* [PATCH 1/9] arm64: cpufeature: make cpus_have_cap() noinstr-safe
2022-09-01 15:13 [PATCH 0/9] arm64: alternatives: improvements Mark Rutland
@ 2022-09-01 15:13 ` Mark Rutland
2022-09-01 15:13 ` [PATCH 2/9] arm64: alternatives: kvm: prepare for cap changes Mark Rutland
` (7 subsequent siblings)
8 siblings, 0 replies; 20+ messages in thread
From: Mark Rutland @ 2022-09-01 15:13 UTC (permalink / raw)
To: linux-arm-kernel
Cc: ardb, catalin.marinas, james.morse, joey.gouly, mark.rutland, maz, will
Currently it isn't safe to use cpus_have_cap() from noinstr code as
test_bit() is explicitly instrumented, and were cpus_have_cap() placed
out-of-line, cpus_have_cap() itself could be instrumented.
Make cpus_have_cap() noinstr safe by marking it __always_inline and
using arch_test_bit().
Aside from the prevention of instrumentation, there should be no
functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
---
arch/arm64/include/asm/cpufeature.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h
index fd7d75a275f6c..630c337c77746 100644
--- a/arch/arm64/include/asm/cpufeature.h
+++ b/arch/arm64/include/asm/cpufeature.h
@@ -448,11 +448,11 @@ static __always_inline bool system_capabilities_finalized(void)
*
* Before the capability is detected, this returns false.
*/
-static inline bool cpus_have_cap(unsigned int num)
+static __always_inline bool cpus_have_cap(unsigned int num)
{
if (num >= ARM64_NCAPS)
return false;
- return test_bit(num, cpu_hwcaps);
+ return arch_test_bit(num, cpu_hwcaps);
}
/*
--
2.30.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 20+ messages in thread* [PATCH 2/9] arm64: alternatives: kvm: prepare for cap changes
2022-09-01 15:13 [PATCH 0/9] arm64: alternatives: improvements Mark Rutland
2022-09-01 15:13 ` [PATCH 1/9] arm64: cpufeature: make cpus_have_cap() noinstr-safe Mark Rutland
@ 2022-09-01 15:13 ` Mark Rutland
2022-09-01 15:13 ` [PATCH 3/9] arm64: alternatives: proton-pack: " Mark Rutland
` (6 subsequent siblings)
8 siblings, 0 replies; 20+ messages in thread
From: Mark Rutland @ 2022-09-01 15:13 UTC (permalink / raw)
To: linux-arm-kernel
Cc: ardb, catalin.marinas, james.morse, joey.gouly, mark.rutland, maz, will
The KVM patching callbacks use cpus_have_final_cap() internally within
has_vhe(), and subsequent patches will make it invalid to call
cpus_have_final_cap() before alternatives patching has completed, and
will mean that cpus_have_const_cap() will always fall back to dynamic
checks prior to alternatives patching.
In preparation for said change, this patch modifies the KVM patching
callbacks to use cpus_have_cap() directly. This is not subject to
patching, and will dynamically check the cpu_hwcaps array, which is
functionally equivalent to the existing behaviour.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
---
arch/arm64/kvm/va_layout.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/kvm/va_layout.c b/arch/arm64/kvm/va_layout.c
index acdb7b3cc97d6..91b22a014610b 100644
--- a/arch/arm64/kvm/va_layout.c
+++ b/arch/arm64/kvm/va_layout.c
@@ -169,7 +169,7 @@ void __init kvm_update_va_mask(struct alt_instr *alt,
* dictates it and we don't have any spare bits in the
* address), NOP everything after masking the kernel VA.
*/
- if (has_vhe() || (!tag_val && i > 0)) {
+ if (cpus_have_cap(ARM64_HAS_VIRT_HOST_EXTN) || (!tag_val && i > 0)) {
updptr[i] = cpu_to_le32(aarch64_insn_gen_nop());
continue;
}
@@ -193,7 +193,8 @@ void kvm_patch_vector_branch(struct alt_instr *alt,
BUG_ON(nr_inst != 4);
- if (!cpus_have_const_cap(ARM64_SPECTRE_V3A) || WARN_ON_ONCE(has_vhe()))
+ if (!cpus_have_cap(ARM64_SPECTRE_V3A) ||
+ WARN_ON_ONCE(cpus_have_cap(ARM64_HAS_VIRT_HOST_EXTN)))
return;
/*
--
2.30.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 20+ messages in thread* [PATCH 3/9] arm64: alternatives: proton-pack: prepare for cap changes
2022-09-01 15:13 [PATCH 0/9] arm64: alternatives: improvements Mark Rutland
2022-09-01 15:13 ` [PATCH 1/9] arm64: cpufeature: make cpus_have_cap() noinstr-safe Mark Rutland
2022-09-01 15:13 ` [PATCH 2/9] arm64: alternatives: kvm: prepare for cap changes Mark Rutland
@ 2022-09-01 15:13 ` Mark Rutland
2022-09-02 16:19 ` Joey Gouly
2022-09-01 15:13 ` [PATCH 4/9] arm64: alternatives: hoist print out of __apply_alternatives() Mark Rutland
` (5 subsequent siblings)
8 siblings, 1 reply; 20+ messages in thread
From: Mark Rutland @ 2022-09-01 15:13 UTC (permalink / raw)
To: linux-arm-kernel
Cc: ardb, catalin.marinas, james.morse, joey.gouly, mark.rutland, maz, will
The spectre patching callbacks use cpus_have_final_cap() internally within
has_vhe(), and subsequent patches will make it invalid to call
cpus_have_final_cap() before alternatives patching has completed.
In preparation for said change, this patch modifies the spectre patching
callbacks use cpus_have_cap(). This is not subject to patching, and will
dynamically check the cpu_hwcaps array, which is functionally equivalent
to the existing behaviour.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
---
arch/arm64/kernel/proton-pack.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c
index 40be3a7c2c531..8010b2e573fac 100644
--- a/arch/arm64/kernel/proton-pack.c
+++ b/arch/arm64/kernel/proton-pack.c
@@ -586,7 +586,7 @@ void __init spectre_v4_patch_fw_mitigation_enable(struct alt_instr *alt,
if (spectre_v4_mitigations_off())
return;
- if (cpus_have_final_cap(ARM64_SSBS))
+ if (cpus_have_cap(ARM64_SSBS))
return;
if (spectre_v4_mitigations_dynamic())
--
2.30.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH 3/9] arm64: alternatives: proton-pack: prepare for cap changes
2022-09-01 15:13 ` [PATCH 3/9] arm64: alternatives: proton-pack: " Mark Rutland
@ 2022-09-02 16:19 ` Joey Gouly
2022-09-05 8:46 ` Mark Rutland
0 siblings, 1 reply; 20+ messages in thread
From: Joey Gouly @ 2022-09-02 16:19 UTC (permalink / raw)
To: Mark Rutland
Cc: linux-arm-kernel, ardb, catalin.marinas, james.morse, maz, will, nd
Hi Mark,
On Thu, Sep 01, 2022 at 04:13:57PM +0100, Mark Rutland wrote:
> The spectre patching callbacks use cpus_have_final_cap() internally within
> has_vhe(), and subsequent patches will make it invalid to call
Remove the 'internally within has_vhe()', a copy paste error from the previous
commit.
Otherwise:
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
> cpus_have_final_cap() before alternatives patching has completed.
>
> In preparation for said change, this patch modifies the spectre patching
> callbacks use cpus_have_cap(). This is not subject to patching, and will
> dynamically check the cpu_hwcaps array, which is functionally equivalent
> to the existing behaviour.
>
> There should be no functional change as a result of this patch.
>
> Signed-off-by: Mark Rutland <mark.rutland@arm.com>
> Cc: Ard Biesheuvel <ardb@kernel.org>
> Cc: Catalin Marinas <catalin.marinas@arm.com>
> Cc: James Morse <james.morse@arm.com>
> Cc: Joey Gouly <joey.gouly@arm.com>
> Cc: Marc Zyngier <maz@kernel.org>
> Cc: Will Deacon <will@kernel.org>
> ---
> arch/arm64/kernel/proton-pack.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c
> index 40be3a7c2c531..8010b2e573fac 100644
> --- a/arch/arm64/kernel/proton-pack.c
> +++ b/arch/arm64/kernel/proton-pack.c
> @@ -586,7 +586,7 @@ void __init spectre_v4_patch_fw_mitigation_enable(struct alt_instr *alt,
> if (spectre_v4_mitigations_off())
> return;
>
> - if (cpus_have_final_cap(ARM64_SSBS))
> + if (cpus_have_cap(ARM64_SSBS))
> return;
>
> if (spectre_v4_mitigations_dynamic())
> --
> 2.30.2
>
>
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 3/9] arm64: alternatives: proton-pack: prepare for cap changes
2022-09-02 16:19 ` Joey Gouly
@ 2022-09-05 8:46 ` Mark Rutland
0 siblings, 0 replies; 20+ messages in thread
From: Mark Rutland @ 2022-09-05 8:46 UTC (permalink / raw)
To: Joey Gouly
Cc: linux-arm-kernel, ardb, catalin.marinas, james.morse, maz, will, nd
On Fri, Sep 02, 2022 at 05:19:32PM +0100, Joey Gouly wrote:
> Hi Mark,
>
> On Thu, Sep 01, 2022 at 04:13:57PM +0100, Mark Rutland wrote:
> > The spectre patching callbacks use cpus_have_final_cap() internally within
> > has_vhe(), and subsequent patches will make it invalid to call
>
> Remove the 'internally within has_vhe()', a copy paste error from the previous
> commit.
Whoops; removed.
> Otherwise:
>
> Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Thanks!
Mark.
>
> > cpus_have_final_cap() before alternatives patching has completed.
> >
> > In preparation for said change, this patch modifies the spectre patching
> > callbacks use cpus_have_cap(). This is not subject to patching, and will
> > dynamically check the cpu_hwcaps array, which is functionally equivalent
> > to the existing behaviour.
> >
> > There should be no functional change as a result of this patch.
> >
> > Signed-off-by: Mark Rutland <mark.rutland@arm.com>
> > Cc: Ard Biesheuvel <ardb@kernel.org>
> > Cc: Catalin Marinas <catalin.marinas@arm.com>
> > Cc: James Morse <james.morse@arm.com>
> > Cc: Joey Gouly <joey.gouly@arm.com>
> > Cc: Marc Zyngier <maz@kernel.org>
> > Cc: Will Deacon <will@kernel.org>
> > ---
> > arch/arm64/kernel/proton-pack.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c
> > index 40be3a7c2c531..8010b2e573fac 100644
> > --- a/arch/arm64/kernel/proton-pack.c
> > +++ b/arch/arm64/kernel/proton-pack.c
> > @@ -586,7 +586,7 @@ void __init spectre_v4_patch_fw_mitigation_enable(struct alt_instr *alt,
> > if (spectre_v4_mitigations_off())
> > return;
> >
> > - if (cpus_have_final_cap(ARM64_SSBS))
> > + if (cpus_have_cap(ARM64_SSBS))
> > return;
> >
> > if (spectre_v4_mitigations_dynamic())
> > --
> > 2.30.2
> >
> >
> > _______________________________________________
> > linux-arm-kernel mailing list
> > linux-arm-kernel@lists.infradead.org
> > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH 4/9] arm64: alternatives: hoist print out of __apply_alternatives()
2022-09-01 15:13 [PATCH 0/9] arm64: alternatives: improvements Mark Rutland
` (2 preceding siblings ...)
