[PATCH v3 0/3] kasan: add memory corruption identification support for hw tag-based kasan

[PATCH v3 0/3] kasan: add memory corruption identification support for hw tag-based kasan

[Kuan-Ying Lee]

Add memory corruption identification for hardware tag-based KASAN mode.

Changes since v3:
 - Preserve Copyright from hw_tags.c/sw_tags.c and
   report_sw_tags.c/report_hw_tags.c
 - Make non-trivial change in kasan sw tag-based mode

Changes since v2:
 - Thanks for Marco's Suggestion
 - Rename the CONFIG_KASAN_SW_TAGS_IDENTIFY
 - Integrate tag-based kasan common part
 - Rebase to latest linux-next

Kuan-Ying Lee (3):
  kasan: rename CONFIG_KASAN_SW_TAGS_IDENTIFY to
    CONFIG_KASAN_TAGS_IDENTIFY
  kasan: integrate the common part of two KASAN tag-based modes
  kasan: add memory corruption identification support for hardware
    tag-based mode

 lib/Kconfig.kasan         |  4 +--
 mm/kasan/Makefile         |  4 +--
 mm/kasan/hw_tags.c        | 22 ---------------
 mm/kasan/kasan.h          |  4 +--
 mm/kasan/report_hw_tags.c |  6 +---
 mm/kasan/report_sw_tags.c | 46 +-----------------------------
 mm/kasan/report_tags.h    | 55 ++++++++++++++++++++++++++++++++++++
 mm/kasan/sw_tags.c        | 41 ---------------------------
 mm/kasan/tags.c           | 59 +++++++++++++++++++++++++++++++++++++++
 9 files changed, 122 insertions(+), 119 deletions(-)
 create mode 100644 mm/kasan/report_tags.h
 create mode 100644 mm/kasan/tags.c

-- 
2.18.0
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v3 1/3] kasan: rename CONFIG_KASAN_SW_TAGS_IDENTIFY to CONFIG_KASAN_TAGS_IDENTIFY

[Kuan-Ying Lee]

This patch renames CONFIG_KASAN_SW_TAGS_IDENTIFY to
CONFIG_KASAN_TAGS_IDENTIFY in order to be compatible
with hardware tag-based mode.

Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
Suggested-by: Marco Elver <elver@google.com>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andrey Konovalov <andreyknvl@gmail.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
---
 lib/Kconfig.kasan         | 2 +-
 mm/kasan/kasan.h          | 4 ++--
 mm/kasan/report_sw_tags.c | 2 +-
 mm/kasan/sw_tags.c        | 4 ++--
 4 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan
index cffc2ebbf185..6f5d48832139 100644
--- a/lib/Kconfig.kasan
+++ b/lib/Kconfig.kasan
@@ -155,7 +155,7 @@ config KASAN_STACK
 	  CONFIG_COMPILE_TEST.	On gcc it is assumed to always be safe
 	  to use and enabled by default.
 
-config KASAN_SW_TAGS_IDENTIFY
+config KASAN_TAGS_IDENTIFY
 	bool "Enable memory corruption identification"
 	depends on KASAN_SW_TAGS
 	help
diff --git a/mm/kasan/kasan.h b/mm/kasan/kasan.h
index 8f450bc28045..b0fc9a1eb7e3 100644
--- a/mm/kasan/kasan.h
+++ b/mm/kasan/kasan.h
@@ -153,7 +153,7 @@ struct kasan_track {
 	depot_stack_handle_t stack;
 };
 
-#ifdef CONFIG_KASAN_SW_TAGS_IDENTIFY
+#ifdef CONFIG_KASAN_TAGS_IDENTIFY
 #define KASAN_NR_FREE_STACKS 5
 #else
 #define KASAN_NR_FREE_STACKS 1
@@ -170,7 +170,7 @@ struct kasan_alloc_meta {
 #else
 	struct kasan_track free_track[KASAN_NR_FREE_STACKS];
 #endif
-#ifdef CONFIG_KASAN_SW_TAGS_IDENTIFY
+#ifdef CONFIG_KASAN_TAGS_IDENTIFY
 	u8 free_pointer_tag[KASAN_NR_FREE_STACKS];
 	u8 free_track_idx;
 #endif
diff --git a/mm/kasan/report_sw_tags.c b/mm/kasan/report_sw_tags.c
index 3d20d3451d9e..821a14a19a92 100644
--- a/mm/kasan/report_sw_tags.c
+++ b/mm/kasan/report_sw_tags.c
@@ -31,7 +31,7 @@
 
 const char *kasan_get_bug_type(struct kasan_access_info *info)
 {
-#ifdef CONFIG_KASAN_SW_TAGS_IDENTIFY
+#ifdef CONFIG_KASAN_TAGS_IDENTIFY
 	struct kasan_alloc_meta *alloc_meta;
 	struct kmem_cache *cache;
 	struct page *page;
diff --git a/mm/kasan/sw_tags.c b/mm/kasan/sw_tags.c
index 9362938abbfa..dd05e6c801fa 100644
--- a/mm/kasan/sw_tags.c
+++ b/mm/kasan/sw_tags.c
@@ -177,7 +177,7 @@ void kasan_set_free_info(struct kmem_cache *cache,
 	if (!alloc_meta)
 		return;
 
