[GIT pull] x86/pti updates for 4.16

From: Thomas Gleixner
Date: Mon Feb 26 2018 - 08:25:19 EST


Linus,

please pull the latest x86-pti-for-linus git tree from:

git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86-pti-for-linus

Yet another pile of melted spectrum related changes:

- Sanitize the array_index_nospec protection mechanism: Remove the
overengineered array_index_nospec_mask_check() magic and allow
const-qualified types as index to avoid temporary storage in a non-const
local variable.

- Make the microcode loader more robust by properly propagating error
codes. Provide information about new feature bits after micro code was
updated so administrators can act upon.

- Optimizations of the entry ASM code which reduce code footprint and make
the code simpler and faster.

- Fix the {pmd,pud}_{set,clear}_flags() implementations to work properly
on paravirt kernels by removing the address translation operations.

- Revert the harmful vmexit_fill_RSB() optimization

- Use IBRS around firmware calls

- Teach objtool about retpolines and add annotations for indirect jumps
and calls.

- Explicitely disable jumplabel patching in __init code and handle
patching failures proper instead of silently ignoring them.

- Remove indirect paravirt calls for writing the speculation control MSR
as these calls are obviously proving the same attack vector which is
tried to be mitigated.

- A few small fixes which address build issues with recent compiler and
assembler versions.

Thanks,

tglx

------------------>
Borislav Petkov (3):
x86/microcode: Propagate return value from updating functions
x86/CPU: Add a microcode loader callback
x86/CPU: Check CPU feature bits after microcode upgrade

Dan Williams (2):
nospec: Kill array_index_nospec_mask_check()
nospec: Include <asm/barrier.h> dependency

David Woodhouse (3):
Revert "x86/retpoline: Simplify vmexit_fill_RSB()"
x86/speculation: Use IBRS if available before calling into firmware
x86/retpoline: Support retpoline builds with Clang

Dominik Brodowski (8):
x86/entry: Reduce the code footprint of the 'idtentry' macro
x86/entry/64: Use 'xorl' for faster register clearing
x86/entry/64: Move PUSH_AND_CLEAR_REGS from interrupt macro to helper function
x86/entry/64: Move ENTER_IRQ_STACK from interrupt macro to interrupt_entry
x86/entry/64: Move the switch_to_thread_stack() call to interrupt_entry()
x86/entry/64: Remove 'interrupt' macro
x86/entry/64: Move ASM_CLAC to interrupt_entry()
x86/entry/64: Open-code switch_to_thread_stack()

Ingo Molnar (1):
x86/speculation: Move firmware_restrict_branch_speculation_*() from C to CPP

Jan Beulich (5):
x86/mm: Fix {pmd,pud}_{set,clear}_flags()
x86/asm: Improve how GEN_*_SUFFIXED_RMWcc() specify clobbers
x86/IO-APIC: Avoid warning in 32-bit builds
x86/LDT: Avoid warning in 32-bit builds with older gcc
x86-64/realmode: Add instruction suffix

Jann Horn (1):
x86/mm: Remove stale comment about KMEMCHECK

Josh Poimboeuf (4):
jump_label: Explicitly disable jump labels in __init code
jump_label: Warn on failed jump_label patching attempt
extable: Make init_kernel_text() global
x86/entry/64: Simplify ENCODE_FRAME_POINTER

Paolo Bonzini (2):
KVM/x86: Remove indirect MSR op calls from SPEC_CTRL
KVM/VMX: Optimize vmx_vcpu_run() and svm_vcpu_run() by marking the RDMSR path as unlikely()

Peter Zijlstra (8):
x86/speculation, objtool: Annotate indirect calls/jumps for objtool
x86/paravirt, objtool: Annotate indirect calls
x86/boot, objtool: Annotate indirect jump in secondary_startup_64()
x86/mm/sme, objtool: Annotate indirect call in sme_encrypt_execute()
objtool: Use existing global variables for options
objtool: Add retpoline validation
objtool: Add module specific retpoline rules
objtool, retpolines: Integrate objtool with retpoline support more closely

Rasmus Villemoes (1):
nospec: Allow index argument to have const-qualified type


Makefile | 5 ++
arch/x86/Kconfig | 1 +
arch/x86/Makefile | 7 +-
arch/x86/entry/calling.h | 34 ++++----
arch/x86/entry/entry_32.S | 3 +-
arch/x86/entry/entry_64.S | 153 ++++++++++++++++++++--------------
arch/x86/entry/entry_64_compat.S | 71 +++++++++-------
arch/x86/include/asm/apm.h | 6 ++
arch/x86/include/asm/asm-prototypes.h | 3 -
arch/x86/include/asm/cpufeatures.h | 1 +
arch/x86/include/asm/efi.h | 17 +++-
arch/x86/include/asm/microcode.h | 9 +-
arch/x86/include/asm/mmu_context.h | 1 +
arch/x86/include/asm/nospec-branch.h | 138 +++++++++++++++++++++++++-----
arch/x86/include/asm/paravirt.h | 17 +++-
arch/x86/include/asm/paravirt_types.h | 5 +-
arch/x86/include/asm/pgtable.h | 8 +-
arch/x86/include/asm/pgtable_types.h | 10 +++
arch/x86/include/asm/processor.h | 1 +
arch/x86/include/asm/refcount.h | 4 +-
arch/x86/include/asm/rmwcc.h | 16 ++--
arch/x86/kernel/apic/io_apic.c | 2 +-
arch/x86/kernel/cpu/bugs.c | 12 ++-
arch/x86/kernel/cpu/common.c | 30 +++++++
arch/x86/kernel/cpu/microcode/amd.c | 10 +--
arch/x86/kernel/cpu/microcode/core.c | 39 +++++----
arch/x86/kernel/cpu/microcode/intel.c | 10 +--
arch/x86/kernel/head_64.S | 2 +
arch/x86/kernel/unwind_orc.c | 3 +-
arch/x86/kvm/svm.c | 9 +-
arch/x86/kvm/vmx.c | 9 +-
arch/x86/lib/Makefile | 1 -
arch/x86/lib/retpoline.S | 56 -------------
arch/x86/mm/fault.c | 4 -
arch/x86/mm/mem_encrypt_boot.S | 2 +
arch/x86/realmode/rm/trampoline_64.S | 2 +-
include/linux/compiler-clang.h | 5 ++
include/linux/compiler-gcc.h | 4 +
include/linux/init.h | 8 +-
include/linux/jump_label.h | 3 +
include/linux/kernel.h | 1 +
include/linux/nospec.h | 26 +-----
init/main.c | 2 +
kernel/extable.c | 2 +-
kernel/jump_label.c | 27 ++++--
scripts/Makefile.build | 8 ++
tools/objtool/builtin-check.c | 6 +-
tools/objtool/builtin-orc.c | 6 +-
tools/objtool/builtin.h | 5 ++
tools/objtool/check.c | 100 +++++++++++++++++++++-
tools/objtool/check.h | 3 +-
51 files changed, 604 insertions(+), 303 deletions(-)

diff --git a/Makefile b/Makefile
index 79ad2bfa24b6..3dfce4d2f25d 100644
--- a/Makefile
+++ b/Makefile
@@ -489,6 +489,11 @@ KBUILD_CFLAGS += $(CLANG_TARGET) $(CLANG_GCC_TC)
KBUILD_AFLAGS += $(CLANG_TARGET) $(CLANG_GCC_TC)
endif

+RETPOLINE_CFLAGS_GCC := -mindirect-branch=thunk-extern -mindirect-branch-register
+RETPOLINE_CFLAGS_CLANG := -mretpoline-external-thunk
+RETPOLINE_CFLAGS := $(call cc-option,$(RETPOLINE_CFLAGS_GCC),$(call cc-option,$(RETPOLINE_CFLAGS_CLANG)))
+export RETPOLINE_CFLAGS
+
ifeq ($(config-targets),1)
# ===========================================================================
# *config targets only - make sure prerequisites are updated, and descend
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 63bf349b2b24..c1aed6c0e413 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -436,6 +436,7 @@ config GOLDFISH
config RETPOLINE
bool "Avoid speculative indirect branches in kernel"
default y
+ select STACK_VALIDATION if HAVE_STACK_VALIDATION
help
Compile kernel with the retpoline compiler options to guard against
kernel-to-user data leaks by avoiding speculative indirect
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index fad55160dcb9..498c1b812300 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -232,10 +232,9 @@ KBUILD_CFLAGS += -fno-asynchronous-unwind-tables

# Avoid indirect branches in kernel to deal with Spectre
ifdef CONFIG_RETPOLINE
- RETPOLINE_CFLAGS += $(call cc-option,-mindirect-branch=thunk-extern -mindirect-branch-register)
- ifneq ($(RETPOLINE_CFLAGS),)
- KBUILD_CFLAGS += $(RETPOLINE_CFLAGS) -DRETPOLINE
- endif
+ifneq ($(RETPOLINE_CFLAGS),)
+ KBUILD_CFLAGS += $(RETPOLINE_CFLAGS) -DRETPOLINE
+endif
endif

archscripts: scripts_basic
diff --git a/arch/x86/entry/calling.h b/arch/x86/entry/calling.h
index dce7092ab24a..be63330c5511 100644
--- a/arch/x86/entry/calling.h
+++ b/arch/x86/entry/calling.h
@@ -97,7 +97,7 @@ For 32-bit we have the following conventions - kernel is built with

#define SIZEOF_PTREGS 21*8

-.macro PUSH_AND_CLEAR_REGS rdx=%rdx rax=%rax
+.macro PUSH_AND_CLEAR_REGS rdx=%rdx rax=%rax save_ret=0
/*
* Push registers and sanitize registers of values that a
* speculation attack might otherwise want to exploit. The
@@ -105,32 +105,41 @@ For 32-bit we have the following conventions - kernel is built with
* could be put to use in a speculative execution gadget.
* Interleave XOR with PUSH for better uop scheduling:
*/
+ .if \save_ret
+ pushq %rsi /* pt_regs->si */
+ movq 8(%rsp), %rsi /* temporarily store the return address in %rsi */
+ movq %rdi, 8(%rsp) /* pt_regs->di (overwriting original return address) */
+ .else
pushq %rdi /* pt_regs->di */
pushq %rsi /* pt_regs->si */
+ .endif
pushq \rdx /* pt_regs->dx */
pushq %rcx /* pt_regs->cx */
pushq \rax /* pt_regs->ax */
pushq %r8 /* pt_regs->r8 */
- xorq %r8, %r8 /* nospec r8 */
+ xorl %r8d, %r8d /* nospec r8 */
pushq %r9 /* pt_regs->r9 */
- xorq %r9, %r9 /* nospec r9 */
+ xorl %r9d, %r9d /* nospec r9 */
pushq %r10 /* pt_regs->r10 */
- xorq %r10, %r10 /* nospec r10 */
+ xorl %r10d, %r10d /* nospec r10 */
pushq %r11 /* pt_regs->r11 */
- xorq %r11, %r11 /* nospec r11*/
+ xorl %r11d, %r11d /* nospec r11*/
pushq %rbx /* pt_regs->rbx */
xorl %ebx, %ebx /* nospec rbx*/
pushq %rbp /* pt_regs->rbp */
xorl %ebp, %ebp /* nospec rbp*/
pushq %r12 /* pt_regs->r12 */
- xorq %r12, %r12 /* nospec r12*/
+ xorl %r12d, %r12d /* nospec r12*/
pushq %r13 /* pt_regs->r13 */
- xorq %r13, %r13 /* nospec r13*/
+ xorl %r13d, %r13d /* nospec r13*/
pushq %r14 /* pt_regs->r14 */
- xorq %r14, %r14 /* nospec r14*/
+ xorl %r14d, %r14d /* nospec r14*/
pushq %r15 /* pt_regs->r15 */
- xorq %r15, %r15 /* nospec r15*/
+ xorl %r15d, %r15d /* nospec r15*/
UNWIND_HINT_REGS
+ .if \save_ret
+ pushq %rsi /* return address on top of stack */
+ .endif
.endm