2022-09-01 15:13 ` [PATCH 3/9] arm64: alternatives: proton-pack: " Mark Rutland
@ 2022-09-01 15:13 ` Mark Rutland
2022-09-01 15:13 ` [PATCH 5/9] arm64: alternatives: make alt_region const Mark Rutland
` (4 subsequent siblings)
8 siblings, 0 replies; 20+ messages in thread
From: Mark Rutland @ 2022-09-01 15:13 UTC (permalink / raw)
To: linux-arm-kernel
Cc: ardb, catalin.marinas, james.morse, joey.gouly, mark.rutland, maz, will
Printing in the middle of __apply_alternatives() is potentially unsafe
and not all that helpful given these days we practically always patch
*something*.
Hoist the print out of __apply_alternatives(), and add separate prints
to __apply_alternatives() and apply_alternatives_all(), which will make
it easier to spot if either patching call goes wrong.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
---
arch/arm64/kernel/alternative.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
index 9bcaa5eacf16c..d94d97cb4a0bf 100644
--- a/arch/arm64/kernel/alternative.c
+++ b/arch/arm64/kernel/alternative.c
@@ -156,8 +156,6 @@ static void __nocfi __apply_alternatives(struct alt_region *region, bool is_modu
else
BUG_ON(alt->alt_len != alt->orig_len);
- pr_info_once("patching kernel code\n");
-
origptr = ALT_ORIG_PTR(alt);
updptr = is_module ? origptr : lm_alias(origptr);
nr_inst = alt->orig_len / AARCH64_INSN_SIZE;
@@ -225,6 +223,8 @@ static int __apply_alternatives_multi_stop(void *unused)
void __init apply_alternatives_all(void)
{
+ pr_info("applying system-wide alternatives\n");
+
/* better not try code patching on a live SMP system */
stop_machine(__apply_alternatives_multi_stop, NULL, cpu_online_mask);
}
@@ -244,6 +244,8 @@ void __init apply_boot_alternatives(void)
/* If called on non-boot cpu things could go wrong */
WARN_ON(smp_processor_id() != 0);
+ pr_info("applying boot alternatives\n");
+
__apply_alternatives(®ion, false, &boot_capabilities[0]);
}
--
2.30.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 20+ messages in thread* [PATCH 5/9] arm64: alternatives: make alt_region const
2022-09-01 15:13 [PATCH 0/9] arm64: alternatives: improvements Mark Rutland
` (3 preceding siblings ...)
2022-09-01 15:13 ` [PATCH 4/9] arm64: alternatives: hoist print out of __apply_alternatives() Mark Rutland
@ 2022-09-01 15:13 ` Mark Rutland
2022-09-06 15:18 ` Ard Biesheuvel
2022-09-01 15:14 ` [PATCH 6/9] arm64: alternatives: have callbacks take a cap Mark Rutland
` (3 subsequent siblings)
8 siblings, 1 reply; 20+ messages in thread
From: Mark Rutland @ 2022-09-01 15:13 UTC (permalink / raw)
To: linux-arm-kernel
Cc: ardb, catalin.marinas, james.morse, joey.gouly, mark.rutland, maz, will
We never alter a struct alt_region after creation, and we open-code the
bounds of the kernel alternatives region in two functions.
This patch adds a shared struct alt_region, and marks the alt regions as
const to prevent unintentional modification.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
---
arch/arm64/kernel/alternative.c | 26 ++++++++++++--------------
1 file changed, 12 insertions(+), 14 deletions(-)
diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
index d94d97cb4a0bf..2e18c9c0f612b 100644
--- a/arch/arm64/kernel/alternative.c
+++ b/arch/arm64/kernel/alternative.c
@@ -133,8 +133,9 @@ static void clean_dcache_range_nopatch(u64 start, u64 end)
} while (cur += d_size, cur < end);
}
-static void __nocfi __apply_alternatives(struct alt_region *region, bool is_module,
- unsigned long *feature_mask)
+static void __nocfi __apply_alternatives(const struct alt_region *region,
+ bool is_module,
+ unsigned long *feature_mask)
{
struct alt_instr *alt;
__le32 *origptr, *updptr;
@@ -190,17 +191,17 @@ static void __nocfi __apply_alternatives(struct alt_region *region, bool is_modu
}
}
+static const struct alt_region kernel_alternatives = {
+ .begin = (struct alt_instr *)__alt_instructions,
+ .end = (struct alt_instr *)__alt_instructions_end,
+};
+
/*
* We might be patching the stop_machine state machine, so implement a
* really simple polling protocol here.
*/
static int __apply_alternatives_multi_stop(void *unused)
{
- struct alt_region region = {
- .begin = (struct alt_instr *)__alt_instructions,
- .end = (struct alt_instr *)__alt_instructions_end,
- };
-
/* We always have a CPU 0 at this point (__init) */
if (smp_processor_id()) {
while (!all_alternatives_applied)
@@ -213,7 +214,8 @@ static int __apply_alternatives_multi_stop(void *unused)
ARM64_NPATCHABLE);
BUG_ON(all_alternatives_applied);
- __apply_alternatives(®ion, false, remaining_capabilities);
+ __apply_alternatives(&kernel_alternatives, false,
+ remaining_capabilities);
/* Barriers provided by the cache flushing */
all_alternatives_applied = 1;
}
@@ -236,17 +238,13 @@ void __init apply_alternatives_all(void)
*/
void __init apply_boot_alternatives(void)
{
- struct alt_region region = {
- .begin = (struct alt_instr *)__alt_instructions,
- .end = (struct alt_instr *)__alt_instructions_end,
- };
-
/* If called on non-boot cpu things could go wrong */
WARN_ON(smp_processor_id() != 0);
pr_info("applying boot alternatives\n");
- __apply_alternatives(®ion, false, &boot_capabilities[0]);
+ __apply_alternatives(&kernel_alternatives, false,
+ &boot_capabilities[0]);
}
#ifdef CONFIG_MODULES
--
2.30.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH 5/9] arm64: alternatives: make alt_region const
2022-09-01 15:13 ` [PATCH 5/9] arm64: alternatives: make alt_region const Mark Rutland
@ 2022-09-06 15:18 ` Ard Biesheuvel
2022-09-12 9:31 ` Mark Rutland
0 siblings, 1 reply; 20+ messages in thread
From: Ard Biesheuvel @ 2022-09-06 15:18 UTC (permalink / raw)
To: Mark Rutland
Cc: linux-arm-kernel, catalin.marinas, james.morse, joey.gouly, maz, will
Hi Mark,
On Thu, 1 Sept 2022 at 17:14, Mark Rutland <mark.rutland@arm.com> wrote:
>
> We never alter a struct alt_region after creation, and we open-code the
> bounds of the kernel alternatives region in two functions.
>
> This patch adds a shared struct alt_region, and marks the alt regions as
> const to prevent unintentional modification.
>
What is the point of struct alt_region? Can we just get rid of it
entirely? It seems its only purpose is to carry a <start, end> tuple
that could easily be converted into start and end arguments to
__apply_alternatives().
> There should be no functional change as a result of this patch.
>
> Signed-off-by: Mark Rutland <mark.rutland@arm.com>
> Cc: Ard Biesheuvel <ardb@kernel.org>
> Cc: Catalin Marinas <catalin.marinas@arm.com>
> Cc: James Morse <james.morse@arm.com>
> Cc: Joey Gouly <joey.gouly@arm.com>
> Cc: Marc Zyngier <maz@kernel.org>
> Cc: Will Deacon <will@kernel.org>
> ---
> arch/arm64/kernel/alternative.c | 26 ++++++++++++--------------
> 1 file changed, 12 insertions(+), 14 deletions(-)
>
> diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
> index d94d97cb4a0bf..2e18c9c0f612b 100644
> --- a/arch/arm64/kernel/alternative.c
> +++ b/arch/arm64/kernel/alternative.c
> @@ -133,8 +133,9 @@ static void clean_dcache_range_nopatch(u64 start, u64 end)
> } while (cur += d_size, cur < end);
> }
>
> -static void __nocfi __apply_alternatives(struct alt_region *region, bool is_module,
> - unsigned long *feature_mask)
> +static void __nocfi __apply_alternatives(const struct alt_region *region,
> + bool is_module,
> + unsigned long *feature_mask)
> {
> struct alt_instr *alt;
> __le32 *origptr, *updptr;
> @@ -190,17 +191,17 @@ static void __nocfi __apply_alternatives(struct alt_region *region, bool is_modu
> }
> }
>
> +static const struct alt_region kernel_alternatives = {
> + .begin = (struct alt_instr *)__alt_instructions,
> + .end = (struct alt_instr *)__alt_instructions_end,
> +};
> +
> /*
> * We might be patching the stop_machine state machine, so implement a
> * really simple polling protocol here.
> */
> static int __apply_alternatives_multi_stop(void *unused)
> {
> - struct alt_region region = {
> - .begin = (struct alt_instr *)__alt_instructions,
> - .end = (struct alt_instr *)__alt_instructions_end,
> - };
> -
> /* We always have a CPU 0 at this point (__init) */
> if (smp_processor_id()) {
> while (!all_alternatives_applied)
> @@ -213,7 +214,8 @@ static int __apply_alternatives_multi_stop(void *unused)
> ARM64_NPATCHABLE);
>
> BUG_ON(all_alternatives_applied);
> - __apply_alternatives(®ion, false, remaining_capabilities);
> + __apply_alternatives(&kernel_alternatives, false,
> + remaining_capabilities);
> /* Barriers provided by the cache flushing */
> all_alternatives_applied = 1;
> }
> @@ -236,17 +238,13 @@ void __init apply_alternatives_all(void)
> */
> void __init apply_boot_alternatives(void)
> {
> - struct alt_region region = {
> - .begin = (struct alt_instr *)__alt_instructions,
> - .end = (struct alt_instr *)__alt_instructions_end,
> - };
> -
> /* If called on non-boot cpu things could go wrong */
> WARN_ON(smp_processor_id() != 0);
>
> pr_info("applying boot alternatives\n");
>
> - __apply_alternatives(®ion, false, &boot_capabilities[0]);
> + __apply_alternatives(&kernel_alternatives, false,
> + &boot_capabilities[0]);
> }
>
> #ifdef CONFIG_MODULES
> --
> 2.30.2
>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH 5/9] arm64: alternatives: make alt_region const
2022-09-06 15:18 ` Ard Biesheuvel
@ 2022-09-12 9:31 ` Mark Rutland
2022-09-12 10:13 ` Ard Biesheuvel
0 siblings, 1 reply; 20+ messages in thread
From: Mark Rutland @ 2022-09-12 9:31 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: linux-arm-kernel, catalin.marinas, james.morse, joey.gouly, maz, will
On Tue, Sep 06, 2022 at 05:18:54PM +0200, Ard Biesheuvel wrote:
> Hi Mark,
>
> On Thu, 1 Sept 2022 at 17:14, Mark Rutland <mark.rutland@arm.com> wrote:
> >
> > We never alter a struct alt_region after creation, and we open-code the
> > bounds of the kernel alternatives region in two functions.
