paddr & PMD_MASK, paddr_next,
-					     E820_TYPE_RESERVED_KERN) &&
 			    !e820__mapped_any(paddr & PMD_MASK, paddr_next,
 					     E820_TYPE_ACPI))
 				set_pmd_init(pmd, __pmd(0), init);
@@ -614,8 +610,6 @@ phys_pud_init(pud_t *pud_page, unsigned long paddr, unsigned long paddr_end,
 			if (!after_bootmem &&
 			    !e820__mapped_any(paddr & PUD_MASK, paddr_next,
 					     E820_TYPE_RAM) &&
-			    !e820__mapped_any(paddr & PUD_MASK, paddr_next,
-					     E820_TYPE_RESERVED_KERN) &&
 			    !e820__mapped_any(paddr & PUD_MASK, paddr_next,
 					     E820_TYPE_ACPI))
 				set_pud_init(pud, __pud(0), init);
@@ -703,8 +697,6 @@ phys_p4d_init(p4d_t *p4d_page, unsigned long paddr, unsigned long paddr_end,
 			if (!after_bootmem &&
 			    !e820__mapped_any(paddr & P4D_MASK, paddr_next,
 					     E820_TYPE_RAM) &&
-			    !e820__mapped_any(paddr & P4D_MASK, paddr_next,
-					     E820_TYPE_RESERVED_KERN) &&
 			    !e820__mapped_any(paddr & P4D_MASK, paddr_next,
 					     E820_TYPE_ACPI))
 				set_p4d_init(p4d, __p4d(0), init);
-- 
2.47.2

[tip: x86/boot] x86/e820: Drop obsolete E820_TYPE_RESERVED_KERN and related code

[tip-bot2 for Mike Rapoport (Microsoft)]

The following commit has been merged into the x86/boot branch of tip:

Commit-ID:     efe659ac014647eb048d62475e55cce42d8951d7
Gitweb:        https://git.kernel.org/tip/efe659ac014647eb048d62475e55cce42d8951d7
Author:        Mike Rapoport (Microsoft) <rppt@kernel.org>
AuthorDate:    Fri, 14 Feb 2025 11:06:51 +02:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Fri, 21 Feb 2025 16:05:00 +01:00

x86/e820: Drop obsolete E820_TYPE_RESERVED_KERN and related code

E820_TYPE_RESERVED_KERN is a relict from the ancient history that was used
to early reserve setup_data, see:

  28bb22379513 ("x86: move reserve_setup_data to setup.c")

Nowadays setup_data is anyway reserved in memblock and there is no point in
carrying E820_TYPE_RESERVED_KERN that behaves exactly like E820_TYPE_RAM
but only complicates the code.

A bonus for removing E820_TYPE_RESERVED_KERN is a small but measurable
speedup of 20 microseconds in init_mem_mappings() on a VM with 32GB or RAM.

Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20250214090651.3331663-5-rppt@kernel.org
---
 arch/x86/include/asm/e820/api.h   |  1 +-
 arch/x86/include/asm/e820/types.h |  9 +----
 arch/x86/kernel/e820.c            | 65 +-----------------------------
 arch/x86/kernel/setup.c           |  1 +-
 arch/x86/kernel/tboot.c           |  3 +-
 arch/x86/mm/init_64.c             |  8 +----
 6 files changed, 4 insertions(+), 83 deletions(-)

diff --git a/arch/x86/include/asm/e820/api.h b/arch/x86/include/asm/e820/api.h
index 2e74a7f..c83645d 100644
--- a/arch/x86/include/asm/e820/api.h
+++ b/arch/x86/include/asm/e820/api.h
@@ -29,7 +29,6 @@ extern unsigned long e820__end_of_low_ram_pfn(void);
 extern u64  e820__memblock_alloc_reserved(u64 size, u64 align);
 extern void e820__memblock_setup(void);
 