.macro POP_REGS pop_rdi=1 skip_r11rcx=0
@@ -172,12 +181,7 @@ For 32-bit we have the following conventions - kernel is built with
*/
.macro ENCODE_FRAME_POINTER ptregs_offset=0
#ifdef CONFIG_FRAME_POINTER
- .if \ptregs_offset
- leaq \ptregs_offset(%rsp), %rbp
- .else
- mov %rsp, %rbp
- .endif
- orq $0x1, %rbp
+ leaq 1+\ptregs_offset(%rsp), %rbp
#endif
.endm

diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index 16c2c022540d..6ad064c8cf35 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -252,8 +252,7 @@ ENTRY(__switch_to_asm)
* exist, overwrite the RSB with entries which capture
* speculative execution to prevent attack.
*/
- /* Clobbers %ebx */
- FILL_RETURN_BUFFER RSB_CLEAR_LOOPS, X86_FEATURE_RSB_CTXSW
+ FILL_RETURN_BUFFER %ebx, RSB_CLEAR_LOOPS, X86_FEATURE_RSB_CTXSW
#endif

/* restore callee-saved registers */
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index 8971bd64d515..d5c7f18f79ac 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -364,8 +364,7 @@ ENTRY(__switch_to_asm)
* exist, overwrite the RSB with entries which capture
* speculative execution to prevent attack.
*/
- /* Clobbers %rbx */
- FILL_RETURN_BUFFER RSB_CLEAR_LOOPS, X86_FEATURE_RSB_CTXSW
+ FILL_RETURN_BUFFER %r12, RSB_CLEAR_LOOPS, X86_FEATURE_RSB_CTXSW
#endif

/* restore callee-saved registers */
@@ -449,9 +448,19 @@ END(irq_entries_start)
*
* The invariant is that, if irq_count != -1, then the IRQ stack is in use.
*/
-.macro ENTER_IRQ_STACK regs=1 old_rsp
+.macro ENTER_IRQ_STACK regs=1 old_rsp save_ret=0
DEBUG_ENTRY_ASSERT_IRQS_OFF
+
+ .if \save_ret
+ /*
+ * If save_ret is set, the original stack contains one additional
+ * entry -- the return address. Therefore, move the address one
+ * entry below %rsp to \old_rsp.
+ */
+ leaq 8(%rsp), \old_rsp
+ .else
movq %rsp, \old_rsp
+ .endif

.if \regs
UNWIND_HINT_REGS base=\old_rsp
@@ -497,6 +506,15 @@ END(irq_entries_start)
.if \regs
UNWIND_HINT_REGS indirect=1
.endif
+
+ .if \save_ret
+ /*
+ * Push the return address to the stack. This return address can
+ * be found at the "real" original RSP, which was offset by 8 at
+ * the beginning of this macro.
+ */
+ pushq -8(\old_rsp)
+ .endif
.endm

/*
@@ -520,27 +538,65 @@ END(irq_entries_start)
.endm

/*
- * Interrupt entry/exit.
- *
- * Interrupt entry points save only callee clobbered registers in fast path.
+ * Interrupt entry helper function.
*
- * Entry runs with interrupts off.
+ * Entry runs with interrupts off. Stack layout at entry:
+ * +----------------------------------------------------+
+ * | regs->ss |
+ * | regs->rsp |
+ * | regs->eflags |
+ * | regs->cs |
+ * | regs->ip |
+ * +----------------------------------------------------+
+ * | regs->orig_ax = ~(interrupt number) |
+ * +----------------------------------------------------+
+ * | return address |
+ * +----------------------------------------------------+
*/
-
-/* 0(%rsp): ~(interrupt number) */
- .macro interrupt func
+ENTRY(interrupt_entry)
+ UNWIND_HINT_FUNC
+ ASM_CLAC
cld

- testb $3, CS-ORIG_RAX(%rsp)
+ testb $3, CS-ORIG_RAX+8(%rsp)
jz 1f
SWAPGS
- call switch_to_thread_stack
+
+ /*
+ * Switch to the thread stack. The IRET frame and orig_ax are
+ * on the stack, as well as the return address. RDI..R12 are
+ * not (yet) on the stack and space has not (yet) been
+ * allocated for them.
+ */
+ pushq %rdi
+
+ /* Need to switch before accessing the thread stack. */
+ SWITCH_TO_KERNEL_CR3 scratch_reg=%rdi
+ movq %rsp, %rdi
+ movq PER_CPU_VAR(cpu_current_top_of_stack), %rsp
+
+ /*
+ * We have RDI, return address, and orig_ax on the stack on
+ * top of the IRET frame. That means offset=24
+ */
+ UNWIND_HINT_IRET_REGS base=%rdi offset=24
+
+ pushq 7*8(%rdi) /* regs->ss */
+ pushq 6*8(%rdi) /* regs->rsp */
+ pushq 5*8(%rdi) /* regs->eflags */
+ pushq 4*8(%rdi) /* regs->cs */
+ pushq 3*8(%rdi) /* regs->ip */
+ pushq 2*8(%rdi) /* regs->orig_ax */
+ pushq 8(%rdi) /* return address */
+ UNWIND_HINT_FUNC
+
+ movq (%rdi), %rdi
1:

- PUSH_AND_CLEAR_REGS
- ENCODE_FRAME_POINTER
+ PUSH_AND_CLEAR_REGS save_ret=1
+ ENCODE_FRAME_POINTER 8

- testb $3, CS(%rsp)
+ testb $3, CS+8(%rsp)
jz 1f

/*
@@ -548,7 +604,7 @@ END(irq_entries_start)
*
* We need to tell lockdep that IRQs are off. We can't do this until
* we fix gsbase, and we should do it before enter_from_user_mode
- * (which can take locks). Since TRACE_IRQS_OFF idempotent,
+ * (which can take locks). Since TRACE_IRQS_OFF is idempotent,
* the simplest way to handle it is to just call it twice if
* we enter from user mode. There's no reason to optimize this since
* TRACE_IRQS_OFF is a no-op if lockdep is off.
@@ -558,12 +614,15 @@ END(irq_entries_start)
CALL_enter_from_user_mode

1:
- ENTER_IRQ_STACK old_rsp=%rdi
+ ENTER_IRQ_STACK old_rsp=%rdi save_ret=1
/* We entered an interrupt context - irqs are off: */
TRACE_IRQS_OFF

- call \func /* rdi points to pt_regs */
- .endm
+ ret
+END(interrupt_entry)
+
+
+/* Interrupt entry/exit. */

/*
* The interrupt stubs push (~vector+0x80) onto the stack and
@@ -571,9 +630,10 @@ END(irq_entries_start)
*/
.p2align CONFIG_X86_L1_CACHE_SHIFT
common_interrupt:
- ASM_CLAC
addq $-0x80, (%rsp) /* Adjust vector to [-256, -1] range */
- interrupt do_IRQ
+ call interrupt_entry
+ UNWIND_HINT_REGS indirect=1
+ call do_IRQ /* rdi points to pt_regs */
/* 0(%rsp): old RSP */
ret_from_intr:
DISABLE_INTERRUPTS(CLBR_ANY)
@@ -766,10 +826,11 @@ END(common_interrupt)
.macro apicinterrupt3 num sym do_sym
ENTRY(\sym)
UNWIND_HINT_IRET_REGS
- ASM_CLAC
pushq $~(\num)
.Lcommon_\sym:
- interrupt \do_sym
+ call interrupt_entry
+ UNWIND_HINT_REGS indirect=1
+ call \do_sym /* rdi points to pt_regs */
jmp ret_from_intr
END(\sym)
.endm
@@ -832,34 +893,6 @@ apicinterrupt IRQ_WORK_VECTOR irq_work_interrupt smp_irq_work_interrupt
*/
#define CPU_TSS_IST(x) PER_CPU_VAR(cpu_tss_rw) + (TSS_ist + ((x) - 1) * 8)

-/*
- * Switch to the thread stack. This is called with the IRET frame and
- * orig_ax on the stack. (That is, RDI..R12 are not on the stack and
- * space has not been allocated for them.)
- */
-ENTRY(switch_to_thread_stack)
- UNWIND_HINT_FUNC
-
- pushq %rdi
- /* Need to switch before accessing the thread stack. */
- SWITCH_TO_KERNEL_CR3 scratch_reg=%rdi
- movq %rsp, %rdi
- movq PER_CPU_VAR(cpu_current_top_of_stack), %rsp
- UNWIND_HINT sp_offset=16 sp_reg=ORC_REG_DI
-
- pushq 7*8(%rdi) /* regs->ss */
- pushq 6*8(%rdi) /* regs->rsp */
- pushq 5*8(%rdi) /* regs->eflags */
- pushq 4*8(%rdi) /* regs->cs */
- pushq 3*8(%rdi) /* regs->ip */
- pushq 2*8(%rdi) /* regs->orig_ax */
- pushq 8(%rdi) /* return address */
- UNWIND_HINT_FUNC
-
- movq (%rdi), %rdi
- ret
-END(switch_to_thread_stack)
-
.macro idtentry sym do_sym has_error_code:req paranoid=0 shift_ist=-1
ENTRY(\sym)
UNWIND_HINT_IRET_REGS offset=\has_error_code*8
@@ -875,12 +908,8 @@ ENTRY(\sym)
pushq $-1 /* ORIG_RAX: no syscall to restart */
.endif

- /* Save all registers in pt_regs */
- PUSH_AND_CLEAR_REGS
- ENCODE_FRAME_POINTER
-
.if \paranoid < 2
- testb $3, CS(%rsp) /* If coming from userspace, switch stacks */
+ testb $3, CS-ORIG_RAX(%rsp) /* If coming from userspace, switch stacks */
jnz .Lfrom_usermode_switch_stack_\@
.endif

@@ -1130,13 +1159,15 @@ idtentry machine_check do_mce has_error_code=0 paranoid=1
#endif

/*
- * Switch gs if needed.
+ * Save all registers in pt_regs, and switch gs if needed.
* Use slow, but surefire "are we in kernel?" check.
* Return: ebx=0: need swapgs on exit, ebx=1: otherwise
*/
ENTRY(paranoid_entry)
UNWIND_HINT_FUNC
cld
+ PUSH_AND_CLEAR_REGS save_ret=1
+ ENCODE_FRAME_POINTER 8
movl $1, %ebx
movl $MSR_GS_BASE, %ecx
rdmsr
@@ -1181,12 +1212,14 @@ ENTRY(paranoid_exit)
END(paranoid_exit)

/*
- * Switch gs if needed.
+ * Save all registers in pt_regs, and switch GS if needed.
* Return: EBX=0: came from user mode; EBX=1: otherwise
*/
ENTRY(error_entry)
- UNWIND_HINT_REGS offset=8
+ UNWIND_HINT_FUNC
cld
+ PUSH_AND_CLEAR_REGS save_ret=1
+ ENCODE_FRAME_POINTER 8
testb $3, CS+8(%rsp)
jz .Lerror_kernelspace

@@ -1577,8 +1610,6 @@ end_repeat_nmi:
* frame to point back to repeat_nmi.
*/
pushq $-1 /* ORIG_RAX: no syscall to restart */
- PUSH_AND_CLEAR_REGS
- ENCODE_FRAME_POINTER

