From: Andrey Konovalov <andreyknvl@google.com>
To: Dmitry Vyukov <dvyukov@google.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will.deacon@arm.com>,
Vincenzo Frascino <vincenzo.frascino@arm.com>,
kasan-dev <kasan-dev@googlegroups.com>,
Andrey Ryabinin <aryabinin@virtuozzo.com>,
Alexander Potapenko <glider@google.com>,
Marco Elver <elver@google.com>,
Evgenii Stepanov <eugenis@google.com>,
Elena Petrova <lenaptr@google.com>,
Branislav Rankov <Branislav.Rankov@arm.com>,
Kevin Brodsky <kevin.brodsky@arm.com>,
Andrew Morton <akpm@linux-foundation.org>,
Linux ARM <linux-arm-kernel@lists.infradead.org>,
Linux-MM <linux-mm@kvack.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v5 08/40] arm64: mte: Switch GCR_EL1 in kernel entry and exit
Date: Thu, 29 Oct 2020 17:52:11 +0100 [thread overview]
Message-ID: <CAAeHK+yARu1Qh46DPbfs4h37yHPAVgx6r8r=86L6jfHD3up8-g@mail.gmail.com> (raw)
In-Reply-To: <CACT4Y+ZyaqdYic_K6Mj9RcvO+23OQ0q2Pe-c3YS1zMW4j1woQw@mail.gmail.com>
On Wed, Oct 28, 2020 at 11:07 AM Dmitry Vyukov <dvyukov@google.com> wrote:
>
> On Mon, Oct 12, 2020 at 10:45 PM Andrey Konovalov <andreyknvl@google.com> wrote:
> >
> > From: Vincenzo Frascino <vincenzo.frascino@arm.com>
> >
> > When MTE is present, the GCR_EL1 register contains the tags mask that
> > allows to exclude tags from the random generation via the IRG instruction.
> >
> > With the introduction of the new Tag-Based KASAN API that provides a
> > mechanism to reserve tags for special reasons, the MTE implementation
> > has to make sure that the GCR_EL1 setting for the kernel does not affect
> > the userspace processes and viceversa.
> >
> > Save and restore the kernel/user mask in GCR_EL1 in kernel entry and exit.
> >
> > Signed-off-by: Vincenzo Frascino <vincenzo.frascino@arm.com>
> > Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
> > Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
> > ---
> > Change-Id: I0081cba5ace27a9111bebb239075c9a466af4c84
> > ---
> > arch/arm64/include/asm/mte-def.h | 1 -
> > arch/arm64/include/asm/mte-kasan.h | 6 +++++
> > arch/arm64/include/asm/mte.h | 2 ++
> > arch/arm64/kernel/asm-offsets.c | 3 +++
> > arch/arm64/kernel/cpufeature.c | 3 +++
> > arch/arm64/kernel/entry.S | 41 ++++++++++++++++++++++++++++++
> > arch/arm64/kernel/mte.c | 22 +++++++++++++---
> > 7 files changed, 74 insertions(+), 4 deletions(-)
> >
> > diff --git a/arch/arm64/include/asm/mte-def.h b/arch/arm64/include/asm/mte-def.h
> > index 8401ac5840c7..2d73a1612f09 100644
> > --- a/arch/arm64/include/asm/mte-def.h
> > +++ b/arch/arm64/include/asm/mte-def.h
> > @@ -10,6 +10,5 @@
> > #define MTE_TAG_SHIFT 56
> > #define MTE_TAG_SIZE 4
> > #define MTE_TAG_MASK GENMASK((MTE_TAG_SHIFT + (MTE_TAG_SIZE - 1)), MTE_TAG_SHIFT)
> > -#define MTE_TAG_MAX (MTE_TAG_MASK >> MTE_TAG_SHIFT)
> >
> > #endif /* __ASM_MTE_DEF_H */
> > diff --git a/arch/arm64/include/asm/mte-kasan.h b/arch/arm64/include/asm/mte-kasan.h
> > index 3a70fb1807fd..a4c61b926d4a 100644
> > --- a/arch/arm64/include/asm/mte-kasan.h
> > +++ b/arch/arm64/include/asm/mte-kasan.h
> > @@ -29,6 +29,8 @@ u8 mte_get_mem_tag(void *addr);
> > u8 mte_get_random_tag(void);
> > void *mte_set_mem_tag_range(void *addr, size_t size, u8 tag);
> >
> > +void mte_init_tags(u64 max_tag);
>
> This should be marked as __init?
