Merge tag 'kvm-riscv-fixes-6.6-1' of https://github.com/kvm-riscv/linux into HEAD

KVM/riscv fixes for 6.6, take #1

- Fix KVM_GET_REG_LIST API for ISA_EXT registers
- Fix reading ISA_EXT register of a missing extension
- Fix ISA_EXT register handling in get-reg-list test
- Fix filtering of AIA registers in get-reg-list test

29 files changed, 274 insertions, 100 deletions

diff --git a/arch/parisc/include/asm/cache.h b/arch/parisc/include/asm/cache.h
index e23d06b51a20..2a60d7a72f1f 100644
--- a/arch/parisc/include/asm/cache.h
+++ b/arch/parisc/include/asm/cache.h
@@ -37,6 +37,7 @@ extern int split_tlb;
 extern int dcache_stride;
 extern int icache_stride;
 extern struct pdc_cache_info cache_info;
+extern struct pdc_btlb_info btlb_info;
 void parisc_setup_cache_timing(void);
 
 #define pdtlb(sr, addr)	asm volatile("pdtlb 0(%%sr%0,%1)" \
diff --git a/arch/parisc/include/asm/mckinley.h b/arch/parisc/include/asm/mckinley.h
deleted file mode 100644
index 1314390b9034..000000000000
--- a/arch/parisc/include/asm/mckinley.h
+++ /dev/null
@@ -1,8 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef ASM_PARISC_MCKINLEY_H
-#define ASM_PARISC_MCKINLEY_H
-
-/* declared in arch/parisc/kernel/setup.c */
-extern struct proc_dir_entry * proc_mckinley_root;
-
-#endif /*ASM_PARISC_MCKINLEY_H*/
diff --git a/arch/parisc/include/asm/pdc.h b/arch/parisc/include/asm/pdc.h
index 269b9a159f01..5d2d9737e579 100644
--- a/arch/parisc/include/asm/pdc.h
+++ b/arch/parisc/include/asm/pdc.h
@@ -44,10 +44,11 @@ int pdc_model_capabilities(unsigned long *capabilities);
 int pdc_model_platform_info(char *orig_prod_num, char *current_prod_num, char *serial_no);
 int pdc_cache_info(struct pdc_cache_info *cache);
 int pdc_spaceid_bits(unsigned long *space_bits);
-#ifndef CONFIG_PA20
 int pdc_btlb_info(struct pdc_btlb_info *btlb);
+int pdc_btlb_insert(unsigned long long vpage, unsigned long physpage, unsigned long len,
+                    unsigned long entry_info, unsigned long slot);
+int pdc_btlb_purge_all(void);
 int pdc_mem_map_hpa(struct pdc_memory_map *r_addr, struct pdc_module_path *mod_path);
-#endif /* !CONFIG_PA20 */
 int pdc_pim_toc11(struct pdc_toc_pim_11 *ret);
 int pdc_pim_toc20(struct pdc_toc_pim_20 *ret);
 int pdc_lan_station_id(char *lan_addr, unsigned long net_hpa);
diff --git a/arch/parisc/include/asm/processor.h b/arch/parisc/include/asm/processor.h
index d77c43d32974..ff6cbdb6903b 100644
--- a/arch/parisc/include/asm/processor.h
+++ b/arch/parisc/include/asm/processor.h
@@ -310,6 +310,7 @@ extern void do_syscall_trace_exit(struct pt_regs *);
 struct seq_file;
 extern void early_trap_init(void);
 extern void collect_boot_cpu_data(void);
+extern void btlb_init_per_cpu(void);
 extern int show_cpuinfo (struct seq_file *m, void *v);
 
 /* driver code in driver/parisc */
diff --git a/arch/parisc/include/asm/ropes.h b/arch/parisc/include/asm/ropes.h
index fd96706c7234..e2d2d7e9bfde 100644
--- a/arch/parisc/include/asm/ropes.h
+++ b/arch/parisc/include/asm/ropes.h
@@ -29,7 +29,7 @@
 struct ioc {
 	void __iomem	*ioc_hpa;	/* I/O MMU base address */
 	char		*res_map;	/* resource map, bit == pdir entry */
-	u64		*pdir_base;	/* physical base address */
+	__le64		*pdir_base;	/* physical base address */
 	unsigned long	ibase;		/* pdir IOV Space base - shared w/lba_pci */
 	unsigned long	imask;		/* pdir IOV Space mask - shared w/lba_pci */
 #ifdef ZX1_SUPPORT
@@ -86,6 +86,9 @@ struct sba_device {
 	struct ioc		ioc[MAX_IOC];
 };
 
+/* list of SBA's in system, see drivers/parisc/sba_iommu.c */
+extern struct sba_device *sba_list;
+
 #define ASTRO_RUNWAY_PORT	0x582
 #define IKE_MERCED_PORT		0x803
 #define REO_MERCED_PORT		0x804
@@ -110,7 +113,7 @@ static inline int IS_PLUTO(struct parisc_device *d) {
 