> >
> > This patch adds a shared struct alt_region, and marks the alt regions as
> > const to prevent unintentional modification.
> >
>
> What is the point of struct alt_region? Can we just get rid of it
> entirely? It seems its only purpose is to carry a <start, end> tuple
> that could easily be converted into start and end arguments to
> __apply_alternatives().
We could right now, but I'm intending to fix some noinstr issues (and make
debugging easier) by splitting the alternatives sanity-checking & patching into
distinct initcalls (alnog with some extra debug), and having the structure for
the common definition is quite nice to avoid open-coding the start and end
value in a bunch of places.
So I'd prefer to keep it for now, but I can follow up and delete it if the
above doesn't turn out to need it, if that's ok?
Mark.
> > There should be no functional change as a result of this patch.
> >
> > Signed-off-by: Mark Rutland <mark.rutland@arm.com>
> > Cc: Ard Biesheuvel <ardb@kernel.org>
> > Cc: Catalin Marinas <catalin.marinas@arm.com>
> > Cc: James Morse <james.morse@arm.com>
> > Cc: Joey Gouly <joey.gouly@arm.com>
> > Cc: Marc Zyngier <maz@kernel.org>
> > Cc: Will Deacon <will@kernel.org>
> > ---
> > arch/arm64/kernel/alternative.c | 26 ++++++++++++--------------
> > 1 file changed, 12 insertions(+), 14 deletions(-)
> >
> > diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
> > index d94d97cb4a0bf..2e18c9c0f612b 100644
> > --- a/arch/arm64/kernel/alternative.c
> > +++ b/arch/arm64/kernel/alternative.c
> > @@ -133,8 +133,9 @@ static void clean_dcache_range_nopatch(u64 start, u64 end)
> > } while (cur += d_size, cur < end);
> > }
> >
> > -static void __nocfi __apply_alternatives(struct alt_region *region, bool is_module,
> > - unsigned long *feature_mask)
> > +static void __nocfi __apply_alternatives(const struct alt_region *region,
> > + bool is_module,
> > + unsigned long *feature_mask)
> > {
> > struct alt_instr *alt;
> > __le32 *origptr, *updptr;
> > @@ -190,17 +191,17 @@ static void __nocfi __apply_alternatives(struct alt_region *region, bool is_modu
> > }
> > }
> >
> > +static const struct alt_region kernel_alternatives = {
> > + .begin = (struct alt_instr *)__alt_instructions,
> > + .end = (struct alt_instr *)__alt_instructions_end,
> > +};
> > +
> > /*
> > * We might be patching the stop_machine state machine, so implement a
> > * really simple polling protocol here.
> > */
> > static int __apply_alternatives_multi_stop(void *unused)
> > {
> > - struct alt_region region = {
> > - .begin = (struct alt_instr *)__alt_instructions,
> > - .end = (struct alt_instr *)__alt_instructions_end,
> > - };
> > -
> > /* We always have a CPU 0 at this point (__init) */
> > if (smp_processor_id()) {
> > while (!all_alternatives_applied)
> > @@ -213,7 +214,8 @@ static int __apply_alternatives_multi_stop(void *unused)
> > ARM64_NPATCHABLE);
> >
> > BUG_ON(all_alternatives_applied);
> > - __apply_alternatives(®ion, false, remaining_capabilities);
> > + __apply_alternatives(&kernel_alternatives, false,
> > + remaining_capabilities);
> > /* Barriers provided by the cache flushing */
> > all_alternatives_applied = 1;
> > }
> > @@ -236,17 +238,13 @@ void __init apply_alternatives_all(void)
> > */
> > void __init apply_boot_alternatives(void)
> > {
> > - struct alt_region region = {
> > - .begin = (struct alt_instr *)__alt_instructions,
> > - .end = (struct alt_instr *)__alt_instructions_end,
> > - };
> > -
> > /* If called on non-boot cpu things could go wrong */
> > WARN_ON(smp_processor_id() != 0);
> >
> > pr_info("applying boot alternatives\n");
> >
> > - __apply_alternatives(®ion, false, &boot_capabilities[0]);
> > + __apply_alternatives(&kernel_alternatives, false,
> > + &boot_capabilities[0]);
> > }
> >
> > #ifdef CONFIG_MODULES
> > --
> > 2.30.2
> >
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH 5/9] arm64: alternatives: make alt_region const
2022-09-12 9:31 ` Mark Rutland
@ 2022-09-12 10:13 ` Ard Biesheuvel
2022-09-12 12:13 ` Mark Rutland
0 siblings, 1 reply; 20+ messages in thread
From: Ard Biesheuvel @ 2022-09-12 10:13 UTC (permalink / raw)
To: Mark Rutland
Cc: linux-arm-kernel, catalin.marinas, james.morse, joey.gouly, maz, will
On Mon, 12 Sept 2022 at 10:32, Mark Rutland <mark.rutland@arm.com> wrote:
>
> On Tue, Sep 06, 2022 at 05:18:54PM +0200, Ard Biesheuvel wrote:
> > Hi Mark,
> >
> > On Thu, 1 Sept 2022 at 17:14, Mark Rutland <mark.rutland@arm.com> wrote:
> > >
> > > We never alter a struct alt_region after creation, and we open-code the
> > > bounds of the kernel alternatives region in two functions.
> > >
> > > This patch adds a shared struct alt_region, and marks the alt regions as
> > > const to prevent unintentional modification.
> > >
> >
> > What is the point of struct alt_region? Can we just get rid of it
> > entirely? It seems its only purpose is to carry a <start, end> tuple
> > that could easily be converted into start and end arguments to
> > __apply_alternatives().
>
> We could right now, but I'm intending to fix some noinstr issues (and make
> debugging easier) by splitting the alternatives sanity-checking & patching into
> distinct initcalls (alnog with some extra debug), and having the structure for
> the common definition is quite nice to avoid open-coding the start and end
> value in a bunch of places.
>
> So I'd prefer to keep it for now, but I can follow up and delete it if the
> above doesn't turn out to need it, if that's ok?
>
Yeah, that's fine. The above wasn't really clear to me from the
context you provided.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 5/9] arm64: alternatives: make alt_region const
2022-09-12 10:13 ` Ard Biesheuvel
@ 2022-09-12 12:13 ` Mark Rutland
0 siblings, 0 replies; 20+ messages in thread
From: Mark Rutland @ 2022-09-12 12:13 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: linux-arm-kernel, catalin.marinas, james.morse, joey.gouly, maz, will
On Mon, Sep 12, 2022 at 11:13:12AM +0100, Ard Biesheuvel wrote:
> On Mon, 12 Sept 2022 at 10:32, Mark Rutland <mark.rutland@arm.com> wrote:
> >
> > On Tue, Sep 06, 2022 at 05:18:54PM +0200, Ard Biesheuvel wrote:
> > > Hi Mark,
> > >
> > > On Thu, 1 Sept 2022 at 17:14, Mark Rutland <mark.rutland@arm.com> wrote:
> > > >
> > > > We never alter a struct alt_region after creation, and we open-code the
> > > > bounds of the kernel alternatives region in two functions.
> > > >
> > > > This patch adds a shared struct alt_region, and marks the alt regions as
> > > > const to prevent unintentional modification.
> > > >
> > >
> > > What is the point of struct alt_region? Can we just get rid of it
> > > entirely? It seems its only purpose is to carry a <start, end> tuple
> > > that could easily be converted into start and end arguments to
> > > __apply_alternatives().
> >
> > We could right now, but I'm intending to fix some noinstr issues (and make
> > debugging easier) by splitting the alternatives sanity-checking & patching into
> > distinct initcalls (alnog with some extra debug), and having the structure for
> > the common definition is quite nice to avoid open-coding the start and end
> > value in a bunch of places.
> >
> > So I'd prefer to keep it for now, but I can follow up and delete it if the
> > above doesn't turn out to need it, if that's ok?
> >
>
> Yeah, that's fine. The above wasn't really clear to me from the
> context you provided.
True; I'll add a note to the commit message.
Thanks,
Mark.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH 6/9] arm64: alternatives: have callbacks take a cap
2022-09-01 15:13 [PATCH 0/9] arm64: alternatives: improvements Mark Rutland
` (4 preceding siblings ...)
2022-09-01 15:13 ` [PATCH 5/9] arm64: alternatives: make alt_region const Mark Rutland
@ 2022-09-01 15:14 ` Mark Rutland
2022-09-02 15:54 ` Joey Gouly
2022-09-01 15:14 ` [PATCH 7/9] arm64: alternatives: add alternative_has_feature_*() Mark Rutland
` (2 subsequent siblings)
8 siblings, 1 reply; 20+ messages in thread
From: Mark Rutland @ 2022-09-01 15:14 UTC (permalink / raw)
To: linux-arm-kernel
Cc: ardb, catalin.marinas, james.morse, joey.gouly, mark.rutland, maz, will
Today, callback alternatives are special-cased within
__apply_alternatives(), and are applied alongside patching for system
capabilities as ARM64_NCAPS is not part of the boot_capabilities feature
mask.
This special-casing is less than ideal. Giving special meaning to
ARM64_NCAPS for this requires some structures and loops to use
ARM64_NCAPS + 1 (AKA ARM64_NPATCHABLE), while others use ARM64_NCAPS.
It's also not immediately clear callback alternatives are only applied
when applying alternatives for system-wide features.
To make this a bit clearer, changes the way that callback alternatives
are identified to remove the special-casing of ARM64_NCAPS, and to allow
callback alternatives to be associated with a cpucap as with all other
alternatives.
New cpucaps, ARM64_ALWAYS_BOOT and ARM64_ALWAYS_SYSTEM are added which
are always detected alongside boot cpu capabilities and system
capabilities respectively. All existing callback alternatives are made
to use ARM64_ALWAYS_SYSTEM, and so will be patched at the same point
during the boot flow as before.
