Re: [PATCH v2 33/37] kasan, arm64: implement HW_TAGS runtime

From: Andrey Konovalov
Date: Fri Sep 18 2020 - 11:00:20 EST

Next message: zhe.he: "[PATCH] SUNRPC: Flush dcache only when receiving more seeking"
Previous message: Srinivasa Rao Mandadapu: "[PATCH 1/2] arm64: dts: qcom: sc7180: Update lpass cpu node for audio over dp"
In reply to: Marco Elver: "Re: [PATCH v2 33/37] kasan, arm64: implement HW_TAGS runtime"
Next in thread: Marco Elver: "Re: [PATCH v2 33/37] kasan, arm64: implement HW_TAGS runtime"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Fri, Sep 18, 2020 at 2:52 PM Marco Elver <elver@xxxxxxxxxx> wrote:
>
> [ Sorry for the additional email on this patch; trying to consolidate
> comments now. ]
>
> On Tue, Sep 15, 2020 at 11:16PM +0200, Andrey Konovalov wrote:
> > Provide implementation of KASAN functions required for the hardware
> > tag-based mode. Those include core functions for memory and pointer
> > tagging (tags_hw.c) and bug reporting (report_tags_hw.c). Also adapt
> > common KASAN code to support the new mode.
> >
> > Signed-off-by: Andrey Konovalov <andreyknvl@xxxxxxxxxx>
> > Signed-off-by: Vincenzo Frascino <vincenzo.frascino@xxxxxxx>
> > ---
> > Change-Id: I8a8689ba098174a4d0ef3f1d008178387c80ee1c
> > ---
> > arch/arm64/include/asm/memory.h | 4 +-
> > arch/arm64/kernel/setup.c | 1 -
> > include/linux/kasan.h | 6 +--
> > include/linux/mm.h | 2 +-
> > include/linux/page-flags-layout.h | 2 +-
> > mm/kasan/Makefile | 5 ++
> > mm/kasan/common.c | 14 +++---
> > mm/kasan/kasan.h | 17 +++++--
> > mm/kasan/report_tags_hw.c | 47 +++++++++++++++++++
> > mm/kasan/report_tags_sw.c | 2 +-
> > mm/kasan/shadow.c | 2 +-
> > mm/kasan/tags_hw.c | 78 +++++++++++++++++++++++++++++++
> > mm/kasan/tags_sw.c | 2 +-
> > 13 files changed, 162 insertions(+), 20 deletions(-)
> > create mode 100644 mm/kasan/report_tags_hw.c
> > create mode 100644 mm/kasan/tags_hw.c
> [...]
> > diff --git a/mm/kasan/common.c b/mm/kasan/common.c
> > index 41c7f1105eaa..412a23d1546b 100644
> > --- a/mm/kasan/common.c
> > +++ b/mm/kasan/common.c
> > @@ -118,7 +118,7 @@ void kasan_free_pages(struct page *page, unsigned int order)
> > */
> > static inline unsigned int optimal_redzone(unsigned int object_size)
> > {
> > - if (IS_ENABLED(CONFIG_KASAN_SW_TAGS))
> > + if (!IS_ENABLED(CONFIG_KASAN_GENERIC))
> > return 0;
> >
> > return
> > @@ -183,14 +183,14 @@ size_t kasan_metadata_size(struct kmem_cache *cache)
> > struct kasan_alloc_meta *get_alloc_info(struct kmem_cache *cache,
> > const void *object)
> > {
> > - return (void *)object + cache->kasan_info.alloc_meta_offset;
> > + return (void *)reset_tag(object) + cache->kasan_info.alloc_meta_offset;
> > }
> >
> > struct kasan_free_meta *get_free_info(struct kmem_cache *cache,
> > const void *object)
> > {
> > BUILD_BUG_ON(sizeof(struct kasan_free_meta) > 32);
> > - return (void *)object + cache->kasan_info.free_meta_offset;
> > + return (void *)reset_tag(object) + cache->kasan_info.free_meta_offset;
> > }
> >
> > void kasan_poison_slab(struct page *page)
> > @@ -272,7 +272,8 @@ void * __must_check kasan_init_slab_obj(struct kmem_cache *cache,
> > alloc_info = get_alloc_info(cache, object);
> > __memset(alloc_info, 0, sizeof(*alloc_info));
>
> Suggested edit below (assuming the line-break wasn't intentional; this