-#ifdef CONFIG_KASAN_SW_TAGS_IDENTIFY
+#ifdef CONFIG_KASAN_TAGS_IDENTIFY
 	idx = alloc_meta->free_track_idx;
 	alloc_meta->free_pointer_tag[idx] = tag;
 	alloc_meta->free_track_idx = (idx + 1) % KASAN_NR_FREE_STACKS;
@@ -196,7 +196,7 @@ struct kasan_track *kasan_get_free_track(struct kmem_cache *cache,
 	if (!alloc_meta)
 		return NULL;
 
-#ifdef CONFIG_KASAN_SW_TAGS_IDENTIFY
+#ifdef CONFIG_KASAN_TAGS_IDENTIFY
 	for (i = 0; i < KASAN_NR_FREE_STACKS; i++) {
 		if (alloc_meta->free_pointer_tag[i] == tag)
 			break;
-- 
2.18.0
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 1/3] kasan: rename CONFIG_KASAN_SW_TAGS_IDENTIFY to CONFIG_KASAN_TAGS_IDENTIFY

[Alexander Potapenko]

On Sun, Jun 20, 2021 at 1:48 PM Kuan-Ying Lee
<Kuan-Ying.Lee@mediatek.com> wrote:
>
> This patch renames CONFIG_KASAN_SW_TAGS_IDENTIFY to
> CONFIG_KASAN_TAGS_IDENTIFY in order to be compatible
> with hardware tag-based mode.
>
> Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
> Suggested-by: Marco Elver <elver@google.com>
> Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
> Cc: Alexander Potapenko <glider@google.com>
> Cc: Andrey Konovalov <andreyknvl@gmail.com>
> Cc: Dmitry Vyukov <dvyukov@google.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Alexander Potapenko <glider@google.com>

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 1/3] kasan: rename CONFIG_KASAN_SW_TAGS_IDENTIFY to CONFIG_KASAN_TAGS_IDENTIFY

[Andrey Konovalov]

On Sun, Jun 20, 2021 at 2:48 PM Kuan-Ying Lee
<Kuan-Ying.Lee@mediatek.com> wrote:
>
> This patch renames CONFIG_KASAN_SW_TAGS_IDENTIFY to
> CONFIG_KASAN_TAGS_IDENTIFY in order to be compatible
> with hardware tag-based mode.
>
> Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
> Suggested-by: Marco Elver <elver@google.com>
> Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
> Cc: Alexander Potapenko <glider@google.com>
> Cc: Andrey Konovalov <andreyknvl@gmail.com>
> Cc: Dmitry Vyukov <dvyukov@google.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> ---
>  lib/Kconfig.kasan         | 2 +-
>  mm/kasan/kasan.h          | 4 ++--
>  mm/kasan/report_sw_tags.c | 2 +-
>  mm/kasan/sw_tags.c        | 4 ++--
>  4 files changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan
> index cffc2ebbf185..6f5d48832139 100644
> --- a/lib/Kconfig.kasan
> +++ b/lib/Kconfig.kasan
> @@ -155,7 +155,7 @@ config KASAN_STACK
>           CONFIG_COMPILE_TEST.  On gcc it is assumed to always be safe
>           to use and enabled by default.
>
> -config KASAN_SW_TAGS_IDENTIFY
> +config KASAN_TAGS_IDENTIFY
>         bool "Enable memory corruption identification"
>         depends on KASAN_SW_TAGS
>         help
> diff --git a/mm/kasan/kasan.h b/mm/kasan/kasan.h
> index 8f450bc28045..b0fc9a1eb7e3 100644
> --- a/mm/kasan/kasan.h
> +++ b/mm/kasan/kasan.h
> @@ -153,7 +153,7 @@ struct kasan_track {
>         depot_stack_handle_t stack;
>  };
>
> -#ifdef CONFIG_KASAN_SW_TAGS_IDENTIFY
> +#ifdef CONFIG_KASAN_TAGS_IDENTIFY
>  #define KASAN_NR_FREE_STACKS 5
>  #else
>  #define KASAN_NR_FREE_STACKS 1
> @@ -170,7 +170,7 @@ struct kasan_alloc_meta {
>  #else
>         struct kasan_track free_track[KASAN_NR_FREE_STACKS];
>  #endif
> -#ifdef CONFIG_KASAN_SW_TAGS_IDENTIFY
> +#ifdef CONFIG_KASAN_TAGS_IDENTIFY
>         u8 free_pointer_tag[KASAN_NR_FREE_STACKS];
>         u8 free_track_idx;
>  #endif
> diff --git a/mm/kasan/report_sw_tags.c b/mm/kasan/report_sw_tags.c
> index 3d20d3451d9e..821a14a19a92 100644
> --- a/mm/kasan/report_sw_tags.c
> +++ b/mm/kasan/report_sw_tags.c
> @@ -31,7 +31,7 @@
>
>  const char *kasan_get_bug_type(struct kasan_access_info *info)
>  {
> -#ifdef CONFIG_KASAN_SW_TAGS_IDENTIFY
> +#ifdef CONFIG_KASAN_TAGS_IDENTIFY
>         struct kasan_alloc_meta *alloc_meta;
>         struct kmem_cache *cache;
>         struct page *page;
> diff --git a/mm/kasan/sw_tags.c b/mm/kasan/sw_tags.c
> index 9362938abbfa..dd05e6c801fa 100644
> --- a/mm/kasan/sw_tags.c
> +++ b/mm/kasan/sw_tags.c
> @@ -177,7 +177,7 @@ void kasan_set_free_info(struct kmem_cache *cache,
>         if (!alloc_meta)
>                 return;
>
> -#ifdef CONFIG_KASAN_SW_TAGS_IDENTIFY
> +#ifdef CONFIG_KASAN_TAGS_IDENTIFY
>         idx = alloc_meta->free_track_idx;
>         alloc_meta->free_pointer_tag[idx] = tag;
>         alloc_meta->free_track_idx = (idx + 1) % KASAN_NR_FREE_STACKS;
> @@ -196,7 +196,7 @@ struct kasan_track *kasan_get_free_track(struct kmem_cache *cache,
>         if (!alloc_meta)
>                 return NULL;
>
> -#ifdef CONFIG_KASAN_SW_TAGS_IDENTIFY
> +#ifdef CONFIG_KASAN_TAGS_IDENTIFY
>         for (i = 0; i < KASAN_NR_FREE_STACKS; i++) {
>                 if (alloc_meta->free_pointer_tag[i] == tag)
>                         break;
> --
> 2.18.0
>

