From a3590d71a1acef850864f19bff2f37f56b2d4f00 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Wed, 30 Oct 2024 00:02:02 +0000
Subject: [PATCH 01/16] kselftest/arm64: Increase frequency of signal delivery
 in fp-stress

Currently we only deliver signals to the processes being tested about once
a second, meaning that the signal code paths are subject to relatively
little stress. Increase this frequency substantially to 25ms intervals,
along with some minor refactoring to make this more readily tuneable and
maintain the 1s logging interval. This interval was chosen based on some
experimentation with emulated platforms to avoid causing so much extra load
that the test starts to run into the 45s limit for selftests or generally
completely disconnect the timeout numbers from the

We could increase this if we moved the signal generation out of the main
supervisor thread, though we should also consider that he percentage of
time that we spend interacting with the floating point state is also a
consideration.

Suggested-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20241030-arm64-fp-stress-interval-v2-1-bd3cef48c22c@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/fp/fp-stress.c | 26 +++++++++++---------
 1 file changed, 15 insertions(+), 11 deletions(-)

diff --git a/tools/testing/selftests/arm64/fp/fp-stress.c b/tools/testing/selftests/arm64/fp/fp-stress.c
index 13958e645afc..b81bc0842f17 100644
--- a/tools/testing/selftests/arm64/fp/fp-stress.c
+++ b/tools/testing/selftests/arm64/fp/fp-stress.c
@@ -28,6 +28,9 @@
 
 #define MAX_VLS 16
 
+#define SIGNAL_INTERVAL_MS 25
+#define LOG_INTERVALS (1000 / SIGNAL_INTERVAL_MS)
+
 struct child_data {
 	char *name, *output;
 	pid_t pid;
@@ -448,7 +451,7 @@ static const struct option options[] = {
 int main(int argc, char **argv)
 {
 	int ret;
-	int timeout = 10;
+	int timeout = 10 * (1000 / SIGNAL_INTERVAL_MS);
 	int cpus, i, j, c;
 	int sve_vl_count, sme_vl_count;
 	bool all_children_started = false;
@@ -504,7 +507,7 @@ int main(int argc, char **argv)
 		       have_sme2 ? "present" : "absent");
 
 	if (timeout > 0)
-		ksft_print_msg("Will run for %ds\n", timeout);
+		ksft_print_msg("Will run for %d\n", timeout);
 	else
 		ksft_print_msg("Will run until terminated\n");
 
@@ -577,14 +580,14 @@ int main(int argc, char **argv)
 			break;
 
 		/*
-		 * Timeout is counted in seconds with no output, the
-		 * tests print during startup then are silent when
-		 * running so this should ensure they all ran enough
-		 * to install the signal handler, this is especially
-		 * useful in emulation where we will both be slow and
-		 * likely to have a large set of VLs.
+		 * Timeout is counted in poll intervals with no
+		 * output, the tests print during startup then are
+		 * silent when running so this should ensure they all
+		 * ran enough to install the signal handler, this is
+		 * especially useful in emulation where we will both
+		 * be slow and likely to have a large set of VLs.
 		 */
-		ret = epoll_wait(epoll_fd, evs, tests, 1000);
+		ret = epoll_wait(epoll_fd, evs, tests, SIGNAL_INTERVAL_MS);
 		if (ret < 0) {
 			if (errno == EINTR)
 				continue;
@@ -624,8 +627,9 @@ int main(int argc, char **argv)
 			all_children_started = true;
 		}
 
-		ksft_print_msg("Sending signals, timeout remaining: %d\n",
-			       timeout);
+		if ((timeout % LOG_INTERVALS) == 0)
+			ksft_print_msg("Sending signals, timeout remaining: %d\n",
+				       timeout);
 
 		for (i = 0; i < num_children; i++)
 			child_tickle(&children[i]);
-- 
2.50.1


From 161e9925053cafa83a1eb265a001b6917dfafa29 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Wed, 30 Oct 2024 00:02:03 +0000
Subject: [PATCH 02/16] kselftest/arm64: Poll less often while waiting for
 fp-stress children

While fp-stress is waiting for children to start it doesn't send any
signals to them so there is no need for it to have as short an epoll()
timeout as it does when the children are all running. We do still want to
have some timeout so that we can log diagnostics about missing children but
this can be relatively large. On emulated platforms the overhead of running
the supervisor process is quite high, especially during the process of
execing the test binaries.

Implement a longer epoll() timeout during the setup phase, using a 5s
timeout while waiting for children and switching  to the signal raise
interval when all the children are started and we start sending signals.

