Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 fixes from Ingo Molnar:
"Misc fixes and updates:

- a handful of MDS documentation/comment updates

- a cleanup related to hweight interfaces

- a SEV guest fix for large pages

- a kprobes LTO fix

- and a final cleanup commit for vDSO HPET support removal"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/speculation/mds: Improve CPU buffer clear documentation
x86/speculation/mds: Revert CPU buffer clear on double fault exit
x86/kconfig: Disable CONFIG_GENERIC_HWEIGHT and remove __HAVE_ARCH_SW_HWEIGHT
x86/mm: Do not use set_{pud, pmd}_safe() when splitting a large page
x86/kprobes: Make trampoline_handler() global and visible
x86/vdso: Remove hpet_page from vDSO

Documentation/x86/mds.rst

···
    mds_user_clear.
 
    The mitigation is invoked in prepare_exit_to_usermode() which covers
-   most of the kernel to user space transitions. There are a few exceptions
-   which are not invoking prepare_exit_to_usermode() on return to user
-   space. These exceptions use the paranoid exit code.
+   all but one of the kernel to user space transitions.  The exception
+   is when we return from a Non Maskable Interrupt (NMI), which is
+   handled directly in do_nmi().
 
-   - Non Maskable Interrupt (NMI):
-
-     Access to sensible data like keys, credentials in the NMI context is
-     mostly theoretical: The CPU can do prefetching or execute a
-     misspeculated code path and thereby fetching data which might end up
-     leaking through a buffer.
-
-     But for mounting other attacks the kernel stack address of the task is
-     already valuable information. So in full mitigation mode, the NMI is
-     mitigated on the return from do_nmi() to provide almost complete
-     coverage.
-
-   - Double fault (#DF):
-
-     A double fault is usually fatal, but the ESPFIX workaround, which can
-     be triggered from user space through modify_ldt(2) is a recoverable
-     double fault. #DF uses the paranoid exit path, so explicit mitigation
-     in the double fault handler is required.
-
-   - Machine Check Exception (#MC):
-
-     Another corner case is a #MC which hits between the CPU buffer clear
-     invocation and the actual return to user. As this still is in kernel
-     space it takes the paranoid exit path which does not clear the CPU
-     buffers. So the #MC handler repopulates the buffers to some
-     extent. Machine checks are not reliably controllable and the window is
-     extremly small so mitigation would just tick a checkbox that this
-     theoretical corner case is covered. To keep the amount of special
-     cases small, ignore #MC.
-
-   - Debug Exception (#DB):
-
-     This takes the paranoid exit path only when the INT1 breakpoint is in
-     kernel space. #DB on a user space address takes the regular exit path,
-     so no extra mitigation required.
+   (The reason that NMI is special is that prepare_exit_to_usermode() can
+    enable IRQs.  In NMI context, NMIs are blocked, and we don't want to
+    enable IRQs with NMIs blocked.)
 
 
 2. C-State transition

arch/x86/Kconfig

···
 config GENERIC_BUG_RELATIVE_POINTERS
 	bool
 
-config GENERIC_HWEIGHT
-	def_bool y
-
 config ARCH_MAY_HAVE_PC_FDC
 	def_bool y
 	depends on ISA_DMA_API

arch/x86/entry/vdso/vdso2c.c

···
 enum {
 	sym_vvar_start,
 	sym_vvar_page,
-	sym_hpet_page,
 	sym_pvclock_page,
 	sym_hvclock_page,
 };
 
 const int special_pages[] = {
 	sym_vvar_page,
-	sym_hpet_page,
 	sym_pvclock_page,
 	sym_hvclock_page,
 };
···
 struct vdso_sym required_syms[] = {
 	[sym_vvar_start] = {"vvar_start", true},
 	[sym_vvar_page] = {"vvar_page", true},
-	[sym_hpet_page] = {"hpet_page", true},
 	[sym_pvclock_page] = {"pvclock_page", true},
 	[sym_hvclock_page] = {"hvclock_page", true},
 	{"VDSO32_NOTE_MASK", true},