/*
* Use paranoid_entry to handle SWAPGS, but no need to use paranoid_exit
diff --git a/arch/x86/entry/entry_64_compat.S b/arch/x86/entry/entry_64_compat.S
index fd65e016e413..e811dd9c5e99 100644
--- a/arch/x86/entry/entry_64_compat.S
+++ b/arch/x86/entry/entry_64_compat.S
@@ -85,25 +85,25 @@ ENTRY(entry_SYSENTER_compat)
pushq %rcx /* pt_regs->cx */
pushq $-ENOSYS /* pt_regs->ax */
pushq $0 /* pt_regs->r8 = 0 */
- xorq %r8, %r8 /* nospec r8 */
+ xorl %r8d, %r8d /* nospec r8 */
pushq $0 /* pt_regs->r9 = 0 */
- xorq %r9, %r9 /* nospec r9 */
+ xorl %r9d, %r9d /* nospec r9 */
pushq $0 /* pt_regs->r10 = 0 */
- xorq %r10, %r10 /* nospec r10 */
+ xorl %r10d, %r10d /* nospec r10 */
pushq $0 /* pt_regs->r11 = 0 */
- xorq %r11, %r11 /* nospec r11 */
+ xorl %r11d, %r11d /* nospec r11 */
pushq %rbx /* pt_regs->rbx */
xorl %ebx, %ebx /* nospec rbx */
pushq %rbp /* pt_regs->rbp (will be overwritten) */
xorl %ebp, %ebp /* nospec rbp */
pushq $0 /* pt_regs->r12 = 0 */
- xorq %r12, %r12 /* nospec r12 */
+ xorl %r12d, %r12d /* nospec r12 */
pushq $0 /* pt_regs->r13 = 0 */
- xorq %r13, %r13 /* nospec r13 */
+ xorl %r13d, %r13d /* nospec r13 */
pushq $0 /* pt_regs->r14 = 0 */
- xorq %r14, %r14 /* nospec r14 */
+ xorl %r14d, %r14d /* nospec r14 */
pushq $0 /* pt_regs->r15 = 0 */
- xorq %r15, %r15 /* nospec r15 */
+ xorl %r15d, %r15d /* nospec r15 */
cld

/*
@@ -224,25 +224,25 @@ GLOBAL(entry_SYSCALL_compat_after_hwframe)
pushq %rbp /* pt_regs->cx (stashed in bp) */
pushq $-ENOSYS /* pt_regs->ax */
pushq $0 /* pt_regs->r8 = 0 */
- xorq %r8, %r8 /* nospec r8 */
+ xorl %r8d, %r8d /* nospec r8 */
pushq $0 /* pt_regs->r9 = 0 */
- xorq %r9, %r9 /* nospec r9 */
+ xorl %r9d, %r9d /* nospec r9 */
pushq $0 /* pt_regs->r10 = 0 */
- xorq %r10, %r10 /* nospec r10 */
+ xorl %r10d, %r10d /* nospec r10 */
pushq $0 /* pt_regs->r11 = 0 */
- xorq %r11, %r11 /* nospec r11 */
+ xorl %r11d, %r11d /* nospec r11 */
pushq %rbx /* pt_regs->rbx */
xorl %ebx, %ebx /* nospec rbx */
pushq %rbp /* pt_regs->rbp (will be overwritten) */
xorl %ebp, %ebp /* nospec rbp */
pushq $0 /* pt_regs->r12 = 0 */
- xorq %r12, %r12 /* nospec r12 */
+ xorl %r12d, %r12d /* nospec r12 */
pushq $0 /* pt_regs->r13 = 0 */
- xorq %r13, %r13 /* nospec r13 */
+ xorl %r13d, %r13d /* nospec r13 */
pushq $0 /* pt_regs->r14 = 0 */
- xorq %r14, %r14 /* nospec r14 */
+ xorl %r14d, %r14d /* nospec r14 */
pushq $0 /* pt_regs->r15 = 0 */
- xorq %r15, %r15 /* nospec r15 */
+ xorl %r15d, %r15d /* nospec r15 */

/*
* User mode is traced as though IRQs are on, and SYSENTER
@@ -298,9 +298,9 @@ sysret32_from_system_call:
*/
SWITCH_TO_USER_CR3_NOSTACK scratch_reg=%r8 scratch_reg2=%r9

- xorq %r8, %r8
- xorq %r9, %r9
- xorq %r10, %r10
+ xorl %r8d, %r8d
+ xorl %r9d, %r9d
+ xorl %r10d, %r10d
swapgs
sysretl
END(entry_SYSCALL_compat)
@@ -347,10 +347,23 @@ ENTRY(entry_INT80_compat)
*/
movl %eax, %eax

+ /* switch to thread stack expects orig_ax and rdi to be pushed */
pushq %rax /* pt_regs->orig_ax */
+ pushq %rdi /* pt_regs->di */
+
+ /* Need to switch before accessing the thread stack. */
+ SWITCH_TO_KERNEL_CR3 scratch_reg=%rdi
+ movq %rsp, %rdi
+ movq PER_CPU_VAR(cpu_current_top_of_stack), %rsp
+
+ pushq 6*8(%rdi) /* regs->ss */
+ pushq 5*8(%rdi) /* regs->rsp */
+ pushq 4*8(%rdi) /* regs->eflags */
+ pushq 3*8(%rdi) /* regs->cs */
+ pushq 2*8(%rdi) /* regs->ip */
+ pushq 1*8(%rdi) /* regs->orig_ax */

- /* switch to thread stack expects orig_ax to be pushed */
- call switch_to_thread_stack
+ movq (%rdi), %rdi /* restore %rdi */

pushq %rdi /* pt_regs->di */
pushq %rsi /* pt_regs->si */
@@ -358,25 +371,25 @@ ENTRY(entry_INT80_compat)
pushq %rcx /* pt_regs->cx */
pushq $-ENOSYS /* pt_regs->ax */
pushq $0 /* pt_regs->r8 = 0 */
- xorq %r8, %r8 /* nospec r8 */
+ xorl %r8d, %r8d /* nospec r8 */
pushq $0 /* pt_regs->r9 = 0 */
- xorq %r9, %r9 /* nospec r9 */
+ xorl %r9d, %r9d /* nospec r9 */
pushq $0 /* pt_regs->r10 = 0 */
- xorq %r10, %r10 /* nospec r10 */
+ xorl %r10d, %r10d /* nospec r10 */
pushq $0 /* pt_regs->r11 = 0 */
- xorq %r11, %r11 /* nospec r11 */
+ xorl %r11d, %r11d /* nospec r11 */
pushq %rbx /* pt_regs->rbx */
xorl %ebx, %ebx /* nospec rbx */
pushq %rbp /* pt_regs->rbp */
xorl %ebp, %ebp /* nospec rbp */
pushq %r12 /* pt_regs->r12 */
- xorq %r12, %r12 /* nospec r12 */
+ xorl %r12d, %r12d /* nospec r12 */
pushq %r13 /* pt_regs->r13 */
- xorq %r13, %r13 /* nospec r13 */
+ xorl %r13d, %r13d /* nospec r13 */
pushq %r14 /* pt_regs->r14 */
- xorq %r14, %r14 /* nospec r14 */
+ xorl %r14d, %r14d /* nospec r14 */
pushq %r15 /* pt_regs->r15 */
- xorq %r15, %r15 /* nospec r15 */
+ xorl %r15d, %r15d /* nospec r15 */
cld

/*
diff --git a/arch/x86/include/asm/apm.h b/arch/x86/include/asm/apm.h
index 4d4015ddcf26..c356098b6fb9 100644
--- a/arch/x86/include/asm/apm.h
+++ b/arch/x86/include/asm/apm.h
@@ -7,6 +7,8 @@
#ifndef _ASM_X86_MACH_DEFAULT_APM_H
#define _ASM_X86_MACH_DEFAULT_APM_H

+#include <asm/nospec-branch.h>
+
#ifdef APM_ZERO_SEGS
# define APM_DO_ZERO_SEGS \
"pushl %%ds\n\t" \
@@ -32,6 +34,7 @@ static inline void apm_bios_call_asm(u32 func, u32 ebx_in, u32 ecx_in,
* N.B. We do NOT need a cld after the BIOS call
* because we always save and restore the flags.
*/
+ firmware_restrict_branch_speculation_start();
__asm__ __volatile__(APM_DO_ZERO_SEGS
"pushl %%edi\n\t"
"pushl %%ebp\n\t"
@@ -44,6 +47,7 @@ static inline void apm_bios_call_asm(u32 func, u32 ebx_in, u32 ecx_in,
"=S" (*esi)
: "a" (func), "b" (ebx_in), "c" (ecx_in)
: "memory", "cc");
+ firmware_restrict_branch_speculation_end();
}

static inline bool apm_bios_call_simple_asm(u32 func, u32 ebx_in,
@@ -56,6 +60,7 @@ static inline bool apm_bios_call_simple_asm(u32 func, u32 ebx_in,
* N.B. We do NOT need a cld after the BIOS call
* because we always save and restore the flags.
*/
+ firmware_restrict_branch_speculation_start();
__asm__ __volatile__(APM_DO_ZERO_SEGS
"pushl %%edi\n\t"
"pushl %%ebp\n\t"
@@ -68,6 +73,7 @@ static inline bool apm_bios_call_simple_asm(u32 func, u32 ebx_in,
"=S" (si)
: "a" (func), "b" (ebx_in), "c" (ecx_in)
: "memory", "cc");
+ firmware_restrict_branch_speculation_end();
return error;
}

diff --git a/arch/x86/include/asm/asm-prototypes.h b/arch/x86/include/asm/asm-prototypes.h
index 4d111616524b..1908214b9125 100644
--- a/arch/x86/include/asm/asm-prototypes.h
+++ b/arch/x86/include/asm/asm-prototypes.h
@@ -38,7 +38,4 @@ INDIRECT_THUNK(dx)
INDIRECT_THUNK(si)
INDIRECT_THUNK(di)
INDIRECT_THUNK(bp)
-asmlinkage void __fill_rsb(void);
-asmlinkage void __clear_rsb(void);
-
#endif /* CONFIG_RETPOLINE */
diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
index 0dfe4d3f74e2..f41079da38c5 100644
--- a/arch/x86/include/asm/cpufeatures.h
+++ b/arch/x86/include/asm/cpufeatures.h
@@ -213,6 +213,7 @@
#define X86_FEATURE_SEV ( 7*32+20) /* AMD Secure Encrypted Virtualization */

#define X86_FEATURE_USE_IBPB ( 7*32+21) /* "" Indirect Branch Prediction Barrier enabled */
+#define X86_FEATURE_USE_IBRS_FW ( 7*32+22) /* "" Use IBRS during runtime firmware calls */

/* Virtualization flags: Linux defined, word 8 */
#define X86_FEATURE_TPR_SHADOW ( 8*32+ 0) /* Intel TPR Shadow */
diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h
index 85f6ccb80b91..a399c1ebf6f0 100644
--- a/arch/x86/include/asm/efi.h
+++ b/arch/x86/include/asm/efi.h
@@ -6,6 +6,7 @@
#include <asm/pgtable.h>
#include <asm/processor-flags.h>
#include <asm/tlb.h>
+#include <asm/nospec-branch.h>

/*
* We map the EFI regions needed for runtime services non-contiguously,
@@ -36,8 +37,18 @@

extern asmlinkage unsigned long efi_call_phys(void *, ...);