Reviewed-by: Andrey Konovalov <andreyknvl@gmail.com>

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v3 2/3] kasan: integrate the common part of two KASAN tag-based modes

[Kuan-Ying Lee]

1. Move kasan_get_free_track() and kasan_set_free_info()
   into tags.c
2. Move kasan_get_bug_type() to header file

Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
Suggested-by: Marco Elver <elver@google.com>
Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andrey Konovalov <andreyknvl@gmail.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
---
 mm/kasan/Makefile         |  4 +--
 mm/kasan/hw_tags.c        | 22 ---------------
 mm/kasan/report_hw_tags.c |  6 +---
 mm/kasan/report_sw_tags.c | 46 +-----------------------------
 mm/kasan/report_tags.h    | 55 ++++++++++++++++++++++++++++++++++++
 mm/kasan/sw_tags.c        | 41 ---------------------------
 mm/kasan/tags.c           | 59 +++++++++++++++++++++++++++++++++++++++
 7 files changed, 118 insertions(+), 115 deletions(-)
 create mode 100644 mm/kasan/report_tags.h
 create mode 100644 mm/kasan/tags.c

diff --git a/mm/kasan/Makefile b/mm/kasan/Makefile
index 9fe39a66388a..634de6c1da9b 100644
--- a/mm/kasan/Makefile
+++ b/mm/kasan/Makefile
@@ -37,5 +37,5 @@ CFLAGS_sw_tags.o := $(CC_FLAGS_KASAN_RUNTIME)
 
 obj-$(CONFIG_KASAN) := common.o report.o
 obj-$(CONFIG_KASAN_GENERIC) += init.o generic.o report_generic.o shadow.o quarantine.o
-obj-$(CONFIG_KASAN_HW_TAGS) += hw_tags.o report_hw_tags.o
-obj-$(CONFIG_KASAN_SW_TAGS) += init.o report_sw_tags.o shadow.o sw_tags.o
+obj-$(CONFIG_KASAN_HW_TAGS) += hw_tags.o report_hw_tags.o tags.o
+obj-$(CONFIG_KASAN_SW_TAGS) += init.o report_sw_tags.o shadow.o sw_tags.o tags.o
diff --git a/mm/kasan/hw_tags.c b/mm/kasan/hw_tags.c
index ed5e5b833d61..4ea8c368b5b8 100644
--- a/mm/kasan/hw_tags.c
+++ b/mm/kasan/hw_tags.c
@@ -216,28 +216,6 @@ void __init kasan_init_hw_tags(void)
 	pr_info("KernelAddressSanitizer initialized\n");
 }
 
-void kasan_set_free_info(struct kmem_cache *cache,
-				void *object, u8 tag)
-{
-	struct kasan_alloc_meta *alloc_meta;
-
-	alloc_meta = kasan_get_alloc_meta(cache, object);
-	if (alloc_meta)
-		kasan_set_track(&alloc_meta->free_track[0], GFP_NOWAIT);
-}
-
-struct kasan_track *kasan_get_free_track(struct kmem_cache *cache,
-				void *object, u8 tag)
-{
-	struct kasan_alloc_meta *alloc_meta;
-
-	alloc_meta = kasan_get_alloc_meta(cache, object);
-	if (!alloc_meta)
-		return NULL;
-
-	return &alloc_meta->free_track[0];
-}
-
 void kasan_alloc_pages(struct page *page, unsigned int order, gfp_t flags)
 {
 	/*
diff --git a/mm/kasan/report_hw_tags.c b/mm/kasan/report_hw_tags.c
index 42b2168755d6..ef5e7378f3aa 100644
--- a/mm/kasan/report_hw_tags.c
+++ b/mm/kasan/report_hw_tags.c
@@ -14,11 +14,7 @@
 #include <linux/types.h>
 
 #include "kasan.h"
-
-const char *kasan_get_bug_type(struct kasan_access_info *info)
-{
-	return "invalid-access";
-}
+#include "report_tags.h"
 
 void *kasan_find_first_bad_addr(void *addr, size_t size)
 {
diff --git a/mm/kasan/report_sw_tags.c b/mm/kasan/report_sw_tags.c
index 821a14a19a92..d965a170083e 100644
--- a/mm/kasan/report_sw_tags.c
+++ b/mm/kasan/report_sw_tags.c
@@ -26,51 +26,7 @@
 