-extern void e820__reserve_setup_data(void);
 extern void e820__finish_early_params(void);
 extern void e820__reserve_resources(void);
 extern void e820__reserve_resources_late(void);
diff --git a/arch/x86/include/asm/e820/types.h b/arch/x86/include/asm/e820/types.h
index 314f75d..80c4a72 100644
--- a/arch/x86/include/asm/e820/types.h
+++ b/arch/x86/include/asm/e820/types.h
@@ -35,15 +35,6 @@ enum e820_type {
 	 * marking it with the IORES_DESC_SOFT_RESERVED designation.
 	 */
 	E820_TYPE_SOFT_RESERVED	= 0xefffffff,
-
-	/*
-	 * Reserved RAM used by the kernel itself if
-	 * CONFIG_INTEL_TXT=y is enabled, memory of this type
-	 * will be included in the S3 integrity calculation
-	 * and so should not include any memory that the BIOS
-	 * might alter over the S3 transition:
-	 */
-	E820_TYPE_RESERVED_KERN	= 128,
 };
 
 /*
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 8d8bd03..9fb67ab 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -187,8 +187,7 @@ void __init e820__range_add(u64 start, u64 size, enum e820_type type)
 static void __init e820_print_type(enum e820_type type)
 {
 	switch (type) {
-	case E820_TYPE_RAM:		/* Fall through: */
-	case E820_TYPE_RESERVED_KERN:	pr_cont("usable");			break;
+	case E820_TYPE_RAM:		pr_cont("usable");			break;
 	case E820_TYPE_RESERVED:	pr_cont("reserved");			break;
 	case E820_TYPE_SOFT_RESERVED:	pr_cont("soft reserved");		break;
 	case E820_TYPE_ACPI:		pr_cont("ACPI data");			break;
@@ -764,7 +763,7 @@ void __init e820__register_nosave_regions(unsigned long limit_pfn)
 
 		pfn = PFN_DOWN(entry->addr + entry->size);
 
-		if (entry->type != E820_TYPE_RAM && entry->type != E820_TYPE_RESERVED_KERN)
+		if (entry->type != E820_TYPE_RAM)
 			register_nosave_region(PFN_UP(entry->addr), pfn);
 
 		if (pfn >= limit_pfn)
@@ -991,60 +990,6 @@ static int __init parse_memmap_opt(char *str)
 early_param("memmap", parse_memmap_opt);
 
