[PATCH 18/21] x86/entry/64: Create a per-CPU SYSCALL entry trampoline

From: Ingo Molnar
Date: Mon Nov 27 2017 - 05:47:26 EST

From: Andy Lutomirski <luto@xxxxxxxxxx>

Handling SYSCALL is tricky: the SYSCALL handler is entered with every
single register (except FLAGS), including RSP, live. It somehow needs
to set RSP to point to a valid stack, which means it needs to save the
user RSP somewhere and find its own stack pointer. The canonical way
to do this is with SWAPGS, which lets us access percpu data using the
%gs prefix.

With KAISER-like pagetable switching, this is problematic. Without a
scratch register, switching CR3 is impossible, so %gs-based percpu
memory would need to be mapped in the user pagetables. Doing that
without information leaks is difficult or impossible.

Instead, use a different sneaky trick. Map a copy of the first part
of the SYSCALL asm at a different address for each CPU. Now RIP
varies depending on the CPU, so we can use RIP-relative memory access
to access percpu memory. By putting the relevant information (one
scratch slot and the stack address) at a constant offset relative to
RIP, we can make SYSCALL work without relying on %gs.

A nice thing about this approach is that we can easily switch it on
and off if we want pagetable switching to be configurable.

The compat variant of SYSCALL doesn't have this problem in the first
place -- there are plenty of scratch registers, since we don't care
about preserving r8-r15. This patch therefore doesn't touch SYSCALL32
at all.

XXX: Whenever we settle how KAISER gets turned on and off, we should do
the same to this.

Signed-off-by: Andy Lutomirski <luto@xxxxxxxxxx>
Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Reviewed-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Reviewed-by: Borislav Petkov <bpetkov@xxxxxxx>
Cc: Brian Gerst <brgerst@xxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
Cc: Denys Vlasenko <dvlasenk@xxxxxxxxxx>
Cc: H. Peter Anvin <hpa@xxxxxxxxx>
Cc: Josh Poimboeuf <jpoimboe@xxxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxx>
Link: https://lkml.kernel.org/r/b95ccae0a5a2f090c901e49fce7c9e8ff6acd40d.1511497875.git.luto@xxxxxxxxxx
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
---
arch/x86/entry/entry_64.S | 48 +++++++++++++++++++++++++++++++++++++++++++
arch/x86/include/asm/fixmap.h | 2 ++
arch/x86/kernel/asm-offsets.c | 1 +
arch/x86/kernel/cpu/common.c | 15 +++++++++++++-
arch/x86/kernel/vmlinux.lds.S | 9 ++++++++
5 files changed, 74 insertions(+), 1 deletion(-)

diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index 426b8c669d6a..5e674b6362e3 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -140,6 +140,54 @@ END(native_usergs_sysret64)
* with them due to bugs in both AMD and Intel CPUs.
*/

+ .pushsection .entry_trampoline, "ax"
+
+/*
+ * The code in here gets remapped into cpu_entry_area's trampoline. This means
+ * that the assembler and linker have the wrong idea as to where this code
+ * lives (and, in fact, it's mapped more than once, so it's not even at a
+ * fixed address). So we can't reference any symbols outside the entry
+ * trampoline and expect it to work.
+ *
+ * Instead, we carefully abuse %rip-relative addressing.
+ * _entry_trampoline(%rip) refers to the start of the remapped) entry
+ * trampoline. We can thus find cpu_entry_area with this macro:
+ */
+
+#define CPU_ENTRY_AREA \
+ _entry_trampoline - CPU_ENTRY_AREA_entry_trampoline(%rip)
+
+/* The top word of the SYSENTER stack is hot and is usable as scratch space. */
+#define RSP_SCRATCH CPU_ENTRY_AREA_tss + CPU_TSS_SYSENTER_stack + \
+ SIZEOF_SYSENTER_stack - 8 + CPU_ENTRY_AREA
+
+ENTRY(entry_SYSCALL_64_trampoline)
+ UNWIND_HINT_EMPTY
+ swapgs
+
+ /* Stash the user RSP. */
+ movq %rsp, RSP_SCRATCH
+
+ /* Load the top of the task stack into RSP */
+ movq CPU_ENTRY_AREA_tss + TSS_sp1 + CPU_ENTRY_AREA, %rsp
+
+ /* Start building the simulated IRET frame. */
+ pushq $__USER_DS /* pt_regs->ss */
+ pushq RSP_SCRATCH /* pt_regs->sp */
+ pushq %r11 /* pt_regs->flags */
+ pushq $__USER_CS /* pt_regs->cs */
+ pushq %rcx /* pt_regs->ip */
+
+ /*
+ * x86 lacks a near absolute jump, and we can't jump to the real
+ * entry text with a relative jump, so we fake it using retq.
+ */
+ pushq $entry_SYSCALL_64_after_hwframe
+ retq
+END(entry_SYSCALL_64_trampoline)
+
+ .popsection
+
ENTRY(entry_SYSCALL_64)
UNWIND_HINT_EMPTY
/*
diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h
index c8aaf4b56699..17be88e73c92 100644
--- a/arch/x86/include/asm/fixmap.h
+++ b/arch/x86/include/asm/fixmap.h
@@ -61,6 +61,8 @@ struct cpu_entry_area {
* of the TSS region.
*/
struct tss_struct tss;
+
+ char entry_trampoline[PAGE_SIZE];
};

