This helps avoid avoid a potentially large stack allocation.
When building with:
$ make CC=clang arch/x86/ CFLAGS=-Wframe-larger-than=1000
The following warning is observed:
arch/x86/kernel/kvm.c:494:13: warning: stack frame size of 1064 bytes in
function 'kvm_send_ipi_mask_allbutself' [-Wframe-larger-than=]
static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int
vector)
^
Debugging with:
https://github.com/ClangBuiltLinux/frame-larger-than
via:
$ python3 frame_larger_than.py arch/x86/kernel/kvm.o \
kvm_send_ipi_mask_allbutself
points to the stack allocated `struct cpumask newmask` in
`kvm_send_ipi_mask_allbutself`. The size of a `struct cpumask` is
potentially large, as it's CONFIG_NR_CPUS divided by BITS_PER_LONG for
the target architecture. CONFIG_NR_CPUS for X86_64 can be as high as
8192, making a single instance of a `struct cpumask` 1024 B.
Signed-off-by: Nick Desaulniers <[email protected]>
---
arch/x86/kernel/kvm.c | 10 ++++++----
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 32ef1ee733b7..d41c0a0d62a2 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -494,13 +494,15 @@ static void kvm_send_ipi_mask(const struct cpumask *mask, int vector)
static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int vector)
{
unsigned int this_cpu = smp_processor_id();
- struct cpumask new_mask;
+ struct cpumask *new_mask;
const struct cpumask *local_mask;
- cpumask_copy(&new_mask, mask);
- cpumask_clear_cpu(this_cpu, &new_mask);
- local_mask = &new_mask;
+ new_mask = kmalloc(sizeof(*new_mask), GFP_KERNEL);
+ cpumask_copy(new_mask, mask);
+ cpumask_clear_cpu(this_cpu, new_mask);
+ local_mask = new_mask;
__send_ipi_mask(local_mask, vector);
+ kfree(new_mask);
}
/*
--
2.17.1
On Mon, Jan 27, 2020 at 12:09 AM Peter Zijlstra <[email protected]> wrote:
>
> On Sun, Jan 26, 2020 at 11:16:02PM -0800, Nick Desaulniers wrote:
> > This helps avoid avoid a potentially large stack allocation.
> >
> > When building with:
> > $ make CC=clang arch/x86/ CFLAGS=-Wframe-larger-than=1000
> > The following warning is observed:
> > arch/x86/kernel/kvm.c:494:13: warning: stack frame size of 1064 bytes in
> > function 'kvm_send_ipi_mask_allbutself' [-Wframe-larger-than=]
> > static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int
> > vector)
> > ^
> > Debugging with:
> > https://github.com/ClangBuiltLinux/frame-larger-than
> > via:
> > $ python3 frame_larger_than.py arch/x86/kernel/kvm.o \
> > kvm_send_ipi_mask_allbutself
> > points to the stack allocated `struct cpumask newmask` in
> > `kvm_send_ipi_mask_allbutself`. The size of a `struct cpumask` is
> > potentially large, as it's CONFIG_NR_CPUS divided by BITS_PER_LONG for
> > the target architecture. CONFIG_NR_CPUS for X86_64 can be as high as
> > 8192, making a single instance of a `struct cpumask` 1024 B.
> >
> > Signed-off-by: Nick Desaulniers <[email protected]>
> > ---
> > arch/x86/kernel/kvm.c | 10 ++++++----
> > 1 file changed, 6 insertions(+), 4 deletions(-)
> >
> > diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> > index 32ef1ee733b7..d41c0a0d62a2 100644
> > --- a/arch/x86/kernel/kvm.c
> > +++ b/arch/x86/kernel/kvm.c
> > @@ -494,13 +494,15 @@ static void kvm_send_ipi_mask(const struct cpumask *mask, int vector)
> > static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int vector)
> > {
> > unsigned int this_cpu = smp_processor_id();
> > - struct cpumask new_mask;
>
> Right, on stack cpumask is definitely dodgy.
>
> > + struct cpumask *new_mask;
> > const struct cpumask *local_mask;
> >
> > - cpumask_copy(&new_mask, mask);
> > - cpumask_clear_cpu(this_cpu, &new_mask);
> > - local_mask = &new_mask;
> > + new_mask = kmalloc(sizeof(*new_mask), GFP_KERNEL);
Probably should check for allocation failure, d'oh!