 #define SBA_PDIR_VALID_BIT	0x8000000000000000ULL
 
-#define SBA_AGPGART_COOKIE	0x0000badbadc0ffeeULL
+#define SBA_AGPGART_COOKIE	(__force __le64) 0x0000badbadc0ffeeULL
 
 #define SBA_FUNC_ID	0x0000	/* function id */
 #define SBA_FCLASS	0x0008	/* function class, bist, header, rev... */
diff --git a/arch/parisc/include/asm/shmparam.h b/arch/parisc/include/asm/shmparam.h
index 74f74e4d35b7..5a95b0f62b87 100644
--- a/arch/parisc/include/asm/shmparam.h
+++ b/arch/parisc/include/asm/shmparam.h
@@ -2,6 +2,21 @@
 #ifndef _ASMPARISC_SHMPARAM_H
 #define _ASMPARISC_SHMPARAM_H
 
+/*
+ * PA-RISC uses virtually indexed & physically tagged (VIPT) caches
+ * which has strict requirements when two pages to the same physical
+ * address are accessed through different mappings. Read the section
+ * "Address Aliasing" in the arch docs for more detail:
+ * PA-RISC 1.1 (page 3-6):
+ * https://parisc.wiki.kernel.org/images-parisc/6/68/Pa11_acd.pdf
+ * PA-RISC 2.0 (page F-5):
+ * https://parisc.wiki.kernel.org/images-parisc/7/73/Parisc2.0.pdf
+ *
+ * For Linux we allow kernel and userspace to map pages on page size
+ * granularity (SHMLBA) but have to ensure that, if two pages are
+ * mapped to the same physical address, the virtual and physical
+ * addresses modulo SHM_COLOUR are identical.
+ */
 #define SHMLBA	   PAGE_SIZE	/* attach addr a multiple of this */
 #define SHM_COLOUR 0x00400000	/* shared mappings colouring */
 
diff --git a/arch/parisc/kernel/asm-offsets.c b/arch/parisc/kernel/asm-offsets.c
index 94652e13c260..757816a7bd4b 100644
--- a/arch/parisc/kernel/asm-offsets.c
+++ b/arch/parisc/kernel/asm-offsets.c
@@ -275,6 +275,8 @@ int main(void)
 	 * and kernel data on physical huge pages */
 #ifdef CONFIG_HUGETLB_PAGE
 	DEFINE(HUGEPAGE_SIZE, 1UL << REAL_HPAGE_SHIFT);
+#elif !defined(CONFIG_64BIT)
+	DEFINE(HUGEPAGE_SIZE, 4*1024*1024);
 #else
 	DEFINE(HUGEPAGE_SIZE, PAGE_SIZE);
 #endif
diff --git a/arch/parisc/kernel/cache.c b/arch/parisc/kernel/cache.c
index 442109a48940..268d90a9325b 100644
--- a/arch/parisc/kernel/cache.c
+++ b/arch/parisc/kernel/cache.c
@@ -58,7 +58,7 @@ int pa_serialize_tlb_flushes __ro_after_init;
 
 struct pdc_cache_info cache_info __ro_after_init;
 #ifndef CONFIG_PA20
-static struct pdc_btlb_info btlb_info __ro_after_init;
+struct pdc_btlb_info btlb_info __ro_after_init;
 #endif
 
 DEFINE_STATIC_KEY_TRUE(parisc_has_cache);
@@ -264,12 +264,6 @@ parisc_cache_init(void)
 	icache_stride = CAFL_STRIDE(cache_info.ic_conf);
 #undef CAFL_STRIDE
 
-#ifndef CONFIG_PA20
-	if (pdc_btlb_info(&btlb_info) < 0) {
-		memset(&btlb_info, 0, sizeof btlb_info);
-	}
-#endif
-
 	if ((boot_cpu_data.pdc.capabilities & PDC_MODEL_NVA_MASK) ==
 						PDC_MODEL_NVA_UNSUPPORTED) {
 		printk(KERN_WARNING "parisc_cache_init: Only equivalent aliasing supported!\n");
diff --git a/arch/parisc/kernel/drivers.c b/arch/parisc/kernel/drivers.c
index 8f4b77648491..ed8b75948061 100644
--- a/arch/parisc/kernel/drivers.c
+++ b/arch/parisc/kernel/drivers.c
@@ -925,9 +925,9 @@ static __init void qemu_header(void)
 	pr_info("#define PARISC_MODEL \"%s\"\n\n",
 			boot_cpu_data.pdc.sys_model_name);
 