arch/x86/include/asm/arch_hweight.h

···
 #define REG_OUT "a"
 #endif
 
-#define __HAVE_ARCH_SW_HWEIGHT
-
 static __always_inline unsigned int __arch_hweight32(unsigned int w)
 {
 	unsigned int res;

arch/x86/include/asm/vdso.h

···
 	long sym_vvar_start;  /* Negative offset to the vvar area */
 
 	long sym_vvar_page;
-	long sym_hpet_page;
 	long sym_pvclock_page;
 	long sym_hvclock_page;
 	long sym_VDSO32_NOTE_MASK;

arch/x86/kernel/kprobes/core.c

···
 /*
  * Called from kretprobe_trampoline
  */
-static __used void *trampoline_handler(struct pt_regs *regs)
+__used __visible void *trampoline_handler(struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb;
 	struct kretprobe_instance *ri = NULL;

arch/x86/kernel/traps.c

···
 #include <asm/alternative.h>
 #include <asm/fpu/xstate.h>
 #include <asm/trace/mpx.h>
-#include <asm/nospec-branch.h>
 #include <asm/mpx.h>
 #include <asm/vm86.h>
 #include <asm/umip.h>
···
 		regs->ip = (unsigned long)general_protection;
 		regs->sp = (unsigned long)&gpregs->orig_ax;
 
-		/*
-		 * This situation can be triggered by userspace via
-		 * modify_ldt(2) and the return does not take the regular
-		 * user space exit, so a CPU buffer clear is required when
-		 * MDS mitigation is enabled.
-		 */
-		mds_user_clear_cpu_buffers();
 		return;
 	}
 #endif

arch/x86/mm/init_64.c

···
 
 #include "ident_map.c"
 
+#define DEFINE_POPULATE(fname, type1, type2, init)		\
+static inline void fname##_init(struct mm_struct *mm,		\
+		type1##_t *arg1, type2##_t *arg2, bool init)	\
+{								\
+	if (init)						\
+		fname##_safe(mm, arg1, arg2);			\
+	else							\
+		fname(mm, arg1, arg2);				\
+}
+
+DEFINE_POPULATE(p4d_populate, p4d, pud, init)
+DEFINE_POPULATE(pgd_populate, pgd, p4d, init)
+DEFINE_POPULATE(pud_populate, pud, pmd, init)
+DEFINE_POPULATE(pmd_populate_kernel, pmd, pte, init)
+
+#define DEFINE_ENTRY(type1, type2, init)			\
+static inline void set_##type1##_init(type1##_t *arg1,		\
+			type2##_t arg2, bool init)		\
+{								\
+	if (init)						\
+		set_##type1##_safe(arg1, arg2);			\
+	else							\
+		set_##type1(arg1, arg2);			\
+}
+
+DEFINE_ENTRY(p4d, p4d, init)
+DEFINE_ENTRY(pud, pud, init)
+DEFINE_ENTRY(pmd, pmd, init)
+DEFINE_ENTRY(pte, pte, init)
+
+
 /*
  * NOTE: pagetable_init alloc all the fixmap pagetables contiguous on the
  * physical space so we can cache the place of the first one and move
···
  */
 static unsigned long __meminit
 phys_pte_init(pte_t *pte_page, unsigned long paddr, unsigned long paddr_end,
-	      pgprot_t prot)
+	      pgprot_t prot, bool init)
 {
 	unsigned long pages = 0, paddr_next;
 	unsigned long paddr_last = paddr_end;
···
 					     E820_TYPE_RAM) &&
 			    !e820__mapped_any(paddr & PAGE_MASK, paddr_next,
 					     E820_TYPE_RESERVED_KERN))
-				set_pte_safe(pte, __pte(0));
+				set_pte_init(pte, __pte(0), init);
 			continue;
 		}
 