 #include <asm/sections.h>
 
-#include "kasan.h"
-#include "../slab.h"
-
-const char *kasan_get_bug_type(struct kasan_access_info *info)
-{
-#ifdef CONFIG_KASAN_TAGS_IDENTIFY
-	struct kasan_alloc_meta *alloc_meta;
-	struct kmem_cache *cache;
-	struct page *page;
-	const void *addr;
-	void *object;
-	u8 tag;
-	int i;
-
-	tag = get_tag(info->access_addr);
-	addr = kasan_reset_tag(info->access_addr);
-	page = kasan_addr_to_page(addr);
-	if (page && PageSlab(page)) {
-		cache = page->slab_cache;
-		object = nearest_obj(cache, page, (void *)addr);
-		alloc_meta = kasan_get_alloc_meta(cache, object);
-
-		if (alloc_meta) {
-			for (i = 0; i < KASAN_NR_FREE_STACKS; i++) {
-				if (alloc_meta->free_pointer_tag[i] == tag)
-					return "use-after-free";
-			}
-		}
-		return "out-of-bounds";
-	}
-
-#endif
-	/*
-	 * If access_size is a negative number, then it has reason to be
-	 * defined as out-of-bounds bug type.
-	 *
-	 * Casting negative numbers to size_t would indeed turn up as
-	 * a large size_t and its value will be larger than ULONG_MAX/2,
-	 * so that this can qualify as out-of-bounds.
-	 */
-	if (info->access_addr + info->access_size < info->access_addr)
-		return "out-of-bounds";
-
-	return "invalid-access";
-}
+#include "report_tags.h"
 
 void *kasan_find_first_bad_addr(void *addr, size_t size)
 {
diff --git a/mm/kasan/report_tags.h b/mm/kasan/report_tags.h
new file mode 100644
index 000000000000..1cb872177904
--- /dev/null
+++ b/mm/kasan/report_tags.h
@@ -0,0 +1,55 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (c) 2014 Samsung Electronics Co., Ltd.
+ * Copyright (c) 2020 Google, Inc.
+ */
+#ifndef __MM_KASAN_REPORT_TAGS_H
+#define __MM_KASAN_REPORT_TAGS_H
+
+#include "kasan.h"
+#include "../slab.h"
+
+const char *kasan_get_bug_type(struct kasan_access_info *info)
+{
+#ifdef CONFIG_KASAN_TAGS_IDENTIFY
+	struct kasan_alloc_meta *alloc_meta;
+	struct kmem_cache *cache;
+	struct page *page;
+	const void *addr;
+	void *object;
+	u8 tag;
+	int i;
+
+	tag = get_tag(info->access_addr);
+	addr = kasan_reset_tag(info->access_addr);
+	page = kasan_addr_to_page(addr);
+	if (page && PageSlab(page)) {
+		cache = page->slab_cache;
+		object = nearest_obj(cache, page, (void *)addr);
+		alloc_meta = kasan_get_alloc_meta(cache, object);
+
+		if (alloc_meta) {
+			for (i = 0; i < KASAN_NR_FREE_STACKS; i++) {
+				if (alloc_meta->free_pointer_tag[i] == tag)
+					return "use-after-free";
+			}
+		}
+		return "out-of-bounds";
+	}
+#endif
+
+	/*
+	 * If access_size is a negative number, then it has reason to be
+	 * defined as out-of-bounds bug type.
+	 *
+	 * Casting negative numbers to size_t would indeed turn up as
+	 * a large size_t and its value will be larger than ULONG_MAX/2,
+	 * so that this can qualify as out-of-bounds.
+	 */
+	if (info->access_addr + info->access_size < info->access_addr)
+		return "out-of-bounds";
+
+	return "invalid-access";
+}
+
+#endif
diff --git a/mm/kasan/sw_tags.c b/mm/kasan/sw_tags.c
index dd05e6c801fa..bd3f540feb47 100644
--- a/mm/kasan/sw_tags.c
+++ b/mm/kasan/sw_tags.c
@@ -167,47 +167,6 @@ void __hwasan_tag_memory(unsigned long addr, u8 tag, unsigned long size)
 }
 EXPORT_SYMBOL(__hwasan_tag_memory);
 