> should still be within checkpatch.pl's 100 col limit):
> ------
> - if (IS_ENABLED(CONFIG_KASAN_SW_TAGS) ||
> - IS_ENABLED(CONFIG_KASAN_HW_TAGS))
> + if (IS_ENABLED(CONFIG_KASAN_SW_TAGS) || IS_ENABLED(CONFIG_KASAN_HW_TAGS))
> object = set_tag(object,
> assign_tag(cache, object, true, false));
>
> @@ -343,8 +342,7 @@ static void *__kasan_kmalloc(struct kmem_cache *cache, const void *object,
> redzone_end = round_up((unsigned long)object + cache->object_size,
> KASAN_GRANULE_SIZE);
>
> - if (IS_ENABLED(CONFIG_KASAN_SW_TAGS) ||
> - IS_ENABLED(CONFIG_KASAN_HW_TAGS))
> + if (IS_ENABLED(CONFIG_KASAN_SW_TAGS) || IS_ENABLED(CONFIG_KASAN_HW_TAGS))
> tag = assign_tag(cache, object, false, keep_tag);
> ------
>
> > - if (IS_ENABLED(CONFIG_KASAN_SW_TAGS))
> > + if (IS_ENABLED(CONFIG_KASAN_SW_TAGS) ||
> > + IS_ENABLED(CONFIG_KASAN_HW_TAGS))
> > object = set_tag(object,
> > assign_tag(cache, object, true, false));
> >
> > @@ -342,10 +343,11 @@ static void *__kasan_kmalloc(struct kmem_cache *cache, const void *object,
> > redzone_end = round_up((unsigned long)object + cache->object_size,
> > KASAN_GRANULE_SIZE);
> >
> > - if (IS_ENABLED(CONFIG_KASAN_SW_TAGS))
> > + if (IS_ENABLED(CONFIG_KASAN_SW_TAGS) ||
> > + IS_ENABLED(CONFIG_KASAN_HW_TAGS))
> > tag = assign_tag(cache, object, false, keep_tag);
> >
> > - /* Tag is ignored in set_tag without CONFIG_KASAN_SW_TAGS */
> > + /* Tag is ignored in set_tag without CONFIG_KASAN_SW/HW_TAGS */
> > kasan_unpoison_memory(set_tag(object, tag), size);
> > kasan_poison_memory((void *)redzone_start, redzone_end - redzone_start,
> > KASAN_KMALLOC_REDZONE);
> [...]
> > diff --git a/mm/kasan/report_tags_hw.c b/mm/kasan/report_tags_hw.c
> > new file mode 100644
> > index 000000000000..c2f73c46279a
> > --- /dev/null
> > +++ b/mm/kasan/report_tags_hw.c
> > @@ -0,0 +1,47 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * This file contains hardware tag-based KASAN specific error reporting code.
> > + *
> > + * Copyright (c) 2020 Google, Inc.
> > + * Author: Andrey Konovalov <andreyknvl@xxxxxxxxxx>
> > + *
> > + * This program is free software; you can redistribute it and/or modify
> > + * it under the terms of the GNU General Public License version 2 as
> > + * published by the Free Software Foundation.
> > + *
>
> I do not think we put the "This program is ..." preamble in new files
> anymore. It should be covered by SPDX tag above.
>
> > + */
> > +
> > +#include <linux/kasan.h>
> > +#include <linux/kernel.h>
> > +#include <linux/memory.h>
> > +#include <linux/mm.h>
> > +#include <linux/string.h>
> > +#include <linux/types.h>
> [...]
> > diff --git a/mm/kasan/shadow.c b/mm/kasan/shadow.c
> > index 4888084ecdfc..ca69726adf8f 100644
> > --- a/mm/kasan/shadow.c
> > +++ b/mm/kasan/shadow.c
> > @@ -111,7 +111,7 @@ void kasan_unpoison_memory(const void *address, size_t size)
> >
> > if (IS_ENABLED(CONFIG_KASAN_SW_TAGS))
> > *shadow = tag;
> > - else
> > + else /* CONFIG_KASAN_GENERIC */
> > *shadow = size & KASAN_GRANULE_MASK;
> > }
> > }
> > diff --git a/mm/kasan/tags_hw.c b/mm/kasan/tags_hw.c
> > new file mode 100644
> > index 000000000000..c93d43379e39
> > --- /dev/null
> > +++ b/mm/kasan/tags_hw.c
> > @@ -0,0 +1,78 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * This file contains core hardware tag-based KASAN code.
