[PATCH 0/2] KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs

[PATCH 0/2] KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs

[Mark Brown]

The set_id_regs selftest lacks coverag for ID_AA64ISR3_EL1 which has
several features exposed to KVM guests in it.  Add coverage, and while
we're here adjust the test to improve maintainability a bit.  

The test will fail without the recently applied change adding FEAT_LSFE:

   https://lore.kernel.org/r/175829303126.1764550.939188785634158487.b4-ty@kernel.org

Signed-off-by: Mark Brown <broonie@kernel.org>
---
Mark Brown (2):
      KVM: arm64: selftests: Remove a duplicate register listing in set_id_regs
      KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs

 tools/testing/selftests/kvm/arm64/set_id_regs.c | 22 ++++++++++++++--------
 1 file changed, 14 insertions(+), 8 deletions(-)
---
base-commit: 5db15c998c390efbe5c82f6cda77cb896a3a6a3e
change-id: 20250919-kvm-arm64-id-aa64isar3-el1-f0bd8ab3d36b

Best regards,
--  
Mark Brown <broonie@kernel.org>

[PATCH 1/2] KVM: arm64: selftests: Remove a duplicate register listing in set_id_regs

[Mark Brown]

Currently we list the main set of registers with bits we test three
times, once in the test_regs array which is used at runtime, once in the
guest code and once in a list of ARRAY_SIZE() operations we use to tell
kselftest how many tests we plan to execute. This is needlessly fiddly,
when adding new registers as the test_cnt calculation is formatted with
two registers per line. Instead count the number of bitfields in the
register arrays at runtime.

The existing code subtracts ARRAY_SIZE(test_regs) from the number of
tests to account for the terminating FTR_REG_END entries in the per
register arrays, the new code accounts for this when enumerating.

Signed-off-by: Mark Brown <broonie@kernel.org>
---
 tools/testing/selftests/kvm/arm64/set_id_regs.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/kvm/arm64/set_id_regs.c b/tools/testing/selftests/kvm/arm64/set_id_regs.c
index d3bf9204409c..bfb70926272d 100644
--- a/tools/testing/selftests/kvm/arm64/set_id_regs.c
+++ b/tools/testing/selftests/kvm/arm64/set_id_regs.c
@@ -763,7 +763,7 @@ int main(void)
 	struct kvm_vm *vm;
 	bool aarch64_only;
 	uint64_t val, el0;
-	int test_cnt;
+	int test_cnt, i, j;
 
 	TEST_REQUIRE(kvm_has_cap(KVM_CAP_ARM_SUPPORTED_REG_MASK_RANGES));
 	TEST_REQUIRE(kvm_has_cap(KVM_CAP_ARM_WRITABLE_IMP_ID_REGS));
@@ -779,13 +779,10 @@ int main(void)
 
 	ksft_print_header();
 
-	test_cnt = ARRAY_SIZE(ftr_id_aa64dfr0_el1) + ARRAY_SIZE(ftr_id_dfr0_el1) +
-		   ARRAY_SIZE(ftr_id_aa64isar0_el1) + ARRAY_SIZE(ftr_id_aa64isar1_el1) +
-		   ARRAY_SIZE(ftr_id_aa64isar2_el1) + ARRAY_SIZE(ftr_id_aa64pfr0_el1) +
-		   ARRAY_SIZE(ftr_id_aa64pfr1_el1) + ARRAY_SIZE(ftr_id_aa64mmfr0_el1) +
-		   ARRAY_SIZE(ftr_id_aa64mmfr1_el1) + ARRAY_SIZE(ftr_id_aa64mmfr2_el1) +
-		   ARRAY_SIZE(ftr_id_aa64mmfr3_el1) + ARRAY_SIZE(ftr_id_aa64zfr0_el1) -
-		   ARRAY_SIZE(test_regs) + 3 + MPAM_IDREG_TEST + MTE_IDREG_TEST;
+	test_cnt = 3 + MPAM_IDREG_TEST + MTE_IDREG_TEST;
+	for (i = 0; i < ARRAY_SIZE(test_regs); i++)
+		for (j = 0; test_regs[i].ftr_bits[j].type != FTR_END; j++)
+			test_cnt++;
 
 	ksft_set_plan(test_cnt);
 

-- 
2.47.2

[PATCH 2/2] KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs

[Mark Brown]

We have a couple of writable bitfields in ID_AA64ISAR3_EL1 but the
set_id_regs selftest does not cover this register at all, add coverage.