Makes sense. I'll add this change into the other series together with
the patch that marks kasan_init_tags() as __init.
>
>
> > #else /* CONFIG_ARM64_MTE */
> >
> > static inline u8 mte_get_ptr_tag(void *ptr)
> > @@ -49,6 +51,10 @@ static inline void *mte_set_mem_tag_range(void *addr, size_t size, u8 tag)
> > return addr;
> > }
> >
> > +static inline void mte_init_tags(u64 max_tag)
> > +{
> > +}
> > +
> > #endif /* CONFIG_ARM64_MTE */
> >
> > #endif /* __ASSEMBLY__ */
> > diff --git a/arch/arm64/include/asm/mte.h b/arch/arm64/include/asm/mte.h
> > index cf1cd181dcb2..d02aff9f493d 100644
> > --- a/arch/arm64/include/asm/mte.h
> > +++ b/arch/arm64/include/asm/mte.h
> > @@ -18,6 +18,8 @@
> >
> > #include <asm/pgtable-types.h>
> >
> > +extern u64 gcr_kernel_excl;
> > +
> > void mte_clear_page_tags(void *addr);
> > unsigned long mte_copy_tags_from_user(void *to, const void __user *from,
> > unsigned long n);
> > diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c
> > index 7d32fc959b1a..dfe6ed8446ac 100644
> > --- a/arch/arm64/kernel/asm-offsets.c
> > +++ b/arch/arm64/kernel/asm-offsets.c
> > @@ -47,6 +47,9 @@ int main(void)
> > #ifdef CONFIG_ARM64_PTR_AUTH
> > DEFINE(THREAD_KEYS_USER, offsetof(struct task_struct, thread.keys_user));
> > DEFINE(THREAD_KEYS_KERNEL, offsetof(struct task_struct, thread.keys_kernel));
> > +#endif
> > +#ifdef CONFIG_ARM64_MTE
> > + DEFINE(THREAD_GCR_EL1_USER, offsetof(struct task_struct, thread.gcr_user_excl));
> > #endif
> > BLANK();
> > DEFINE(S_X0, offsetof(struct pt_regs, regs[0]));
> > diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
> > index eca06b8c74db..e76634ad5bc7 100644
> > --- a/arch/arm64/kernel/cpufeature.c
> > +++ b/arch/arm64/kernel/cpufeature.c
> > @@ -1721,6 +1721,9 @@ static void cpu_enable_mte(struct arm64_cpu_capabilities const *cap)
> >
> > /* Enable in-kernel MTE only if KASAN_HW_TAGS is enabled */
> > if (IS_ENABLED(CONFIG_KASAN_HW_TAGS)) {
> > + /* Enable the kernel exclude mask for random tags generation */
> > + write_sysreg_s(SYS_GCR_EL1_RRND | gcr_kernel_excl, SYS_GCR_EL1);
> > +
> > /* Enable MTE Sync Mode for EL1 */
> > sysreg_clear_set(sctlr_el1, SCTLR_ELx_TCF_MASK, SCTLR_ELx_TCF_SYNC);
> > isb();
> > diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S
> > index ff34461524d4..eeaac91021bf 100644
> > --- a/arch/arm64/kernel/entry.S
> > +++ b/arch/arm64/kernel/entry.S
> > @@ -175,6 +175,43 @@ alternative_else_nop_endif
> > #endif
> > .endm
> >
> > + .macro mte_set_gcr, tmp, tmp2
> > +#ifdef CONFIG_ARM64_MTE
> > + /*
> > + * Calculate and set the exclude mask preserving
> > + * the RRND (bit[16]) setting.