 /*
- * Reserve all entries from the bootloader's extensible data nodes list,
- * because if present we are going to use it later on to fetch e820
- * entries from it:
- */
-void __init e820__reserve_setup_data(void)
-{
-	struct setup_indirect *indirect;
-	struct setup_data *data;
-	u64 pa_data, pa_next;
-	u32 len;
-
-	pa_data = boot_params.hdr.setup_data;
-	if (!pa_data)
-		return;
-
-	while (pa_data) {
-		data = early_memremap(pa_data, sizeof(*data));
-		if (!data) {
-			pr_warn("e820: failed to memremap setup_data entry\n");
-			return;
-		}
-
-		len = sizeof(*data);
-		pa_next = data->next;
-
-		e820__range_update(pa_data, sizeof(*data)+data->len, E820_TYPE_RAM, E820_TYPE_RESERVED_KERN);
-
-		if (data->type == SETUP_INDIRECT) {
-			len += data->len;
-			early_memunmap(data, sizeof(*data));
-			data = early_memremap(pa_data, len);
-			if (!data) {
-				pr_warn("e820: failed to memremap indirect setup_data\n");
-				return;
-			}
-
-			indirect = (struct setup_indirect *)data->data;
-
-			if (indirect->type != SETUP_INDIRECT)
-				e820__range_update(indirect->addr, indirect->len,
-						   E820_TYPE_RAM, E820_TYPE_RESERVED_KERN);
-		}
-
-		pa_data = pa_next;
-		early_memunmap(data, len);
-	}
-
-	e820__update_table(e820_table);
-
-	pr_info("extended physical RAM map:\n");
-	e820__print_table("reserve setup_data");
-}
-
-/*
  * Called after parse_early_param(), after early parameters (such as mem=)
  * have been processed, in which case we already have an E820 table filled in
  * via the parameter callback function(s), but it's not sorted and printed yet:
@@ -1063,7 +1008,6 @@ void __init e820__finish_early_params(void)
 static const char *__init e820_type_to_string(struct e820_entry *entry)
 {
 	switch (entry->type) {
-	case E820_TYPE_RESERVED_KERN:	/* Fall-through: */
 	case E820_TYPE_RAM:		return "System RAM";
 	case E820_TYPE_ACPI:		return "ACPI Tables";
 	case E820_TYPE_NVS:		return "ACPI Non-volatile Storage";
@@ -1079,7 +1023,6 @@ static const char *__init e820_type_to_string(struct e820_entry *entry)
 static unsigned long __init e820_type_to_iomem_type(struct e820_entry *entry)
 {
 	switch (entry->type) {
-	case E820_TYPE_RESERVED_KERN:	/* Fall-through: */
 	case E820_TYPE_RAM:		return IORESOURCE_SYSTEM_RAM;
 	case E820_TYPE_ACPI:		/* Fall-through: */
 	case E820_TYPE_NVS:		/* Fall-through: */
@@ -1101,7 +1044,6 @@ static unsigned long __init e820_type_to_iores_desc(struct e820_entry *entry)
 	case E820_TYPE_PRAM:		return IORES_DESC_PERSISTENT_MEMORY_LEGACY;
 	case E820_TYPE_RESERVED:	return IORES_DESC_RESERVED;
 	case E820_TYPE_SOFT_RESERVED:	return IORES_DESC_SOFT_RESERVED;
-	case E820_TYPE_RESERVED_KERN:	/* Fall-through: */
 	case E820_TYPE_RAM:		/* Fall-through: */
 	case E820_TYPE_UNUSABLE:	/* Fall-through: */
 	default:			return IORES_DESC_NONE;
@@ -1124,7 +1066,6 @@ static bool __init do_mark_busy(enum e820_type type, struct resource *res)
 	case E820_TYPE_PRAM:
 	case E820_TYPE_PMEM:
 		return false;
-	case E820_TYPE_RESERVED_KERN:
 	case E820_TYPE_RAM:
 	case E820_TYPE_ACPI:
 	case E820_TYPE_NVS:
@@ -1353,7 +1294,7 @@ void __init e820__memblock_setup(void)
 		if (entry->type == E820_TYPE_SOFT_RESERVED)
 			memblock_reserve(entry->addr, entry->size);
 
-		if (entry->type != E820_TYPE_RAM && entry->type != E820_TYPE_RESERVED_KERN)
+		if (entry->type != E820_TYPE_RAM)
 			continue;
 
 		memblock_add(entry->addr, entry->size);
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 6fb9a85..46d92d0 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -895,7 +895,6 @@ void __init setup_arch(char **cmdline_p)
 		setup_clear_cpu_cap(X86_FEATURE_APIC);
 	}
 
-	e820__reserve_setup_data();
 	e820__finish_early_params();
 
 	if (efi_enabled(EFI_BOOT))
diff --git a/arch/x86/kernel/tboot.c b/arch/x86/kernel/tboot.c
index 4c1bcb6..46b8f1f 100644
--- a/arch/x86/kernel/tboot.c
+++ b/arch/x86/kernel/tboot.c
@@ -200,8 +200,7 @@ static int tboot_setup_sleep(void)
 	tboot->num_mac_regions = 0;
 
 	for (i = 0; i < e820_table->nr_entries; i++) {
-		if ((e820_table->entries[i].type != E820_TYPE_RAM)
-		 && (e820_table->entries[i].type != E820_TYPE_RESERVED_KERN))
+		if (e820_table->entries[i].type != E820_TYPE_RAM)
 			continue;
 