Signed-off-by: Mark Brown <broonie@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20241030-arm64-fp-stress-interval-v2-2-bd3cef48c22c@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/fp/fp-stress.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/arm64/fp/fp-stress.c b/tools/testing/selftests/arm64/fp/fp-stress.c
index b81bc0842f17..ad867ff9687a 100644
--- a/tools/testing/selftests/arm64/fp/fp-stress.c
+++ b/tools/testing/selftests/arm64/fp/fp-stress.c
@@ -452,6 +452,7 @@ int main(int argc, char **argv)
 {
 	int ret;
 	int timeout = 10 * (1000 / SIGNAL_INTERVAL_MS);
+	int poll_interval = 5000;
 	int cpus, i, j, c;
 	int sve_vl_count, sme_vl_count;
 	bool all_children_started = false;
@@ -587,7 +588,7 @@ int main(int argc, char **argv)
 		 * especially useful in emulation where we will both
 		 * be slow and likely to have a large set of VLs.
 		 */
-		ret = epoll_wait(epoll_fd, evs, tests, SIGNAL_INTERVAL_MS);
+		ret = epoll_wait(epoll_fd, evs, tests, poll_interval);
 		if (ret < 0) {
 			if (errno == EINTR)
 				continue;
@@ -625,6 +626,7 @@ int main(int argc, char **argv)
 			}
 
 			all_children_started = true;
+			poll_interval = SIGNAL_INTERVAL_MS;
 		}
 
 		if ((timeout % LOG_INTERVALS) == 0)
-- 
2.50.1


From 94de486e4215b83de8a5e908b4b8f7d3979c3b0e Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Thu, 7 Nov 2024 01:39:20 +0000
Subject: [PATCH 03/16] kselftest/arm64: Correct misleading comments on
 fp-stress irritators

The comments in the handlers for the irritator signal in the test threads
for fp-stress suggest that the irritator will corrupt the register state
observed by the main thread but this is not the case, instead the FPSIMD
and SVE irritators (which are the only ones that are implemented) modify
the current register state which is expected to be overwritten on return
from the handler by the saved register state. Update the comment to reflect
what the handler is actually doing.

Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241107-arm64-fp-stress-irritator-v2-1-c4b9622e36ee@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/fp/fpsimd-test.S | 3 +--
 tools/testing/selftests/arm64/fp/sve-test.S    | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/arm64/fp/fpsimd-test.S b/tools/testing/selftests/arm64/fp/fpsimd-test.S
index 8b960d01ed2e..bdfb7cf2e4ec 100644
--- a/tools/testing/selftests/arm64/fp/fpsimd-test.S
+++ b/tools/testing/selftests/arm64/fp/fpsimd-test.S
@@ -134,8 +134,7 @@ function check_vreg
 	b	memcmp
 endfunction
 
-// Any SVE register modified here can cause corruption in the main
-// thread -- but *only* the registers modified here.
+// Modify live register state, the signal return will undo our changes
 function irritator_handler
 	// Increment the irritation signal count (x23):
 	ldr	x0, [x2, #ucontext_regs + 8 * 23]
diff --git a/tools/testing/selftests/arm64/fp/sve-test.S b/tools/testing/selftests/arm64/fp/sve-test.S
index fff60e2a25ad..e3c0d585684d 100644
--- a/tools/testing/selftests/arm64/fp/sve-test.S
+++ b/tools/testing/selftests/arm64/fp/sve-test.S
@@ -291,8 +291,7 @@ function check_ffr
 #endif
 endfunction
 
-// Any SVE register modified here can cause corruption in the main
-// thread -- but *only* the registers modified here.
+// Modify live register state, the signal return will undo our changes
 function irritator_handler
 	// Increment the irritation signal count (x23):
 	ldr	x0, [x2, #ucontext_regs + 8 * 23]
-- 
2.50.1


From ffca567fef9c4661c792ca48a2cdd038b2df4887 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Thu, 7 Nov 2024 01:39:21 +0000
Subject: [PATCH 04/16] kselftest/arm64: Remove unused ADRs from irritator
 handlers

The irritator handlers for the fp-stress test programs all use ADR to load
an address into x0 which is then not referenced. Remove these ADRs as they
just cause confusion.

Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241107-arm64-fp-stress-irritator-v2-2-c4b9622e36ee@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/fp/fpsimd-test.S | 1 -
 tools/testing/selftests/arm64/fp/sve-test.S    | 1 -
 tools/testing/selftests/arm64/fp/za-test.S     | 1 -
 tools/testing/selftests/arm64/fp/zt-test.S     | 1 -
 4 files changed, 4 deletions(-)

diff --git a/tools/testing/selftests/arm64/fp/fpsimd-test.S b/tools/testing/selftests/arm64/fp/fpsimd-test.S
index bdfb7cf2e4ec..9977ffdd758a 100644
--- a/tools/testing/selftests/arm64/fp/fpsimd-test.S
+++ b/tools/testing/selftests/arm64/fp/fpsimd-test.S
@@ -142,7 +142,6 @@ function irritator_handler
 	str	x0, [x2, #ucontext_regs + 8 * 23]
 
 	// Corrupt some random V-regs
-	adr	x0, .text + (irritator_handler - .text) / 16 * 16
 	movi	v0.8b, #7
 	movi	v9.16b, #9
 	movi	v31.8b, #31
diff --git a/tools/testing/selftests/arm64/fp/sve-test.S b/tools/testing/selftests/arm64/fp/sve-test.S
index e3c0d585684d..f1fb9745c681 100644
--- a/tools/testing/selftests/arm64/fp/sve-test.S
+++ b/tools/testing/selftests/arm64/fp/sve-test.S
@@ -299,7 +299,6 @@ function irritator_handler
 	str	x0, [x2, #ucontext_regs + 8 * 23]
 
 	// Corrupt some random Z-regs
-	adr	x0, .text + (irritator_handler - .text) / 16 * 16
 	movi	v0.8b, #1
 	movi	v9.16b, #2
 	movi	v31.8b, #3
diff --git a/tools/testing/selftests/arm64/fp/za-test.S b/tools/testing/selftests/arm64/fp/za-test.S
index 095b45531640..1ee0ec36766d 100644
--- a/tools/testing/selftests/arm64/fp/za-test.S
+++ b/tools/testing/selftests/arm64/fp/za-test.S
@@ -158,7 +158,6 @@ function irritator_handler
 
 	// Corrupt some random ZA data
 #if 0
-	adr	x0, .text + (irritator_handler - .text) / 16 * 16
 	movi	v0.8b, #1
 	movi	v9.16b, #2
 	movi	v31.8b, #3
diff --git a/tools/testing/selftests/arm64/fp/zt-test.S b/tools/testing/selftests/arm64/fp/zt-test.S
index b5c81e81a379..ade9c98abcda 100644
--- a/tools/testing/selftests/arm64/fp/zt-test.S
+++ b/tools/testing/selftests/arm64/fp/zt-test.S
@@ -127,7 +127,6 @@ function irritator_handler
 
 	// Corrupt some random ZT data
 #if 0
-	adr	x0, .text + (irritator_handler - .text) / 16 * 16
 	movi	v0.8b, #1
 	movi	v9.16b, #2
 	movi	v31.8b, #3
-- 
2.50.1


From d65f27d240bb897ac5a4a3fb7557724b3717e022 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Thu, 7 Nov 2024 01:39:23 +0000
Subject: [PATCH 05/16] kselftest/arm64: Implement irritators for ZA and ZT

Currently we don't use the irritator signal in our floating point stress
tests so when we added ZA and ZT stress tests we didn't actually bother
implementing any actual action in the handlers, we just counted the signal
deliveries. In preparation for using the irritators let's implement them,
just trivially SMSTOP and SMSTART to reset all bits in the register to 0.

Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241107-arm64-fp-stress-irritator-v2-4-c4b9622e36ee@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/fp/za-test.S | 12 ++++--------
 tools/testing/selftests/arm64/fp/zt-test.S | 12 ++++--------
 2 files changed, 8 insertions(+), 16 deletions(-)

diff --git a/tools/testing/selftests/arm64/fp/za-test.S b/tools/testing/selftests/arm64/fp/za-test.S
index 1ee0ec36766d..f902e6ef9077 100644
--- a/tools/testing/selftests/arm64/fp/za-test.S
+++ b/tools/testing/selftests/arm64/fp/za-test.S
@@ -148,20 +148,16 @@ function check_za
 	b	memcmp
 endfunction
 
-// Any SME register modified here can cause corruption in the main
-// thread -- but *only* the locations modified here.
+// Modify the live SME register state, signal return will undo our changes
 function irritator_handler
 	// Increment the irritation signal count (x23):
 	ldr	x0, [x2, #ucontext_regs + 8 * 23]
 	add	x0, x0, #1
 	str	x0, [x2, #ucontext_regs + 8 * 23]
 