> > + cpumask_copy(new_mask, mask);
> > + cpumask_clear_cpu(this_cpu, new_mask);
> > + local_mask = new_mask;
> > __send_ipi_mask(local_mask, vector);
> > + kfree(new_mask);
> > }
>
> One alternative approach is adding the inverse of cpu_bit_bitmap. I'm
> not entirely sure how often we need the all-but-self mask, but ISTR
> there were other places too.
On Sun, Jan 26, 2020 at 11:16:02PM -0800, Nick Desaulniers wrote:
> This helps avoid avoid a potentially large stack allocation.
>
> When building with:
> $ make CC=clang arch/x86/ CFLAGS=-Wframe-larger-than=1000
> The following warning is observed:
> arch/x86/kernel/kvm.c:494:13: warning: stack frame size of 1064 bytes in
> function 'kvm_send_ipi_mask_allbutself' [-Wframe-larger-than=]
> static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int
> vector)
> ^
> Debugging with:
> https://github.com/ClangBuiltLinux/frame-larger-than
> via:
> $ python3 frame_larger_than.py arch/x86/kernel/kvm.o \
> kvm_send_ipi_mask_allbutself
> points to the stack allocated `struct cpumask newmask` in
> `kvm_send_ipi_mask_allbutself`. The size of a `struct cpumask` is
> potentially large, as it's CONFIG_NR_CPUS divided by BITS_PER_LONG for
> the target architecture. CONFIG_NR_CPUS for X86_64 can be as high as
> 8192, making a single instance of a `struct cpumask` 1024 B.
>
> Signed-off-by: Nick Desaulniers <[email protected]>
> ---
> arch/x86/kernel/kvm.c | 10 ++++++----
> 1 file changed, 6 insertions(+), 4 deletions(-)
>
> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> index 32ef1ee733b7..d41c0a0d62a2 100644
> --- a/arch/x86/kernel/kvm.c
> +++ b/arch/x86/kernel/kvm.c
> @@ -494,13 +494,15 @@ static void kvm_send_ipi_mask(const struct cpumask *mask, int vector)
> static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int vector)
> {
> unsigned int this_cpu = smp_processor_id();
> - struct cpumask new_mask;
Right, on stack cpumask is definitely dodgy.
> + struct cpumask *new_mask;
> const struct cpumask *local_mask;
>
> - cpumask_copy(&new_mask, mask);
> - cpumask_clear_cpu(this_cpu, &new_mask);
> - local_mask = &new_mask;
> + new_mask = kmalloc(sizeof(*new_mask), GFP_KERNEL);
> + cpumask_copy(new_mask, mask);
> + cpumask_clear_cpu(this_cpu, new_mask);
> + local_mask = new_mask;
> __send_ipi_mask(local_mask, vector);
> + kfree(new_mask);
> }
One alternative approach is adding the inverse of cpu_bit_bitmap. I'm
not entirely sure how often we need the all-but-self mask, but ISTR
there were other places too.
Nick Desaulniers <[email protected]> writes:
> This helps avoid avoid a potentially large stack allocation.
>
> When building with:
> $ make CC=clang arch/x86/ CFLAGS=-Wframe-larger-than=1000
> The following warning is observed:
> arch/x86/kernel/kvm.c:494:13: warning: stack frame size of 1064 bytes in
> function 'kvm_send_ipi_mask_allbutself' [-Wframe-larger-than=]
> static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int
> vector)
> ^
> Debugging with:
> https://github.com/ClangBuiltLinux/frame-larger-than
> via:
> $ python3 frame_larger_than.py arch/x86/kernel/kvm.o \
> kvm_send_ipi_mask_allbutself
> points to the stack allocated `struct cpumask newmask` in
> `kvm_send_ipi_mask_allbutself`. The size of a `struct cpumask` is
> potentially large, as it's CONFIG_NR_CPUS divided by BITS_PER_LONG for
> the target architecture. CONFIG_NR_CPUS for X86_64 can be as high as
> 8192, making a single instance of a `struct cpumask` 1024 B.