Subsequent patches will make more use of these new cpucaps.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
---
arch/arm64/include/asm/alternative-macros.h | 18 ++++++++-----
arch/arm64/include/asm/assembler.h | 10 ++++----
arch/arm64/include/asm/cpufeature.h | 4 +--
arch/arm64/include/asm/kvm_mmu.h | 5 ++--
arch/arm64/kernel/alternative.c | 28 ++++++++++++---------
arch/arm64/kernel/cpufeature.c | 19 ++++++++++++--
arch/arm64/kernel/entry.S | 8 +++---
arch/arm64/kvm/hyp/hyp-entry.S | 4 +--
arch/arm64/tools/cpucaps | 2 ++
9 files changed, 62 insertions(+), 36 deletions(-)
diff --git a/arch/arm64/include/asm/alternative-macros.h b/arch/arm64/include/asm/alternative-macros.h
index 7e157ab6cd505..189c31be163ce 100644
--- a/arch/arm64/include/asm/alternative-macros.h
+++ b/arch/arm64/include/asm/alternative-macros.h
@@ -2,10 +2,16 @@
#ifndef __ASM_ALTERNATIVE_MACROS_H
#define __ASM_ALTERNATIVE_MACROS_H
+#include <linux/const.h>
+
#include <asm/cpucaps.h>
#include <asm/insn-def.h>
-#define ARM64_CB_PATCH ARM64_NCAPS
+#define ARM64_CB_BIT (UL(1) << 15)
+
+#if ARM64_NCAPS >= ARM64_CB_BIT
+#error "cpucaps have overflown ARM64_CB_BIT"
+#endif
#ifndef __ASSEMBLY__
@@ -73,8 +79,8 @@
#define _ALTERNATIVE_CFG(oldinstr, newinstr, feature, cfg, ...) \
__ALTERNATIVE_CFG(oldinstr, newinstr, feature, IS_ENABLED(cfg))
-#define ALTERNATIVE_CB(oldinstr, cb) \
- __ALTERNATIVE_CFG_CB(oldinstr, ARM64_CB_PATCH, 1, cb)
+#define ALTERNATIVE_CB(oldinstr, feature, cb) \
+ __ALTERNATIVE_CFG_CB(oldinstr, ARM64_CB_BIT | (feature), 1, cb)
#else
#include <asm/assembler.h>
@@ -82,7 +88,7 @@
.macro altinstruction_entry orig_offset alt_offset feature orig_len alt_len
.word \orig_offset - .
.word \alt_offset - .
- .hword \feature
+ .hword (\feature)
.byte \orig_len
.byte \alt_len
.endm
@@ -141,10 +147,10 @@
661:
.endm
-.macro alternative_cb cb
+.macro alternative_cb cap, cb
.set .Lasm_alt_mode, 0
.pushsection .altinstructions, "a"
- altinstruction_entry 661f, \cb, ARM64_CB_PATCH, 662f-661f, 0
+ altinstruction_entry 661f, \cb, ARM64_CB_BIT | \cap, 662f-661f, 0
.popsection
661:
.endm
diff --git a/arch/arm64/include/asm/assembler.h b/arch/arm64/include/asm/assembler.h
index 5846145be523c..d851e27334395 100644
--- a/arch/arm64/include/asm/assembler.h
+++ b/arch/arm64/include/asm/assembler.h
@@ -293,7 +293,7 @@ alternative_endif
alternative_if_not ARM64_KVM_PROTECTED_MODE
ASM_BUG()
alternative_else_nop_endif
-alternative_cb kvm_compute_final_ctr_el0
+alternative_cb ARM64_ALWAYS_SYSTEM, kvm_compute_final_ctr_el0
movz \reg, #0
movk \reg, #0, lsl #16
movk \reg, #0, lsl #32
@@ -877,7 +877,7 @@ alternative_endif
.macro __mitigate_spectre_bhb_loop tmp
#ifdef CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY
-alternative_cb spectre_bhb_patch_loop_iter
+alternative_cb ARM64_ALWAYS_SYSTEM, spectre_bhb_patch_loop_iter
mov \tmp, #32 // Patched to correct the immediate
alternative_cb_end
.Lspectre_bhb_loop\@:
@@ -890,7 +890,7 @@ alternative_cb_end
.macro mitigate_spectre_bhb_loop tmp
#ifdef CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY
-alternative_cb spectre_bhb_patch_loop_mitigation_enable
+alternative_cb ARM64_ALWAYS_SYSTEM, spectre_bhb_patch_loop_mitigation_enable
b .L_spectre_bhb_loop_done\@ // Patched to NOP
alternative_cb_end
__mitigate_spectre_bhb_loop \tmp
@@ -904,7 +904,7 @@ alternative_cb_end
stp x0, x1, [sp, #-16]!
stp x2, x3, [sp, #-16]!
mov w0, #ARM_SMCCC_ARCH_WORKAROUND_3
-alternative_cb smccc_patch_fw_mitigation_conduit
+alternative_cb ARM64_ALWAYS_SYSTEM, smccc_patch_fw_mitigation_conduit
nop // Patched to SMC/HVC #0
alternative_cb_end
ldp x2, x3, [sp], #16
@@ -914,7 +914,7 @@ alternative_cb_end
.macro mitigate_spectre_bhb_clear_insn
#ifdef CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY
-alternative_cb spectre_bhb_patch_clearbhb
+alternative_cb ARM64_ALWAYS_SYSTEM, spectre_bhb_patch_clearbhb
/* Patched to NOP when not supported */
clearbhb
isb
diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h
index 630c337c77746..f5852ad6e8845 100644
--- a/arch/arm64/include/asm/cpufeature.h
+++ b/arch/arm64/include/asm/cpufeature.h
@@ -422,9 +422,7 @@ extern DECLARE_BITMAP(cpu_hwcaps, ARM64_NCAPS);
extern struct static_key_false cpu_hwcap_keys[ARM64_NCAPS];
extern struct static_key_false arm64_const_caps_ready;
-/* ARM64 CAPS + alternative_cb */
-#define ARM64_NPATCHABLE (ARM64_NCAPS + 1)
-extern DECLARE_BITMAP(boot_capabilities, ARM64_NPATCHABLE);
+extern DECLARE_BITMAP(boot_capabilities, ARM64_NCAPS);
#define for_each_available_cap(cap) \
for_each_set_bit(cap, cpu_hwcaps, ARM64_NCAPS)
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index b208da3bebec8..7784081088e78 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -63,7 +63,7 @@
* specific registers encoded in the instructions).
*/
.macro kern_hyp_va reg
-alternative_cb kvm_update_va_mask
+alternative_cb ARM64_ALWAYS_SYSTEM, kvm_update_va_mask
and \reg, \reg, #1 /* mask with va_mask */
ror \reg, \reg, #1 /* rotate to the first tag bit */
add \reg, \reg, #0 /* insert the low 12 bits of the tag */
@@ -97,7 +97,7 @@ alternative_cb_end
hyp_pa \reg, \tmp
/* Load kimage_voffset. */
-alternative_cb kvm_get_kimage_voffset
+alternative_cb ARM64_ALWAYS_SYSTEM, kvm_get_kimage_voffset
movz \tmp, #0
movk \tmp, #0, lsl #16
movk \tmp, #0, lsl #32
@@ -131,6 +131,7 @@ static __always_inline unsigned long __kern_hyp_va(unsigned long v)
"add %0, %0, #0\n"
"add %0, %0, #0, lsl 12\n"
"ror %0, %0, #63\n",
+ ARM64_ALWAYS_SYSTEM,
kvm_update_va_mask)
: "+r" (v));
return v;
diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
index 2e18c9c0f612b..da706c9f9a9a5 100644
--- a/arch/arm64/kernel/alternative.c
+++ b/arch/arm64/kernel/alternative.c
@@ -21,6 +21,11 @@
#define ALT_ORIG_PTR(a) __ALT_PTR(a, orig_offset)
#define ALT_REPL_PTR(a) __ALT_PTR(a, alt_offset)
+#define ALT_CAP(a) ((a)->cpufeature & ~ARM64_CB_BIT)
+#define ALT_HAS_CB(a) ((a)->cpufeature & ARM64_CB_BIT)
+
+#define ALT_NR_INST(a) ((a)->orig_len / AARCH64_INSN_SIZE)
+
/* Volatile, as we may be patching the guts of READ_ONCE() */
static volatile int all_alternatives_applied;
@@ -143,16 +148,15 @@ static void __nocfi __apply_alternatives(const struct alt_region *region,
for (alt = region->begin; alt < region->end; alt++) {
int nr_inst;
+ int cap = ALT_CAP(alt);
- if (!test_bit(alt->cpufeature, feature_mask))
+ if (!test_bit(cap, feature_mask))
continue;
- /* Use ARM64_CB_PATCH as an unconditional patch */
- if (alt->cpufeature < ARM64_CB_PATCH &&
- !cpus_have_cap(alt->cpufeature))
+ if (!cpus_have_cap(cap))
continue;
- if (alt->cpufeature == ARM64_CB_PATCH)
+ if (ALT_HAS_CB(alt))
BUG_ON(alt->alt_len != 0);
else
BUG_ON(alt->alt_len != alt->orig_len);
@@ -161,10 +165,10 @@ static void __nocfi __apply_alternatives(const struct alt_region *region,
updptr = is_module ? origptr : lm_alias(origptr);
nr_inst = alt->orig_len / AARCH64_INSN_SIZE;
- if (alt->cpufeature < ARM64_CB_PATCH)
- alt_cb = patch_alternative;
- else
+ if (ALT_HAS_CB(alt))
alt_cb = ALT_REPL_PTR(alt);
+ else
+ alt_cb = patch_alternative;
alt_cb(alt, origptr, updptr, nr_inst);
@@ -208,10 +212,10 @@ static int __apply_alternatives_multi_stop(void *unused)
cpu_relax();
isb();
} else {
- DECLARE_BITMAP(remaining_capabilities, ARM64_NPATCHABLE);
+ DECLARE_BITMAP(remaining_capabilities, ARM64_NCAPS);
bitmap_complement(remaining_capabilities, boot_capabilities,
- ARM64_NPATCHABLE);
+ ARM64_NCAPS);
BUG_ON(all_alternatives_applied);
__apply_alternatives(&kernel_alternatives, false,
@@ -254,9 +258,9 @@ void apply_alternatives_module(void *start, size_t length)
.begin = start,
.end = start + length,
};
- DECLARE_BITMAP(all_capabilities, ARM64_NPATCHABLE);
+ DECLARE_BITMAP(all_capabilities, ARM64_NCAPS);
- bitmap_fill(all_capabilities, ARM64_NPATCHABLE);
+ bitmap_fill(all_capabilities, ARM64_NCAPS);
__apply_alternatives(®ion, true, &all_capabilities[0]);
}
diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
index af4de817d7123..68a0545285a1d 100644
--- a/arch/arm64/kernel/cpufeature.c
+++ b/arch/arm64/kernel/cpufeature.c
@@ -108,8 +108,7 @@ DECLARE_BITMAP(cpu_hwcaps, ARM64_NCAPS);
EXPORT_SYMBOL(cpu_hwcaps);
static struct arm64_cpu_capabilities const __ro_after_init *cpu_hwcaps_ptrs[ARM64_NCAPS];
-/* Need also bit for ARM64_CB_PATCH */
-DECLARE_BITMAP(boot_capabilities, ARM64_NPATCHABLE);
+DECLARE_BITMAP(boot_capabilities, ARM64_NCAPS);
bool arm64_use_ng_mappings = false;
EXPORT_SYMBOL(arm64_use_ng_mappings);
@@ -1391,6 +1390,12 @@ u64 __read_sysreg_by_encoding(u32 sys_id)
#include <linux/irqchip/arm-gic-v3.h>
+static bool
+has_always(const struct arm64_cpu_capabilities *entry, int scope)
+{
+ return true;
+}
+
static bool
feature_matches(u64 reg, const struct arm64_cpu_capabilities *entry)
{
@@ -2087,6 +2092,16 @@ cpucap_panic_on_conflict(const struct arm64_cpu_capabilities *cap)
}
static const struct arm64_cpu_capabilities arm64_features[] = {
+ {
+ .capability = ARM64_ALWAYS_BOOT,
+ .type = ARM64_CPUCAP_BOOT_CPU_FEATURE,
+ .matches = has_always,
+ },
+ {
+ .capability = ARM64_ALWAYS_SYSTEM,
+ .type = ARM64_CPUCAP_SYSTEM_FEATURE,
+ .matches = has_always,
+ },
{
.desc = "GIC system register CPU interface",
.capability = ARM64_HAS_SYSREG_GIC_CPUIF,
diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S
index 2d73b3e793b2b..e28137d64b768 100644
--- a/arch/arm64/kernel/entry.S
+++ b/arch/arm64/kernel/entry.S
@@ -114,7 +114,7 @@
* them if required.