> > + */
> > + mrs_s \tmp2, SYS_GCR_EL1
> > + bfi \tmp2, \tmp, #0, #16
> > + msr_s SYS_GCR_EL1, \tmp2
> > + isb
> > +#endif
> > + .endm
> > +
> > + .macro mte_set_kernel_gcr, tmp, tmp2
> > +#ifdef CONFIG_KASAN_HW_TAGS
> > +alternative_if_not ARM64_MTE
> > + b 1f
> > +alternative_else_nop_endif
> > + ldr_l \tmp, gcr_kernel_excl
> > +
> > + mte_set_gcr \tmp, \tmp2
> > +1:
> > +#endif
> > + .endm
> > +
> > + .macro mte_set_user_gcr, tsk, tmp, tmp2
> > +#ifdef CONFIG_ARM64_MTE
> > +alternative_if_not ARM64_MTE
> > + b 1f
> > +alternative_else_nop_endif
> > + ldr \tmp, [\tsk, #THREAD_GCR_EL1_USER]
> > +
> > + mte_set_gcr \tmp, \tmp2
> > +1:
> > +#endif
> > + .endm
> > +
> > .macro kernel_entry, el, regsize = 64
> > .if \regsize == 32
> > mov w0, w0 // zero upper 32 bits of x0
> > @@ -214,6 +251,8 @@ alternative_else_nop_endif
> >
> > ptrauth_keys_install_kernel tsk, x20, x22, x23
> >
> > + mte_set_kernel_gcr x22, x23
> > +
> > scs_load tsk, x20
> > .else
> > add x21, sp, #S_FRAME_SIZE
> > @@ -332,6 +371,8 @@ alternative_else_nop_endif
> > /* No kernel C function calls after this as user keys are set. */
> > ptrauth_keys_install_user tsk, x0, x1, x2
> >
> > + mte_set_user_gcr tsk, x0, x1
> > +
> > apply_ssbd 0, x0, x1
> > .endif
> >
> > diff --git a/arch/arm64/kernel/mte.c b/arch/arm64/kernel/mte.c
> > index a9f03be75cef..ca8206b7f9a6 100644
> > --- a/arch/arm64/kernel/mte.c
> > +++ b/arch/arm64/kernel/mte.c
> > @@ -23,6 +23,8 @@
> > #include <asm/ptrace.h>
> > #include <asm/sysreg.h>
> >
> > +u64 gcr_kernel_excl __ro_after_init;
> > +
> > static void mte_sync_page_tags(struct page *page, pte_t *ptep, bool check_swap)
> > {
> > pte_t old_pte = READ_ONCE(*ptep);
> > @@ -121,6 +123,17 @@ void *mte_set_mem_tag_range(void *addr, size_t size, u8 tag)
> > return ptr;
> > }
> >
> > +void mte_init_tags(u64 max_tag)
> > +{
> > + /*
> > + * The format of the tags in KASAN is 0xFF and in MTE is 0xF.
> > + * This conversion is required to extract the MTE tag from a KASAN one.
> > + */
> > + u64 incl = GENMASK(FIELD_GET(MTE_TAG_MASK >> MTE_TAG_SHIFT, max_tag), 0);
> > +
> > + gcr_kernel_excl = ~incl & SYS_GCR_EL1_EXCL_MASK;
> > +}
> > +
> > static void update_sctlr_el1_tcf0(u64 tcf0)
> > {
> > /* ISB required for the kernel uaccess routines */
> > @@ -156,7 +169,11 @@ static void update_gcr_el1_excl(u64 excl)
> > static void set_gcr_el1_excl(u64 excl)
> > {
> > current->thread.gcr_user_excl = excl;
> > - update_gcr_el1_excl(excl);
> > +
> > + /*
> > + * SYS_GCR_EL1 will be set to current->thread.gcr_user_excl value
> > + * by mte_set_user_gcr() in kernel_exit,
> > + */
> > }
> >
> > void flush_mte_state(void)
> > @@ -182,7 +199,6 @@ void mte_thread_switch(struct task_struct *next)
> > /* avoid expensive SCTLR_EL1 accesses if no change */
> > if (current->thread.sctlr_tcf0 != next->thread.sctlr_tcf0)
> > update_sctlr_el1_tcf0(next->thread.sctlr_tcf0);
> > - update_gcr_el1_excl(next->thread.gcr_user_excl);
> > }
> >
> > void mte_suspend_exit(void)
> > @@ -190,7 +206,7 @@ void mte_suspend_exit(void)
> > if (!system_supports_mte())
> > return;
> >
> > - update_gcr_el1_excl(current->thread.gcr_user_excl);
> > + update_gcr_el1_excl(gcr_kernel_excl);
> > }
> >
> > long set_mte_ctrl(struct task_struct *task, unsigned long arg)
> > --
> > 2.28.0.1011.ga647a8990f-goog
> >
next prev parent reply other threads:[~2020-10-29 16:53 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-12 20:44 [PATCH v5 00/40] kasan: add hardware tag-based mode for arm64 Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 01/40] arm64: Enable armv8.5-a asm-arch option Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 02/40] arm64: mte: Add in-kernel MTE helpers Andrey Konovalov
2020-10-28 11:28 ` Dmitry Vyukov
2020-10-29 16:50 ` Andrey Konovalov
2020-11-04 17:49 ` Vincenzo Frascino
2020-10-12 20:44 ` [PATCH v5 03/40] arm64: mte: Reset the page tag in page->flags Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 04/40] arm64: kasan: Add arch layer for memory tagging helpers Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 05/40] arm64: mte: Add in-kernel tag fault handler Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 06/40] arm64: kasan: Enable in-kernel MTE Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 07/40] arm64: mte: Convert gcr_user into an exclude mask Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 08/40] arm64: mte: Switch GCR_EL1 in kernel entry and exit Andrey Konovalov