+	#define p ((unsigned long *)&boot_cpu_data.pdc.model)
 	pr_info("#define PARISC_PDC_MODEL 0x%lx, 0x%lx, 0x%lx, "
 		"0x%lx, 0x%lx, 0x%lx, 0x%lx, 0x%lx, 0x%lx\n\n",
-	#define p ((unsigned long *)&boot_cpu_data.pdc.model)
 		p[0], p[1], p[2], p[3], p[4], p[5], p[6], p[7], p[8]);
 	#undef p
 
diff --git a/arch/parisc/kernel/firmware.c b/arch/parisc/kernel/firmware.c
index 8f37e75f2fb9..81078abec521 100644
--- a/arch/parisc/kernel/firmware.c
+++ b/arch/parisc/kernel/firmware.c
@@ -687,7 +687,6 @@ int pdc_spaceid_bits(unsigned long *space_bits)
 	return retval;
 }
 
-#ifndef CONFIG_PA20
 /**
  * pdc_btlb_info - Return block TLB information.
  * @btlb: The return buffer.
@@ -696,18 +695,51 @@ int pdc_spaceid_bits(unsigned long *space_bits)
  */
 int pdc_btlb_info(struct pdc_btlb_info *btlb) 
 {
-        int retval;
+	int retval;
 	unsigned long flags;
 
-        spin_lock_irqsave(&pdc_lock, flags);
-        retval = mem_pdc_call(PDC_BLOCK_TLB, PDC_BTLB_INFO, __pa(pdc_result), 0);
-        memcpy(btlb, pdc_result, sizeof(*btlb));
-        spin_unlock_irqrestore(&pdc_lock, flags);
+	if (IS_ENABLED(CONFIG_PA20))
+		return PDC_BAD_PROC;
 
-        if(retval < 0) {
-                btlb->max_size = 0;
-        }
-        return retval;
+	spin_lock_irqsave(&pdc_lock, flags);
+	retval = mem_pdc_call(PDC_BLOCK_TLB, PDC_BTLB_INFO, __pa(pdc_result), 0);
+	memcpy(btlb, pdc_result, sizeof(*btlb));
+	spin_unlock_irqrestore(&pdc_lock, flags);
+
+	if(retval < 0) {
+		btlb->max_size = 0;
+	}
+	return retval;
+}
+
+int pdc_btlb_insert(unsigned long long vpage, unsigned long physpage, unsigned long len,
+		    unsigned long entry_info, unsigned long slot)
+{
+	int retval;
+	unsigned long flags;
+
+	if (IS_ENABLED(CONFIG_PA20))
+		return PDC_BAD_PROC;
+
+	spin_lock_irqsave(&pdc_lock, flags);
+	retval = mem_pdc_call(PDC_BLOCK_TLB, PDC_BTLB_INSERT, (unsigned long) (vpage >> 32),
+			      (unsigned long) vpage, physpage, len, entry_info, slot);
+	spin_unlock_irqrestore(&pdc_lock, flags);
+	return retval;
+}
+
+int pdc_btlb_purge_all(void)
+{
+	int retval;
+	unsigned long flags;
+
+	if (IS_ENABLED(CONFIG_PA20))
+		return PDC_BAD_PROC;
+
+	spin_lock_irqsave(&pdc_lock, flags);
+	retval = mem_pdc_call(PDC_BLOCK_TLB, PDC_BTLB_PURGE_ALL);
+	spin_unlock_irqrestore(&pdc_lock, flags);
+	return retval;
 }
 
 /**
@@ -728,6 +760,9 @@ int pdc_mem_map_hpa(struct pdc_memory_map *address,
         int retval;
 	unsigned long flags;
 
+	if (IS_ENABLED(CONFIG_PA20))
+		return PDC_BAD_PROC;
+
         spin_lock_irqsave(&pdc_lock, flags);
         memcpy(pdc_result2, mod_path, sizeof(*mod_path));
         retval = mem_pdc_call(PDC_MEM_MAP, PDC_MEM_MAP_HPA, __pa(pdc_result),
@@ -737,7 +772,6 @@ int pdc_mem_map_hpa(struct pdc_memory_map *address,
 
         return retval;
 }
-#endif	/* !CONFIG_PA20 */
 
 /**
  * pdc_lan_station_id - Get the LAN address.
diff --git a/arch/parisc/kernel/head.S b/arch/parisc/kernel/head.S
index fd15fd4bbb61..a171bf3c6b31 100644
--- a/arch/parisc/kernel/head.S
+++ b/arch/parisc/kernel/head.S
@@ -180,10 +180,10 @@ $pgt_fill_loop:
 	std		%dp,0x18(%r10)
 #endif
 