-void kasan_set_free_info(struct kmem_cache *cache,
-				void *object, u8 tag)
-{
-	struct kasan_alloc_meta *alloc_meta;
-	u8 idx = 0;
-
-	alloc_meta = kasan_get_alloc_meta(cache, object);
-	if (!alloc_meta)
-		return;
-
-#ifdef CONFIG_KASAN_TAGS_IDENTIFY
-	idx = alloc_meta->free_track_idx;
-	alloc_meta->free_pointer_tag[idx] = tag;
-	alloc_meta->free_track_idx = (idx + 1) % KASAN_NR_FREE_STACKS;
-#endif
-
-	kasan_set_track(&alloc_meta->free_track[idx], GFP_NOWAIT);
-}
-
-struct kasan_track *kasan_get_free_track(struct kmem_cache *cache,
-				void *object, u8 tag)
-{
-	struct kasan_alloc_meta *alloc_meta;
-	int i = 0;
-
-	alloc_meta = kasan_get_alloc_meta(cache, object);
-	if (!alloc_meta)
-		return NULL;
-
-#ifdef CONFIG_KASAN_TAGS_IDENTIFY
-	for (i = 0; i < KASAN_NR_FREE_STACKS; i++) {
-		if (alloc_meta->free_pointer_tag[i] == tag)
-			break;
-	}
-	if (i == KASAN_NR_FREE_STACKS)
-		i = alloc_meta->free_track_idx;
-#endif
-
-	return &alloc_meta->free_track[i];
-}
-
 void kasan_tag_mismatch(unsigned long addr, unsigned long access_info,
 			unsigned long ret_ip)
 {
diff --git a/mm/kasan/tags.c b/mm/kasan/tags.c
new file mode 100644
index 000000000000..8f48b9502a17
--- /dev/null
+++ b/mm/kasan/tags.c
@@ -0,0 +1,59 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * This file contains common tag-based KASAN code.
+ *
+ * Copyright (c) 2018 Google, Inc.
+ * Copyright (c) 2020 Google, Inc.
+ */
+
+#include <linux/init.h>
+#include <linux/kasan.h>
+#include <linux/kernel.h>
+#include <linux/memory.h>
+#include <linux/mm.h>
+#include <linux/static_key.h>
+#include <linux/string.h>
+#include <linux/types.h>
+
+#include "kasan.h"
+
+void kasan_set_free_info(struct kmem_cache *cache,
+				void *object, u8 tag)
+{
+	struct kasan_alloc_meta *alloc_meta;
+	u8 idx = 0;
+
+	alloc_meta = kasan_get_alloc_meta(cache, object);
+	if (!alloc_meta)
+		return;
+
+#ifdef CONFIG_KASAN_TAGS_IDENTIFY
+	idx = alloc_meta->free_track_idx;
+	alloc_meta->free_pointer_tag[idx] = tag;
+	alloc_meta->free_track_idx = (idx + 1) % KASAN_NR_FREE_STACKS;
+#endif
+
+	kasan_set_track(&alloc_meta->free_track[idx], GFP_NOWAIT);
+}
+
+struct kasan_track *kasan_get_free_track(struct kmem_cache *cache,
+				void *object, u8 tag)
+{
+	struct kasan_alloc_meta *alloc_meta;
+	int i = 0;
+
+	alloc_meta = kasan_get_alloc_meta(cache, object);
+	if (!alloc_meta)
+		return NULL;
+
+#ifdef CONFIG_KASAN_TAGS_IDENTIFY
+	for (i = 0; i < KASAN_NR_FREE_STACKS; i++) {
+		if (alloc_meta->free_pointer_tag[i] == tag)
+			break;
+	}
+	if (i == KASAN_NR_FREE_STACKS)
+		i = alloc_meta->free_track_idx;
+#endif
+
+	return &alloc_meta->free_track[i];
+}
-- 
2.18.0
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 2/3] kasan: integrate the common part of two KASAN tag-based modes

[Alexander Potapenko]

> diff --git a/mm/kasan/report_tags.h b/mm/kasan/report_tags.h
> new file mode 100644
> index 000000000000..1cb872177904
> --- /dev/null
> +++ b/mm/kasan/report_tags.h
Why don't you make it a C file instead?

> +const char *kasan_get_bug_type(struct kasan_access_info *info)
If this function has to be in the header, it should be declared as
static inline.
But I don't think it has to be there in the first place.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 2/3] kasan: integrate the common part of two KASAN tag-based modes

[Marco Elver]

On Tue, 22 Jun 2021 at 11:28, Alexander Potapenko <glider@google.com> wrote:
>
> > diff --git a/mm/kasan/report_tags.h b/mm/kasan/report_tags.h
> > new file mode 100644
> > index 000000000000..1cb872177904
> > --- /dev/null
> > +++ b/mm/kasan/report_tags.h
> Why don't you make it a C file instead?

Yes, good point. report_{hw,sw}_tags.c essentially define it, but it's
called by report.c code.

I think I suggested to make it a header first because there were still
parts that were report_{hw,sw}_tags.c specific, and the helper
function would be used by those 2 to build their version of
kasan_get_bug_type(), but that doesn't seem to be the case anymore.

> > +const char *kasan_get_bug_type(struct kasan_access_info *info)
> If this function has to be in the header, it should be declared as
> static inline.
> But I don't think it has to be there in the first place.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 2/3] kasan: integrate the common part of two KASAN tag-based modes

[Andrey Konovalov]

On Sun, Jun 20, 2021 at 2:48 PM Kuan-Ying Lee
<Kuan-Ying.Lee@mediatek.com> wrote:
>
> 1. Move kasan_get_free_track() and kasan_set_free_info()
>    into tags.c

Please mention that the patch doesn't only move but also combines
these functions for SW_TAGS and HW_TAGS modes.

> --- /dev/null
> +++ b/mm/kasan/report_tags.h
> @@ -0,0 +1,55 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/*
> + * Copyright (c) 2014 Samsung Electronics Co., Ltd.
> + * Copyright (c) 2020 Google, Inc.
> + */
> +#ifndef __MM_KASAN_REPORT_TAGS_H
> +#define __MM_KASAN_REPORT_TAGS_H
> +
> +#include "kasan.h"
> +#include "../slab.h"
> +
> +const char *kasan_get_bug_type(struct kasan_access_info *info)

As mentioned by Alex, don't put this implementation into a header. Put
it into report_tags.c. The declaration is already in kasan.h.