>
> Signed-off-by: Nick Desaulniers <[email protected]>
> ---
> arch/x86/kernel/kvm.c | 10 ++++++----
> 1 file changed, 6 insertions(+), 4 deletions(-)
>
> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> index 32ef1ee733b7..d41c0a0d62a2 100644
> --- a/arch/x86/kernel/kvm.c
> +++ b/arch/x86/kernel/kvm.c
> @@ -494,13 +494,15 @@ static void kvm_send_ipi_mask(const struct cpumask *mask, int vector)
> static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int vector)
> {
> unsigned int this_cpu = smp_processor_id();
> - struct cpumask new_mask;
> + struct cpumask *new_mask;
> const struct cpumask *local_mask;
>
> - cpumask_copy(&new_mask, mask);
> - cpumask_clear_cpu(this_cpu, &new_mask);
> - local_mask = &new_mask;
> + new_mask = kmalloc(sizeof(*new_mask), GFP_KERNEL);
You could've used alloc_cpumask_var() instead, however, I think that
memory allocation on this path is undesireable. We can always
pre-allocate 1 cpumask variable per cpu and use it every time, e.g. we
do this for Hyper-V.
> + cpumask_copy(new_mask, mask);
> + cpumask_clear_cpu(this_cpu, new_mask);
> + local_mask = new_mask;
> __send_ipi_mask(local_mask, vector);
> + kfree(new_mask);
> }
>
> /*
--
Vitaly
Hi Nick,
On Mon, 27 Jan 2020 at 15:16, Nick Desaulniers
<[email protected]> wrote:
>
> This helps avoid avoid a potentially large stack allocation.
>
> When building with:
> $ make CC=clang arch/x86/ CFLAGS=-Wframe-larger-than=1000
> The following warning is observed:
> arch/x86/kernel/kvm.c:494:13: warning: stack frame size of 1064 bytes in
> function 'kvm_send_ipi_mask_allbutself' [-Wframe-larger-than=]
> static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int
> vector)
> ^
> Debugging with:
> https://github.com/ClangBuiltLinux/frame-larger-than
> via:
> $ python3 frame_larger_than.py arch/x86/kernel/kvm.o \
> kvm_send_ipi_mask_allbutself
> points to the stack allocated `struct cpumask newmask` in
> `kvm_send_ipi_mask_allbutself`. The size of a `struct cpumask` is
> potentially large, as it's CONFIG_NR_CPUS divided by BITS_PER_LONG for
> the target architecture. CONFIG_NR_CPUS for X86_64 can be as high as
> 8192, making a single instance of a `struct cpumask` 1024 B.
Could you help test the below untested patch?
From 867753e2fa27906f15df7902ba1bce7f9cef6ebe Mon Sep 17 00:00:00 2001
From: Wanpeng Li <[email protected]>
Date: Mon, 3 Feb 2020 16:26:35 +0800
Subject: [PATCH] KVM: Pre-allocate 1 cpumask variable per cpu for both
pv tlb and pv ipis
Reported-by: Nick Desaulniers <[email protected]>
Signed-off-by: Wanpeng Li <[email protected]>
---
arch/x86/kernel/kvm.c | 33 +++++++++++++++++++++------------
1 file changed, 21 insertions(+), 12 deletions(-)
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 81045aab..b1e8efa 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -425,6 +425,8 @@ static void __init sev_map_percpu_data(void)
}
}
+static DEFINE_PER_CPU(cpumask_var_t, __pv_cpu_mask);
+
#ifdef CONFIG_SMP
#define KVM_IPI_CLUSTER_SIZE (2 * BITS_PER_LONG)
@@ -490,12 +492,12 @@ static void kvm_send_ipi_mask(const struct
cpumask *mask, int vector)
static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask,
int vector)
{
unsigned int this_cpu = smp_processor_id();
- struct cpumask new_mask;
+ struct cpumask *new_mask = this_cpu_cpumask_var_ptr(__pv_cpu_mask);
const struct cpumask *local_mask;
- cpumask_copy(&new_mask, mask);
- cpumask_clear_cpu(this_cpu, &new_mask);
- local_mask = &new_mask;
+ cpumask_copy(new_mask, mask);
+ cpumask_clear_cpu(this_cpu, new_mask);
+ local_mask = new_mask;
__send_ipi_mask(local_mask, vector);
}
@@ -575,7 +577,6 @@ static void __init kvm_apf_trap_init(void)
update_intr_gate(X86_TRAP_PF, async_page_fault);
}
-static DEFINE_PER_CPU(cpumask_var_t, __pv_tlb_mask);
static void kvm_flush_tlb_others(const struct cpumask *cpumask,
const struct flush_tlb_info *info)
@@ -583,7 +584,7 @@ static void kvm_flush_tlb_others(const struct