-#ifdef CONFIG_64BIT
-	/* Get PDCE_PROC for monarch CPU. */
 #define MEM_PDC_LO 0x388
 #define MEM_PDC_HI 0x35C
+#ifdef CONFIG_64BIT
+	/* Get PDCE_PROC for monarch CPU. */
 	ldw             MEM_PDC_LO(%r0),%r3
 	ldw             MEM_PDC_HI(%r0),%r10
 	depd            %r10, 31, 32, %r3        /* move to upper word */
@@ -269,7 +269,17 @@ stext_pdc_ret:
 	tovirt_r1	%r6
 	mtctl		%r6,%cr30		/* restore task thread info */
 #endif
-	
+
+#ifndef CONFIG_64BIT
+	/* clear all BTLBs */
+	ldi		PDC_BLOCK_TLB,%arg0
+	load32          PA(stext_pdc_btlb_ret), %rp
+	ldw             MEM_PDC_LO(%r0),%r3
+	bv              (%r3)
+	ldi		PDC_BTLB_PURGE_ALL,%arg1
+stext_pdc_btlb_ret:
+#endif
+
 	/* PARANOID: clear user scratch/user space SR's */
 	mtsp	%r0,%sr0
 	mtsp	%r0,%sr1
diff --git a/arch/parisc/kernel/irq.c b/arch/parisc/kernel/irq.c
index 12c4d4104ade..2f81bfd4f15e 100644
--- a/arch/parisc/kernel/irq.c
+++ b/arch/parisc/kernel/irq.c
@@ -365,7 +365,7 @@ union irq_stack_union {
 	volatile unsigned int lock[1];
 };
 
-DEFINE_PER_CPU(union irq_stack_union, irq_stack_union) = {
+static DEFINE_PER_CPU(union irq_stack_union, irq_stack_union) = {
 		.slock = { 1,1,1,1 },
 	};
 #endif
diff --git a/arch/parisc/kernel/processor.c b/arch/parisc/kernel/processor.c
index a0e2d37c5b3b..1fc89fa2c2d2 100644
--- a/arch/parisc/kernel/processor.c
+++ b/arch/parisc/kernel/processor.c
@@ -368,6 +368,8 @@ int init_per_cpu(int cpunum)
 	/* FUTURE: Enable Performance Monitor : ccr bit 0x20 */
 	init_percpu_prof(cpunum);
 
+	btlb_init_per_cpu();
+
 	return ret;
 }
 
diff --git a/arch/parisc/kernel/vmlinux.lds.S b/arch/parisc/kernel/vmlinux.lds.S
index 1aaa2ca09800..58694d1989c2 100644
--- a/arch/parisc/kernel/vmlinux.lds.S
+++ b/arch/parisc/kernel/vmlinux.lds.S
@@ -154,6 +154,7 @@ SECTIONS
 	}
 
 	/* End of data section */
+	. = ALIGN(PAGE_SIZE);
 	_edata = .;
 
 	/* BSS */
diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index a088c243edea..a2a3e89f2d9a 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -32,6 +32,7 @@
 #include <asm/sections.h>
 #include <asm/msgbuf.h>
 #include <asm/sparsemem.h>
+#include <asm/asm-offsets.h>
 
 extern int  data_start;
 extern void parisc_kernel_start(void);	/* Kernel entry point in head.S */
@@ -720,6 +721,77 @@ void __init paging_init(void)
 	parisc_bootmem_free();
 }
 
+static void alloc_btlb(unsigned long start, unsigned long end, int *slot,
+			unsigned long entry_info)
+{
+	const int slot_max = btlb_info.fixed_range_info.num_comb;
+	int min_num_pages = btlb_info.min_size;
+	unsigned long size;
+
+	/* map at minimum 4 pages */
+	if (min_num_pages < 4)
+		min_num_pages = 4;
+
+	size = HUGEPAGE_SIZE;
+	while (start < end && *slot < slot_max && size >= PAGE_SIZE) {
+		/* starting address must have same alignment as size! */
+		/* if correctly aligned and fits in double size, increase */
+		if (((start & (2 * size - 1)) == 0) &&
+		    (end - start) >= (2 * size)) {
+			size <<= 1;
+			continue;
+		}
+		/* if current size alignment is too big, try smaller size */
+		if ((start & (size - 1)) != 0) {
+			size >>= 1;
+			continue;
+		}
+		if ((end - start) >= size) {
+			if ((size >> PAGE_SHIFT) >= min_num_pages)
+				pdc_btlb_insert(start >> PAGE_SHIFT, __pa(start) >> PAGE_SHIFT,
+					size >> PAGE_SHIFT, entry_info, *slot);
+			(*slot)++;
+			start += size;
+			continue;
+		}
+		size /= 2;
+		continue;
+	}
+}
+
+void btlb_init_per_cpu(void)
+{
+	unsigned long s, t, e;
+	int slot;
+
+	/* BTLBs are not available on 64-bit CPUs */
+	if (IS_ENABLED(CONFIG_PA20))
+		return;
+	else if (pdc_btlb_info(&btlb_info) < 0) {
+		memset(&btlb_info, 0, sizeof btlb_info);
+	}
+
+	/* insert BLTLBs for code and data segments */
+	s = (uintptr_t) dereference_function_descriptor(&_stext);
+	e = (uintptr_t) dereference_function_descriptor(&_etext);
+	t = (uintptr_t) dereference_function_descriptor(&_sdata);
+	BUG_ON(t != e);
+
+	/* code segments */
+	slot = 0;
+	alloc_btlb(s, e, &slot, 0x13800000);
+
+	/* sanity check */
+	t = (uintptr_t) dereference_function_descriptor(&_edata);
+	e = (uintptr_t) dereference_function_descriptor(&__bss_start);
+	BUG_ON(t != e);
+
+	/* data segments */
+	s = (uintptr_t) dereference_function_descriptor(&_sdata);
+	e = (uintptr_t) dereference_function_descriptor(&__bss_stop);
+	alloc_btlb(s, e, &slot, 0x11800000);
+}
+
 #ifdef CONFIG_PA20
 
