Re: [PATCH v14 7/8] KVM: arm64: use CASLT instruction for swapping guest descriptor

[Marc Zyngier <maz@kernel.org>]

On Wed, 25 Feb 2026 18:27:07 +0000,
Yeoreum Yun <yeoreum.yun@arm.com> wrote:
> 
> Use the CASLT instruction to swap the guest descriptor when FEAT_LSUI
> is enabled, avoiding the need to clear the PAN bit.
> 
> Signed-off-by: Yeoreum Yun <yeoreum.yun@arm.com>
> ---
>  arch/arm64/include/asm/cpucaps.h |  2 ++
>  arch/arm64/include/asm/futex.h   | 17 +----------------
>  arch/arm64/include/asm/lsui.h    | 27 +++++++++++++++++++++++++++
>  arch/arm64/kvm/at.c              | 30 +++++++++++++++++++++++++++++-
>  4 files changed, 59 insertions(+), 17 deletions(-)
>  create mode 100644 arch/arm64/include/asm/lsui.h
> 
> diff --git a/arch/arm64/include/asm/cpucaps.h b/arch/arm64/include/asm/cpucaps.h
> index 177c691914f8..6e3da333442e 100644
> --- a/arch/arm64/include/asm/cpucaps.h
> +++ b/arch/arm64/include/asm/cpucaps.h
> @@ -71,6 +71,8 @@ cpucap_is_possible(const unsigned int cap)
>  		return true;
>  	case ARM64_HAS_PMUV3:
>  		return IS_ENABLED(CONFIG_HW_PERF_EVENTS);
> +	case ARM64_HAS_LSUI:
> +		return IS_ENABLED(CONFIG_ARM64_LSUI);
>  	}
>  
>  	return true;

It would make more sense to move this hunk to the first patch, where
you deal with features and capabilities, instead of having this in a
random KVM-specific patch.

> diff --git a/arch/arm64/include/asm/futex.h b/arch/arm64/include/asm/futex.h
> index b579e9d0964d..6779c4ad927f 100644
> --- a/arch/arm64/include/asm/futex.h
> +++ b/arch/arm64/include/asm/futex.h
> @@ -7,11 +7,9 @@
>  
>  #include <linux/futex.h>
>  #include <linux/uaccess.h>
> -#include <linux/stringify.h>
>  
> -#include <asm/alternative.h>
> -#include <asm/alternative-macros.h>
>  #include <asm/errno.h>
> +#include <asm/lsui.h>
>  
>  #define FUTEX_MAX_LOOPS	128 /* What's the largest number you can think of? */
>  
> @@ -91,8 +89,6 @@ __llsc_futex_cmpxchg(u32 __user *uaddr, u32 oldval, u32 newval, u32 *oval)
>  
>  #ifdef CONFIG_ARM64_LSUI
>  
> -#define __LSUI_PREAMBLE	".arch_extension lsui\n"
> -
>  #define LSUI_FUTEX_ATOMIC_OP(op, asm_op)				\
>  static __always_inline int						\
>  __lsui_futex_atomic_##op(int oparg, u32 __user *uaddr, int *oval)	\
> @@ -235,17 +231,6 @@ __lsui_futex_cmpxchg(u32 __user *uaddr, u32 oldval, u32 newval, u32 *oval)
>  {
>  	return __lsui_cmpxchg32(uaddr, oldval, newval, oval);
>  }
> -
> -#define __lsui_llsc_body(op, ...)					\
> -({									\
> -	alternative_has_cap_unlikely(ARM64_HAS_LSUI) ?			\
> -		__lsui_##op(__VA_ARGS__) : __llsc_##op(__VA_ARGS__);	\
> -})
> -
> -#else	/* CONFIG_ARM64_LSUI */
> -
> -#define __lsui_llsc_body(op, ...)	__llsc_##op(__VA_ARGS__)
> -
>  #endif	/* CONFIG_ARM64_LSUI */
>  
>  
> diff --git a/arch/arm64/include/asm/lsui.h b/arch/arm64/include/asm/lsui.h
> new file mode 100644
> index 000000000000..8f0d81953eb6
> --- /dev/null
> +++ b/arch/arm64/include/asm/lsui.h
> @@ -0,0 +1,27 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +#ifndef __ASM_LSUI_H
> +#define __ASM_LSUI_H
> +
> +#include <linux/compiler_types.h>
> +#include <linux/stringify.h>
> +#include <asm/alternative.h>
> +#include <asm/alternative-macros.h>
> +#include <asm/cpucaps.h>
> +
> +#define __LSUI_PREAMBLE	".arch_extension lsui\n"
> +
> +#ifdef CONFIG_ARM64_LSUI
> +
> +#define __lsui_llsc_body(op, ...)					\
> +({									\
> +	alternative_has_cap_unlikely(ARM64_HAS_LSUI) ?			\
> +		__lsui_##op(__VA_ARGS__) : __llsc_##op(__VA_ARGS__);	\
> +})
> +
> +#else	/* CONFIG_ARM64_LSUI */
> +
> +#define __lsui_llsc_body(op, ...)	__llsc_##op(__VA_ARGS__)
> +
> +#endif	/* CONFIG_ARM64_LSUI */
> +
> +#endif	/* __ASM_LSUI_H */