-	// Corrupt some random ZA data
-#if 0
-	movi	v0.8b, #1
-	movi	v9.16b, #2
-	movi	v31.8b, #3
-#endif
+	// This will reset ZA to all bits 0
+	smstop
+	smstart
 
 	ret
 endfunction
diff --git a/tools/testing/selftests/arm64/fp/zt-test.S b/tools/testing/selftests/arm64/fp/zt-test.S
index ade9c98abcda..c96cb7c2ad4b 100644
--- a/tools/testing/selftests/arm64/fp/zt-test.S
+++ b/tools/testing/selftests/arm64/fp/zt-test.S
@@ -117,20 +117,16 @@ function check_zt
 	b	memcmp
 endfunction
 
-// Any SME register modified here can cause corruption in the main
-// thread -- but *only* the locations modified here.
+// Modify the live SME register state, signal return will undo our changes
 function irritator_handler
 	// Increment the irritation signal count (x23):
 	ldr	x0, [x2, #ucontext_regs + 8 * 23]
 	add	x0, x0, #1
 	str	x0, [x2, #ucontext_regs + 8 * 23]
 
-	// Corrupt some random ZT data
-#if 0
-	movi	v0.8b, #1
-	movi	v9.16b, #2
-	movi	v31.8b, #3
-#endif
+	// This will reset ZT to all bits 0
+	smstop
+	smstart
 
 	ret
 endfunction
-- 
2.50.1


From 7368debf275aa2419365ce47da00922ca51c6094 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Thu, 7 Nov 2024 01:39:24 +0000
Subject: [PATCH 06/16] kselftest/arm64: Provide a SIGUSR1 handler in the
 kernel mode FP stress test

The other stress test programs provide a SIGUSR1 handler which modifies the
live register state in order to validate that signal context is being
restored during signal return. While we can't usefully do this when testing
kernel mode FP usage provide a handler for SIGUSR1 which just counts the
number of signals like we do for SIGUSR2, allowing fp-stress to treat all
the test programs uniformly.

Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241107-arm64-fp-stress-irritator-v2-5-c4b9622e36ee@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/fp/kernel-test.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/tools/testing/selftests/arm64/fp/kernel-test.c b/tools/testing/selftests/arm64/fp/kernel-test.c
index e8da3b4cbd23..859345379044 100644
--- a/tools/testing/selftests/arm64/fp/kernel-test.c
+++ b/tools/testing/selftests/arm64/fp/kernel-test.c
@@ -267,6 +267,10 @@ int main(void)
 		       strerror(errno), errno);
 
 	sa.sa_sigaction = handle_kick_signal;
+	ret = sigaction(SIGUSR1, &sa, NULL);
+	if (ret < 0)
+		printf("Failed to install SIGUSR1 handler: %s (%d)\n",
+		       strerror(errno), errno);
 	ret = sigaction(SIGUSR2, &sa, NULL);
 	if (ret < 0)
 		printf("Failed to install SIGUSR2 handler: %s (%d)\n",
-- 
2.50.1


From ead1c35ce3b3c766443a82b56ee343cfe7ee8305 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Thu, 7 Nov 2024 01:39:25 +0000
Subject: [PATCH 07/16] kselftest/arm64: Test signal handler state modification
 in fp-stress

Currently in fp-stress we test signal delivery to the test threads by
sending SIGUSR2 which simply counts how many signals are delivered. The
test programs now also all have a SIGUSR1 handler which for the threads
doing userspace testing additionally modifies the floating point register
state in the signal handler, verifying that when we return the saved
register state is restored from the signal context as expected. Switch over
to triggering that to validate that we are restoring as expected.

Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241107-arm64-fp-stress-irritator-v2-6-c4b9622e36ee@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/fp/fp-stress.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/arm64/fp/fp-stress.c b/tools/testing/selftests/arm64/fp/fp-stress.c
index ad867ff9687a..74e23208b94c 100644
--- a/tools/testing/selftests/arm64/fp/fp-stress.c
+++ b/tools/testing/selftests/arm64/fp/fp-stress.c
@@ -223,7 +223,7 @@ static void child_output(struct child_data *child, uint32_t events,
 static void child_tickle(struct child_data *child)
 {
 	if (child->output_seen && !child->exited)
-		kill(child->pid, SIGUSR2);
+		kill(child->pid, SIGUSR1);
 }
 
 static void child_stop(struct child_data *child)
-- 
2.50.1


From ae465d9ca192f582cf4932e628a25f9625a8bf83 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Fri, 8 Nov 2024 15:20:46 +0000
Subject: [PATCH 08/16] kselftest/arm64: Fix build with stricter assemblers

While some assemblers (including the LLVM assembler I mostly use) will
happily accept SMSTART as an instruction by default others, specifically
gas, require that any architecture extensions be explicitly enabled.
The assembler SME test programs use manually encoded helpers for the new
instructions but no SMSTART helper is defined, only SM and ZA specific
variants.  Unfortunately the irritators that were just added use plain
SMSTART so on stricter assemblers these fail to build:

za-test.S:160: Error: selected processor does not support `smstart'

Switch to using SMSTART ZA via the manually encoded smstart_za macro we
already have defined.