 /*
diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h
index e2ecd01bfac7..b55b434f0059 100644
--- a/arch/riscv/include/asm/errata_list.h
+++ b/arch/riscv/include/asm/errata_list.h
@@ -105,7 +105,7 @@ asm volatile(ALTERNATIVE(						\
  * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
  *   0000001    01001      rs1       000      00000  0001011
  * dcache.cva rs1 (clean, virtual address)
- *   0000001    00100      rs1       000      00000  0001011
+ *   0000001    00101      rs1       000      00000  0001011
  *
  * dcache.cipa rs1 (clean then invalidate, physical address)
  * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
@@ -118,7 +118,7 @@ asm volatile(ALTERNATIVE(						\
  *   0000000    11001     00000      000      00000  0001011
  */
 #define THEAD_inval_A0	".long 0x0265000b"
-#define THEAD_clean_A0	".long 0x0245000b"
+#define THEAD_clean_A0	".long 0x0255000b"
 #define THEAD_flush_A0	".long 0x0275000b"
 #define THEAD_SYNC_S	".long 0x0190000b"
 
diff --git a/arch/riscv/kernel/elf_kexec.c b/arch/riscv/kernel/elf_kexec.c
index f4099059ed8f..e60fbd8660c4 100644
--- a/arch/riscv/kernel/elf_kexec.c
+++ b/arch/riscv/kernel/elf_kexec.c
@@ -98,7 +98,13 @@ static int elf_find_pbase(struct kimage *image, unsigned long kernel_len,
 	kbuf.image = image;
 	kbuf.buf_min = lowest_paddr;
 	kbuf.buf_max = ULONG_MAX;
-	kbuf.buf_align = PAGE_SIZE;
+
+	/*
+	 * Current riscv boot protocol requires 2MB alignment for
+	 * RV64 and 4MB alignment for RV32
+	 *
+	 */
+	kbuf.buf_align = PMD_SIZE;
 	kbuf.mem = KEXEC_BUF_MEM_UNKNOWN;
 	kbuf.memsz = ALIGN(kernel_len, PAGE_SIZE);
 	kbuf.top_down = false;
diff --git a/arch/riscv/kvm/vcpu_onereg.c b/arch/riscv/kvm/vcpu_onereg.c
index 1b7e9fa265cb..b7e0e03c69b1 100644
--- a/arch/riscv/kvm/vcpu_onereg.c
+++ b/arch/riscv/kvm/vcpu_onereg.c
@@ -460,8 +460,11 @@ static int riscv_vcpu_get_isa_ext_single(struct kvm_vcpu *vcpu,
 	    reg_num >= ARRAY_SIZE(kvm_isa_ext_arr))
 		return -ENOENT;
 
-	*reg_val = 0;
 	host_isa_ext = kvm_isa_ext_arr[reg_num];
+	if (!__riscv_isa_extension_available(NULL, host_isa_ext))
+		return -ENOENT;
+
+	*reg_val = 0;
 	if (__riscv_isa_extension_available(vcpu->arch.isa, host_isa_ext))
 		*reg_val = 1; /* Mark the given extension as available */
 
@@ -842,7 +845,7 @@ static int copy_isa_ext_reg_indices(const struct kvm_vcpu *vcpu,
 		u64 reg = KVM_REG_RISCV | size | KVM_REG_RISCV_ISA_EXT | i;
 
 		isa_ext = kvm_isa_ext_arr[i];
-		if (!__riscv_isa_extension_available(vcpu->arch.isa, isa_ext))
+		if (!__riscv_isa_extension_available(NULL, isa_ext))
 			continue;
 
 		if (uindices) {
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 982b777eadc7..66bfabae8814 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1945,6 +1945,7 @@ config EFI
 	select UCS2_STRING
 	select EFI_RUNTIME_WRAPPERS
 	select ARCH_USE_MEMREMAP_PROT
+	select EFI_RUNTIME_MAP if KEXEC_CORE
 	help
 	  This enables the kernel to use EFI runtime services that are
 	  available (such as the EFI variable services).
@@ -2020,7 +2021,6 @@ config EFI_MAX_FAKE_MEM
 config EFI_RUNTIME_MAP
 	bool "Export EFI runtime maps to sysfs" if EXPERT
 	depends on EFI
-	default KEXEC_CORE
 	help
 	  Export EFI runtime memory regions to /sys/firmware/efi/runtime-map.
 	  That memory map is required by the 2nd kernel to set up EFI virtual
diff --git a/arch/x86/boot/compressed/ident_map_64.c b/arch/x86/boot/compressed/ident_map_64.c
index bcc956c17872..08f93b0401bb 100644
--- a/arch/x86/boot/compressed/ident_map_64.c
+++ b/arch/x86/boot/compressed/ident_map_64.c
@@ -59,6 +59,14 @@ static void *alloc_pgt_page(void *context)
 		return NULL;
 	}
 