Similarly, fold this into the patch that introduces FEAT_LSUI support
for futexes (#5) so that the code is in its final position from the
beginning. This will avoid churn that makes the patches pointlessly
hard to follow, since this change is unrelated to KVM.

> diff --git a/arch/arm64/kvm/at.c b/arch/arm64/kvm/at.c
> index 885bd5bb2f41..fd3c5749e853 100644
> --- a/arch/arm64/kvm/at.c
> +++ b/arch/arm64/kvm/at.c
> @@ -9,6 +9,7 @@
>  #include <asm/esr.h>
>  #include <asm/kvm_hyp.h>
>  #include <asm/kvm_mmu.h>
> +#include <asm/lsui.h>
>  
>  static void fail_s1_walk(struct s1_walk_result *wr, u8 fst, bool s1ptw)
>  {
> @@ -1704,6 +1705,31 @@ int __kvm_find_s1_desc_level(struct kvm_vcpu *vcpu, u64 va, u64 ipa, int *level)
>  	}
>  }
>  
> +static int __lsui_swap_desc(u64 __user *ptep, u64 old, u64 new)
> +{
> +	u64 tmp = old;
> +	int ret = 0;
> +
> +	uaccess_ttbr0_enable();

Why do we need this? If FEAT_LSUI is present, than FEAT_PAN is also
present. And since PAN support not a compilation option anymore, we
should be able to rely on PAN being enabled.

Or am I missing something? If so, please document why we require it.

> +
> +	asm volatile(__LSUI_PREAMBLE
> +		     "1: caslt	%[old], %[new], %[addr]\n"
> +		     "2:\n"
> +		     _ASM_EXTABLE_UACCESS_ERR(1b, 2b, %w[ret])
> +		     : [old] "+r" (old), [addr] "+Q" (*ptep), [ret] "+r" (ret)
> +		     : [new] "r" (new)
> +		     : "memory");
> +
> +	uaccess_ttbr0_disable();
> +
> +	if (ret)
> +		return ret;
> +	if (tmp != old)
> +		return -EAGAIN;
> +
> +	return ret;
> +}
> +
>  static int __lse_swap_desc(u64 __user *ptep, u64 old, u64 new)
>  {
>  	u64 tmp = old;
> @@ -1779,7 +1805,9 @@ int __kvm_at_swap_desc(struct kvm *kvm, gpa_t ipa, u64 old, u64 new)
>  		return -EPERM;
>  
>  	ptep = (u64 __user *)hva + offset;
> -	if (cpus_have_final_cap(ARM64_HAS_LSE_ATOMICS))
> +	if (cpucap_is_possible(ARM64_HAS_LSUI) && cpus_have_final_cap(ARM64_HAS_LSUI))
> +		r = __lsui_swap_desc(ptep, old, new);
> +	else if (cpus_have_final_cap(ARM64_HAS_LSE_ATOMICS))
>  		r = __lse_swap_desc(ptep, old, new);
>  	else
>  		r = __llsc_swap_desc(ptep, old, new);

Thanks,

	M.

-- 
Without deviation from the norm, progress is not possible.