 		add_mac_region(e820_table->entries[i].addr, e820_table->entries[i].size);
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 01ea7c6..519aa53 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -469,8 +469,6 @@ phys_pte_init(pte_t *pte_page, unsigned long paddr, unsigned long paddr_end,
 			    !e820__mapped_any(paddr & PAGE_MASK, paddr_next,
 					     E820_TYPE_RAM) &&
 			    !e820__mapped_any(paddr & PAGE_MASK, paddr_next,
-					     E820_TYPE_RESERVED_KERN) &&
-			    !e820__mapped_any(paddr & PAGE_MASK, paddr_next,
 					     E820_TYPE_ACPI))
 				set_pte_init(pte, __pte(0), init);
 			continue;
@@ -526,8 +524,6 @@ phys_pmd_init(pmd_t *pmd_page, unsigned long paddr, unsigned long paddr_end,
 			    !e820__mapped_any(paddr & PMD_MASK, paddr_next,
 					     E820_TYPE_RAM) &&
 			    !e820__mapped_any(paddr & PMD_MASK, paddr_next,
-					     E820_TYPE_RESERVED_KERN) &&
-			    !e820__mapped_any(paddr & PMD_MASK, paddr_next,
 					     E820_TYPE_ACPI))
 				set_pmd_init(pmd, __pmd(0), init);
 			continue;
@@ -615,8 +611,6 @@ phys_pud_init(pud_t *pud_page, unsigned long paddr, unsigned long paddr_end,
 			    !e820__mapped_any(paddr & PUD_MASK, paddr_next,
 					     E820_TYPE_RAM) &&
 			    !e820__mapped_any(paddr & PUD_MASK, paddr_next,
-					     E820_TYPE_RESERVED_KERN) &&
-			    !e820__mapped_any(paddr & PUD_MASK, paddr_next,
 					     E820_TYPE_ACPI))
 				set_pud_init(pud, __pud(0), init);
 			continue;
@@ -704,8 +698,6 @@ phys_p4d_init(p4d_t *p4d_page, unsigned long paddr, unsigned long paddr_end,
 			    !e820__mapped_any(paddr & P4D_MASK, paddr_next,
 					     E820_TYPE_RAM) &&
 			    !e820__mapped_any(paddr & P4D_MASK, paddr_next,
-					     E820_TYPE_RESERVED_KERN) &&
-			    !e820__mapped_any(paddr & P4D_MASK, paddr_next,
 					     E820_TYPE_ACPI))
 				set_p4d_init(p4d, __p4d(0), init);
 			continue;

[tip: x86/boot] x86/boot: Split kernel resources setup into the setup_kernel_resources() helper function

[tip-bot2 for Mike Rapoport (Microsoft)]

The following commit has been merged into the x86/boot branch of tip:

Commit-ID:     297fb82ebaad50e1c40aab7ac61897bf372842ba
Gitweb:        https://git.kernel.org/tip/297fb82ebaad50e1c40aab7ac61897bf372842ba
Author:        Mike Rapoport (Microsoft) <rppt@kernel.org>
AuthorDate:    Fri, 14 Feb 2025 11:06:49 +02:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Fri, 21 Feb 2025 16:05:00 +01:00

x86/boot: Split kernel resources setup into the setup_kernel_resources() helper function

Makes setup_arch() a bit easier to comprehend.

No functional changes.

Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20250214090651.3331663-3-rppt@kernel.org
---
 arch/x86/kernel/setup.c | 36 ++++++++++++++++++++++--------------
 1 file changed, 22 insertions(+), 14 deletions(-)

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 4baadea..3d95946 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -527,6 +527,23 @@ void __init reserve_standard_io_resources(void)
 
 }
 
+static void __init setup_kernel_resources(void)
+{
+	code_resource.start = __pa_symbol(_text);
+	code_resource.end = __pa_symbol(_etext)-1;
+	rodata_resource.start = __pa_symbol(__start_rodata);
+	rodata_resource.end = __pa_symbol(__end_rodata)-1;
+	data_resource.start = __pa_symbol(_sdata);
+	data_resource.end = __pa_symbol(_edata)-1;
+	bss_resource.start = __pa_symbol(__bss_start);
+	bss_resource.end = __pa_symbol(__bss_stop)-1;
+
+	insert_resource(&iomem_resource, &code_resource);
+	insert_resource(&iomem_resource, &rodata_resource);
+	insert_resource(&iomem_resource, &data_resource);
+	insert_resource(&iomem_resource, &bss_resource);
+}
+
 static bool __init snb_gfx_workaround_needed(void)
 {
 #ifdef CONFIG_PCI
@@ -845,15 +862,6 @@ void __init setup_arch(char **cmdline_p)
 		root_mountflags &= ~MS_RDONLY;
 	setup_initial_init_mm(_text, _etext, _edata, (void *)_brk_end);
 
-	code_resource.start = __pa_symbol(_text);
-	code_resource.end = __pa_symbol(_etext)-1;
-	rodata_resource.start = __pa_symbol(__start_rodata);
-	rodata_resource.end = __pa_symbol(__end_rodata)-1;
-	data_resource.start = __pa_symbol(_sdata);
-	data_resource.end = __pa_symbol(_edata)-1;
-	bss_resource.start = __pa_symbol(__bss_start);
-	bss_resource.end = __pa_symbol(__bss_stop)-1;
-
 	/*
 	 * x86_configure_nx() is called before parse_early_param() to detect
 	 * whether hardware doesn't support NX (so that the early EHCI debug
@@ -897,11 +905,11 @@ void __init setup_arch(char **cmdline_p)
 	tsc_early_init();
 	x86_init.resources.probe_roms();
 
-	/* after parse_early_param, so could debug it */
-	insert_resource(&iomem_resource, &code_resource);
-	insert_resource(&iomem_resource, &rodata_resource);
-	insert_resource(&iomem_resource, &data_resource);
-	insert_resource(&iomem_resource, &bss_resource);
+	/*
+	 * Add resources for kernel text and data to the iomem_resource.
+	 * Do it after parse_early_param, so it can be debugged.
+	 */
+	setup_kernel_resources();
 
 	e820_add_kernel_range();
 	trim_bios_range();

[tip: x86/boot] x86/boot: Split parsing of boot_params into the parse_boot_params() helper function

[tip-bot2 for Mike Rapoport (Microsoft)]

The following commit has been merged into the x86/boot branch of tip:

Commit-ID:     d45dd0a9b27ee8be7792ae186bc33ed700144224
Gitweb:        https://git.kernel.org/tip/d45dd0a9b27ee8be7792ae186bc33ed700144224
Author:        Mike Rapoport (Microsoft) <rppt@kernel.org>
AuthorDate:    Fri, 14 Feb 2025 11:06:50 +02:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Fri, 21 Feb 2025 16:05:00 +01:00

x86/boot: Split parsing of boot_params into the parse_boot_params() helper function

Makes setup_arch() a bit easier to comprehend.

No functional changes.

Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20250214090651.3331663-4-rppt@kernel.org
---
 arch/x86/kernel/setup.c | 72 ++++++++++++++++++++++------------------
 1 file changed, 41 insertions(+), 31 deletions(-)

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 3d95946..6fb9a85 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -429,6 +429,46 @@ static void __init parse_setup_data(void)
 	}
 }
 