cpumask *cpumask,
u8 state;
int cpu;
struct kvm_steal_time *src;
- struct cpumask *flushmask = this_cpu_cpumask_var_ptr(__pv_tlb_mask);
+ struct cpumask *flushmask = this_cpu_cpumask_var_ptr(__pv_cpu_mask);
cpumask_copy(flushmask, cpumask);
/*
@@ -624,6 +625,7 @@ static void __init kvm_guest_init(void)
kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) {
pv_ops.mmu.flush_tlb_others = kvm_flush_tlb_others;
pv_ops.mmu.tlb_remove_table = tlb_remove_table;
+ pr_info("KVM setup pv remote TLB flush\n");
}
if (kvm_para_has_feature(KVM_FEATURE_PV_EOI))
@@ -732,23 +734,30 @@ static __init int activate_jump_labels(void)
}
arch_initcall(activate_jump_labels);
-static __init int kvm_setup_pv_tlb_flush(void)
+static __init int kvm_alloc_cpumask(void)
{
int cpu;
+ bool alloc = false;
if (kvm_para_has_feature(KVM_FEATURE_PV_TLB_FLUSH) &&
!kvm_para_has_hint(KVM_HINTS_REALTIME) &&
- kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) {
+ kvm_para_has_feature(KVM_FEATURE_STEAL_TIME))
+ alloc = true;
+
+#if defined(CONFIG_SMP)
+ if (!alloc && kvm_para_has_feature(KVM_FEATURE_PV_SEND_IPI))
+ alloc = true;
+#endif
+
+ if (alloc)
for_each_possible_cpu(cpu) {
- zalloc_cpumask_var_node(per_cpu_ptr(&__pv_tlb_mask, cpu),
+ zalloc_cpumask_var_node(per_cpu_ptr(&__pv_cpu_mask, cpu),
GFP_KERNEL, cpu_to_node(cpu));
}
- pr_info("KVM setup pv remote TLB flush\n");
- }
return 0;
}
-arch_initcall(kvm_setup_pv_tlb_flush);
+arch_initcall(kvm_alloc_cpumask);
#ifdef CONFIG_PARAVIRT_SPINLOCKS
--
1.8.3.1
On Mon, Feb 3, 2020 at 8:32 AM Wanpeng Li <[email protected]> wrote:
>
> Hi Nick,
> On Mon, 27 Jan 2020 at 15:16, Nick Desaulniers
> <[email protected]> wrote:
> >
> > This helps avoid avoid a potentially large stack allocation.
> >
> > When building with:
> > $ make CC=clang arch/x86/ CFLAGS=-Wframe-larger-than=1000
> > The following warning is observed:
> > arch/x86/kernel/kvm.c:494:13: warning: stack frame size of 1064 bytes in
> > function 'kvm_send_ipi_mask_allbutself' [-Wframe-larger-than=]
> > static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int
> > vector)
> > ^
> > Debugging with:
> > https://github.com/ClangBuiltLinux/frame-larger-than
> > via:
> > $ python3 frame_larger_than.py arch/x86/kernel/kvm.o \
> > kvm_send_ipi_mask_allbutself
> > points to the stack allocated `struct cpumask newmask` in
> > `kvm_send_ipi_mask_allbutself`. The size of a `struct cpumask` is
> > potentially large, as it's CONFIG_NR_CPUS divided by BITS_PER_LONG for
> > the target architecture. CONFIG_NR_CPUS for X86_64 can be as high as
> > 8192, making a single instance of a `struct cpumask` 1024 B.
>
> Could you help test the below untested patch?
>
> From 867753e2fa27906f15df7902ba1bce7f9cef6ebe Mon Sep 17 00:00:00 2001
> From: Wanpeng Li <[email protected]>
> Date: Mon, 3 Feb 2020 16:26:35 +0800
> Subject: [PATCH] KVM: Pre-allocate 1 cpumask variable per cpu for both
> pv tlb and pv ipis
>
> Reported-by: Nick Desaulniers <[email protected]>
> Signed-off-by: Wanpeng Li <[email protected]>
> ---
> arch/x86/kernel/kvm.c | 33 +++++++++++++++++++++------------
> 1 file changed, 21 insertions(+), 12 deletions(-)
>
> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> index 81045aab..b1e8efa 100644
> --- a/arch/x86/kernel/kvm.c
> +++ b/arch/x86/kernel/kvm.c
> @@ -425,6 +425,8 @@ static void __init sev_map_percpu_data(void)
> }
> }
>
> +static DEFINE_PER_CPU(cpumask_var_t, __pv_cpu_mask);
> +
> #ifdef CONFIG_SMP
> #define KVM_IPI_CLUSTER_SIZE (2 * BITS_PER_LONG)
>
> @@ -490,12 +492,12 @@ static void kvm_send_ipi_mask(const struct
> cpumask *mask, int vector)
> static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask,
> int vector)
> {
> unsigned int this_cpu = smp_processor_id();
> - struct cpumask new_mask;
> + struct cpumask *new_mask = this_cpu_cpumask_var_ptr(__pv_cpu_mask);
Yes, this should help reduce the stack usage, thanks.