> > + *
> > + * Copyright (c) 2020 Google, Inc.
> > + * Author: Andrey Konovalov <andreyknvl@xxxxxxxxxx>
> > + *
> > + * This program is free software; you can redistribute it and/or modify
> > + * it under the terms of the GNU General Public License version 2 as
> > + * published by the Free Software Foundation.
> > + *
>
> I do not think we put the "This program is ..." preamble in new files
> anymore. It should be covered by SPDX tag above.
>
> > + */
> > +
> > +#include <linux/kasan.h>
> > +#include <linux/kernel.h>
> > +#include <linux/memory.h>
> > +#include <linux/mm.h>
> > +#include <linux/string.h>
> > +#include <linux/types.h>
> > +
> > +#include "kasan.h"
> > +
> > +void kasan_init_tags(void)
> > +{
> > + init_tags(KASAN_TAG_MAX);
> > +}
> > +
> > +void *kasan_reset_tag(const void *addr)
> > +{
> > + return reset_tag(addr);
> > +}
> > +
>
> To help readability, would this edit be ok?
> ------
> void kasan_poison_memory(const void *address, size_t size, u8 value)
> {
> - set_mem_tag_range(reset_tag(address),
> - round_up(size, KASAN_GRANULE_SIZE), value);
> + set_mem_tag_range(reset_tag(address), round_up(size, KASAN_GRANULE_SIZE), value);
> }
>
> void kasan_unpoison_memory(const void *address, size_t size)
> {
> - set_mem_tag_range(reset_tag(address),
> - round_up(size, KASAN_GRANULE_SIZE), get_tag(address));
> + set_mem_tag_range(reset_tag(address), round_up(size, KASAN_GRANULE_SIZE), get_tag(address));
> }
> ------
>
> > +void kasan_poison_memory(const void *address, size_t size, u8 value)
> > +{
> > + set_mem_tag_range(reset_tag(address),
> > + round_up(size, KASAN_GRANULE_SIZE), value);
> > +}
> > +
> > +void kasan_unpoison_memory(const void *address, size_t size)
> > +{
> > + set_mem_tag_range(reset_tag(address),
> > + round_up(size, KASAN_GRANULE_SIZE), get_tag(address));
> > +}
> > +
> > +u8 random_tag(void)
> > +{
> > + return get_random_tag();
> > +}
> > +
> > +bool check_invalid_free(void *addr)
> > +{
> > + u8 ptr_tag = get_tag(addr);
> > + u8 mem_tag = get_mem_tag(addr);
> > +
>
>
> Why not just:
> ------
> - if (shadow_byte == KASAN_TAG_INVALID)
> - return true;
> - if (tag != KASAN_TAG_KERNEL && tag != shadow_byte)
> - return true;
> - return false;
> + return shadow_byte == KASAN_TAG_INVALID ||
> + (tag != KASAN_TAG_KERNEL && tag != shadow_byte);
> }
> ------
>
> > + if (mem_tag == KASAN_TAG_INVALID)
> > + return true;
> > + if (ptr_tag != KASAN_TAG_KERNEL && ptr_tag != mem_tag)
> > + return true;
> > + return false;
> > +}
> > +

Will fix all these in v3, thanks!

Next message: zhe.he: "[PATCH] SUNRPC: Flush dcache only when receiving more seeking"
Previous message: Srinivasa Rao Mandadapu: "[PATCH 1/2] arm64: dts: qcom: sc7180: Update lpass cpu node for audio over dp"
In reply to: Marco Elver: "Re: [PATCH v2 33/37] kasan, arm64: implement HW_TAGS runtime"
Next in thread: Marco Elver: "Re: [PATCH v2 33/37] kasan, arm64: implement HW_TAGS runtime"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]