+/*
+ * Translate the fields of 'struct boot_param' into global variables
+ * representing these parameters.
+ */
+static void __init parse_boot_params(void)
+{
+	ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
+	screen_info = boot_params.screen_info;
+	edid_info = boot_params.edid_info;
+#ifdef CONFIG_X86_32
+	apm_info.bios = boot_params.apm_bios_info;
+	ist_info = boot_params.ist_info;
+#endif
+	saved_video_mode = boot_params.hdr.vid_mode;
+	bootloader_type = boot_params.hdr.type_of_loader;
+	if ((bootloader_type >> 4) == 0xe) {
+		bootloader_type &= 0xf;
+		bootloader_type |= (boot_params.hdr.ext_loader_type+0x10) << 4;
+	}
+	bootloader_version  = bootloader_type & 0xf;
+	bootloader_version |= boot_params.hdr.ext_loader_ver << 4;
+
+#ifdef CONFIG_BLK_DEV_RAM
+	rd_image_start = boot_params.hdr.ram_size & RAMDISK_IMAGE_START_MASK;
+#endif
+#ifdef CONFIG_EFI
+	if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
+		     EFI32_LOADER_SIGNATURE, 4)) {
+		set_bit(EFI_BOOT, &efi.flags);
+	} else if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
+		     EFI64_LOADER_SIGNATURE, 4)) {
+		set_bit(EFI_BOOT, &efi.flags);
+		set_bit(EFI_64BIT, &efi.flags);
+	}
+#endif
+
+	if (!boot_params.hdr.root_flags)
+		root_mountflags &= ~MS_RDONLY;
+}
+
 static void __init memblock_x86_reserve_range_setup_data(void)
 {
 	struct setup_indirect *indirect;
@@ -806,35 +846,7 @@ void __init setup_arch(char **cmdline_p)
 
 	setup_olpc_ofw_pgd();
 
-	ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
-	screen_info = boot_params.screen_info;
-	edid_info = boot_params.edid_info;
-#ifdef CONFIG_X86_32
-	apm_info.bios = boot_params.apm_bios_info;
-	ist_info = boot_params.ist_info;
-#endif
-	saved_video_mode = boot_params.hdr.vid_mode;
-	bootloader_type = boot_params.hdr.type_of_loader;
-	if ((bootloader_type >> 4) == 0xe) {
-		bootloader_type &= 0xf;
-		bootloader_type |= (boot_params.hdr.ext_loader_type+0x10) << 4;
-	}
-	bootloader_version  = bootloader_type & 0xf;
-	bootloader_version |= boot_params.hdr.ext_loader_ver << 4;
-
-#ifdef CONFIG_BLK_DEV_RAM
-	rd_image_start = boot_params.hdr.ram_size & RAMDISK_IMAGE_START_MASK;
-#endif
-#ifdef CONFIG_EFI
-	if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
-		     EFI32_LOADER_SIGNATURE, 4)) {
-		set_bit(EFI_BOOT, &efi.flags);
-	} else if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
-		     EFI64_LOADER_SIGNATURE, 4)) {
-		set_bit(EFI_BOOT, &efi.flags);
-		set_bit(EFI_64BIT, &efi.flags);
-	}
-#endif
+	parse_boot_params();
 
 	x86_init.oem.arch_setup();
 
@@ -858,8 +870,6 @@ void __init setup_arch(char **cmdline_p)
 
 	copy_edd();
 
-	if (!boot_params.hdr.root_flags)
-		root_mountflags &= ~MS_RDONLY;
 	setup_initial_init_mm(_text, _etext, _edata, (void *)_brk_end);
 
 	/*

[tip: x86/boot] x86/boot: Move setting of memblock parameters to e820__memblock_setup()

[tip-bot2 for Mike Rapoport (Microsoft)]

The following commit has been merged into the x86/boot branch of tip:

Commit-ID:     2d6bff31399b2443aba230cd20f120acb2d4eeb1
Gitweb:        https://git.kernel.org/tip/2d6bff31399b2443aba230cd20f120acb2d4eeb1
Author:        Mike Rapoport (Microsoft) <rppt@kernel.org>
AuthorDate:    Fri, 14 Feb 2025 11:06:48 +02:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Fri, 21 Feb 2025 16:05:00 +01:00

x86/boot: Move setting of memblock parameters to e820__memblock_setup()

Changing memblock parameters, namely bottom_up and allocation upper
limit does not have any effect before memblock initialization in
e820__memblock_setup().