-#define arch_efi_call_virt_setup() kernel_fpu_begin()
-#define arch_efi_call_virt_teardown() kernel_fpu_end()
+#define arch_efi_call_virt_setup() \
+({ \
+ kernel_fpu_begin(); \
+ firmware_restrict_branch_speculation_start(); \
+})
+
+#define arch_efi_call_virt_teardown() \
+({ \
+ firmware_restrict_branch_speculation_end(); \
+ kernel_fpu_end(); \
+})
+

/*
* Wrap all the virtual calls in a way that forces the parameters on the stack.
@@ -73,6 +84,7 @@ struct efi_scratch {
efi_sync_low_kernel_mappings(); \
preempt_disable(); \
__kernel_fpu_begin(); \
+ firmware_restrict_branch_speculation_start(); \
\
if (efi_scratch.use_pgd) { \
efi_scratch.prev_cr3 = __read_cr3(); \
@@ -91,6 +103,7 @@ struct efi_scratch {
__flush_tlb_all(); \
} \
\
+ firmware_restrict_branch_speculation_end(); \
__kernel_fpu_end(); \
preempt_enable(); \
})
diff --git a/arch/x86/include/asm/microcode.h b/arch/x86/include/asm/microcode.h
index 55520cec8b27..7fb1047d61c7 100644
--- a/arch/x86/include/asm/microcode.h
+++ b/arch/x86/include/asm/microcode.h
@@ -37,7 +37,12 @@ struct cpu_signature {

struct device;

-enum ucode_state { UCODE_ERROR, UCODE_OK, UCODE_NFOUND };
+enum ucode_state {
+ UCODE_OK = 0,
+ UCODE_UPDATED,
+ UCODE_NFOUND,
+ UCODE_ERROR,
+};

struct microcode_ops {
enum ucode_state (*request_microcode_user) (int cpu,
@@ -54,7 +59,7 @@ struct microcode_ops {
* are being called.
* See also the "Synchronization" section in microcode_core.c.
*/
- int (*apply_microcode) (int cpu);
+ enum ucode_state (*apply_microcode) (int cpu);
int (*collect_cpu_info) (int cpu, struct cpu_signature *csig);
};

diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
index c931b88982a0..1de72ce514cd 100644
--- a/arch/x86/include/asm/mmu_context.h
+++ b/arch/x86/include/asm/mmu_context.h
@@ -74,6 +74,7 @@ static inline void *ldt_slot_va(int slot)
return (void *)(LDT_BASE_ADDR + LDT_SLOT_STRIDE * slot);
#else
BUG();
+ return (void *)fix_to_virt(FIX_HOLE);
#endif
}

diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h
index 76b058533e47..b7063cfa19f9 100644
--- a/arch/x86/include/asm/nospec-branch.h
+++ b/arch/x86/include/asm/nospec-branch.h
@@ -8,6 +8,50 @@
#include <asm/cpufeatures.h>
#include <asm/msr-index.h>

+/*
+ * Fill the CPU return stack buffer.
+ *
+ * Each entry in the RSB, if used for a speculative 'ret', contains an
+ * infinite 'pause; lfence; jmp' loop to capture speculative execution.
+ *
+ * This is required in various cases for retpoline and IBRS-based
+ * mitigations for the Spectre variant 2 vulnerability. Sometimes to
+ * eliminate potentially bogus entries from the RSB, and sometimes
+ * purely to ensure that it doesn't get empty, which on some CPUs would
+ * allow predictions from other (unwanted!) sources to be used.
+ *
+ * We define a CPP macro such that it can be used from both .S files and
+ * inline assembly. It's possible to do a .macro and then include that
+ * from C via asm(".include <asm/nospec-branch.h>") but let's not go there.
+ */
+
+#define RSB_CLEAR_LOOPS 32 /* To forcibly overwrite all entries */
+#define RSB_FILL_LOOPS 16 /* To avoid underflow */
+
+/*
+ * Google experimented with loop-unrolling and this turned out to be
+ * the optimal version â two calls, each with their own speculation
+ * trap should their return address end up getting used, in a loop.
+ */
+#define __FILL_RETURN_BUFFER(reg, nr, sp) \
+ mov $(nr/2), reg; \
+771: \
+ call 772f; \
+773: /* speculation trap */ \
+ pause; \
+ lfence; \
+ jmp 773b; \
+772: \
+ call 774f; \
+775: /* speculation trap */ \
+ pause; \
+ lfence; \
+ jmp 775b; \
+774: \
+ dec reg; \
+ jnz 771b; \
+ add $(BITS_PER_LONG/8) * nr, sp;
+
#ifdef __ASSEMBLY__

/*
@@ -23,6 +67,18 @@
.popsection
.endm

+/*
+ * This should be used immediately before an indirect jump/call. It tells
+ * objtool the subsequent indirect jump/call is vouched safe for retpoline
+ * builds.
+ */
+.macro ANNOTATE_RETPOLINE_SAFE
+ .Lannotate_\@:
+ .pushsection .discard.retpoline_safe
+ _ASM_PTR .Lannotate_\@
+ .popsection
+.endm
+
/*
* These are the bare retpoline primitives for indirect jmp and call.
* Do not use these directly; they only exist to make the ALTERNATIVE
@@ -59,9 +115,9 @@
.macro JMP_NOSPEC reg:req
#ifdef CONFIG_RETPOLINE
ANNOTATE_NOSPEC_ALTERNATIVE
- ALTERNATIVE_2 __stringify(jmp *\reg), \
+ ALTERNATIVE_2 __stringify(ANNOTATE_RETPOLINE_SAFE; jmp *\reg), \
__stringify(RETPOLINE_JMP \reg), X86_FEATURE_RETPOLINE, \
- __stringify(lfence; jmp *\reg), X86_FEATURE_RETPOLINE_AMD
+ __stringify(lfence; ANNOTATE_RETPOLINE_SAFE; jmp *\reg), X86_FEATURE_RETPOLINE_AMD
#else
jmp *\reg
#endif
@@ -70,18 +126,25 @@
.macro CALL_NOSPEC reg:req
#ifdef CONFIG_RETPOLINE
ANNOTATE_NOSPEC_ALTERNATIVE
- ALTERNATIVE_2 __stringify(call *\reg), \
+ ALTERNATIVE_2 __stringify(ANNOTATE_RETPOLINE_SAFE; call *\reg), \
__stringify(RETPOLINE_CALL \reg), X86_FEATURE_RETPOLINE,\
- __stringify(lfence; call *\reg), X86_FEATURE_RETPOLINE_AMD
+ __stringify(lfence; ANNOTATE_RETPOLINE_SAFE; call *\reg), X86_FEATURE_RETPOLINE_AMD
#else
call *\reg
#endif
.endm

-/* This clobbers the BX register */
-.macro FILL_RETURN_BUFFER nr:req ftr:req
+ /*
+ * A simpler FILL_RETURN_BUFFER macro. Don't make people use the CPP
+ * monstrosity above, manually.
+ */
+.macro FILL_RETURN_BUFFER reg:req nr:req ftr:req
#ifdef CONFIG_RETPOLINE
- ALTERNATIVE "", "call __clear_rsb", \ftr
+ ANNOTATE_NOSPEC_ALTERNATIVE
+ ALTERNATIVE "jmp .Lskip_rsb_\@", \
+ __stringify(__FILL_RETURN_BUFFER(\reg,\nr,%_ASM_SP)) \
+ \ftr
+.Lskip_rsb_\@:
#endif
.endm

@@ -93,6 +156,12 @@
".long 999b - .\n\t" \
".popsection\n\t"

+#define ANNOTATE_RETPOLINE_SAFE \
+ "999:\n\t" \
+ ".pushsection .discard.retpoline_safe\n\t" \
+ _ASM_PTR " 999b\n\t" \
+ ".popsection\n\t"
+
#if defined(CONFIG_X86_64) && defined(RETPOLINE)

/*
@@ -102,6 +171,7 @@
# define CALL_NOSPEC \
ANNOTATE_NOSPEC_ALTERNATIVE \
ALTERNATIVE( \
+ ANNOTATE_RETPOLINE_SAFE \
"call *%[thunk_target]\n", \
"call __x86_indirect_thunk_%V[thunk_target]\n", \
X86_FEATURE_RETPOLINE)
@@ -156,25 +226,53 @@ extern char __indirect_thunk_end[];
static inline void vmexit_fill_RSB(void)
{
#ifdef CONFIG_RETPOLINE
- alternative_input("",
- "call __fill_rsb",
- X86_FEATURE_RETPOLINE,
- ASM_NO_INPUT_CLOBBER(_ASM_BX, "memory"));
+ unsigned long loops;
+
+ asm volatile (ANNOTATE_NOSPEC_ALTERNATIVE
+ ALTERNATIVE("jmp 910f",
+ __stringify(__FILL_RETURN_BUFFER(%0, RSB_CLEAR_LOOPS, %1)),
+ X86_FEATURE_RETPOLINE)
+ "910:"
+ : "=r" (loops), ASM_CALL_CONSTRAINT
+ : : "memory" );
#endif
}

+#define alternative_msr_write(_msr, _val, _feature) \
+ asm volatile(ALTERNATIVE("", \
+ "movl %[msr], %%ecx\n\t" \
+ "movl %[val], %%eax\n\t" \
+ "movl $0, %%edx\n\t" \
+ "wrmsr", \
+ _feature) \
+ : : [msr] "i" (_msr), [val] "i" (_val) \
+ : "eax", "ecx", "edx", "memory")
+
static inline void indirect_branch_prediction_barrier(void)
{
- asm volatile(ALTERNATIVE("",
- "movl %[msr], %%ecx\n\t"
- "movl %[val], %%eax\n\t"
- "movl $0, %%edx\n\t"
- "wrmsr",
- X86_FEATURE_USE_IBPB)
- : : [msr] "i" (MSR_IA32_PRED_CMD),
- [val] "i" (PRED_CMD_IBPB)
- : "eax", "ecx", "edx", "memory");
+ alternative_msr_write(MSR_IA32_PRED_CMD, PRED_CMD_IBPB,
+ X86_FEATURE_USE_IBPB);
}

+/*
+ * With retpoline, we must use IBRS to restrict branch prediction
+ * before calling into firmware.
+ *
+ * (Implemented as CPP macros due to header hell.)
+ */
+#define firmware_restrict_branch_speculation_start() \
+do { \
+ preempt_disable(); \
+ alternative_msr_write(MSR_IA32_SPEC_CTRL, SPEC_CTRL_IBRS, \
+ X86_FEATURE_USE_IBRS_FW); \
+} while (0)
+
+#define firmware_restrict_branch_speculation_end() \
+do { \
+ alternative_msr_write(MSR_IA32_SPEC_CTRL, 0, \
+ X86_FEATURE_USE_IBRS_FW); \
+ preempt_enable(); \
+} while (0)
+
#endif /* __ASSEMBLY__ */
#endif /* _ASM_X86_NOSPEC_BRANCH_H_ */
diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
index 554841fab717..c83a2f418cea 100644
--- a/arch/x86/include/asm/paravirt.h
+++ b/arch/x86/include/asm/paravirt.h
@@ -7,6 +7,7 @@
#ifdef CONFIG_PARAVIRT
#include <asm/pgtable_types.h>
#include <asm/asm.h>
+#include <asm/nospec-branch.h>

#include <asm/paravirt_types.h>

@@ -879,23 +880,27 @@ extern void default_banner(void);