···
 			pr_info("   pte=%p addr=%lx pte=%016lx\n", pte, paddr,
 				pfn_pte(paddr >> PAGE_SHIFT, PAGE_KERNEL).pte);
 		pages++;
-		set_pte_safe(pte, pfn_pte(paddr >> PAGE_SHIFT, prot));
+		set_pte_init(pte, pfn_pte(paddr >> PAGE_SHIFT, prot), init);
 		paddr_last = (paddr & PAGE_MASK) + PAGE_SIZE;
 	}
 
···
  */
 static unsigned long __meminit
 phys_pmd_init(pmd_t *pmd_page, unsigned long paddr, unsigned long paddr_end,
-	      unsigned long page_size_mask, pgprot_t prot)
+	      unsigned long page_size_mask, pgprot_t prot, bool init)
 {
 	unsigned long pages = 0, paddr_next;
 	unsigned long paddr_last = paddr_end;
···
 					     E820_TYPE_RAM) &&
 			    !e820__mapped_any(paddr & PMD_MASK, paddr_next,
 					     E820_TYPE_RESERVED_KERN))
-				set_pmd_safe(pmd, __pmd(0));
+				set_pmd_init(pmd, __pmd(0), init);
 			continue;
 		}
 
···
 				spin_lock(&init_mm.page_table_lock);
 				pte = (pte_t *)pmd_page_vaddr(*pmd);
 				paddr_last = phys_pte_init(pte, paddr,
-							   paddr_end, prot);
+							   paddr_end, prot,
+							   init);
 				spin_unlock(&init_mm.page_table_lock);
 				continue;
 			}
···
 		if (page_size_mask & (1<<PG_LEVEL_2M)) {
 			pages++;
 			spin_lock(&init_mm.page_table_lock);
-			set_pte_safe((pte_t *)pmd,
-				pfn_pte((paddr & PMD_MASK) >> PAGE_SHIFT,
-					__pgprot(pgprot_val(prot) | _PAGE_PSE)));
+			set_pte_init((pte_t *)pmd,
+				     pfn_pte((paddr & PMD_MASK) >> PAGE_SHIFT,
+					     __pgprot(pgprot_val(prot) | _PAGE_PSE)),
+				     init);
 			spin_unlock(&init_mm.page_table_lock);
 			paddr_last = paddr_next;
 			continue;
 		}
 
 		pte = alloc_low_page();
-		paddr_last = phys_pte_init(pte, paddr, paddr_end, new_prot);
+		paddr_last = phys_pte_init(pte, paddr, paddr_end, new_prot, init);
 
 		spin_lock(&init_mm.page_table_lock);
-		pmd_populate_kernel_safe(&init_mm, pmd, pte);
+		pmd_populate_kernel_init(&init_mm, pmd, pte, init);
 		spin_unlock(&init_mm.page_table_lock);
 	}
 	update_page_count(PG_LEVEL_2M, pages);
···
  */
 static unsigned long __meminit
 phys_pud_init(pud_t *pud_page, unsigned long paddr, unsigned long paddr_end,
-	      unsigned long page_size_mask)
+	      unsigned long page_size_mask, bool init)
 {
 	unsigned long pages = 0, paddr_next;
 	unsigned long paddr_last = paddr_end;
···
 					     E820_TYPE_RAM) &&
 			    !e820__mapped_any(paddr & PUD_MASK, paddr_next,
 					     E820_TYPE_RESERVED_KERN))
-				set_pud_safe(pud, __pud(0));
+				set_pud_init(pud, __pud(0), init);
 			continue;
 		}
 
···
 				paddr_last = phys_pmd_init(pmd, paddr,
 							   paddr_end,
 							   page_size_mask,
-							   prot);
+							   prot, init);
 				continue;
 			}
 			/*
···
 		if (page_size_mask & (1<<PG_LEVEL_1G)) {
 			pages++;
 			spin_lock(&init_mm.page_table_lock);
-			set_pte_safe((pte_t *)pud,
-				pfn_pte((paddr & PUD_MASK) >> PAGE_SHIFT,
-					PAGE_KERNEL_LARGE));
+			set_pte_init((pte_t *)pud,
+				     pfn_pte((paddr & PUD_MASK) >> PAGE_SHIFT,
+					     PAGE_KERNEL_LARGE),
+				     init);
 			spin_unlock(&init_mm.page_table_lock);
 			paddr_last = paddr_next;
 			continue;
···
 
 		pmd = alloc_low_page();
 		paddr_last = phys_pmd_init(pmd, paddr, paddr_end,
-					   page_size_mask, prot);
+					   page_size_mask, prot, init);
 
 		spin_lock(&init_mm.page_table_lock);
-		pud_populate_safe(&init_mm, pud, pmd);
+		pud_populate_init(&init_mm, pud, pmd, init);
 		spin_unlock(&init_mm.page_table_lock);
 	}
 
···
 
 static unsigned long __meminit
 phys_p4d_init(p4d_t *p4d_page, unsigned long paddr, unsigned long paddr_end,
-	      unsigned long page_size_mask)
+	      unsigned long page_size_mask, bool init)
 {
 	unsigned long paddr_next, paddr_last = paddr_end;
 	unsigned long vaddr = (unsigned long)__va(paddr);
 	int i = p4d_index(vaddr);
 
 	if (!pgtable_l5_enabled())
-		return phys_pud_init((pud_t *) p4d_page, paddr, paddr_end, page_size_mask);
+		return phys_pud_init((pud_t *) p4d_page, paddr, paddr_end,
+				     page_size_mask, init);
 
 	for (; i < PTRS_PER_P4D; i++, paddr = paddr_next) {
 		p4d_t *p4d;
···
 					     E820_TYPE_RAM) &&
 			    !e820__mapped_any(paddr & P4D_MASK, paddr_next,
 					     E820_TYPE_RESERVED_KERN))
-				set_p4d_safe(p4d, __p4d(0));
+				set_p4d_init(p4d, __p4d(0), init);
 			continue;
 		}
 
 		if (!p4d_none(*p4d)) {
 			pud = pud_offset(p4d, 0);
-			paddr_last = phys_pud_init(pud, paddr,
-					paddr_end,
-					page_size_mask);
+			paddr_last = phys_pud_init(pud, paddr, paddr_end,
+						   page_size_mask, init);
 			continue;
 		}
 
 		pud = alloc_low_page();
 		paddr_last = phys_pud_init(pud, paddr, paddr_end,
-					   page_size_mask);
+					   page_size_mask, init);
 
 		spin_lock(&init_mm.page_table_lock);
-		p4d_populate_safe(&init_mm, p4d, pud);
+		p4d_populate_init(&init_mm, p4d, pud, init);
 		spin_unlock(&init_mm.page_table_lock);
 	}
 
 	return paddr_last;
 }
 
-/*
- * Create page table mapping for the physical memory for specific physical
- * addresses. The virtual and physical addresses have to be aligned on PMD level
- * down. It returns the last physical address mapped.
- */
-unsigned long __meminit
-kernel_physical_mapping_init(unsigned long paddr_start,
-			     unsigned long paddr_end,
-			     unsigned long page_size_mask)
+static unsigned long __meminit
+__kernel_physical_mapping_init(unsigned long paddr_start,
+			       unsigned long paddr_end,
+			       unsigned long page_size_mask,
+			       bool init)
 {
 	bool pgd_changed = false;
 	unsigned long vaddr, vaddr_start, vaddr_end, vaddr_next, paddr_last;
···
 			p4d = (p4d_t *)pgd_page_vaddr(*pgd);
 			paddr_last = phys_p4d_init(p4d, __pa(vaddr),
 						   __pa(vaddr_end),
-						   page_size_mask);
+						   page_size_mask,
+						   init);
 			continue;
 		}
 