Signed-off-by: Mark Brown <broonie@kernel.org>
---
 tools/testing/selftests/kvm/arm64/set_id_regs.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/tools/testing/selftests/kvm/arm64/set_id_regs.c b/tools/testing/selftests/kvm/arm64/set_id_regs.c
index bfb70926272d..c7c38b1a1f10 100644
--- a/tools/testing/selftests/kvm/arm64/set_id_regs.c
+++ b/tools/testing/selftests/kvm/arm64/set_id_regs.c
@@ -125,6 +125,13 @@ static const struct reg_ftr_bits ftr_id_aa64isar2_el1[] = {
 	REG_FTR_END,
 };
 
+static const struct reg_ftr_bits ftr_id_aa64isar3_el1[] = {
+	REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64ISAR3_EL1, FPRCVT, 0),
+	REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64ISAR3_EL1, LSFE, 0),
+	REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64ISAR3_EL1, FAMINMAX, 0),
+	REG_FTR_END,
+};
+
 static const struct reg_ftr_bits ftr_id_aa64pfr0_el1[] = {
 	REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, CSV3, 0),
 	REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, CSV2, 0),
@@ -221,6 +228,7 @@ static struct test_feature_reg test_regs[] = {
 	TEST_REG(SYS_ID_AA64ISAR0_EL1, ftr_id_aa64isar0_el1),
 	TEST_REG(SYS_ID_AA64ISAR1_EL1, ftr_id_aa64isar1_el1),
 	TEST_REG(SYS_ID_AA64ISAR2_EL1, ftr_id_aa64isar2_el1),
+	TEST_REG(SYS_ID_AA64ISAR3_EL1, ftr_id_aa64isar3_el1),
 	TEST_REG(SYS_ID_AA64PFR0_EL1, ftr_id_aa64pfr0_el1),
 	TEST_REG(SYS_ID_AA64PFR1_EL1, ftr_id_aa64pfr1_el1),
 	TEST_REG(SYS_ID_AA64MMFR0_EL1, ftr_id_aa64mmfr0_el1),
@@ -239,6 +247,7 @@ static void guest_code(void)
 	GUEST_REG_SYNC(SYS_ID_AA64ISAR0_EL1);
 	GUEST_REG_SYNC(SYS_ID_AA64ISAR1_EL1);
 	GUEST_REG_SYNC(SYS_ID_AA64ISAR2_EL1);
+	GUEST_REG_SYNC(SYS_ID_AA64ISAR3_EL1);
 	GUEST_REG_SYNC(SYS_ID_AA64PFR0_EL1);
 	GUEST_REG_SYNC(SYS_ID_AA64MMFR0_EL1);
 	GUEST_REG_SYNC(SYS_ID_AA64MMFR1_EL1);

-- 
2.47.2

Re: [PATCH 2/2] KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs

[Zenghui Yu]

On 2025/9/21 03:52, Mark Brown wrote:
> We have a couple of writable bitfields in ID_AA64ISAR3_EL1 but the
> set_id_regs selftest does not cover this register at all, add coverage.
> 
> Signed-off-by: Mark Brown <broonie@kernel.org>
> ---
>  tools/testing/selftests/kvm/arm64/set_id_regs.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> diff --git a/tools/testing/selftests/kvm/arm64/set_id_regs.c b/tools/testing/selftests/kvm/arm64/set_id_regs.c
> index bfb70926272d..c7c38b1a1f10 100644
> --- a/tools/testing/selftests/kvm/arm64/set_id_regs.c
> +++ b/tools/testing/selftests/kvm/arm64/set_id_regs.c
> @@ -125,6 +125,13 @@ static const struct reg_ftr_bits ftr_id_aa64isar2_el1[] = {
>  	REG_FTR_END,
>  };
>  
> +static const struct reg_ftr_bits ftr_id_aa64isar3_el1[] = {
> +	REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64ISAR3_EL1, FPRCVT, 0),
> +	REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64ISAR3_EL1, LSFE, 0),
> +	REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64ISAR3_EL1, FAMINMAX, 0),
> +	REG_FTR_END,
> +};
> +
>  static const struct reg_ftr_bits ftr_id_aa64pfr0_el1[] = {
>  	REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, CSV3, 0),
>  	REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, CSV2, 0),
> @@ -221,6 +228,7 @@ static struct test_feature_reg test_regs[] = {
>  	TEST_REG(SYS_ID_AA64ISAR0_EL1, ftr_id_aa64isar0_el1),
>  	TEST_REG(SYS_ID_AA64ISAR1_EL1, ftr_id_aa64isar1_el1),
>  	TEST_REG(SYS_ID_AA64ISAR2_EL1, ftr_id_aa64isar2_el1),
> +	TEST_REG(SYS_ID_AA64ISAR3_EL1, ftr_id_aa64isar3_el1),
>  	TEST_REG(SYS_ID_AA64PFR0_EL1, ftr_id_aa64pfr0_el1),
>  	TEST_REG(SYS_ID_AA64PFR1_EL1, ftr_id_aa64pfr1_el1),
>  	TEST_REG(SYS_ID_AA64MMFR0_EL1, ftr_id_aa64mmfr0_el1),
> @@ -239,6 +247,7 @@ static void guest_code(void)
>  	GUEST_REG_SYNC(SYS_ID_AA64ISAR0_EL1);
>  	GUEST_REG_SYNC(SYS_ID_AA64ISAR1_EL1);
>  	GUEST_REG_SYNC(SYS_ID_AA64ISAR2_EL1);
> +	GUEST_REG_SYNC(SYS_ID_AA64ISAR3_EL1);
>  	GUEST_REG_SYNC(SYS_ID_AA64PFR0_EL1);
>  	GUEST_REG_SYNC(SYS_ID_AA64MMFR0_EL1);
>  	GUEST_REG_SYNC(SYS_ID_AA64MMFR1_EL1);