+	/* Consumed more tables than expected? */
+	if (pages->pgt_buf_offset == BOOT_PGT_SIZE_WARN) {
+		debug_putstr("pgt_buf running low in " __FILE__ "\n");
+		debug_putstr("Need to raise BOOT_PGT_SIZE?\n");
+		debug_putaddr(pages->pgt_buf_offset);
+		debug_putaddr(pages->pgt_buf_size);
+	}
+
 	entry = pages->pgt_buf + pages->pgt_buf_offset;
 	pages->pgt_buf_offset += PAGE_SIZE;
 
diff --git a/arch/x86/include/asm/boot.h b/arch/x86/include/asm/boot.h
index 4ae14339cb8c..b3a7cfb0d99e 100644
--- a/arch/x86/include/asm/boot.h
+++ b/arch/x86/include/asm/boot.h
@@ -40,23 +40,40 @@
 #ifdef CONFIG_X86_64
 # define BOOT_STACK_SIZE	0x4000
 
+/*
+ * Used by decompressor's startup_32() to allocate page tables for identity
+ * mapping of the 4G of RAM in 4-level paging mode:
+ * - 1 level4 table;
+ * - 1 level3 table;
+ * - 4 level2 table that maps everything with 2M pages;
+ *
+ * The additional level5 table needed for 5-level paging is allocated from
+ * trampoline_32bit memory.
+ */
 # define BOOT_INIT_PGT_SIZE	(6*4096)
-# ifdef CONFIG_RANDOMIZE_BASE
+
 /*
- * Assuming all cross the 512GB boundary:
- * 1 page for level4
- * (2+2)*4 pages for kernel, param, cmd_line, and randomized kernel
- * 2 pages for first 2M (video RAM: CONFIG_X86_VERBOSE_BOOTUP).
- * Total is 19 pages.
+ * Total number of page tables kernel_add_identity_map() can allocate,
+ * including page tables consumed by startup_32().
+ *
+ * Worst-case scenario:
+ *  - 5-level paging needs 1 level5 table;
+ *  - KASLR needs to map kernel, boot_params, cmdline and randomized kernel,
+ *    assuming all of them cross 256T boundary:
+ *    + 4*2 level4 table;
+ *    + 4*2 level3 table;
+ *    + 4*2 level2 table;
+ *  - X86_VERBOSE_BOOTUP needs to map the first 2M (video RAM):
+ *    + 1 level4 table;
+ *    + 1 level3 table;
+ *    + 1 level2 table;
+ * Total: 28 tables
+ *
+ * Add 4 spare table in case decompressor touches anything beyond what is
+ * accounted above. Warn if it happens.
  */
-#  ifdef CONFIG_X86_VERBOSE_BOOTUP
-#   define BOOT_PGT_SIZE	(19*4096)
-#  else /* !CONFIG_X86_VERBOSE_BOOTUP */
-#   define BOOT_PGT_SIZE	(17*4096)
-#  endif
-# else /* !CONFIG_RANDOMIZE_BASE */
-#  define BOOT_PGT_SIZE		BOOT_INIT_PGT_SIZE
-# endif
+# define BOOT_PGT_SIZE_WARN	(28*4096)
+# define BOOT_PGT_SIZE		(32*4096)
 
 #else /* !CONFIG_X86_64 */
 # define BOOT_STACK_SIZE	0x1000
diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h
index b0994ae3bc23..c4555b269a1b 100644
--- a/arch/x86/include/asm/efi.h
+++ b/arch/x86/include/asm/efi.h
@@ -91,19 +91,6 @@ static inline void efi_fpu_end(void)
 
 #ifdef CONFIG_X86_32
 #define EFI_X86_KERNEL_ALLOC_LIMIT		(SZ_512M - 1)
-
-#define arch_efi_call_virt_setup()					\
-({									\
-	efi_fpu_begin();						\
-	firmware_restrict_branch_speculation_start();			\
-})
-
-#define arch_efi_call_virt_teardown()					\
-({									\
-	firmware_restrict_branch_speculation_end();			\
-	efi_fpu_end();							\
-})
-
 #else /* !CONFIG_X86_32 */
 #define EFI_X86_KERNEL_ALLOC_LIMIT		EFI_ALLOC_LIMIT
 