*/
.macro apply_ssbd, state, tmp1, tmp2
-alternative_cb spectre_v4_patch_fw_mitigation_enable
+alternative_cb ARM64_ALWAYS_SYSTEM, spectre_v4_patch_fw_mitigation_enable
b .L__asm_ssbd_skip\@ // Patched to NOP
alternative_cb_end
ldr_this_cpu \tmp2, arm64_ssbd_callback_required, \tmp1
@@ -123,7 +123,7 @@ alternative_cb_end
tbnz \tmp2, #TIF_SSBD, .L__asm_ssbd_skip\@
mov w0, #ARM_SMCCC_ARCH_WORKAROUND_2
mov w1, #\state
-alternative_cb smccc_patch_fw_mitigation_conduit
+alternative_cb ARM64_ALWAYS_SYSTEM, smccc_patch_fw_mitigation_conduit
nop // Patched to SMC/HVC #0
alternative_cb_end
.L__asm_ssbd_skip\@:
@@ -175,7 +175,7 @@ alternative_else_nop_endif
.macro mte_set_kernel_gcr, tmp, tmp2
#ifdef CONFIG_KASAN_HW_TAGS
-alternative_cb kasan_hw_tags_enable
+alternative_cb ARM64_ALWAYS_SYSTEM, kasan_hw_tags_enable
b 1f
alternative_cb_end
mov \tmp, KERNEL_GCR_EL1
@@ -186,7 +186,7 @@ alternative_cb_end
.macro mte_set_user_gcr, tsk, tmp, tmp2
#ifdef CONFIG_KASAN_HW_TAGS
-alternative_cb kasan_hw_tags_enable
+alternative_cb ARM64_ALWAYS_SYSTEM, kasan_hw_tags_enable
b 1f
alternative_cb_end
ldr \tmp, [\tsk, #THREAD_MTE_CTRL]
diff --git a/arch/arm64/kvm/hyp/hyp-entry.S b/arch/arm64/kvm/hyp/hyp-entry.S
index 7839d075729b1..8f3f93fa119ed 100644
--- a/arch/arm64/kvm/hyp/hyp-entry.S
+++ b/arch/arm64/kvm/hyp/hyp-entry.S
@@ -196,7 +196,7 @@ SYM_CODE_END(__kvm_hyp_vector)
sub sp, sp, #(8 * 4)
stp x2, x3, [sp, #(8 * 0)]
stp x0, x1, [sp, #(8 * 2)]
- alternative_cb spectre_bhb_patch_wa3
+ alternative_cb ARM64_ALWAYS_SYSTEM, spectre_bhb_patch_wa3
/* Patched to mov WA3 when supported */
mov w0, #ARM_SMCCC_ARCH_WORKAROUND_1
alternative_cb_end
@@ -216,7 +216,7 @@ SYM_CODE_END(__kvm_hyp_vector)
mitigate_spectre_bhb_clear_insn
.endif
.if \indirect != 0
- alternative_cb kvm_patch_vector_branch
+ alternative_cb ARM64_ALWAYS_SYSTEM, kvm_patch_vector_branch
/*
* For ARM64_SPECTRE_V3A configurations, these NOPs get replaced with:
*
diff --git a/arch/arm64/tools/cpucaps b/arch/arm64/tools/cpucaps
index 63b2484ce6c3d..ff882ec175e73 100644
--- a/arch/arm64/tools/cpucaps
+++ b/arch/arm64/tools/cpucaps
@@ -2,6 +2,8 @@
#
# Internal CPU capabilities constants, keep this list sorted
+ALWAYS_BOOT
+ALWAYS_SYSTEM
BTI
# Unreliable: use system_supports_32bit_el0() instead.
HAS_32BIT_EL0_DO_NOT_USE
--
2.30.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH 6/9] arm64: alternatives: have callbacks take a cap
2022-09-01 15:14 ` [PATCH 6/9] arm64: alternatives: have callbacks take a cap Mark Rutland
@ 2022-09-02 15:54 ` Joey Gouly
2022-09-05 8:48 ` Mark Rutland
0 siblings, 1 reply; 20+ messages in thread
From: Joey Gouly @ 2022-09-02 15:54 UTC (permalink / raw)
To: Mark Rutland
Cc: linux-arm-kernel, ardb, catalin.marinas, james.morse, maz, will, nd
Hi Mark,
On Thu, Sep 01, 2022 at 04:14:00PM +0100, Mark Rutland wrote:
> Today, callback alternatives are special-cased within
> __apply_alternatives(), and are applied alongside patching for system
> capabilities as ARM64_NCAPS is not part of the boot_capabilities feature
> mask.
>
> This special-casing is less than ideal. Giving special meaning to
> ARM64_NCAPS for this requires some structures and loops to use
> ARM64_NCAPS + 1 (AKA ARM64_NPATCHABLE), while others use ARM64_NCAPS.
> It's also not immediately clear callback alternatives are only applied
> when applying alternatives for system-wide features.
>
> To make this a bit clearer, changes the way that callback alternatives
> are identified to remove the special-casing of ARM64_NCAPS, and to allow
> callback alternatives to be associated with a cpucap as with all other
> alternatives.
>
> New cpucaps, ARM64_ALWAYS_BOOT and ARM64_ALWAYS_SYSTEM are added which
> are always detected alongside boot cpu capabilities and system
> capabilities respectively. All existing callback alternatives are made
> to use ARM64_ALWAYS_SYSTEM, and so will be patched at the same point
> during the boot flow as before.
>
> Subsequent patches will make more use of these new cpucaps.
>
> There should be no functional change as a result of this patch.
>
> Signed-off-by: Mark Rutland <mark.rutland@arm.com>
> Cc: Ard Biesheuvel <ardb@kernel.org>
> Cc: Catalin Marinas <catalin.marinas@arm.com>
> Cc: James Morse <james.morse@arm.com>
> Cc: Joey Gouly <joey.gouly@arm.com>
> Cc: Marc Zyngier <maz@kernel.org>
> Cc: Will Deacon <will@kernel.org>
> ---
> arch/arm64/include/asm/alternative-macros.h | 18 ++++++++-----
> arch/arm64/include/asm/assembler.h | 10 ++++----
> arch/arm64/include/asm/cpufeature.h | 4 +--
> arch/arm64/include/asm/kvm_mmu.h | 5 ++--
> arch/arm64/kernel/alternative.c | 28 ++++++++++++---------
> arch/arm64/kernel/cpufeature.c | 19 ++++++++++++--
> arch/arm64/kernel/entry.S | 8 +++---
> arch/arm64/kvm/hyp/hyp-entry.S | 4 +--
> arch/arm64/tools/cpucaps | 2 ++
> 9 files changed, 62 insertions(+), 36 deletions(-)
>
[..]
> diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
> index 2e18c9c0f612b..da706c9f9a9a5 100644
> --- a/arch/arm64/kernel/alternative.c
> +++ b/arch/arm64/kernel/alternative.c
> @@ -21,6 +21,11 @@
> #define ALT_ORIG_PTR(a) __ALT_PTR(a, orig_offset)
> #define ALT_REPL_PTR(a) __ALT_PTR(a, alt_offset)
>
> +#define ALT_CAP(a) ((a)->cpufeature & ~ARM64_CB_BIT)
> +#define ALT_HAS_CB(a) ((a)->cpufeature & ARM64_CB_BIT)
> +
> +#define ALT_NR_INST(a) ((a)->orig_len / AARCH64_INSN_SIZE)
You introduced this macro, but don't use it.