Not related to this patch but seems that we forgot to sync several
registers (ID_AA64PFR1, MPIDR, CLIDR) in guest to make sure the guest
had seen the written value.

Zenghui

Re: [PATCH 0/2] KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs

[Oliver Upton]

On Sat, Sep 20, 2025 at 08:51:58PM +0100, Mark Brown wrote:
> The set_id_regs selftest lacks coverag for ID_AA64ISR3_EL1 which has
> several features exposed to KVM guests in it.  Add coverage, and while
> we're here adjust the test to improve maintainability a bit.  
> 
> The test will fail without the recently applied change adding FEAT_LSFE:
> 
>    https://lore.kernel.org/r/175829303126.1764550.939188785634158487.b4-ty@kernel.org
> 
> Signed-off-by: Mark Brown <broonie@kernel.org>
> ---
> Mark Brown (2):
>       KVM: arm64: selftests: Remove a duplicate register listing in set_id_regs
>       KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs

Reviewed-by: Oliver Upton <oliver.upton@linux.dev>

Thanks,
Oliver

Re: [PATCH 0/2] KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs

[Marc Zyngier]

On Sat, 20 Sep 2025 20:51:58 +0100,
Mark Brown <broonie@kernel.org> wrote:
> 
> The set_id_regs selftest lacks coverag for ID_AA64ISR3_EL1 which has
> several features exposed to KVM guests in it.  Add coverage, and while
> we're here adjust the test to improve maintainability a bit.  
> 
> The test will fail without the recently applied change adding FEAT_LSFE:
> 
>    https://lore.kernel.org/r/175829303126.1764550.939188785634158487.b4-ty@kernel.org
> 
> Signed-off-by: Mark Brown <broonie@kernel.org>
> ---
> Mark Brown (2):
>       KVM: arm64: selftests: Remove a duplicate register listing in set_id_regs
>       KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs
> 
>  tools/testing/selftests/kvm/arm64/set_id_regs.c | 22 ++++++++++++++--------
>  1 file changed, 14 insertions(+), 8 deletions(-)
> ---
> base-commit: 5db15c998c390efbe5c82f6cda77cb896a3a6a3e

$ git show 5db15c998c390efbe5c82f6cda77cb896a3a6a3e
fatal: bad object 5db15c998c390efbe5c82f6cda77cb896a3a6a3e

What's the point of indicating a base commit if that's not something I
can find? You should never base any work on top of something that
isn't a stable tag.

	M.

-- 
Jazz isn't dead. It just smells funny.

Re: [PATCH 0/2] KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 772 bytes --]

On Wed, Sep 24, 2025 at 07:29:49PM +0100, Marc Zyngier wrote:
> Mark Brown <broonie@kernel.org> wrote:

> > ---
> > base-commit: 5db15c998c390efbe5c82f6cda77cb896a3a6a3e

> $ git show 5db15c998c390efbe5c82f6cda77cb896a3a6a3e
> fatal: bad object 5db15c998c390efbe5c82f6cda77cb896a3a6a3e

> What's the point of indicating a base commit if that's not something I
> can find? You should never base any work on top of something that
> isn't a stable tag.

This was due to you having applied the LSFE patch but that application
not yet being visible to me at the time I did the rebase (I've noticed
occasional slow mirroring on kernel.org recently, possibly it was that),
I needed to base on the already applied patch so that I could check that
the test binary worked properly.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH 0/2] KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs

[Marc Zyngier]

On Sat, 20 Sep 2025 20:51:58 +0100, Mark Brown wrote:
> The set_id_regs selftest lacks coverag for ID_AA64ISR3_EL1 which has
> several features exposed to KVM guests in it.  Add coverage, and while
> we're here adjust the test to improve maintainability a bit.
> 
> The test will fail without the recently applied change adding FEAT_LSFE:
> 
>    https://lore.kernel.org/r/175829303126.1764550.939188785634158487.b4-ty@kernel.org
> 
> [...]

Applied to next, thanks!

[1/2] KVM: arm64: selftests: Remove a duplicate register listing in set_id_regs
      commit: 5a070fc376babc7efdc8288b97431e43e18f4646
[2/2] KVM: arm64: selftests: Cover ID_AA64ISAR3_EL1 in set_id_regs
      commit: b02a2c060b657296c080cff1b54ee4e9e650811c

Cheers,

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