@@ -116,14 +103,6 @@ extern bool efi_disable_ibt_for_runtime;
 	__efi_call(__VA_ARGS__);					\
 })
 
-#define arch_efi_call_virt_setup()					\
-({									\
-	efi_sync_low_kernel_mappings();					\
-	efi_fpu_begin();						\
-	firmware_restrict_branch_speculation_start();			\
-	efi_enter_mm();							\
-})
-
 #undef arch_efi_call_virt
 #define arch_efi_call_virt(p, f, args...) ({				\
 	u64 ret, ibt = ibt_save(efi_disable_ibt_for_runtime);		\
@@ -132,13 +111,6 @@ extern bool efi_disable_ibt_for_runtime;
 	ret;								\
 })
 
-#define arch_efi_call_virt_teardown()					\
-({									\
-	efi_leave_mm();							\
-	firmware_restrict_branch_speculation_end();			\
-	efi_fpu_end();							\
-})
-
 #ifdef CONFIG_KASAN
 /*
  * CONFIG_KASAN may redefine memset to __memset.  __memset function is present
@@ -168,8 +140,8 @@ extern void efi_delete_dummy_variable(void);
 extern void efi_crash_gracefully_on_page_fault(unsigned long phys_addr);
 extern void efi_free_boot_services(void);
 
-void efi_enter_mm(void);
-void efi_leave_mm(void);
+void arch_efi_call_virt_setup(void);
+void arch_efi_call_virt_teardown(void);
 
 /* kexec external ABI */
 struct efi_setup_data {
diff --git a/arch/x86/include/asm/linkage.h b/arch/x86/include/asm/linkage.h
index 97a3de7892d3..5ff49fd67732 100644
--- a/arch/x86/include/asm/linkage.h
+++ b/arch/x86/include/asm/linkage.h
@@ -8,6 +8,14 @@
 #undef notrace
 #define notrace __attribute__((no_instrument_function))
 
+#ifdef CONFIG_64BIT
+/*
+ * The generic version tends to create spurious ENDBR instructions under
+ * certain conditions.
+ */
+#define _THIS_IP_ ({ unsigned long __here; asm ("lea 0(%%rip), %0" : "=r" (__here)); __here; })
+#endif
+
 #ifdef CONFIG_X86_32
 #define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0)))
 #endif /* CONFIG_X86_32 */
diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index d9f5d7492f83..205cee567629 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -1533,7 +1533,7 @@ static void __init build_socket_tables(void)
 {
 	struct uv_gam_range_entry *gre = uv_gre_table;
 	int nums, numn, nump;
-	int cpu, i, lnid;
+	int i, lnid, apicid;
 	int minsock = _min_socket;
 	int maxsock = _max_socket;
 	int minpnode = _min_pnode;
@@ -1584,15 +1584,14 @@ static void __init build_socket_tables(void)
 
 	/* Set socket -> node values: */
 	lnid = NUMA_NO_NODE;
-	for_each_possible_cpu(cpu) {
-		int nid = cpu_to_node(cpu);
-		int apicid, sockid;
+	for (apicid = 0; apicid < ARRAY_SIZE(__apicid_to_node); apicid++) {
+		int nid = __apicid_to_node[apicid];
+		int sockid;
 
-		if (lnid == nid)
+		if ((nid == NUMA_NO_NODE) || (lnid == nid))
 			continue;
 		lnid = nid;
 
-		apicid = per_cpu(x86_cpu_to_apicid, cpu);
 		sockid = apicid >> uv_cpuid.socketid_shift;
 
 		if (_socket_to_node[sockid - minsock] == SOCK_EMPTY)
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 4e45ff44aa07..48e040618731 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -579,7 +579,6 @@ static bool match_llc(struct cpuinfo_x86 *c, struct cpuinfo_x86 *o)
 }
 
 
-#if defined(CONFIG_SCHED_SMT) || defined(CONFIG_SCHED_CLUSTER) || defined(CONFIG_SCHED_MC)
 static inline int x86_sched_itmt_flags(void)
 {
 	return sysctl_sched_itmt_enabled ? SD_ASYM_PACKING : 0;
@@ -603,7 +602,14 @@ static int x86_cluster_flags(void)
 	return cpu_cluster_flags() | x86_sched_itmt_flags();
 }
 #endif
-#endif
+
+static int x86_die_flags(void)
+{
+	if (cpu_feature_enabled(X86_FEATURE_HYBRID_CPU))
+	       return x86_sched_itmt_flags();
+
+	return 0;
+}
 