#define INTERRUPT_RETURN \
PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_iret), CLBR_NONE, \
- jmp PARA_INDIRECT(pv_cpu_ops+PV_CPU_iret))
+ ANNOTATE_RETPOLINE_SAFE; \
+ jmp PARA_INDIRECT(pv_cpu_ops+PV_CPU_iret);)

#define DISABLE_INTERRUPTS(clobbers) \
PARA_SITE(PARA_PATCH(pv_irq_ops, PV_IRQ_irq_disable), clobbers, \
PV_SAVE_REGS(clobbers | CLBR_CALLEE_SAVE); \
+ ANNOTATE_RETPOLINE_SAFE; \
call PARA_INDIRECT(pv_irq_ops+PV_IRQ_irq_disable); \
PV_RESTORE_REGS(clobbers | CLBR_CALLEE_SAVE);)

#define ENABLE_INTERRUPTS(clobbers) \
PARA_SITE(PARA_PATCH(pv_irq_ops, PV_IRQ_irq_enable), clobbers, \
PV_SAVE_REGS(clobbers | CLBR_CALLEE_SAVE); \
+ ANNOTATE_RETPOLINE_SAFE; \
call PARA_INDIRECT(pv_irq_ops+PV_IRQ_irq_enable); \
PV_RESTORE_REGS(clobbers | CLBR_CALLEE_SAVE);)

#ifdef CONFIG_X86_32
#define GET_CR0_INTO_EAX \
push %ecx; push %edx; \
+ ANNOTATE_RETPOLINE_SAFE; \
call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0); \
pop %edx; pop %ecx
#else /* !CONFIG_X86_32 */
@@ -917,21 +922,25 @@ extern void default_banner(void);
*/
#define SWAPGS \
PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_swapgs), CLBR_NONE, \
- call PARA_INDIRECT(pv_cpu_ops+PV_CPU_swapgs) \
+ ANNOTATE_RETPOLINE_SAFE; \
+ call PARA_INDIRECT(pv_cpu_ops+PV_CPU_swapgs); \
)

#define GET_CR2_INTO_RAX \
- call PARA_INDIRECT(pv_mmu_ops+PV_MMU_read_cr2)
+ ANNOTATE_RETPOLINE_SAFE; \
+ call PARA_INDIRECT(pv_mmu_ops+PV_MMU_read_cr2);

#define USERGS_SYSRET64 \
PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_usergs_sysret64), \
CLBR_NONE, \
- jmp PARA_INDIRECT(pv_cpu_ops+PV_CPU_usergs_sysret64))
+ ANNOTATE_RETPOLINE_SAFE; \
+ jmp PARA_INDIRECT(pv_cpu_ops+PV_CPU_usergs_sysret64);)

#ifdef CONFIG_DEBUG_ENTRY
#define SAVE_FLAGS(clobbers) \
PARA_SITE(PARA_PATCH(pv_irq_ops, PV_IRQ_save_fl), clobbers, \
PV_SAVE_REGS(clobbers | CLBR_CALLEE_SAVE); \
+ ANNOTATE_RETPOLINE_SAFE; \
call PARA_INDIRECT(pv_irq_ops+PV_IRQ_save_fl); \
PV_RESTORE_REGS(clobbers | CLBR_CALLEE_SAVE);)
#endif
diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h
index f624f1f10316..180bc0bff0fb 100644
--- a/arch/x86/include/asm/paravirt_types.h
+++ b/arch/x86/include/asm/paravirt_types.h
@@ -43,6 +43,7 @@
#include <asm/desc_defs.h>
#include <asm/kmap_types.h>
#include <asm/pgtable_types.h>
+#include <asm/nospec-branch.h>

struct page;
struct thread_struct;
@@ -392,7 +393,9 @@ int paravirt_disable_iospace(void);
* offset into the paravirt_patch_template structure, and can therefore be
* freely converted back into a structure offset.
*/
-#define PARAVIRT_CALL "call *%c[paravirt_opptr];"
+#define PARAVIRT_CALL \
+ ANNOTATE_RETPOLINE_SAFE \
+ "call *%c[paravirt_opptr];"

/*
* These macros are intended to wrap calls through one of the paravirt
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index 63c2552b6b65..b444d83cfc95 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -350,14 +350,14 @@ static inline pmd_t pmd_set_flags(pmd_t pmd, pmdval_t set)
{
pmdval_t v = native_pmd_val(pmd);

- return __pmd(v | set);
+ return native_make_pmd(v | set);
}

static inline pmd_t pmd_clear_flags(pmd_t pmd, pmdval_t clear)
{
pmdval_t v = native_pmd_val(pmd);

- return __pmd(v & ~clear);
+ return native_make_pmd(v & ~clear);
}

static inline pmd_t pmd_mkold(pmd_t pmd)
@@ -409,14 +409,14 @@ static inline pud_t pud_set_flags(pud_t pud, pudval_t set)
{
pudval_t v = native_pud_val(pud);

- return __pud(v | set);
+ return native_make_pud(v | set);
}

static inline pud_t pud_clear_flags(pud_t pud, pudval_t clear)
{
pudval_t v = native_pud_val(pud);

- return __pud(v & ~clear);
+ return native_make_pud(v & ~clear);
}

static inline pud_t pud_mkold(pud_t pud)
diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h
index 3696398a9475..246f15b4e64c 100644
--- a/arch/x86/include/asm/pgtable_types.h
+++ b/arch/x86/include/asm/pgtable_types.h
@@ -323,6 +323,11 @@ static inline pudval_t native_pud_val(pud_t pud)
#else
#include <asm-generic/pgtable-nopud.h>

+static inline pud_t native_make_pud(pudval_t val)
+{
+ return (pud_t) { .p4d.pgd = native_make_pgd(val) };
+}
+
static inline pudval_t native_pud_val(pud_t pud)
{
return native_pgd_val(pud.p4d.pgd);
@@ -344,6 +349,11 @@ static inline pmdval_t native_pmd_val(pmd_t pmd)
#else
#include <asm-generic/pgtable-nopmd.h>

+static inline pmd_t native_make_pmd(pmdval_t val)
+{
+ return (pmd_t) { .pud.p4d.pgd = native_make_pgd(val) };
+}
+
static inline pmdval_t native_pmd_val(pmd_t pmd)
{
return native_pgd_val(pmd.pud.p4d.pgd);
diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index 1bd9ed87606f..b0ccd4847a58 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -977,4 +977,5 @@ bool xen_set_default_idle(void);

void stop_this_cpu(void *dummy);
void df_debug(struct pt_regs *regs, long error_code);
+void microcode_check(void);
#endif /* _ASM_X86_PROCESSOR_H */
diff --git a/arch/x86/include/asm/refcount.h b/arch/x86/include/asm/refcount.h
index 4e44250e7d0d..d65171120e90 100644
--- a/arch/x86/include/asm/refcount.h
+++ b/arch/x86/include/asm/refcount.h
@@ -67,13 +67,13 @@ static __always_inline __must_check
bool refcount_sub_and_test(unsigned int i, refcount_t *r)
{
GEN_BINARY_SUFFIXED_RMWcc(LOCK_PREFIX "subl", REFCOUNT_CHECK_LT_ZERO,
- r->refs.counter, "er", i, "%0", e);
+ r->refs.counter, "er", i, "%0", e, "cx");
}

static __always_inline __must_check bool refcount_dec_and_test(refcount_t *r)
{
GEN_UNARY_SUFFIXED_RMWcc(LOCK_PREFIX "decl", REFCOUNT_CHECK_LT_ZERO,
- r->refs.counter, "%0", e);
+ r->refs.counter, "%0", e, "cx");
}

static __always_inline __must_check
diff --git a/arch/x86/include/asm/rmwcc.h b/arch/x86/include/asm/rmwcc.h
index f91c365e57c3..4914a3e7c803 100644
--- a/arch/x86/include/asm/rmwcc.h
+++ b/arch/x86/include/asm/rmwcc.h
@@ -2,8 +2,7 @@
#ifndef _ASM_X86_RMWcc
#define _ASM_X86_RMWcc

-#define __CLOBBERS_MEM "memory"
-#define __CLOBBERS_MEM_CC_CX "memory", "cc", "cx"
+#define __CLOBBERS_MEM(clb...) "memory", ## clb

#if !defined(__GCC_ASM_FLAG_OUTPUTS__) && defined(CC_HAVE_ASM_GOTO)

@@ -40,18 +39,19 @@ do { \
#endif /* defined(__GCC_ASM_FLAG_OUTPUTS__) || !defined(CC_HAVE_ASM_GOTO) */

#define GEN_UNARY_RMWcc(op, var, arg0, cc) \
- __GEN_RMWcc(op " " arg0, var, cc, __CLOBBERS_MEM)
+ __GEN_RMWcc(op " " arg0, var, cc, __CLOBBERS_MEM())

-#define GEN_UNARY_SUFFIXED_RMWcc(op, suffix, var, arg0, cc) \
+#define GEN_UNARY_SUFFIXED_RMWcc(op, suffix, var, arg0, cc, clobbers...)\
__GEN_RMWcc(op " " arg0 "\n\t" suffix, var, cc, \
- __CLOBBERS_MEM_CC_CX)
+ __CLOBBERS_MEM(clobbers))

#define GEN_BINARY_RMWcc(op, var, vcon, val, arg0, cc) \
__GEN_RMWcc(op __BINARY_RMWcc_ARG arg0, var, cc, \
- __CLOBBERS_MEM, vcon (val))
+ __CLOBBERS_MEM(), vcon (val))

-#define GEN_BINARY_SUFFIXED_RMWcc(op, suffix, var, vcon, val, arg0, cc) \
+#define GEN_BINARY_SUFFIXED_RMWcc(op, suffix, var, vcon, val, arg0, cc, \
+ clobbers...) \
__GEN_RMWcc(op __BINARY_RMWcc_ARG arg0 "\n\t" suffix, var, cc, \
- __CLOBBERS_MEM_CC_CX, vcon (val))
+ __CLOBBERS_MEM(clobbers), vcon (val))

#endif /* _ASM_X86_RMWcc */
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 8ad2e410974f..7c5538769f7e 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -1603,7 +1603,7 @@ static void __init delay_with_tsc(void)
do {
rep_nop();
now = rdtsc();
- } while ((now - start) < 40000000000UL / HZ &&
+ } while ((now - start) < 40000000000ULL / HZ &&
time_before_eq(jiffies, end));
}

diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index d71c8b54b696..bfca937bdcc3 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -300,6 +300,15 @@ static void __init spectre_v2_select_mitigation(void)
setup_force_cpu_cap(X86_FEATURE_USE_IBPB);
pr_info("Spectre v2 mitigation: Enabling Indirect Branch Prediction Barrier\n");
}
+
+ /*
+ * Retpoline means the kernel is safe because it has no indirect
+ * branches. But firmware isn't, so use IBRS to protect that.
+ */
+ if (boot_cpu_has(X86_FEATURE_IBRS)) {
+ setup_force_cpu_cap(X86_FEATURE_USE_IBRS_FW);
+ pr_info("Enabling Restricted Speculation for firmware calls\n");
+ }
}

#undef pr_fmt
@@ -326,8 +335,9 @@ ssize_t cpu_show_spectre_v2(struct device *dev, struct device_attribute *attr, c
if (!boot_cpu_has_bug(X86_BUG_SPECTRE_V2))
return sprintf(buf, "Not affected\n");