2020-10-28 10:06 ` Dmitry Vyukov
2020-10-29 16:52 ` Andrey Konovalov [this message]
2020-10-12 20:44 ` [PATCH v5 09/40] arm64: kasan: Align allocations for HW_TAGS Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 10/40] kasan: drop unnecessary GPL text from comment headers Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 11/40] kasan: KASAN_VMALLOC depends on KASAN_GENERIC Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 12/40] kasan: group vmalloc code Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 13/40] kasan: shadow declarations only for software modes Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 14/40] kasan: rename (un)poison_shadow to (un)poison_memory Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 15/40] kasan: rename KASAN_SHADOW_* to KASAN_GRANULE_* Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 16/40] kasan: only build init.c for software modes Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 17/40] kasan: split out shadow.c from common.c Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 18/40] kasan: define KASAN_GRANULE_PAGE Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 19/40] kasan: rename report and tags files Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 20/40] kasan: don't duplicate config dependencies Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 21/40] kasan: hide invalid free check implementation Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 22/40] kasan: decode stack frame only with KASAN_STACK_ENABLE Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 23/40] kasan, arm64: only init shadow for software modes Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 24/40] kasan, arm64: only use kasan_depth " Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 25/40] kasan: rename addr_has_shadow to addr_has_metadata Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 26/40] kasan: rename print_shadow_for_address to print_memory_metadata Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 27/40] kasan: kasan_non_canonical_hook only for software modes Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 28/40] kasan: rename SHADOW layout macros to META Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 29/40] kasan: separate metadata_fetch_row for each mode Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 30/40] kasan, arm64: don't allow SW_TAGS with ARM64_MTE Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 31/40] kasan: introduce CONFIG_KASAN_HW_TAGS Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 32/40] kasan: define KASAN_GRANULE_SIZE for HW_TAGS Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 33/40] kasan, x86, s390: update undef CONFIG_KASAN Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 34/40] kasan, arm64: expand CONFIG_KASAN checks Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 35/40] kasan, arm64: implement HW_TAGS runtime Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 36/40] kasan, arm64: print report from tag fault handler Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 37/40] kasan, mm: reset tags when accessing metadata Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 38/40] kasan, arm64: enable CONFIG_KASAN_HW_TAGS Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 39/40] kasan: add documentation for hardware tag-based mode Andrey Konovalov
2020-10-12 20:44 ` [PATCH v5 40/40] kselftest/arm64: Check GCR_EL1 after context switch Andrey Konovalov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAAeHK+yARu1Qh46DPbfs4h37yHPAVgx6r8r=86L6jfHD3up8-g@mail.gmail.com' \
--to=andreyknvl@google.com \
--cc=Branislav.Rankov@arm.com \
--cc=akpm@linux-foundation.org \
--cc=aryabinin@virtuozzo.com \
--cc=catalin.marinas@arm.com \
--cc=dvyukov@google.com \
--cc=elver@google.com \
--cc=eugenis@google.com \
--cc=glider@google.com \
--cc=kasan-dev@googlegroups.com \
--cc=kevin.brodsky@arm.com \
--cc=lenaptr@google.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=vincenzo.frascino@arm.com \
--cc=will.deacon@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).