 /*
  * Set if a package/die has multiple NUMA nodes inside.
@@ -640,7 +646,7 @@ static void __init build_sched_topology(void)
 	 */
 	if (!x86_has_numa_in_package) {
 		x86_topology[i++] = (struct sched_domain_topology_level){
-			cpu_cpu_mask, SD_INIT_NAME(DIE)
+			cpu_cpu_mask, x86_die_flags, SD_INIT_NAME(DIE)
 		};
 	}
 
diff --git a/arch/x86/lib/putuser.S b/arch/x86/lib/putuser.S
index 1451e0c4ae22..235bbda6fc82 100644
--- a/arch/x86/lib/putuser.S
+++ b/arch/x86/lib/putuser.S
@@ -56,7 +56,6 @@ SYM_FUNC_END(__put_user_1)
 EXPORT_SYMBOL(__put_user_1)
 
 SYM_FUNC_START(__put_user_nocheck_1)
-	ENDBR
 	ASM_STAC
 2:	movb %al,(%_ASM_CX)
 	xor %ecx,%ecx
@@ -76,7 +75,6 @@ SYM_FUNC_END(__put_user_2)
 EXPORT_SYMBOL(__put_user_2)
 
 SYM_FUNC_START(__put_user_nocheck_2)
-	ENDBR
 	ASM_STAC
 4:	movw %ax,(%_ASM_CX)
 	xor %ecx,%ecx
@@ -96,7 +94,6 @@ SYM_FUNC_END(__put_user_4)
 EXPORT_SYMBOL(__put_user_4)
 
 SYM_FUNC_START(__put_user_nocheck_4)
-	ENDBR
 	ASM_STAC
 6:	movl %eax,(%_ASM_CX)
 	xor %ecx,%ecx
@@ -119,7 +116,6 @@ SYM_FUNC_END(__put_user_8)
 EXPORT_SYMBOL(__put_user_8)
 
 SYM_FUNC_START(__put_user_nocheck_8)
-	ENDBR
 	ASM_STAC
 9:	mov %_ASM_AX,(%_ASM_CX)
 #ifdef CONFIG_X86_32
diff --git a/arch/x86/platform/efi/efi_32.c b/arch/x86/platform/efi/efi_32.c
index e06a199423c0..b2cc7b4552a1 100644
--- a/arch/x86/platform/efi/efi_32.c
+++ b/arch/x86/platform/efi/efi_32.c
@@ -140,3 +140,15 @@ void __init efi_runtime_update_mappings(void)
 		}
 	}
 }
+
+void arch_efi_call_virt_setup(void)
+{
+	efi_fpu_begin();
+	firmware_restrict_branch_speculation_start();
+}
+
+void arch_efi_call_virt_teardown(void)
+{
+	firmware_restrict_branch_speculation_end();
+	efi_fpu_end();
+}
diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c
index 77f7ac3668cb..91d31ac422d6 100644
--- a/arch/x86/platform/efi/efi_64.c
+++ b/arch/x86/platform/efi/efi_64.c
@@ -474,19 +474,34 @@ void __init efi_dump_pagetable(void)
  * can not change under us.
  * It should be ensured that there are no concurrent calls to this function.
  */
-void efi_enter_mm(void)
+static void efi_enter_mm(void)
 {
 	efi_prev_mm = current->active_mm;
 	current->active_mm = &efi_mm;
 	switch_mm(efi_prev_mm, &efi_mm, NULL);
 }
 
-void efi_leave_mm(void)
+static void efi_leave_mm(void)
 {
 	current->active_mm = efi_prev_mm;
 	switch_mm(&efi_mm, efi_prev_mm, NULL);
 }
 
+void arch_efi_call_virt_setup(void)
+{
+	efi_sync_low_kernel_mappings();
+	efi_fpu_begin();
+	firmware_restrict_branch_speculation_start();
+	efi_enter_mm();
+}
+
+void arch_efi_call_virt_teardown(void)
+{
+	efi_leave_mm();
+	firmware_restrict_branch_speculation_end();
+	efi_fpu_end();
+}
+
 static DEFINE_SPINLOCK(efi_runtime_lock);
 
 /*
diff --git a/arch/x86/purgatory/Makefile b/arch/x86/purgatory/Makefile
index c2a29be35c01..08aa0f25f12a 100644
--- a/arch/x86/purgatory/Makefile
+++ b/arch/x86/purgatory/Makefile
@@ -19,6 +19,10 @@ CFLAGS_sha256.o := -D__DISABLE_EXPORTS -D__NO_FORTIFY
 # optimization flags.
 KBUILD_CFLAGS := $(filter-out -fprofile-sample-use=% -fprofile-use=%,$(KBUILD_CFLAGS))
 
+# When LTO is enabled, llvm emits many text sections, which is not supported
+# by kexec. Remove -flto=* flags.
+KBUILD_CFLAGS := $(filter-out $(CC_FLAGS_LTO),$(KBUILD_CFLAGS))
+
 # When linking purgatory.ro with -r unresolved symbols are not checked,
 # also link a purgatory.chk binary without -r to check for unresolved symbols.
 PURGATORY_LDFLAGS := -e purgatory_start -z nodefaultlib