> +
> /* Volatile, as we may be patching the guts of READ_ONCE() */
> static volatile int all_alternatives_applied;
>
> @@ -143,16 +148,15 @@ static void __nocfi __apply_alternatives(const struct alt_region *region,
>
> for (alt = region->begin; alt < region->end; alt++) {
> int nr_inst;
> + int cap = ALT_CAP(alt);
>
> - if (!test_bit(alt->cpufeature, feature_mask))
> + if (!test_bit(cap, feature_mask))
> continue;
>
> - /* Use ARM64_CB_PATCH as an unconditional patch */
> - if (alt->cpufeature < ARM64_CB_PATCH &&
> - !cpus_have_cap(alt->cpufeature))
> + if (!cpus_have_cap(cap))
> continue;
>
> - if (alt->cpufeature == ARM64_CB_PATCH)
> + if (ALT_HAS_CB(alt))
> BUG_ON(alt->alt_len != 0);
> else
> BUG_ON(alt->alt_len != alt->orig_len);
> @@ -161,10 +165,10 @@ static void __nocfi __apply_alternatives(const struct alt_region *region,
> updptr = is_module ? origptr : lm_alias(origptr);
> nr_inst = alt->orig_len / AARCH64_INSN_SIZE;
>
> - if (alt->cpufeature < ARM64_CB_PATCH)
> - alt_cb = patch_alternative;
> - else
> + if (ALT_HAS_CB(alt))
> alt_cb = ALT_REPL_PTR(alt);
> + else
> + alt_cb = patch_alternative;
>
> alt_cb(alt, origptr, updptr, nr_inst);
>
> @@ -208,10 +212,10 @@ static int __apply_alternatives_multi_stop(void *unused)
> cpu_relax();
> isb();
> } else {
> - DECLARE_BITMAP(remaining_capabilities, ARM64_NPATCHABLE);
> + DECLARE_BITMAP(remaining_capabilities, ARM64_NCAPS);
>
> bitmap_complement(remaining_capabilities, boot_capabilities,
> - ARM64_NPATCHABLE);
> + ARM64_NCAPS);
>
> BUG_ON(all_alternatives_applied);
> __apply_alternatives(&kernel_alternatives, false,
> @@ -254,9 +258,9 @@ void apply_alternatives_module(void *start, size_t length)
> .begin = start,
> .end = start + length,
> };
> - DECLARE_BITMAP(all_capabilities, ARM64_NPATCHABLE);
> + DECLARE_BITMAP(all_capabilities, ARM64_NCAPS);
>
> - bitmap_fill(all_capabilities, ARM64_NPATCHABLE);
> + bitmap_fill(all_capabilities, ARM64_NCAPS);
>
> __apply_alternatives(®ion, true, &all_capabilities[0]);
> }
Thanks,
Joey
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH 6/9] arm64: alternatives: have callbacks take a cap
2022-09-02 15:54 ` Joey Gouly
@ 2022-09-05 8:48 ` Mark Rutland
0 siblings, 0 replies; 20+ messages in thread
From: Mark Rutland @ 2022-09-05 8:48 UTC (permalink / raw)
To: Joey Gouly
Cc: linux-arm-kernel, ardb, catalin.marinas, james.morse, maz, will, nd
On Fri, Sep 02, 2022 at 04:54:45PM +0100, Joey Gouly wrote:
> Hi Mark,
>
> On Thu, Sep 01, 2022 at 04:14:00PM +0100, Mark Rutland wrote:
> > diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
> > index 2e18c9c0f612b..da706c9f9a9a5 100644
> > --- a/arch/arm64/kernel/alternative.c
> > +++ b/arch/arm64/kernel/alternative.c
> > @@ -21,6 +21,11 @@
> > #define ALT_ORIG_PTR(a) __ALT_PTR(a, orig_offset)
> > #define ALT_REPL_PTR(a) __ALT_PTR(a, alt_offset)
> >
> > +#define ALT_CAP(a) ((a)->cpufeature & ~ARM64_CB_BIT)
> > +#define ALT_HAS_CB(a) ((a)->cpufeature & ARM64_CB_BIT)
> > +
> > +#define ALT_NR_INST(a) ((a)->orig_len / AARCH64_INSN_SIZE)
> You introduced this macro, but don't use it.
Whoops; I'd originally added that for the dumping code, but ended up reworking
things to not use it anyway.
Removed now!
Thanks,
Mark.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH 7/9] arm64: alternatives: add alternative_has_feature_*()
2022-09-01 15:13 [PATCH 0/9] arm64: alternatives: improvements Mark Rutland
` (5 preceding siblings ...)
2022-09-01 15:14 ` [PATCH 6/9] arm64: alternatives: have callbacks take a cap Mark Rutland
@ 2022-09-01 15:14 ` Mark Rutland
2022-09-01 15:14 ` [PATCH 8/9] arm64: alternatives: add shared NOP callback Mark Rutland
2022-09-01 15:14 ` [PATCH 9/9] HACK: arm64: alternatives: dump summary of alternatives Mark Rutland
8 siblings, 0 replies; 20+ messages in thread
From: Mark Rutland @ 2022-09-01 15:14 UTC (permalink / raw)
To: linux-arm-kernel
Cc: ardb, catalin.marinas, james.morse, joey.gouly, mark.rutland, maz, will
Currrently we use a mixture of alternative sequences and static branches
to handle features detected at boot time. For ease of maintenance we
generally prefer to use static branches in C code, but this has a few
downsides:
* Each static branch has metadata in the __jump_table section, which is
not discarded after features are finalized. This wastes some space,
and slows down the patching of other static branches.
* The static branches are patched at a different point in time from the
alternatives, so changes are not atomic. This leaves a transient
period where there could be a mismatch between the behaviour of
alternatives and static branches, which could be problematic for some
features (e.g. pseudo-NMI).
* More (instrumentable) kernel code is executed to patch each static
branch, which can be risky when patching certain features (e.g.
irqflags management for pseudo-NMI).
* When CONFIG_JUMP_LABEL=n, static branches are turned into a load of a
flag and a conditional branch. This means it isn't safe to use such
static branches in an alternative address space (e.g. the NVHE/PKVM
hyp code), where the generated address isn't safe to acccess.
To deal with these issues, this patch introduces new
alternative_has_feature_*() helpers, which work like static branches but
are patched using alternatives. This ensures the patching is performed
at the same time as other alternative patching, allows the metadata to
be freed after patching, and is safe for use in alternative address
spaces.
Note that all supported toolchains have asm goto support, and since
commit:
a0a12c3ed057af57 ("asm goto: eradicate CC_HAS_ASM_GOTO)"
... the CC_HAS_ASM_GOTO Kconfig symbol has been removed, so no feature
check is necessary, and we can always make use of asm goto.
Additionally, note that:
* This has no impact on cpus_have_cap(), which is a dynamic check.
* This has no functional impact on cpus_have_const_cap(). The branches
are patched slightly later than before this patch, but these branches
are not reachable until caps have been finalised.
* It is now invalid to use cpus_have_final_cap() in the window between
feature detection and patching. All existing uses are only expected
after patching anyway, so this should not be a problem.
* The LSE atomics will now be enabled during alternatives patching
rather than immediately before. As the LL/SC an LSE atomics are
functionally equivalent this should not be problematic.
When building defconfig with GCC 12.1.0, the resulting Image is 64KiB
smaller:
| % ls -al Image-*
| -rw-r--r-- 1 mark mark 37108224 Aug 23 09:56 Image-after
| -rw-r--r-- 1 mark mark 37173760 Aug 23 09:54 Image-before
According to bloat-o-meter.pl:
| add/remove: 44/34 grow/shrink: 602/1294 up/down: 39692/-61108 (-21416)
| Function old new delta
| [...]
| Total: Before=16618336, After=16596920, chg -0.13%
| add/remove: 0/2 grow/shrink: 0/0 up/down: 0/-1296 (-1296)
| Data old new delta
| arm64_const_caps_ready 16 - -16
| cpu_hwcap_keys 1280 - -1280
| Total: Before=8987120, After=8985824, chg -0.01%
| add/remove: 0/0 grow/shrink: 0/0 up/down: 0/0 (0)
| RO Data old new delta
| Total: Before=18408, After=18408, chg +0.00%
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
---
arch/arm64/include/asm/alternative-macros.h | 41 +++++++++++++++++++++
arch/arm64/include/asm/cpufeature.h | 7 ++--
arch/arm64/include/asm/lse.h | 5 +--
arch/arm64/kernel/cpufeature.c | 25 -------------
arch/arm64/kernel/image-vars.h | 4 --
5 files changed, 46 insertions(+), 36 deletions(-)
diff --git a/arch/arm64/include/asm/alternative-macros.h b/arch/arm64/include/asm/alternative-macros.h
index 189c31be163ce..eaba9ec127897 100644
--- a/arch/arm64/include/asm/alternative-macros.h
+++ b/arch/arm64/include/asm/alternative-macros.h
@@ -213,4 +213,45 @@ alternative_endif
#define ALTERNATIVE(oldinstr, newinstr, ...) \
_ALTERNATIVE_CFG(oldinstr, newinstr, __VA_ARGS__, 1)
+#ifndef __ASSEMBLY__
+
+#include <linux/bug.h>
+#include <linux/types.h>
+
+static __always_inline bool
+alternative_has_feature_likely(unsigned long feature)
+{
+ BUILD_BUG_ON(feature >= ARM64_NCAPS);
+
+ asm_volatile_goto(
+ ALTERNATIVE("b %l[l_no]", "nop", %[feature])
+ :
+ : [feature] "i" (feature)
+ :
+ : l_no);
+
+ return true;
+l_no:
+ return false;
+}
+
+static __always_inline bool
+alternative_has_feature_unlikely(unsigned long feature)
+{
+ BUILD_BUG_ON(feature >= ARM64_NCAPS);
+
+ asm_volatile_goto(
+ ALTERNATIVE("nop", "b %l[l_yes]", %[feature])
+ :
+ : [feature] "i" (feature)
+ :
+ : l_yes);
+
+ return false;
+l_yes:
+ return true;
+}
+
+#endif /* __ASSEMBLY__ */
+
#endif /* __ASM_ALTERNATIVE_MACROS_H */
diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h
index f5852ad6e8845..ba8a47433b5a2 100644
--- a/arch/arm64/include/asm/cpufeature.h
+++ b/arch/arm64/include/asm/cpufeature.h
@@ -6,6 +6,7 @@
#ifndef __ASM_CPUFEATURE_H
#define __ASM_CPUFEATURE_H
+#include <asm/alternative-macros.h>
#include <asm/cpucaps.h>
#include <asm/cputype.h>
#include <asm/hwcap.h>
@@ -419,8 +420,6 @@ static __always_inline bool is_hyp_code(void)
}
extern DECLARE_BITMAP(cpu_hwcaps, ARM64_NCAPS);
-extern struct static_key_false cpu_hwcap_keys[ARM64_NCAPS];
-extern struct static_key_false arm64_const_caps_ready;
extern DECLARE_BITMAP(boot_capabilities, ARM64_NCAPS);
@@ -438,7 +437,7 @@ unsigned long cpu_get_elf_hwcap2(void);
static __always_inline bool system_capabilities_finalized(void)
{
- return static_branch_likely(&arm64_const_caps_ready);
+ return alternative_has_feature_likely(ARM64_ALWAYS_SYSTEM);
}
/*
@@ -465,7 +464,7 @@ static __always_inline bool __cpus_have_const_cap(int num)
{
if (num >= ARM64_NCAPS)
return false;
- return static_branch_unlikely(&cpu_hwcap_keys[num]);
+ return alternative_has_feature_unlikely(num);
}
/*
diff --git a/arch/arm64/include/asm/lse.h b/arch/arm64/include/asm/lse.h
index 29c85810ae690..c503db8e73b01 100644
--- a/arch/arm64/include/asm/lse.h
+++ b/arch/arm64/include/asm/lse.h
@@ -13,14 +13,13 @@
#include <linux/jump_label.h>
#include <linux/stringify.h>
#include <asm/alternative.h>
+#include <asm/alternative-macros.h>
#include <asm/atomic_lse.h>
#include <asm/cpucaps.h>
-extern struct static_key_false cpu_hwcap_keys[ARM64_NCAPS];
-
static __always_inline bool system_uses_lse_atomics(void)
{
- return static_branch_likely(&cpu_hwcap_keys[ARM64_HAS_LSE_ATOMICS]);
+ return alternative_has_feature_likely(ARM64_HAS_LSE_ATOMICS);
}
#define __lse_ll_sc_body(op, ...) \
diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
index 68a0545285a1d..d9c3879728bcc 100644
--- a/arch/arm64/kernel/cpufeature.c
+++ b/arch/arm64/kernel/cpufeature.c
@@ -133,31 +133,12 @@ DEFINE_STATIC_KEY_FALSE(arm64_mismatched_32bit_el0);
*/
static cpumask_var_t cpu_32bit_el0_mask __cpumask_var_read_mostly;
-/*
- * Flag to indicate if we have computed the system wide
- * capabilities based on the boot time active CPUs. This
- * will be used to determine if a new booting CPU should
- * go through the verification process to make sure that it
- * supports the system capabilities, without using a hotplug
- * notifier. This is also used to decide if we could use
- * the fast path for checking constant CPU caps.