Acked-by: Nick Desaulniers <[email protected]>
> const struct cpumask *local_mask;
>
> - cpumask_copy(&new_mask, mask);
> - cpumask_clear_cpu(this_cpu, &new_mask);
> - local_mask = &new_mask;
> + cpumask_copy(new_mask, mask);
> + cpumask_clear_cpu(this_cpu, new_mask);
> + local_mask = new_mask;
> __send_ipi_mask(local_mask, vector);
> }
>
> @@ -575,7 +577,6 @@ static void __init kvm_apf_trap_init(void)
> update_intr_gate(X86_TRAP_PF, async_page_fault);
> }
>
> -static DEFINE_PER_CPU(cpumask_var_t, __pv_tlb_mask);
>
> static void kvm_flush_tlb_others(const struct cpumask *cpumask,
> const struct flush_tlb_info *info)
> @@ -583,7 +584,7 @@ static void kvm_flush_tlb_others(const struct
> cpumask *cpumask,
> u8 state;
> int cpu;
> struct kvm_steal_time *src;
> - struct cpumask *flushmask = this_cpu_cpumask_var_ptr(__pv_tlb_mask);
> + struct cpumask *flushmask = this_cpu_cpumask_var_ptr(__pv_cpu_mask);
>
> cpumask_copy(flushmask, cpumask);
> /*
> @@ -624,6 +625,7 @@ static void __init kvm_guest_init(void)
> kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) {
> pv_ops.mmu.flush_tlb_others = kvm_flush_tlb_others;
> pv_ops.mmu.tlb_remove_table = tlb_remove_table;
> + pr_info("KVM setup pv remote TLB flush\n");
> }
>
> if (kvm_para_has_feature(KVM_FEATURE_PV_EOI))
> @@ -732,23 +734,30 @@ static __init int activate_jump_labels(void)
> }
> arch_initcall(activate_jump_labels);
>
> -static __init int kvm_setup_pv_tlb_flush(void)
> +static __init int kvm_alloc_cpumask(void)
> {
> int cpu;
> + bool alloc = false;
>
> if (kvm_para_has_feature(KVM_FEATURE_PV_TLB_FLUSH) &&
> !kvm_para_has_hint(KVM_HINTS_REALTIME) &&
> - kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) {
> + kvm_para_has_feature(KVM_FEATURE_STEAL_TIME))
> + alloc = true;
> +
> +#if defined(CONFIG_SMP)
> + if (!alloc && kvm_para_has_feature(KVM_FEATURE_PV_SEND_IPI))
> + alloc = true;
> +#endif
> +
> + if (alloc)
> for_each_possible_cpu(cpu) {
> - zalloc_cpumask_var_node(per_cpu_ptr(&__pv_tlb_mask, cpu),
> + zalloc_cpumask_var_node(per_cpu_ptr(&__pv_cpu_mask, cpu),
> GFP_KERNEL, cpu_to_node(cpu));
> }
> - pr_info("KVM setup pv remote TLB flush\n");
> - }
>
> return 0;
> }
> -arch_initcall(kvm_setup_pv_tlb_flush);
> +arch_initcall(kvm_alloc_cpumask);
>
> #ifdef CONFIG_PARAVIRT_SPINLOCKS
>
> --
> 1.8.3.1
>
> --
> You received this message because you are subscribed to the Google Groups "Clang Built Linux" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
> To view this discussion on the web visit https://groups.google.com/d/msgid/clang-built-linux/CANRm%2BCwK0Cg45mktda9Yz9fsjPCvtuB8O%2Bfma5L3tV725ki1qw%40mail.gmail.com.
--
Thanks,
~Nick Desaulniers