- return sprintf(buf, "%s%s%s\n", spectre_v2_strings[spectre_v2_enabled],
+ return sprintf(buf, "%s%s%s%s\n", spectre_v2_strings[spectre_v2_enabled],
boot_cpu_has(X86_FEATURE_USE_IBPB) ? ", IBPB" : "",
+ boot_cpu_has(X86_FEATURE_USE_IBRS_FW) ? ", IBRS_FW" : "",
spectre_v2_module_string());
}
#endif
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 824aee0117bb..348cf4821240 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -1749,3 +1749,33 @@ static int __init init_cpu_syscore(void)
return 0;
}
core_initcall(init_cpu_syscore);
+
+/*
+ * The microcode loader calls this upon late microcode load to recheck features,
+ * only when microcode has been updated. Caller holds microcode_mutex and CPU
+ * hotplug lock.
+ */
+void microcode_check(void)
+{
+ struct cpuinfo_x86 info;
+
+ perf_check_microcode();
+
+ /* Reload CPUID max function as it might've changed. */
+ info.cpuid_level = cpuid_eax(0);
+
+ /*
+ * Copy all capability leafs to pick up the synthetic ones so that
+ * memcmp() below doesn't fail on that. The ones coming from CPUID will
+ * get overwritten in get_cpu_cap().
+ */
+ memcpy(&info.x86_capability, &boot_cpu_data.x86_capability, sizeof(info.x86_capability));
+
+ get_cpu_cap(&info);
+
+ if (!memcmp(&info.x86_capability, &boot_cpu_data.x86_capability, sizeof(info.x86_capability)))
+ return;
+
+ pr_warn("x86/CPU: CPU features have changed after loading microcode, but might not take effect.\n");
+ pr_warn("x86/CPU: Please consider either early loading through initrd/built-in or a potential BIOS update.\n");
+}
diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c
index 330b8462d426..a998e1a7d46f 100644
--- a/arch/x86/kernel/cpu/microcode/amd.c
+++ b/arch/x86/kernel/cpu/microcode/amd.c
@@ -498,7 +498,7 @@ static unsigned int verify_patch_size(u8 family, u32 patch_size,
return patch_size;
}

-static int apply_microcode_amd(int cpu)
+static enum ucode_state apply_microcode_amd(int cpu)
{
struct cpuinfo_x86 *c = &cpu_data(cpu);
struct microcode_amd *mc_amd;
@@ -512,7 +512,7 @@ static int apply_microcode_amd(int cpu)

p = find_patch(cpu);
if (!p)
- return 0;
+ return UCODE_NFOUND;

mc_amd = p->data;
uci->mc = p->data;
@@ -523,13 +523,13 @@ static int apply_microcode_amd(int cpu)
if (rev >= mc_amd->hdr.patch_id) {
c->microcode = rev;
uci->cpu_sig.rev = rev;
- return 0;
+ return UCODE_OK;
}

if (__apply_microcode_amd(mc_amd)) {
pr_err("CPU%d: update failed for patch_level=0x%08x\n",
cpu, mc_amd->hdr.patch_id);
- return -1;
+ return UCODE_ERROR;
}
pr_info("CPU%d: new patch_level=0x%08x\n", cpu,
mc_amd->hdr.patch_id);
@@ -537,7 +537,7 @@ static int apply_microcode_amd(int cpu)
uci->cpu_sig.rev = mc_amd->hdr.patch_id;
c->microcode = mc_amd->hdr.patch_id;

- return 0;
+ return UCODE_UPDATED;
}

static int install_equiv_cpu_table(const u8 *buf)
diff --git a/arch/x86/kernel/cpu/microcode/core.c b/arch/x86/kernel/cpu/microcode/core.c
index 319dd65f98a2..aa1b9a422f2b 100644
--- a/arch/x86/kernel/cpu/microcode/core.c
+++ b/arch/x86/kernel/cpu/microcode/core.c
@@ -374,7 +374,7 @@ static int collect_cpu_info(int cpu)
}

struct apply_microcode_ctx {
- int err;
+ enum ucode_state err;
};

static void apply_microcode_local(void *arg)
@@ -489,31 +489,30 @@ static void __exit microcode_dev_exit(void)
/* fake device for request_firmware */
static struct platform_device *microcode_pdev;

-static int reload_for_cpu(int cpu)
+static enum ucode_state reload_for_cpu(int cpu)
{
struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
enum ucode_state ustate;
- int err = 0;

if (!uci->valid)
- return err;
+ return UCODE_OK;

ustate = microcode_ops->request_microcode_fw(cpu, &microcode_pdev->dev, true);
- if (ustate == UCODE_OK)
- apply_microcode_on_target(cpu);
- else
- if (ustate == UCODE_ERROR)
- err = -EINVAL;
- return err;
+ if (ustate != UCODE_OK)
+ return ustate;
+
+ return apply_microcode_on_target(cpu);
}

static ssize_t reload_store(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t size)
{
+ enum ucode_state tmp_ret = UCODE_OK;
+ bool do_callback = false;
unsigned long val;
+ ssize_t ret = 0;
int cpu;
- ssize_t ret = 0, tmp_ret;

ret = kstrtoul(buf, 0, &val);
if (ret)
@@ -526,15 +525,21 @@ static ssize_t reload_store(struct device *dev,
mutex_lock(&microcode_mutex);
for_each_online_cpu(cpu) {
tmp_ret = reload_for_cpu(cpu);
- if (tmp_ret != 0)
+ if (tmp_ret > UCODE_NFOUND) {
pr_warn("Error reloading microcode on CPU %d\n", cpu);

- /* save retval of the first encountered reload error */
- if (!ret)
- ret = tmp_ret;
+ /* set retval for the first encountered reload error */
+ if (!ret)
+ ret = -EINVAL;
+ }
+
+ if (tmp_ret == UCODE_UPDATED)
+ do_callback = true;
}
- if (!ret)
- perf_check_microcode();
+
+ if (!ret && do_callback)
+ microcode_check();
+
mutex_unlock(&microcode_mutex);
put_online_cpus();

diff --git a/arch/x86/kernel/cpu/microcode/intel.c b/arch/x86/kernel/cpu/microcode/intel.c
index a15db2b4e0d6..923054a6b760 100644
--- a/arch/x86/kernel/cpu/microcode/intel.c
+++ b/arch/x86/kernel/cpu/microcode/intel.c
@@ -772,7 +772,7 @@ static int collect_cpu_info(int cpu_num, struct cpu_signature *csig)
return 0;
}

-static int apply_microcode_intel(int cpu)
+static enum ucode_state apply_microcode_intel(int cpu)
{
struct microcode_intel *mc;
struct ucode_cpu_info *uci;
@@ -782,7 +782,7 @@ static int apply_microcode_intel(int cpu)

/* We should bind the task to the CPU */
if (WARN_ON(raw_smp_processor_id() != cpu))
- return -1;
+ return UCODE_ERROR;

uci = ucode_cpu_info + cpu;
mc = uci->mc;
@@ -790,7 +790,7 @@ static int apply_microcode_intel(int cpu)
/* Look for a newer patch in our cache: */
mc = find_patch(uci);
if (!mc)
- return 0;
+ return UCODE_NFOUND;
}

/* write microcode via MSR 0x79 */
@@ -801,7 +801,7 @@ static int apply_microcode_intel(int cpu)
if (rev != mc->hdr.rev) {
pr_err("CPU%d update to revision 0x%x failed\n",
cpu, mc->hdr.rev);
- return -1;
+ return UCODE_ERROR;
}

if (rev != prev_rev) {
@@ -818,7 +818,7 @@ static int apply_microcode_intel(int cpu)
uci->cpu_sig.rev = rev;
c->microcode = rev;

- return 0;
+ return UCODE_UPDATED;
}

static enum ucode_state generic_load_microcode(int cpu, void *data, size_t size,
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index 04a625f0fcda..0f545b3cf926 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -23,6 +23,7 @@
#include <asm/nops.h>
#include "../entry/calling.h"
#include <asm/export.h>
+#include <asm/nospec-branch.h>

#ifdef CONFIG_PARAVIRT
#include <asm/asm-offsets.h>
@@ -134,6 +135,7 @@ ENTRY(secondary_startup_64)

/* Ensure I am executing from virtual addresses */
movq $1f, %rax
+ ANNOTATE_RETPOLINE_SAFE
jmp *%rax
1:
UNWIND_HINT_EMPTY
diff --git a/arch/x86/kernel/unwind_orc.c b/arch/x86/kernel/unwind_orc.c
index 1f9188f5357c..feb28fee6cea 100644
--- a/arch/x86/kernel/unwind_orc.c
+++ b/arch/x86/kernel/unwind_orc.c
@@ -5,7 +5,6 @@
#include <asm/unwind.h>
#include <asm/orc_types.h>
#include <asm/orc_lookup.h>
-#include <asm/sections.h>

#define orc_warn(fmt, ...) \
printk_deferred_once(KERN_WARNING pr_fmt("WARNING: " fmt), ##__VA_ARGS__)
@@ -148,7 +147,7 @@ static struct orc_entry *orc_find(unsigned long ip)
}

/* vmlinux .init slow lookup: */
- if (ip >= (unsigned long)_sinittext && ip < (unsigned long)_einittext)
+ if (init_kernel_text(ip))
return __orc_find(__start_orc_unwind_ip, __start_orc_unwind,
__stop_orc_unwind_ip - __start_orc_unwind_ip, ip);

diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index b3e488a74828..24c9521ebc24 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -49,6 +49,7 @@
#include <asm/debugreg.h>
#include <asm/kvm_para.h>
#include <asm/irq_remapping.h>
+#include <asm/microcode.h>
#include <asm/nospec-branch.h>

#include <asm/virtext.h>
@@ -5355,7 +5356,7 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu)
* being speculatively taken.
*/
if (svm->spec_ctrl)
- wrmsrl(MSR_IA32_SPEC_CTRL, svm->spec_ctrl);
+ native_wrmsrl(MSR_IA32_SPEC_CTRL, svm->spec_ctrl);

asm volatile (
"push %%" _ASM_BP "; \n\t"
@@ -5464,11 +5465,11 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu)
* If the L02 MSR bitmap does not intercept the MSR, then we need to
* save it.
*/
- if (!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL))
- rdmsrl(MSR_IA32_SPEC_CTRL, svm->spec_ctrl);
+ if (unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL)))
+ svm->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL);

if (svm->spec_ctrl)
- wrmsrl(MSR_IA32_SPEC_CTRL, 0);
+ native_wrmsrl(MSR_IA32_SPEC_CTRL, 0);

/* Eliminate branch target predictions from guest mode */
vmexit_fill_RSB();
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 3dec126aa302..7f8401d05939 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -51,6 +51,7 @@
#include <asm/apic.h>
#include <asm/irq_remapping.h>
#include <asm/mmu_context.h>
+#include <asm/microcode.h>
#include <asm/nospec-branch.h>

#include "trace.h"
@@ -9452,7 +9453,7 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
* being speculatively taken.
*/
if (vmx->spec_ctrl)
- wrmsrl(MSR_IA32_SPEC_CTRL, vmx->spec_ctrl);
+ native_wrmsrl(MSR_IA32_SPEC_CTRL, vmx->spec_ctrl);

vmx->__launched = vmx->loaded_vmcs->launched;
asm(
@@ -9587,11 +9588,11 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
* If the L02 MSR bitmap does not intercept the MSR, then we need to
* save it.
*/
- if (!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL))
- rdmsrl(MSR_IA32_SPEC_CTRL, vmx->spec_ctrl);
+ if (unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL)))
+ vmx->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL);

if (vmx->spec_ctrl)
- wrmsrl(MSR_IA32_SPEC_CTRL, 0);
+ native_wrmsrl(MSR_IA32_SPEC_CTRL, 0);