> +{
> +#ifdef CONFIG_KASAN_TAGS_IDENTIFY
> +       struct kasan_alloc_meta *alloc_meta;
> +       struct kmem_cache *cache;
> +       struct page *page;
> +       const void *addr;
> +       void *object;
> +       u8 tag;
> +       int i;
> +
> +       tag = get_tag(info->access_addr);
> +       addr = kasan_reset_tag(info->access_addr);
> +       page = kasan_addr_to_page(addr);
> +       if (page && PageSlab(page)) {
> +               cache = page->slab_cache;
> +               object = nearest_obj(cache, page, (void *)addr);
> +               alloc_meta = kasan_get_alloc_meta(cache, object);
> +
> +               if (alloc_meta) {
> +                       for (i = 0; i < KASAN_NR_FREE_STACKS; i++) {
> +                               if (alloc_meta->free_pointer_tag[i] == tag)
> +                                       return "use-after-free";
> +                       }
> +               }
> +               return "out-of-bounds";
> +       }
> +#endif
> +
> +       /*
> +        * If access_size is a negative number, then it has reason to be
> +        * defined as out-of-bounds bug type.
> +        *
> +        * Casting negative numbers to size_t would indeed turn up as
> +        * a large size_t and its value will be larger than ULONG_MAX/2,
> +        * so that this can qualify as out-of-bounds.
> +        */
> +       if (info->access_addr + info->access_size < info->access_addr)
> +               return "out-of-bounds";
> +
> +       return "invalid-access";
> +}

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 2/3] kasan: integrate the common part of two KASAN tag-based modes

[Kuan-Ying Lee]

On Tue, 2021-06-22 at 16:54 +0300, Andrey Konovalov wrote:
> On Sun, Jun 20, 2021 at 2:48 PM Kuan-Ying Lee
> <Kuan-Ying.Lee@mediatek.com> wrote:
> > 
> > 1. Move kasan_get_free_track() and kasan_set_free_info()
> >    into tags.c
> 
> Please mention that the patch doesn't only move but also combines
> these functions for SW_TAGS and HW_TAGS modes.
> 

Got it.

> > --- /dev/null
> > +++ b/mm/kasan/report_tags.h
> > @@ -0,0 +1,55 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +/*
> > + * Copyright (c) 2014 Samsung Electronics Co., Ltd.
> > + * Copyright (c) 2020 Google, Inc.
> > + */
> > +#ifndef __MM_KASAN_REPORT_TAGS_H
> > +#define __MM_KASAN_REPORT_TAGS_H
> > +
> > +#include "kasan.h"
> > +#include "../slab.h"
> > +
> > +const char *kasan_get_bug_type(struct kasan_access_info *info)
> 
> As mentioned by Alex, don't put this implementation into a header.
> Put
> it into report_tags.c. The declaration is already in kasan.h.
> 

Ok. I will refactor in v4.
Thanks for suggestions.

> > +{
> > +#ifdef CONFIG_KASAN_TAGS_IDENTIFY
> > +       struct kasan_alloc_meta *alloc_meta;
> > +       struct kmem_cache *cache;
> > +       struct page *page;
> > +       const void *addr;
> > +       void *object;
> > +       u8 tag;
> > +       int i;
> > +
> > +       tag = get_tag(info->access_addr);
> > +       addr = kasan_reset_tag(info->access_addr);
> > +       page = kasan_addr_to_page(addr);
> > +       if (page && PageSlab(page)) {
> > +               cache = page->slab_cache;
> > +               object = nearest_obj(cache, page, (void *)addr);
> > +               alloc_meta = kasan_get_alloc_meta(cache, object);
> > +
> > +               if (alloc_meta) {
> > +                       for (i = 0; i < KASAN_NR_FREE_STACKS; i++)
> > {
> > +                               if (alloc_meta->free_pointer_tag[i] 
> > == tag)
> > +                                       return "use-after-free";
> > +                       }
> > +               }
> > +               return "out-of-bounds";
> > +       }
> > +#endif
> > +
> > +       /*
> > +        * If access_size is a negative number, then it has reason
> > to be
> > +        * defined as out-of-bounds bug type.
> > +        *
> > +        * Casting negative numbers to size_t would indeed turn up
> > as
> > +        * a large size_t and its value will be larger than
> > ULONG_MAX/2,
> > +        * so that this can qualify as out-of-bounds.
> > +        */
> > +       if (info->access_addr + info->access_size < info-
> > >access_addr)
> > +               return "out-of-bounds";
> > +
> > +       return "invalid-access";
> > +}
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v3 3/3] kasan: add memory corruption identification support for hardware tag-based mode

[Kuan-Ying Lee]

Add memory corruption identification support for hardware tag-based
mode. We store one old free pointer tag and free backtrace.

Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
Suggested-by: Marco Elver <elver@google.com>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andrey Konovalov <andreyknvl@gmail.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
---
 lib/Kconfig.kasan | 2 +-
 mm/kasan/kasan.h  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan
index 6f5d48832139..2cc25792bc2f 100644
--- a/lib/Kconfig.kasan
+++ b/lib/Kconfig.kasan
@@ -157,7 +157,7 @@ config KASAN_STACK
 
 config KASAN_TAGS_IDENTIFY
 	bool "Enable memory corruption identification"
-	depends on KASAN_SW_TAGS
+	depends on KASAN_SW_TAGS || KASAN_HW_TAGS
 	help
 	  This option enables best-effort identification of bug type
 	  (use-after-free or out-of-bounds) at the cost of increased
diff --git a/mm/kasan/kasan.h b/mm/kasan/kasan.h
index b0fc9a1eb7e3..d6f982b8a84e 100644
--- a/mm/kasan/kasan.h
+++ b/mm/kasan/kasan.h
@@ -153,7 +153,7 @@ struct kasan_track {
 	depot_stack_handle_t stack;
 };
 