- */
-DEFINE_STATIC_KEY_FALSE(arm64_const_caps_ready);
-EXPORT_SYMBOL(arm64_const_caps_ready);
-static inline void finalize_system_capabilities(void)
-{
- static_branch_enable(&arm64_const_caps_ready);
-}
-
void dump_cpu_features(void)
{
/* file-wide pr_fmt adds "CPU features: " prefix */
pr_emerg("0x%*pb\n", ARM64_NCAPS, &cpu_hwcaps);
}
-DEFINE_STATIC_KEY_ARRAY_FALSE(cpu_hwcap_keys, ARM64_NCAPS);
-EXPORT_SYMBOL(cpu_hwcap_keys);
-
#define __ARM64_FTR_BITS(SIGNED, VISIBLE, STRICT, TYPE, SHIFT, WIDTH, SAFE_VAL) \
{ \
.sign = SIGNED, \
@@ -2944,9 +2925,6 @@ static void __init enable_cpu_capabilities(u16 scope_mask)
if (!cpus_have_cap(num))
continue;
- /* Ensure cpus_have_const_cap(num) works */
- static_branch_enable(&cpu_hwcap_keys[num]);
-
if (boot_scope && caps->cpu_enable)
/*
* Capabilities with SCOPE_BOOT_CPU scope are finalised
@@ -3268,9 +3246,6 @@ void __init setup_cpu_features(void)
sme_setup();
minsigstksz_setup();
- /* Advertise that we have computed the system capabilities */
- finalize_system_capabilities();
-
/*
* Check for sane CTR_EL0.CWG value.
*/
diff --git a/arch/arm64/kernel/image-vars.h b/arch/arm64/kernel/image-vars.h
index afa69e04e75ed..118973a6ab053 100644
--- a/arch/arm64/kernel/image-vars.h
+++ b/arch/arm64/kernel/image-vars.h
@@ -89,10 +89,6 @@ KVM_NVHE_ALIAS(__icache_flags);
/* VMID bits set by the KVM VMID allocator */
KVM_NVHE_ALIAS(kvm_arm_vmid_bits);
-/* Kernel symbols needed for cpus_have_final/const_caps checks. */
-KVM_NVHE_ALIAS(arm64_const_caps_ready);
-KVM_NVHE_ALIAS(cpu_hwcap_keys);
-
/* Static keys which are set if a vGIC trap should be handled in hyp. */
KVM_NVHE_ALIAS(vgic_v2_cpuif_trap);
KVM_NVHE_ALIAS(vgic_v3_cpuif_trap);
--
2.30.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 20+ messages in thread* [PATCH 8/9] arm64: alternatives: add shared NOP callback
2022-09-01 15:13 [PATCH 0/9] arm64: alternatives: improvements Mark Rutland
` (6 preceding siblings ...)
2022-09-01 15:14 ` [PATCH 7/9] arm64: alternatives: add alternative_has_feature_*() Mark Rutland
@ 2022-09-01 15:14 ` Mark Rutland
2022-09-01 15:14 ` [PATCH 9/9] HACK: arm64: alternatives: dump summary of alternatives Mark Rutland
8 siblings, 0 replies; 20+ messages in thread
From: Mark Rutland @ 2022-09-01 15:14 UTC (permalink / raw)
To: linux-arm-kernel
Cc: ardb, catalin.marinas, james.morse, joey.gouly, mark.rutland, maz, will
For each instance of an alternative, the compiler outputs a distinct
copy of the alternative instructions into a subsection. As the compiler
doesn't have special knowledge of alternatives, it cannot coalesce these
to save space.
In a defconfig kernel built with GCC 12.1.0, there are approximately
10,000 instances of alternative_has_feature_likely(), where the
replacement instruction is always a NOP. As NOPs are
position-independent, we don't need a unique copy per alternative
sequence.
This patch adds a callback to patch an alternative sequence with NOPs,
and make use of this in alternative_has_feature_likely(). So that this
can be used for other sites in future, this is written to patch multiple
instructions up to the original sequence length.
For NVHE, an alias is added to image-vars.h.
For modules, the callback is exported. Note that as modules are loaded
within 2GiB of the kernel, an alt_instr entry in a module can always
refer directly to the callback, and no special handling is necessary.
When building with GCC 12.1.0, the vmlinux is ~158KiB smaller, though
the resulting Image size is unchanged due to alignment constraints and
padding:
| % ls -al vmlinux-*
| -rwxr-xr-x 1 mark mark 134644592 Sep 1 14:52 vmlinux-after
| -rwxr-xr-x 1 mark mark 134486232 Sep 1 14:50 vmlinux-before
| % ls -al Image-*
| -rw-r--r-- 1 mark mark 37108224 Sep 1 14:52 Image-after
| -rw-r--r-- 1 mark mark 37108224 Sep 1 14:50 Image-before
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
---
arch/arm64/include/asm/alternative-macros.h | 2 +-
arch/arm64/kernel/alternative.c | 8 ++++++++
arch/arm64/kernel/image-vars.h | 1 +
3 files changed, 10 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/include/asm/alternative-macros.h b/arch/arm64/include/asm/alternative-macros.h
index eaba9ec127897..4a2a98d6d2227 100644
--- a/arch/arm64/include/asm/alternative-macros.h
+++ b/arch/arm64/include/asm/alternative-macros.h
@@ -224,7 +224,7 @@ alternative_has_feature_likely(unsigned long feature)
BUILD_BUG_ON(feature >= ARM64_NCAPS);
asm_volatile_goto(
- ALTERNATIVE("b %l[l_no]", "nop", %[feature])
+ ALTERNATIVE_CB("b %l[l_no]", %[feature], alt_cb_patch_nops)
:
: [feature] "i" (feature)
:
diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
index da706c9f9a9a5..122c59ce2772b 100644
--- a/arch/arm64/kernel/alternative.c
+++ b/arch/arm64/kernel/alternative.c
@@ -265,3 +265,11 @@ void apply_alternatives_module(void *start, size_t length)
__apply_alternatives(®ion, true, &all_capabilities[0]);
}
#endif
+
+noinstr void alt_cb_patch_nops(struct alt_instr *alt, __le32 *origptr,
+ __le32 *updptr, int nr_inst)
+{
+ for (int i = 0; i < nr_inst; i++)
+ updptr[i] = cpu_to_le32(aarch64_insn_gen_nop());
+}
+EXPORT_SYMBOL(alt_cb_patch_nops);
diff --git a/arch/arm64/kernel/image-vars.h b/arch/arm64/kernel/image-vars.h
index 118973a6ab053..4aaa5f3d1f65f 100644
--- a/arch/arm64/kernel/image-vars.h
+++ b/arch/arm64/kernel/image-vars.h
@@ -73,6 +73,7 @@ KVM_NVHE_ALIAS(spectre_bhb_patch_loop_iter);
KVM_NVHE_ALIAS(spectre_bhb_patch_loop_mitigation_enable);
KVM_NVHE_ALIAS(spectre_bhb_patch_wa3);
KVM_NVHE_ALIAS(spectre_bhb_patch_clearbhb);
+KVM_NVHE_ALIAS(alt_cb_patch_nops);
/* Global kernel state accessed by nVHE hyp code. */
KVM_NVHE_ALIAS(kvm_vgic_global_state);
--
2.30.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 20+ messages in thread* [PATCH 9/9] HACK: arm64: alternatives: dump summary of alternatives
2022-09-01 15:13 [PATCH 0/9] arm64: alternatives: improvements Mark Rutland
` (7 preceding siblings ...)
2022-09-01 15:14 ` [PATCH 8/9] arm64: alternatives: add shared NOP callback Mark Rutland
@ 2022-09-01 15:14 ` Mark Rutland
2022-09-12 12:36 ` Mark Brown
8 siblings, 1 reply; 20+ messages in thread
From: Mark Rutland @ 2022-09-01 15:14 UTC (permalink / raw)
To: linux-arm-kernel
Cc: ardb, catalin.marinas, james.morse, joey.gouly, mark.rutland, maz, will
NOTE: THIS PATCH IS NOT INTENDED FOR UPSTREAM.
To figure out whether it's worth making further changes to alternatives
(e.g. whether it's worth replacing regular entries with callbacks), it
would be useful to know the makeup of alternatives in a given kernel
Image or module.
This patch makes the alternatives code dump a summary of kernel
alternatives at boot time, and module alternatives at module load time.
For example, a defconfig v6.0-rc3+ kernel build with GCC 12.1.0 looks
reports:
| alternatives: Alternatives summary:
| entries: 28000 (336000 bytes)
| regular: 17280
| callback: 10720
| instructions: 32052 (128208 bytes)
| replacements: 20962 ( 83848 bytes)
| alternatives: cpucap 1 => entries: 925 orig: 1295, repl: 0, cb: 925
| alternatives: cpucap 2 => entries: 10 orig: 10, repl: 10, cb: 0
| alternatives: cpucap 4 => entries: 2 orig: 2, repl: 2, cb: 0
| alternatives: cpucap 5 => entries: 49 orig: 142, repl: 142, cb: 0
| alternatives: cpucap 10 => entries: 36 orig: 36, repl: 36, cb: 0
| alternatives: cpucap 11 => entries: 9 orig: 12, repl: 12, cb: 0
| alternatives: cpucap 12 => entries: 3 orig: 6, repl: 6, cb: 0
| alternatives: cpucap 13 => entries: 17 orig: 17, repl: 17, cb: 0
| alternatives: cpucap 14 => entries: 3 orig: 3, repl: 3, cb: 0
| alternatives: cpucap 16 => entries: 1 orig: 1, repl: 1, cb: 0
| alternatives: cpucap 18 => entries: 7 orig: 13, repl: 13, cb: 0
| alternatives: cpucap 19 => entries: 2 orig: 2, repl: 2, cb: 0
| alternatives: cpucap 20 => entries: 17 orig: 17, repl: 17, cb: 0
| alternatives: cpucap 24 => entries: 1128 orig: 1128, repl: 1128, cb: 0
| alternatives: cpucap 26 => entries: 10780 orig: 13953, repl: 4158, cb: 9795
| alternatives: cpucap 27 => entries: 39 orig: 39, repl: 39, cb: 0
| alternatives: cpucap 28 => entries: 4 orig: 8, repl: 8, cb: 0
| alternatives: cpucap 29 => entries: 15 orig: 15, repl: 15, cb: 0
| alternatives: cpucap 30 => entries: 15 orig: 27, repl: 27, cb: 0
| alternatives: cpucap 31 => entries: 3 orig: 3, repl: 3, cb: 0
| alternatives: cpucap 32 => entries: 59 orig: 118, repl: 118, cb: 0
| alternatives: cpucap 33 => entries: 6 orig: 6, repl: 6, cb: 0
| alternatives: cpucap 36 => entries: 20 orig: 20, repl: 20, cb: 0
| alternatives: cpucap 37 => entries: 2727 orig: 2727, repl: 2727, cb: 0
| alternatives: cpucap 38 => entries: 3 orig: 3, repl: 3, cb: 0
| alternatives: cpucap 40 => entries: 25 orig: 29, repl: 29, cb: 0
| alternatives: cpucap 41 => entries: 11 orig: 21, repl: 21, cb: 0
| alternatives: cpucap 42 => entries: 142 orig: 152, repl: 152, cb: 0
| alternatives: cpucap 44 => entries: 63 orig: 63, repl: 63, cb: 0
| alternatives: cpucap 45 => entries: 4 orig: 4, repl: 4, cb: 0
| alternatives: cpucap 46 => entries: 5 orig: 5, repl: 5, cb: 0
| alternatives: cpucap 47 => entries: 2 orig: 2, repl: 2, cb: 0
| alternatives: cpucap 50 => entries: 3 orig: 3, repl: 3, cb: 0
| alternatives: cpucap 51 => entries: 105 orig: 105, repl: 105, cb: 0
| alternatives: cpucap 52 => entries: 57 orig: 59, repl: 59, cb: 0
| alternatives: cpucap 53 => entries: 3 orig: 3, repl: 3, cb: 0
| alternatives: cpucap 54 => entries: 5 orig: 5, repl: 5, cb: 0
| alternatives: cpucap 55 => entries: 1 orig: 1, repl: 1, cb: 0
| alternatives: cpucap 59 => entries: 28 orig: 28, repl: 28, cb: 0
| alternatives: cpucap 60 => entries: 2 orig: 2, repl: 2, cb: 0
| alternatives: cpucap 61 => entries: 1 orig: 1, repl: 1, cb: 0
| alternatives: cpucap 65 => entries: 2 orig: 2, repl: 2, cb: 0
| alternatives: cpucap 68 => entries: 1 orig: 1, repl: 1, cb: 0
| alternatives: cpucap 70 => entries: 1 orig: 1, repl: 1, cb: 0
| alternatives: cpucap 71 => entries: 1 orig: 3, repl: 3, cb: 0
| alternatives: cpucap 72 => entries: 1 orig: 1, repl: 1, cb: 0
| alternatives: cpucap 73 => entries: 32 orig: 32, repl: 32, cb: 0
| alternatives: cpucap 74 => entries: 4 orig: 4, repl: 4, cb: 0
| alternatives: cpucap 75 => entries: 5 orig: 5, repl: 5, cb: 0
| alternatives: cpucap 76 => entries: 11391 orig: 11391, repl: 11391, cb: 0
| alternatives: cpucap 77 => entries: 1 orig: 1, repl: 1, cb: 0
| alternatives: cpucap 78 => entries: 64 orig: 224, repl: 224, cb: 0
| alternatives: cpucap 79 => entries: 141 orig: 282, repl: 282, cb: 0
| alternatives: cpucap 80 => entries: 19 orig: 19, repl: 19, cb: 0
From this, it's worth noting:
* cpucap 1 is ARM64_ALWAYS_SYSTEM.
* cpucap 24 is ARM64_HAS_IRQ_PRIO_MASKING. Due to the existing structure
of the alternatives, alternative entries are created for the
irqflags.h code even when CONFIG_ARM64_PSEUDO_NMI=n, creating ~14KiB
of alt_instr entries, and ~4KiB of replacement instructions.
This could be avoided by reworking the irqflags.h code to use the new
alternative_has_feature_*() helpers.
* cpucap 26 is ARM64_HAS_LSE_ATOMICS, and most entries are using the
shared NOP patcher. The other entries are for inline cmpxchg
sequences.
* cpucap 37 is ARM64_HAS_VIRT_HOST_EXTN.
* cpucap 76 is ARM64_WORKAROUND_DEVICE_LOAD_ACQUIRE, which could be
rewritten to use a callback to patch LDR to LDAR (or vice-versa), were
the insn framework extended, to save ~44KiB of replacement
instructions.
NOTE: THIS PATCH IS NOT INTENDED FOR UPSTREAM.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
---
arch/arm64/kernel/alternative.c | 67 ++++++++++++++++++++++++++++++++-
1 file changed, 65 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
index 122c59ce2772b..2f55d03adbe80 100644
--- a/arch/arm64/kernel/alternative.c
+++ b/arch/arm64/kernel/alternative.c
@@ -138,15 +138,74 @@ static void clean_dcache_range_nopatch(u64 start, u64 end)
} while (cur += d_size, cur < end);
}
+#define for_each_region_alt(region, alt) \
+ for (struct alt_instr *alt = (region)->begin; \
+ (alt) < (region)->end; \
+ (alt)++)
+
+void summarize_alternatives(const struct alt_region *region)
+{
+ unsigned int entries[ARM64_NCAPS] = { 0 };
+ unsigned int orig_len[ARM64_NCAPS] = { 0 };
+ unsigned int repl_len[ARM64_NCAPS] = { 0 };
+ unsigned int callbacks[ARM64_NCAPS] = { 0 };
+
+ unsigned int total_entries = 0;
+ unsigned int total_orig = 0;
+ unsigned int total_repl = 0;
+ unsigned int total_callbacks = 0;
+
+ for_each_region_alt(region, alt) {
+ int cap = ALT_CAP(alt);
+
+ entries[cap]++;
+ total_entries++;
+
+ orig_len[cap] += alt->orig_len;
+ total_orig += alt->orig_len;
+
+ repl_len[cap] += alt->alt_len;
+ total_repl += alt->alt_len;
+
+ if (ALT_HAS_CB(alt)) {
+ callbacks[cap]++;
+ total_callbacks++;
+ }
+ }
+
+ pr_info("Alternatives summary:\n"
+ " entries: %6u (%6zu bytes)\n"
+ " regular: %6d\n"
+ " callback: %6d\n"
+ " instructions: %6u (%6u bytes)\n"
+ " replacements: %6u (%6u bytes)\n",
+ total_entries, total_entries * sizeof (struct alt_instr),
+ total_entries - total_callbacks,
+ total_callbacks,
+ total_orig / AARCH64_INSN_SIZE, total_orig,
+ total_repl / AARCH64_INSN_SIZE, total_repl);
+
+ for (int i = 0; i < ARM64_NCAPS; i++) {
+ if (!entries[i])
+ continue;
+
+ pr_info("cpucap %2d => entries: %5d orig: %5d, repl: %5d, cb: %5d\n",
+ i,
+ entries[i],
+ orig_len[i] / AARCH64_INSN_SIZE,
+ repl_len[i] / AARCH64_INSN_SIZE,
+ callbacks[i]);
+ }
+}
+
static void __nocfi __apply_alternatives(const struct alt_region *region,
bool is_module,
unsigned long *feature_mask)
{
- struct alt_instr *alt;
__le32 *origptr, *updptr;
alternative_cb_t alt_cb;
- for (alt = region->begin; alt < region->end; alt++) {
+ for_each_region_alt(region, alt) {
int nr_inst;
int cap = ALT_CAP(alt);
@@ -245,6 +304,8 @@ void __init apply_boot_alternatives(void)
/* If called on non-boot cpu things could go wrong */
WARN_ON(smp_processor_id() != 0);
+ summarize_alternatives(&kernel_alternatives);
+
pr_info("applying boot alternatives\n");
__apply_alternatives(&kernel_alternatives, false,
@@ -262,6 +323,8 @@ void apply_alternatives_module(void *start, size_t length)
bitmap_fill(all_capabilities, ARM64_NCAPS);
+ summarize_alternatives(®ion);
+
__apply_alternatives(®ion, true, &all_capabilities[0]);
}
#endif
--
2.30.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH 9/9] HACK: arm64: alternatives: dump summary of alternatives
2022-09-01 15:14 ` [PATCH 9/9] HACK: arm64: alternatives: dump summary of alternatives Mark Rutland
@ 2022-09-12 12:36 ` Mark Brown
2022-09-12 16:14 ` Mark Rutland
0 siblings, 1 reply; 20+ messages in thread
From: Mark Brown @ 2022-09-12 12:36 UTC (permalink / raw)
To: Mark Rutland
Cc: linux-arm-kernel, ardb, catalin.marinas, james.morse, joey.gouly,
maz, will
[-- Attachment #1.1: Type: text/plain, Size: 329 bytes --]
On Thu, Sep 01, 2022 at 04:14:03PM +0100, Mark Rutland wrote:
> NOTE: THIS PATCH IS NOT INTENDED FOR UPSTREAM.
>
It seems like a useful thing to have hanging around behind a
Kconfig option - we should probably be keeping an eye on this,
and there's going to be work people do where they'd have to
reinvent the wheel.
[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
[-- Attachment #2: Type: text/plain, Size: 176 bytes --]
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 9/9] HACK: arm64: alternatives: dump summary of alternatives
2022-09-12 12:36 ` Mark Brown
@ 2022-09-12 16:14 ` Mark Rutland
0 siblings, 0 replies; 20+ messages in thread
From: Mark Rutland @ 2022-09-12 16:14 UTC (permalink / raw)
To: Mark Brown
Cc: linux-arm-kernel, ardb, catalin.marinas, james.morse, joey.gouly,
maz, will
On Mon, Sep 12, 2022 at 01:36:18PM +0100, Mark Brown wrote:
> On Thu, Sep 01, 2022 at 04:14:03PM +0100, Mark Rutland wrote:
>
> > NOTE: THIS PATCH IS NOT INTENDED FOR UPSTREAM.
> >
>
> It seems like a useful thing to have hanging around behind a
> Kconfig option - we should probably be keeping an eye on this,
> and there's going to be work people do where they'd have to
> reinvent the wheel.
I'd wanted to rework this to have a cmdline option (and also add some other
dbeug options here), so I'll pull this out of v2 and send it as a follow-up
along with those.
Thanks,
Mark.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 20+ messages in thread