/* Eliminate branch target predictions from guest mode */
vmexit_fill_RSB();
diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile
index 91e9700cc6dc..25a972c61b0a 100644
--- a/arch/x86/lib/Makefile
+++ b/arch/x86/lib/Makefile
@@ -28,7 +28,6 @@ lib-$(CONFIG_INSTRUCTION_DECODER) += insn.o inat.o insn-eval.o
lib-$(CONFIG_RANDOMIZE_BASE) += kaslr.o
lib-$(CONFIG_FUNCTION_ERROR_INJECTION) += error-inject.o
lib-$(CONFIG_RETPOLINE) += retpoline.o
-OBJECT_FILES_NON_STANDARD_retpoline.o :=y

obj-y += msr.o msr-reg.o msr-reg-export.o hweight.o

diff --git a/arch/x86/lib/retpoline.S b/arch/x86/lib/retpoline.S
index 480edc3a5e03..c909961e678a 100644
--- a/arch/x86/lib/retpoline.S
+++ b/arch/x86/lib/retpoline.S
@@ -7,7 +7,6 @@
#include <asm/alternative-asm.h>
#include <asm/export.h>
#include <asm/nospec-branch.h>
-#include <asm/bitsperlong.h>

.macro THUNK reg
.section .text.__x86.indirect_thunk
@@ -47,58 +46,3 @@ GENERATE_THUNK(r13)
GENERATE_THUNK(r14)
GENERATE_THUNK(r15)
#endif
-
-/*
- * Fill the CPU return stack buffer.
- *
- * Each entry in the RSB, if used for a speculative 'ret', contains an
- * infinite 'pause; lfence; jmp' loop to capture speculative execution.
- *
- * This is required in various cases for retpoline and IBRS-based
- * mitigations for the Spectre variant 2 vulnerability. Sometimes to
- * eliminate potentially bogus entries from the RSB, and sometimes
- * purely to ensure that it doesn't get empty, which on some CPUs would
- * allow predictions from other (unwanted!) sources to be used.
- *
- * Google experimented with loop-unrolling and this turned out to be
- * the optimal version - two calls, each with their own speculation
- * trap should their return address end up getting used, in a loop.
- */
-.macro STUFF_RSB nr:req sp:req
- mov $(\nr / 2), %_ASM_BX
- .align 16
-771:
- call 772f
-773: /* speculation trap */
- pause
- lfence
- jmp 773b
- .align 16
-772:
- call 774f
-775: /* speculation trap */
- pause
- lfence
- jmp 775b
- .align 16
-774:
- dec %_ASM_BX
- jnz 771b
- add $((BITS_PER_LONG/8) * \nr), \sp
-.endm
-
-#define RSB_FILL_LOOPS 16 /* To avoid underflow */
-
-ENTRY(__fill_rsb)
- STUFF_RSB RSB_FILL_LOOPS, %_ASM_SP
- ret
-END(__fill_rsb)
-EXPORT_SYMBOL_GPL(__fill_rsb)
-
-#define RSB_CLEAR_LOOPS 32 /* To forcibly overwrite all entries */
-
-ENTRY(__clear_rsb)
- STUFF_RSB RSB_CLEAR_LOOPS, %_ASM_SP
- ret
-END(__clear_rsb)
-EXPORT_SYMBOL_GPL(__clear_rsb)
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 800de815519c..c88573d90f3e 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -1248,10 +1248,6 @@ __do_page_fault(struct pt_regs *regs, unsigned long error_code,
tsk = current;
mm = tsk->mm;

- /*
- * Detect and handle instructions that would cause a page fault for
- * both a tracked kernel page and a userspace page.
- */
prefetchw(&mm->mmap_sem);

if (unlikely(kmmio_fault(regs, address)))
diff --git a/arch/x86/mm/mem_encrypt_boot.S b/arch/x86/mm/mem_encrypt_boot.S
index 01f682cf77a8..40a6085063d6 100644
--- a/arch/x86/mm/mem_encrypt_boot.S
+++ b/arch/x86/mm/mem_encrypt_boot.S
@@ -15,6 +15,7 @@
#include <asm/page.h>
#include <asm/processor-flags.h>
#include <asm/msr-index.h>
+#include <asm/nospec-branch.h>

.text
.code64
@@ -59,6 +60,7 @@ ENTRY(sme_encrypt_execute)
movq %rax, %r8 /* Workarea encryption routine */
addq $PAGE_SIZE, %r8 /* Workarea intermediate copy buffer */

+ ANNOTATE_RETPOLINE_SAFE
call *%rax /* Call the encryption routine */

pop %r12
diff --git a/arch/x86/realmode/rm/trampoline_64.S b/arch/x86/realmode/rm/trampoline_64.S
index de53bd15df5a..24bb7598774e 100644
--- a/arch/x86/realmode/rm/trampoline_64.S
+++ b/arch/x86/realmode/rm/trampoline_64.S
@@ -102,7 +102,7 @@ ENTRY(startup_32)
* don't we'll eventually crash trying to execute encrypted
* instructions.
*/
- bt $TH_FLAGS_SME_ACTIVE_BIT, pa_tr_flags
+ btl $TH_FLAGS_SME_ACTIVE_BIT, pa_tr_flags
jnc .Ldone
movl $MSR_K8_SYSCFG, %ecx
rdmsr
diff --git a/include/linux/compiler-clang.h b/include/linux/compiler-clang.h
index d02a4df3f473..d3f264a5b04d 100644
--- a/include/linux/compiler-clang.h
+++ b/include/linux/compiler-clang.h
@@ -27,3 +27,8 @@
#if __has_feature(address_sanitizer)
#define __SANITIZE_ADDRESS__
#endif
+
+/* Clang doesn't have a way to turn it off per-function, yet. */
+#ifdef __noretpoline
+#undef __noretpoline
+#endif
diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h
index 73bc63e0a1c4..673fbf904fe5 100644
--- a/include/linux/compiler-gcc.h
+++ b/include/linux/compiler-gcc.h
@@ -93,6 +93,10 @@
#define __weak __attribute__((weak))
#define __alias(symbol) __attribute__((alias(#symbol)))

+#ifdef RETPOLINE
+#define __noretpoline __attribute__((indirect_branch("keep")))
+#endif
+
/*
* it doesn't make sense on ARM (currently the only user of __naked)
* to trace naked functions because then mcount is called without
diff --git a/include/linux/init.h b/include/linux/init.h
index 506a98151131..bc27cf03c41e 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -6,10 +6,10 @@
#include <linux/types.h>

/* Built-in __init functions needn't be compiled with retpoline */
-#if defined(RETPOLINE) && !defined(MODULE)
-#define __noretpoline __attribute__((indirect_branch("keep")))
+#if defined(__noretpoline) && !defined(MODULE)
+#define __noinitretpoline __noretpoline
#else
-#define __noretpoline
+#define __noinitretpoline
#endif

/* These macros are used to mark some functions or
@@ -47,7 +47,7 @@

/* These are for everybody (although not all archs will actually
discard it in modules) */
-#define __init __section(.init.text) __cold __latent_entropy __noretpoline
+#define __init __section(.init.text) __cold __latent_entropy __noinitretpoline
#define __initdata __section(.init.data)
#define __initconst __section(.init.rodata)
#define __exitdata __section(.exit.data)
diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h
index b6a29c126cc4..2168cc6b8b30 100644
--- a/include/linux/jump_label.h
+++ b/include/linux/jump_label.h
@@ -151,6 +151,7 @@ extern struct jump_entry __start___jump_table[];
extern struct jump_entry __stop___jump_table[];

extern void jump_label_init(void);
+extern void jump_label_invalidate_init(void);
extern void jump_label_lock(void);
extern void jump_label_unlock(void);
extern void arch_jump_label_transform(struct jump_entry *entry,
@@ -198,6 +199,8 @@ static __always_inline void jump_label_init(void)
static_key_initialized = true;
}

+static inline void jump_label_invalidate_init(void) {}
+
static __always_inline bool static_key_false(struct static_key *key)
{
if (unlikely(static_key_count(key) > 0))
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index ce51455e2adf..3fd291503576 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -472,6 +472,7 @@ extern bool parse_option_str(const char *str, const char *option);
extern char *next_arg(char *args, char **param, char **val);

extern int core_kernel_text(unsigned long addr);
+extern int init_kernel_text(unsigned long addr);
extern int core_kernel_data(unsigned long addr);
extern int __kernel_text_address(unsigned long addr);
extern int kernel_text_address(unsigned long addr);
diff --git a/include/linux/nospec.h b/include/linux/nospec.h
index fbc98e2c8228..e791ebc65c9c 100644
--- a/include/linux/nospec.h
+++ b/include/linux/nospec.h
@@ -5,6 +5,7 @@

#ifndef _LINUX_NOSPEC_H
#define _LINUX_NOSPEC_H
+#include <asm/barrier.h>

/**
* array_index_mask_nospec() - generate a ~0 mask when index < size, 0 otherwise
@@ -29,26 +30,6 @@ static inline unsigned long array_index_mask_nospec(unsigned long index,
}
#endif

-/*
- * Warn developers about inappropriate array_index_nospec() usage.
- *
- * Even if the CPU speculates past the WARN_ONCE branch, the
- * sign bit of @index is taken into account when generating the
- * mask.
- *
- * This warning is compiled out when the compiler can infer that
- * @index and @size are less than LONG_MAX.
- */
-#define array_index_mask_nospec_check(index, size) \
-({ \
- if (WARN_ONCE(index > LONG_MAX || size > LONG_MAX, \
- "array_index_nospec() limited to range of [0, LONG_MAX]\n")) \
- _mask = 0; \
- else \
- _mask = array_index_mask_nospec(index, size); \
- _mask; \
-})
-
/*
* array_index_nospec - sanitize an array index after a bounds check
*
@@ -67,12 +48,11 @@ static inline unsigned long array_index_mask_nospec(unsigned long index,
({ \
typeof(index) _i = (index); \
typeof(size) _s = (size); \
- unsigned long _mask = array_index_mask_nospec_check(_i, _s); \
+ unsigned long _mask = array_index_mask_nospec(_i, _s); \
\
BUILD_BUG_ON(sizeof(_i) > sizeof(long)); \
BUILD_BUG_ON(sizeof(_s) > sizeof(long)); \
\
- _i &= _mask; \
- _i; \
+ (typeof(_i)) (_i & _mask); \
})
#endif /* _LINUX_NOSPEC_H */
diff --git a/init/main.c b/init/main.c
index a8100b954839..969eaf140ef0 100644
--- a/init/main.c
+++ b/init/main.c
@@ -89,6 +89,7 @@
#include <linux/io.h>
#include <linux/cache.h>
#include <linux/rodata_test.h>
+#include <linux/jump_label.h>

#include <asm/io.h>
#include <asm/bugs.h>
@@ -1000,6 +1001,7 @@ static int __ref kernel_init(void *unused)
/* need to finish all async __init code before freeing the memory */
async_synchronize_full();
ftrace_free_init_mem();
+ jump_label_invalidate_init();
free_initmem();
mark_readonly();
system_state = SYSTEM_RUNNING;
diff --git a/kernel/extable.c b/kernel/extable.c
index a17fdb63dc3e..6a5b61ebc66c 100644
--- a/kernel/extable.c
+++ b/kernel/extable.c
@@ -64,7 +64,7 @@ const struct exception_table_entry *search_exception_tables(unsigned long addr)
return e;
}