Move the calls to memblock_set_bottom_up() and memblock_set_current_limit()
to e820__memblock_setup() to group all the memblock initial setup and make
setup_arch() more readable.

No functional changes.

Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20250214090651.3331663-2-rppt@kernel.org
---
 arch/x86/kernel/e820.c  | 30 ++++++++++++++++++++++++++++++
 arch/x86/kernel/setup.c | 25 -------------------------
 2 files changed, 30 insertions(+), 25 deletions(-)

diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 82b96ed..8d8bd03 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -1302,6 +1302,36 @@ void __init e820__memblock_setup(void)
 	int i;
 	u64 end;
 
+#ifdef CONFIG_MEMORY_HOTPLUG
+	/*
+	 * Memory used by the kernel cannot be hot-removed because Linux
+	 * cannot migrate the kernel pages. When memory hotplug is
+	 * enabled, we should prevent memblock from allocating memory
+	 * for the kernel.
+	 *
+	 * ACPI SRAT records all hotpluggable memory ranges. But before
+	 * SRAT is parsed, we don't know about it.
+	 *
+	 * The kernel image is loaded into memory at very early time. We
+	 * cannot prevent this anyway. So on NUMA system, we set any
+	 * node the kernel resides in as un-hotpluggable.
+	 *
+	 * Since on modern servers, one node could have double-digit
+	 * gigabytes memory, we can assume the memory around the kernel
+	 * image is also un-hotpluggable. So before SRAT is parsed, just
+	 * allocate memory near the kernel image to try the best to keep
+	 * the kernel away from hotpluggable memory.
+	 */
+	if (movable_node_is_enabled())
+		memblock_set_bottom_up(true);
+#endif
+
+	/*
+	 * At this point only the first megabyte is mapped for sure, the
+	 * rest of the memory cannot be used for memblock resizing
+	 */
+	memblock_set_current_limit(ISA_END_ADDRESS);
+
 	/*
 	 * The bootstrap memblock region count maximum is 128 entries
 	 * (INIT_MEMBLOCK_REGIONS), but EFI might pass us more E820 entries
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index cebee31..4baadea 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -866,30 +866,6 @@ void __init setup_arch(char **cmdline_p)
 	if (efi_enabled(EFI_BOOT))
 		efi_memblock_x86_reserve_range();
 
-#ifdef CONFIG_MEMORY_HOTPLUG
-	/*
-	 * Memory used by the kernel cannot be hot-removed because Linux
-	 * cannot migrate the kernel pages. When memory hotplug is
-	 * enabled, we should prevent memblock from allocating memory
-	 * for the kernel.
-	 *
-	 * ACPI SRAT records all hotpluggable memory ranges. But before
-	 * SRAT is parsed, we don't know about it.
-	 *
-	 * The kernel image is loaded into memory at very early time. We
-	 * cannot prevent this anyway. So on NUMA system, we set any
-	 * node the kernel resides in as un-hotpluggable.
-	 *
-	 * Since on modern servers, one node could have double-digit
-	 * gigabytes memory, we can assume the memory around the kernel
-	 * image is also un-hotpluggable. So before SRAT is parsed, just
-	 * allocate memory near the kernel image to try the best to keep
-	 * the kernel away from hotpluggable memory.
-	 */
-	if (movable_node_is_enabled())
-		memblock_set_bottom_up(true);
-#endif
-
 	x86_report_nx();
 
 	apic_setup_apic_calls();
@@ -990,7 +966,6 @@ void __init setup_arch(char **cmdline_p)
 
 	cleanup_highmap();
 
-	memblock_set_current_limit(ISA_END_ADDRESS);
 	e820__memblock_setup();
 
 	/*