 		p4d = alloc_low_page();
 		paddr_last = phys_p4d_init(p4d, __pa(vaddr), __pa(vaddr_end),
-					   page_size_mask);
+					   page_size_mask, init);
 
 		spin_lock(&init_mm.page_table_lock);
 		if (pgtable_l5_enabled())
-			pgd_populate_safe(&init_mm, pgd, p4d);
+			pgd_populate_init(&init_mm, pgd, p4d, init);
 		else
-			p4d_populate_safe(&init_mm, p4d_offset(pgd, vaddr), (pud_t *) p4d);
+			p4d_populate_init(&init_mm, p4d_offset(pgd, vaddr),
+					  (pud_t *) p4d, init);
+
 		spin_unlock(&init_mm.page_table_lock);
 		pgd_changed = true;
 	}
···
 		sync_global_pgds(vaddr_start, vaddr_end - 1);
 
 	return paddr_last;
+}
+
+
+/*
+ * Create page table mapping for the physical memory for specific physical
+ * addresses. Note that it can only be used to populate non-present entries.
+ * The virtual and physical addresses have to be aligned on PMD level
+ * down. It returns the last physical address mapped.
+ */
+unsigned long __meminit
+kernel_physical_mapping_init(unsigned long paddr_start,
+			     unsigned long paddr_end,
+			     unsigned long page_size_mask)
+{
+	return __kernel_physical_mapping_init(paddr_start, paddr_end,
+					      page_size_mask, true);
+}
+
+/*
+ * This function is similar to kernel_physical_mapping_init() above with the
+ * exception that it uses set_{pud,pmd}() instead of the set_{pud,pte}_safe()
+ * when updating the mapping. The caller is responsible to flush the TLBs after
+ * the function returns.
+ */
+unsigned long __meminit
+kernel_physical_mapping_change(unsigned long paddr_start,
+			       unsigned long paddr_end,
+			       unsigned long page_size_mask)
+{
+	return __kernel_physical_mapping_init(paddr_start, paddr_end,
+					      page_size_mask, false);
 }
 
 #ifndef CONFIG_NUMA

arch/x86/mm/mem_encrypt.c

···
 		else
 			split_page_size_mask = 1 << PG_LEVEL_2M;
 
-		kernel_physical_mapping_init(__pa(vaddr & pmask),
-					     __pa((vaddr_end & pmask) + psize),
-					     split_page_size_mask);
+		/*
+		 * kernel_physical_mapping_change() does not flush the TLBs, so
+		 * a TLB flush is required after we exit from the for loop.
+		 */
+		kernel_physical_mapping_change(__pa(vaddr & pmask),
+					       __pa((vaddr_end & pmask) + psize),
+					       split_page_size_mask);
 	}
 
 	ret = 0;

arch/x86/mm/mm_internal.h

···
 unsigned long kernel_physical_mapping_init(unsigned long start,
 					     unsigned long end,
 					     unsigned long page_size_mask);
+unsigned long kernel_physical_mapping_change(unsigned long start,
+					     unsigned long end,
+					     unsigned long page_size_mask);
 void zone_sizes_init(void);
 
 extern int after_bootmem;

lib/hweight.c

···
  * The Hamming Weight of a number is the total number of bits set in it.
  */
 
-#ifndef __HAVE_ARCH_SW_HWEIGHT
 unsigned int __sw_hweight32(unsigned int w)
 {
 #ifdef CONFIG_ARCH_HAS_FAST_MULTIPLIER
···
 #endif
 }
 EXPORT_SYMBOL(__sw_hweight32);
-#endif
 
 unsigned int __sw_hweight16(unsigned int w)
 {
···
 }
 EXPORT_SYMBOL(__sw_hweight8);
 
-#ifndef __HAVE_ARCH_SW_HWEIGHT
 unsigned long __sw_hweight64(__u64 w)
 {
 #if BITS_PER_LONG == 32
···
 #endif
 }
 EXPORT_SYMBOL(__sw_hweight64);
-#endif