#define CPU_ENTRY_AREA_PAGES (sizeof(struct cpu_entry_area) / PAGE_SIZE)
diff --git a/arch/x86/kernel/asm-offsets.c b/arch/x86/kernel/asm-offsets.c
index 55858b277cf6..61b1af88ac07 100644
--- a/arch/x86/kernel/asm-offsets.c
+++ b/arch/x86/kernel/asm-offsets.c
@@ -101,4 +101,5 @@ void common(void) {

/* Layout info for cpu_entry_area */
OFFSET(CPU_ENTRY_AREA_tss, cpu_entry_area, tss);
+ OFFSET(CPU_ENTRY_AREA_entry_trampoline, cpu_entry_area, entry_trampoline);
}
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 7c82a8a8bfda..9af1f176cc5b 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -507,6 +507,8 @@ DEFINE_PER_CPU(struct cpu_entry_area *, cpu_entry_area);
static inline void setup_cpu_entry_area(int cpu)
{
#ifdef CONFIG_X86_64
+ extern char _entry_trampoline[];
+
/* On 64-bit systems, we use a read-only fixmap GDT. */
pgprot_t gdt_prot = PAGE_KERNEL_RO;
#else
@@ -553,6 +555,11 @@ static inline void setup_cpu_entry_area(int cpu)
#ifdef CONFIG_X86_32
this_cpu_write(cpu_entry_area, get_cpu_entry_area(cpu));
#endif
+
+#ifdef CONFIG_X86_64
+ __set_fixmap(get_cpu_entry_area_index(cpu, entry_trampoline),
+ __pa_symbol(_entry_trampoline), PAGE_KERNEL_RX);
+#endif
}

/* Load the original GDT from the per-cpu structure */
@@ -1417,10 +1424,16 @@ static DEFINE_PER_CPU_PAGE_ALIGNED(char, exception_stacks
/* May not be marked __init: used by software suspend */
void syscall_init(void)
{
+ extern char _entry_trampoline[];
+ extern char entry_SYSCALL_64_trampoline[];
+
int cpu = smp_processor_id();
+ unsigned long SYSCALL64_entry_trampoline =
+ (unsigned long)get_cpu_entry_area(cpu)->entry_trampoline +
+ (entry_SYSCALL_64_trampoline - _entry_trampoline);

wrmsr(MSR_STAR, 0, (__USER32_CS << 16) | __KERNEL_CS);
- wrmsrl(MSR_LSTAR, (unsigned long)entry_SYSCALL_64);
+ wrmsrl(MSR_LSTAR, SYSCALL64_entry_trampoline);

#ifdef CONFIG_IA32_EMULATION
wrmsrl(MSR_CSTAR, (unsigned long)entry_SYSCALL_compat);
diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S
index a4009fb9be87..d2a8b5a24a44 100644
--- a/arch/x86/kernel/vmlinux.lds.S
+++ b/arch/x86/kernel/vmlinux.lds.S
@@ -107,6 +107,15 @@ SECTIONS
SOFTIRQENTRY_TEXT
*(.fixup)
*(.gnu.warning)
+
+#ifdef CONFIG_X86_64
+ . = ALIGN(PAGE_SIZE);
+ _entry_trampoline = .;
+ *(.entry_trampoline)
+ . = ALIGN(PAGE_SIZE);
+ ASSERT(. - _entry_trampoline == PAGE_SIZE, "entry trampoline is too big");
+#endif
+
/* End of text section */
_etext = .;
} :text = 0x9090
--
2.14.1