-static inline int init_kernel_text(unsigned long addr)
+int init_kernel_text(unsigned long addr)
{
if (addr >= (unsigned long)_sinittext &&
addr < (unsigned long)_einittext)
diff --git a/kernel/jump_label.c b/kernel/jump_label.c
index b4517095db6a..52a0a7af8640 100644
--- a/kernel/jump_label.c
+++ b/kernel/jump_label.c
@@ -366,12 +366,15 @@ static void __jump_label_update(struct static_key *key,
{
for (; (entry < stop) && (jump_entry_key(entry) == key); entry++) {
/*
- * entry->code set to 0 invalidates module init text sections
- * kernel_text_address() verifies we are not in core kernel
- * init code, see jump_label_invalidate_module_init().
+ * An entry->code of 0 indicates an entry which has been
+ * disabled because it was in an init text area.
*/
- if (entry->code && kernel_text_address(entry->code))
- arch_jump_label_transform(entry, jump_label_type(entry));
+ if (entry->code) {
+ if (kernel_text_address(entry->code))
+ arch_jump_label_transform(entry, jump_label_type(entry));
+ else
+ WARN_ONCE(1, "can't patch jump_label at %pS", (void *)entry->code);
+ }
}
}

@@ -417,6 +420,19 @@ void __init jump_label_init(void)
cpus_read_unlock();
}

+/* Disable any jump label entries in __init code */
+void __init jump_label_invalidate_init(void)
+{
+ struct jump_entry *iter_start = __start___jump_table;
+ struct jump_entry *iter_stop = __stop___jump_table;
+ struct jump_entry *iter;
+
+ for (iter = iter_start; iter < iter_stop; iter++) {
+ if (init_kernel_text(iter->code))
+ iter->code = 0;
+ }
+}
+
#ifdef CONFIG_MODULES

static enum jump_label_type jump_label_init_type(struct jump_entry *entry)
@@ -633,6 +649,7 @@ static void jump_label_del_module(struct module *mod)
}
}

+/* Disable any jump label entries in module init code */
static void jump_label_invalidate_module_init(struct module *mod)
{
struct jump_entry *iter_start = mod->jump_entries;
diff --git a/scripts/Makefile.build b/scripts/Makefile.build
index 47cddf32aeba..4f2b25d43ec9 100644
--- a/scripts/Makefile.build
+++ b/scripts/Makefile.build
@@ -256,6 +256,8 @@ __objtool_obj := $(objtree)/tools/objtool/objtool

objtool_args = $(if $(CONFIG_UNWINDER_ORC),orc generate,check)

+objtool_args += $(if $(part-of-module), --module,)
+
ifndef CONFIG_FRAME_POINTER
objtool_args += --no-fp
endif
@@ -264,6 +266,12 @@ objtool_args += --no-unreachable
else
objtool_args += $(call cc-ifversion, -lt, 0405, --no-unreachable)
endif
+ifdef CONFIG_RETPOLINE
+ifneq ($(RETPOLINE_CFLAGS),)
+ objtool_args += --retpoline
+endif
+endif
+

ifdef CONFIG_MODVERSIONS
objtool_o = $(@D)/.tmp_$(@F)
diff --git a/tools/objtool/builtin-check.c b/tools/objtool/builtin-check.c
index 57254f5b2779..694abc628e9b 100644
--- a/tools/objtool/builtin-check.c
+++ b/tools/objtool/builtin-check.c
@@ -29,7 +29,7 @@
#include "builtin.h"
#include "check.h"

-bool no_fp, no_unreachable;
+bool no_fp, no_unreachable, retpoline, module;

static const char * const check_usage[] = {
"objtool check [<options>] file.o",
@@ -39,6 +39,8 @@ static const char * const check_usage[] = {
const struct option check_options[] = {
OPT_BOOLEAN('f', "no-fp", &no_fp, "Skip frame pointer validation"),
OPT_BOOLEAN('u', "no-unreachable", &no_unreachable, "Skip 'unreachable instruction' warnings"),
+ OPT_BOOLEAN('r', "retpoline", &retpoline, "Validate retpoline assumptions"),
+ OPT_BOOLEAN('m', "module", &module, "Indicates the object will be part of a kernel module"),
OPT_END(),
};

@@ -53,5 +55,5 @@ int cmd_check(int argc, const char **argv)

objname = argv[0];

- return check(objname, no_fp, no_unreachable, false);
+ return check(objname, false);
}
diff --git a/tools/objtool/builtin-orc.c b/tools/objtool/builtin-orc.c
index 91e8e19ff5e0..77ea2b97117d 100644
--- a/tools/objtool/builtin-orc.c
+++ b/tools/objtool/builtin-orc.c
@@ -25,7 +25,6 @@
*/

#include <string.h>
-#include <subcmd/parse-options.h>
#include "builtin.h"
#include "check.h"

@@ -36,9 +35,6 @@ static const char *orc_usage[] = {
NULL,
};

-extern const struct option check_options[];
-extern bool no_fp, no_unreachable;
-
int cmd_orc(int argc, const char **argv)
{
const char *objname;
@@ -54,7 +50,7 @@ int cmd_orc(int argc, const char **argv)

objname = argv[0];

- return check(objname, no_fp, no_unreachable, true);
+ return check(objname, true);
}

if (!strcmp(argv[0], "dump")) {
diff --git a/tools/objtool/builtin.h b/tools/objtool/builtin.h
index dd526067fed5..28ff40e19a14 100644
--- a/tools/objtool/builtin.h
+++ b/tools/objtool/builtin.h
@@ -17,6 +17,11 @@
#ifndef _BUILTIN_H
#define _BUILTIN_H

+#include <subcmd/parse-options.h>
+
+extern const struct option check_options[];
+extern bool no_fp, no_unreachable, retpoline, module;
+
extern int cmd_check(int argc, const char **argv);
extern int cmd_orc(int argc, const char **argv);

diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index a8cb69a26576..472e64e95891 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -18,6 +18,7 @@
#include <string.h>
#include <stdlib.h>

+#include "builtin.h"
#include "check.h"
#include "elf.h"
#include "special.h"
@@ -33,7 +34,6 @@ struct alternative {
};

const char *objname;
-static bool no_fp;
struct cfi_state initial_func_cfi;

struct instruction *find_insn(struct objtool_file *file,
@@ -497,6 +497,7 @@ static int add_jump_destinations(struct objtool_file *file)
* disguise, so convert them accordingly.
*/
insn->type = INSN_JUMP_DYNAMIC;
+ insn->retpoline_safe = true;
continue;
} else {
/* sibling call */
@@ -548,7 +549,8 @@ static int add_call_destinations(struct objtool_file *file)
if (!insn->call_dest && !insn->ignore) {
WARN_FUNC("unsupported intra-function call",
insn->sec, insn->offset);
- WARN("If this is a retpoline, please patch it in with alternatives and annotate it with ANNOTATE_NOSPEC_ALTERNATIVE.");
+ if (retpoline)
+ WARN("If this is a retpoline, please patch it in with alternatives and annotate it with ANNOTATE_NOSPEC_ALTERNATIVE.");
return -1;
}

@@ -1108,6 +1110,54 @@ static int read_unwind_hints(struct objtool_file *file)
return 0;
}

+static int read_retpoline_hints(struct objtool_file *file)
+{
+ struct section *sec, *relasec;
+ struct instruction *insn;
+ struct rela *rela;
+ int i;
+
+ sec = find_section_by_name(file->elf, ".discard.retpoline_safe");
+ if (!sec)
+ return 0;
+
+ relasec = sec->rela;
+ if (!relasec) {
+ WARN("missing .rela.discard.retpoline_safe section");
+ return -1;
+ }
+
+ if (sec->len % sizeof(unsigned long)) {
+ WARN("retpoline_safe size mismatch: %d %ld", sec->len, sizeof(unsigned long));
+ return -1;
+ }
+
+ for (i = 0; i < sec->len / sizeof(unsigned long); i++) {
+ rela = find_rela_by_dest(sec, i * sizeof(unsigned long));
+ if (!rela) {
+ WARN("can't find rela for retpoline_safe[%d]", i);
+ return -1;
+ }
+
+ insn = find_insn(file, rela->sym->sec, rela->addend);
+ if (!insn) {
+ WARN("can't find insn for retpoline_safe[%d]", i);
+ return -1;
+ }
+
+ if (insn->type != INSN_JUMP_DYNAMIC &&
+ insn->type != INSN_CALL_DYNAMIC) {
+ WARN_FUNC("retpoline_safe hint not a indirect jump/call",
+ insn->sec, insn->offset);
+ return -1;
+ }
+
+ insn->retpoline_safe = true;
+ }
+
+ return 0;
+}
+
static int decode_sections(struct objtool_file *file)
{
int ret;
@@ -1146,6 +1196,10 @@ static int decode_sections(struct objtool_file *file)
if (ret)
return ret;

+ ret = read_retpoline_hints(file);
+ if (ret)
+ return ret;
+
return 0;
}

@@ -1891,6 +1945,38 @@ static int validate_unwind_hints(struct objtool_file *file)
return warnings;
}

+static int validate_retpoline(struct objtool_file *file)
+{
+ struct instruction *insn;
+ int warnings = 0;
+
+ for_each_insn(file, insn) {
+ if (insn->type != INSN_JUMP_DYNAMIC &&
+ insn->type != INSN_CALL_DYNAMIC)
+ continue;
+
+ if (insn->retpoline_safe)
+ continue;
+
+ /*
+ * .init.text code is ran before userspace and thus doesn't
+ * strictly need retpolines, except for modules which are
+ * loaded late, they very much do need retpoline in their
+ * .init.text
+ */
+ if (!strcmp(insn->sec->name, ".init.text") && !module)
+ continue;
+
+ WARN_FUNC("indirect %s found in RETPOLINE build",
+ insn->sec, insn->offset,
+ insn->type == INSN_JUMP_DYNAMIC ? "jump" : "call");
+
+ warnings++;
+ }
+
+ return warnings;
+}
+
static bool is_kasan_insn(struct instruction *insn)
{
return (insn->type == INSN_CALL &&
@@ -2022,13 +2108,12 @@ static void cleanup(struct objtool_file *file)
elf_close(file->elf);
}

-int check(const char *_objname, bool _no_fp, bool no_unreachable, bool orc)
+int check(const char *_objname, bool orc)
{
struct objtool_file file;
int ret, warnings = 0;

objname = _objname;
- no_fp = _no_fp;

file.elf = elf_open(objname, orc ? O_RDWR : O_RDONLY);
if (!file.elf)
@@ -2052,6 +2137,13 @@ int check(const char *_objname, bool _no_fp, bool no_unreachable, bool orc)
if (list_empty(&file.insn_list))
goto out;

+ if (retpoline) {
+ ret = validate_retpoline(&file);
+ if (ret < 0)
+ return ret;
+ warnings += ret;
+ }
+
ret = validate_functions(&file);
if (ret < 0)
goto out;
diff --git a/tools/objtool/check.h b/tools/objtool/check.h
index 23a1d065cae1..c6b68fcb926f 100644
--- a/tools/objtool/check.h
+++ b/tools/objtool/check.h
@@ -45,6 +45,7 @@ struct instruction {
unsigned char type;
unsigned long immediate;
bool alt_group, visited, dead_end, ignore, hint, save, restore, ignore_alts;
+ bool retpoline_safe;
struct symbol *call_dest;
struct instruction *jump_dest;
struct instruction *first_jump_src;
@@ -63,7 +64,7 @@ struct objtool_file {
bool ignore_unreachables, c_file, hints;
};

-int check(const char *objname, bool no_fp, bool no_unreachable, bool orc);
+int check(const char *objname, bool orc);

struct instruction *find_insn(struct objtool_file *file,
struct section *sec, unsigned long offset);