-#ifdef CONFIG_KASAN_TAGS_IDENTIFY
+#if defined(CONFIG_KASAN_TAGS_IDENTIFY) && defined(CONFIG_KASAN_SW_TAGS)
 #define KASAN_NR_FREE_STACKS 5
 #else
 #define KASAN_NR_FREE_STACKS 1
-- 
2.18.0
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 3/3] kasan: add memory corruption identification support for hardware tag-based mode

[Alexander Potapenko]

On Sun, Jun 20, 2021 at 1:48 PM Kuan-Ying Lee
<Kuan-Ying.Lee@mediatek.com> wrote:
>
> Add memory corruption identification support for hardware tag-based
> mode. We store one old free pointer tag and free backtrace.
>
> Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
> Suggested-by: Marco Elver <elver@google.com>
> Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
> Cc: Alexander Potapenko <glider@google.com>
> Cc: Andrey Konovalov <andreyknvl@gmail.com>
> Cc: Dmitry Vyukov <dvyukov@google.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Alexander Potapenko <glider@google.com>
> ---
>  lib/Kconfig.kasan | 2 +-
>  mm/kasan/kasan.h  | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan
> index 6f5d48832139..2cc25792bc2f 100644
> --- a/lib/Kconfig.kasan
> +++ b/lib/Kconfig.kasan
> @@ -157,7 +157,7 @@ config KASAN_STACK
>
>  config KASAN_TAGS_IDENTIFY
>         bool "Enable memory corruption identification"
> -       depends on KASAN_SW_TAGS
> +       depends on KASAN_SW_TAGS || KASAN_HW_TAGS
>         help
>           This option enables best-effort identification of bug type
>           (use-after-free or out-of-bounds) at the cost of increased
> diff --git a/mm/kasan/kasan.h b/mm/kasan/kasan.h
> index b0fc9a1eb7e3..d6f982b8a84e 100644
> --- a/mm/kasan/kasan.h
> +++ b/mm/kasan/kasan.h
> @@ -153,7 +153,7 @@ struct kasan_track {
>         depot_stack_handle_t stack;
>  };
>
> -#ifdef CONFIG_KASAN_TAGS_IDENTIFY
> +#if defined(CONFIG_KASAN_TAGS_IDENTIFY) && defined(CONFIG_KASAN_SW_TAGS)
>  #define KASAN_NR_FREE_STACKS 5
>  #else
>  #define KASAN_NR_FREE_STACKS 1
> --
> 2.18.0
>
> --
> You received this message because you are subscribed to the Google Groups "kasan-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to kasan-dev+unsubscribe@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/kasan-dev/20210620114756.31304-4-Kuan-Ying.Lee%40mediatek.com.



-- 
Alexander Potapenko
Software Engineer

Google Germany GmbH
Erika-Mann-Straße, 33
80636 München

Geschäftsführer: Paul Manicle, Halimah DeLaine Prado
Registergericht und -nummer: Hamburg, HRB 86891
Sitz der Gesellschaft: Hamburg

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 3/3] kasan: add memory corruption identification support for hardware tag-based mode

[Andrey Konovalov]

On Sun, Jun 20, 2021 at 2:48 PM Kuan-Ying Lee
<Kuan-Ying.Lee@mediatek.com> wrote:
>
> Add memory corruption identification support for hardware tag-based
> mode. We store one old free pointer tag and free backtrace.

Please explain why only one.

> Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
> Suggested-by: Marco Elver <elver@google.com>
> Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
> Cc: Alexander Potapenko <glider@google.com>
> Cc: Andrey Konovalov <andreyknvl@gmail.com>
> Cc: Dmitry Vyukov <dvyukov@google.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> ---
>  lib/Kconfig.kasan | 2 +-
>  mm/kasan/kasan.h  | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan
> index 6f5d48832139..2cc25792bc2f 100644
> --- a/lib/Kconfig.kasan
> +++ b/lib/Kconfig.kasan
> @@ -157,7 +157,7 @@ config KASAN_STACK
>
>  config KASAN_TAGS_IDENTIFY
>         bool "Enable memory corruption identification"
> -       depends on KASAN_SW_TAGS
> +       depends on KASAN_SW_TAGS || KASAN_HW_TAGS
>         help
>           This option enables best-effort identification of bug type
>           (use-after-free or out-of-bounds) at the cost of increased
> diff --git a/mm/kasan/kasan.h b/mm/kasan/kasan.h
> index b0fc9a1eb7e3..d6f982b8a84e 100644
> --- a/mm/kasan/kasan.h
> +++ b/mm/kasan/kasan.h
> @@ -153,7 +153,7 @@ struct kasan_track {
>         depot_stack_handle_t stack;
>  };
>
> -#ifdef CONFIG_KASAN_TAGS_IDENTIFY
> +#if defined(CONFIG_KASAN_TAGS_IDENTIFY) && defined(CONFIG_KASAN_SW_TAGS)
>  #define KASAN_NR_FREE_STACKS 5
>  #else
>  #define KASAN_NR_FREE_STACKS 1
> --
> 2.18.0
>

Other than that:

Reviewed-by: Andrey Konovalov <andreyknvl@gmail.com>

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 3/3] kasan: add memory corruption identification support for hardware tag-based mode

[Kuan-Ying Lee]

On Tue, 2021-06-22 at 16:54 +0300, Andrey Konovalov wrote:
> On Sun, Jun 20, 2021 at 2:48 PM Kuan-Ying Lee
> <Kuan-Ying.Lee@mediatek.com> wrote:
> > 
> > Add memory corruption identification support for hardware tag-based
> > mode. We store one old free pointer tag and free backtrace.
> 
> Please explain why only one.
> 

Got it. I will rewrite this commit message in v4.

> > Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
> > Suggested-by: Marco Elver <elver@google.com>
> > Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
> > Cc: Alexander Potapenko <glider@google.com>
> > Cc: Andrey Konovalov <andreyknvl@gmail.com>
> > Cc: Dmitry Vyukov <dvyukov@google.com>
> > Cc: Andrew Morton <akpm@linux-foundation.org>
> > ---
> >  lib/Kconfig.kasan | 2 +-
> >  mm/kasan/kasan.h  | 2 +-
> >  2 files changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan
> > index 6f5d48832139..2cc25792bc2f 100644
> > --- a/lib/Kconfig.kasan
> > +++ b/lib/Kconfig.kasan
> > @@ -157,7 +157,7 @@ config KASAN_STACK
> > 
> >  config KASAN_TAGS_IDENTIFY
> >         bool "Enable memory corruption identification"
> > -       depends on KASAN_SW_TAGS
> > +       depends on KASAN_SW_TAGS || KASAN_HW_TAGS
> >         help
> >           This option enables best-effort identification of bug
> > type
> >           (use-after-free or out-of-bounds) at the cost of
> > increased
> > diff --git a/mm/kasan/kasan.h b/mm/kasan/kasan.h
> > index b0fc9a1eb7e3..d6f982b8a84e 100644
> > --- a/mm/kasan/kasan.h
> > +++ b/mm/kasan/kasan.h
> > @@ -153,7 +153,7 @@ struct kasan_track {
> >         depot_stack_handle_t stack;
> >  };
> > 
> > -#ifdef CONFIG_KASAN_TAGS_IDENTIFY
> > +#if defined(CONFIG_KASAN_TAGS_IDENTIFY) &&
> > defined(CONFIG_KASAN_SW_TAGS)
> >  #define KASAN_NR_FREE_STACKS 5
> >  #else
> >  #define KASAN_NR_FREE_STACKS 1
> > --
> > 2.18.0
> > 
> 
> Other than that:
> 
> Reviewed-by: Andrey Konovalov <andreyknvl@gmail.com>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 0/3] kasan: add memory corruption identification support for hw tag-based kasan

[Marco Elver]

On Sun, 20 Jun 2021 at 13:48, Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com> wrote:
>
> Add memory corruption identification for hardware tag-based KASAN mode.
>
> Changes since v3:
>  - Preserve Copyright from hw_tags.c/sw_tags.c and
>    report_sw_tags.c/report_hw_tags.c
>  - Make non-trivial change in kasan sw tag-based mode
>
> Changes since v2:
>  - Thanks for Marco's Suggestion
>  - Rename the CONFIG_KASAN_SW_TAGS_IDENTIFY
>  - Integrate tag-based kasan common part
>  - Rebase to latest linux-next
>
> Kuan-Ying Lee (3):
>   kasan: rename CONFIG_KASAN_SW_TAGS_IDENTIFY to
>     CONFIG_KASAN_TAGS_IDENTIFY
>   kasan: integrate the common part of two KASAN tag-based modes
>   kasan: add memory corruption identification support for hardware
>     tag-based mode

I think this looks fine, thank you for your efforts. How did you test
this? Did you run the lib/test_kasan module with both SW_TAGS and
HW_TAGS mode? I was about to run that before adding my Reviewed-by.

Andrey, Alex, if you have time, please have a quick look at the series.

Thanks,
-- Marco

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 0/3] kasan: add memory corruption identification support for hw tag-based kasan

[Kuan-Ying Lee]

On Mon, 2021-06-21 at 14:45 +0200, Marco Elver wrote:
> On Sun, 20 Jun 2021 at 13:48, Kuan-Ying Lee <
> Kuan-Ying.Lee@mediatek.com> wrote:
> > 
> > Add memory corruption identification for hardware tag-based KASAN
> > mode.
> > 
> > Changes since v3:
> >  - Preserve Copyright from hw_tags.c/sw_tags.c and
> >    report_sw_tags.c/report_hw_tags.c
> >  - Make non-trivial change in kasan sw tag-based mode
> > 
> > Changes since v2:
> >  - Thanks for Marco's Suggestion
> >  - Rename the CONFIG_KASAN_SW_TAGS_IDENTIFY
> >  - Integrate tag-based kasan common part
> >  - Rebase to latest linux-next
> > 
> > Kuan-Ying Lee (3):
> >   kasan: rename CONFIG_KASAN_SW_TAGS_IDENTIFY to
> >     CONFIG_KASAN_TAGS_IDENTIFY
> >   kasan: integrate the common part of two KASAN tag-based modes
> >   kasan: add memory corruption identification support for hardware
> >     tag-based mode
> 
> I think this looks fine, thank you for your efforts. How did you test
> this? Did you run the lib/test_kasan module with both SW_TAGS and
> HW_TAGS mode? I was about to run that before adding my Reviewed-by.

Thanks for the reminder.
Yes, I run the lib/test_kasan module with SW_TAGS and HW_TAGS mode. :)

> 
> Andrey, Alex, if you have time, please have a quick look at the
> series.
> 
> Thanks,
> -- Marco
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel