[PATCHv2 12/14] arm64: add basic VMAP_STACK support

From: Mark Rutland <mark.rutland@arm.com>
To: linux-arm-kernel@lists.infradead.org
Cc: ard.biesheuvel@linaro.org, catalin.marinas@arm.com,
	james.morse@arm.com, labbott@redhat.com,
	linux-kernel@vger.kernel.org, luto@amacapital.net,
	mark.rutland@arm.com, matt@codeblueprint.co.uk,
	will.deacon@arm.com, kernel-hardening@lists.openwall.com,
	keescook@chromium.org
Subject: [PATCHv2 12/14] arm64: add basic VMAP_STACK support
Date: Tue, 15 Aug 2017 13:50:47 +0100	[thread overview]
Message-ID: <1502801449-29246-13-git-send-email-mark.rutland@arm.com> (raw)
In-Reply-To: <1502801449-29246-1-git-send-email-mark.rutland@arm.com>

This patch enables arm64 to be built with vmap'd task and IRQ stacks.

As vmap'd stacks are mapped at page granularity, stacks must be a multiple of
PAGE_SIZE. This means that a 64K page kernel must use stacks of at least 64K in
size.

To minimize the increase in Image size, IRQ stacks are dynamically allocated at
boot time, rather than embedding the boot CPU's IRQ stack in the kernel image.

This patch was co-authored by Ard Biesheuvel and Mark Rutland.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Laura Abbott <labbott@redhat.com>
Cc: Will Deacon <will.deacon@arm.com>
---
 arch/arm64/Kconfig              |  1 +
 arch/arm64/include/asm/efi.h    |  7 ++++++-
 arch/arm64/include/asm/memory.h | 23 ++++++++++++++++++++++-
 arch/arm64/kernel/irq.c         | 30 ++++++++++++++++++++++++++++--
 arch/arm64/kernel/vmlinux.lds.S |  2 +-
 5 files changed, 58 insertions(+), 5 deletions(-)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index dfd9086..d66f9db 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -75,6 +75,7 @@ config ARM64
 	select HAVE_ARCH_SECCOMP_FILTER
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE
+	select HAVE_ARCH_VMAP_STACK
 	select HAVE_ARM_SMCCC
 	select HAVE_EBPF_JIT
 	select HAVE_C_RECORDMCOUNT
diff --git a/arch/arm64/include/asm/efi.h b/arch/arm64/include/asm/efi.h
index 0e8cc3b..2b1e5de 100644
--- a/arch/arm64/include/asm/efi.h
+++ b/arch/arm64/include/asm/efi.h
@@ -49,7 +49,12 @@
  */
 #define EFI_FDT_ALIGN	SZ_2M   /* used by allocate_new_fdt_and_exit_boot() */
 
-#define EFI_KIMG_ALIGN	SEGMENT_ALIGN
+/*
+ * In some configurations (e.g. VMAP_STACK && 64K pages), stacks built into the
+ * kernel need greater alignment than we require the segments to be padded to.
+ */
+#define EFI_KIMG_ALIGN	\
+	(SEGMENT_ALIGN > THREAD_ALIGN ? SEGMENT_ALIGN : THREAD_ALIGN)
 
 /* on arm64, the FDT may be located anywhere in system RAM */
 static inline unsigned long efi_get_max_fdt_addr(unsigned long dram_base)
diff --git a/arch/arm64/include/asm/memory.h b/arch/arm64/include/asm/memory.h
index 7fa6ad4..c5cd2c5 100644
--- a/arch/arm64/include/asm/memory.h
+++ b/arch/arm64/include/asm/memory.h
@@ -102,7 +102,17 @@
 #define KASAN_SHADOW_SIZE	(0)
 #endif
 
-#define THREAD_SHIFT		14
+#define MIN_THREAD_SHIFT	14
+
+/*
+ * VMAP'd stacks are allocated at page granularity, so we must ensure that such
+ * stacks are a multiple of page size.
+ */
+#if defined(CONFIG_VMAP_STACK) && (MIN_THREAD_SHIFT < PAGE_SHIFT)
+#define THREAD_SHIFT		PAGE_SHIFT
+#else
+#define THREAD_SHIFT		MIN_THREAD_SHIFT
+#endif
 
 #if THREAD_SHIFT >= PAGE_SHIFT
 #define THREAD_SIZE_ORDER	(THREAD_SHIFT - PAGE_SHIFT)
@@ -110,6 +120,17 @@
 
 #define THREAD_SIZE		(UL(1) << THREAD_SHIFT)
 
+/*
+ * By aligning VMAP'd stacks to 2 * THREAD_SIZE, we can detect overflow by
+ * checking sp & (1 << THREAD_SHIFT), which we can do cheaply in the entry
+ * assembly.
+ */
+#ifdef CONFIG_VMAP_STACK
+#define THREAD_ALIGN		(2 * THREAD_SIZE)
+#else
+#define THREAD_ALIGN		THREAD_SIZE
+#endif
+
 #define IRQ_STACK_SIZE		THREAD_SIZE
 
 /*
diff --git a/arch/arm64/kernel/irq.c b/arch/arm64/kernel/irq.c
index 5141282..713561e 100644
--- a/arch/arm64/kernel/irq.c
+++ b/arch/arm64/kernel/irq.c
@@ -23,15 +23,15 @@
 
 #include <linux/kernel_stat.h>
 #include <linux/irq.h>
+#include <linux/memory.h>
 #include <linux/smp.h>
 #include <linux/init.h>
 #include <linux/irqchip.h>
 #include <linux/seq_file.h>
+#include <linux/vmalloc.h>
 
 unsigned long irq_err_count;
 
-/* irq stack only needs to be 16 byte aligned - not IRQ_STACK_SIZE aligned. */
-DEFINE_PER_CPU(unsigned long [IRQ_STACK_SIZE/sizeof(long)], irq_stack) __aligned(16);
 DEFINE_PER_CPU(unsigned long *, irq_stack_ptr);
 
 int arch_show_interrupts(struct seq_file *p, int prec)
@@ -51,6 +51,31 @@ void __init set_handle_irq(void (*handle_irq)(struct pt_regs *))
 	handle_arch_irq = handle_irq;
 }
 
+#ifdef CONFIG_VMAP_STACK
+static void init_irq_stacks(void)
+{
+	int cpu;
+	unsigned long *p;
+
+	for_each_possible_cpu(cpu) {
+		/*
+		* To ensure that VMAP'd stack overflow detection works
+		* correctly, the IRQ stacks need to have the same
+		* alignment as other stacks.
+		*/
+		p = __vmalloc_node_range(IRQ_STACK_SIZE, THREAD_ALIGN,
+					 VMALLOC_START, VMALLOC_END,
+					 THREADINFO_GFP, PAGE_KERNEL,
+					 0, cpu_to_node(cpu),
+					 __builtin_return_address(0));
+
+		per_cpu(irq_stack_ptr, cpu) = p;
+	}
+}
+#else
+/* irq stack only needs to be 16 byte aligned - not IRQ_STACK_SIZE aligned. */
+DEFINE_PER_CPU_ALIGNED(unsigned long [IRQ_STACK_SIZE/sizeof(long)], irq_stack);
+
 static void init_irq_stacks(void)
 {
 	int cpu;
@@ -58,6 +83,7 @@ static void init_irq_stacks(void)
 	for_each_possible_cpu(cpu)
 		per_cpu(irq_stack_ptr, cpu) = per_cpu(irq_stack, cpu);
 }
+#endif
 
 void __init init_IRQ(void)
 {
diff --git a/arch/arm64/kernel/vmlinux.lds.S b/arch/arm64/kernel/vmlinux.lds.S
index 7156538..fe56c26 100644
--- a/arch/arm64/kernel/vmlinux.lds.S
+++ b/arch/arm64/kernel/vmlinux.lds.S
@@ -176,7 +176,7 @@ SECTIONS
 
 	_data = .;
 	_sdata = .;
-	RW_DATA_SECTION(L1_CACHE_BYTES, PAGE_SIZE, THREAD_SIZE)
+	RW_DATA_SECTION(L1_CACHE_BYTES, PAGE_SIZE, THREAD_ALIGN)
 
 	/*
 	 * Data written with the MMU off but read with the MMU on requires
-- 
1.9.1