Fixes: d65f27d240bb ("kselftest/arm64: Implement irritators for ZA and ZT")
Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241108-arm64-selftest-asm-error-v1-1-7ce27b42a677@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/fp/za-test.S | 2 +-
 tools/testing/selftests/arm64/fp/zt-test.S | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/arm64/fp/za-test.S b/tools/testing/selftests/arm64/fp/za-test.S
index f902e6ef9077..5abcb8f009f8 100644
--- a/tools/testing/selftests/arm64/fp/za-test.S
+++ b/tools/testing/selftests/arm64/fp/za-test.S
@@ -157,7 +157,7 @@ function irritator_handler
 
 	// This will reset ZA to all bits 0
 	smstop
-	smstart
+	smstart_za
 
 	ret
 endfunction
diff --git a/tools/testing/selftests/arm64/fp/zt-test.S b/tools/testing/selftests/arm64/fp/zt-test.S
index c96cb7c2ad4b..7b9de8d2a873 100644
--- a/tools/testing/selftests/arm64/fp/zt-test.S
+++ b/tools/testing/selftests/arm64/fp/zt-test.S
@@ -126,7 +126,7 @@ function irritator_handler
 
 	// This will reset ZT to all bits 0
 	smstop
-	smstart
+	smstart_za
 
 	ret
 endfunction
-- 
2.50.1


From b6bd50dd3b564d50b8cd748de6bae58804ecb768 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 8 Nov 2024 13:49:17 +0000
Subject: [PATCH 09/16] kselftest/arm64: Fix printf() compiler warnings in the
 arm64 fp tests

Lots of incorrect length modifiers, missing arguments or conversion
specifiers. Fix them.

Cc: Shuah Khan <skhan@linuxfoundation.org>
Cc: Mark Brown <broonie@kernel.org>
Reviewed-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241108134920.1233992-2-catalin.marinas@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/fp/sve-ptrace.c | 16 +++++++++-------
 tools/testing/selftests/arm64/fp/za-ptrace.c  |  8 +++++---
 tools/testing/selftests/arm64/fp/zt-ptrace.c  |  8 +++++---
 3 files changed, 19 insertions(+), 13 deletions(-)

diff --git a/tools/testing/selftests/arm64/fp/sve-ptrace.c b/tools/testing/selftests/arm64/fp/sve-ptrace.c
index 6d61992fe8a0..577b6e05e860 100644
--- a/tools/testing/selftests/arm64/fp/sve-ptrace.c
+++ b/tools/testing/selftests/arm64/fp/sve-ptrace.c
@@ -82,10 +82,12 @@ static void fill_buf(char *buf, size_t size)
 static int do_child(void)
 {
 	if (ptrace(PTRACE_TRACEME, -1, NULL, NULL))
-		ksft_exit_fail_msg("PTRACE_TRACEME", strerror(errno));
+		ksft_exit_fail_msg("ptrace(PTRACE_TRACEME) failed: %s (%d)\n",
+				   strerror(errno), errno);
 
 	if (raise(SIGSTOP))
-		ksft_exit_fail_msg("raise(SIGSTOP)", strerror(errno));
+		ksft_exit_fail_msg("raise(SIGSTOP) failed: %s (%d)\n",
+				   strerror(errno), errno);
 
 	return EXIT_SUCCESS;
 }
@@ -340,7 +342,7 @@ static void ptrace_set_sve_get_sve_data(pid_t child,
 	data_size = SVE_PT_SVE_OFFSET + SVE_PT_SVE_SIZE(vq, SVE_PT_REGS_SVE);
 	write_buf = malloc(data_size);
 	if (!write_buf) {
-		ksft_test_result_fail("Error allocating %d byte buffer for %s VL %u\n",
+		ksft_test_result_fail("Error allocating %ld byte buffer for %s VL %u\n",
 				      data_size, type->name, vl);
 		return;
 	}
@@ -441,7 +443,7 @@ static void ptrace_set_sve_get_fpsimd_data(pid_t child,
 	data_size = SVE_PT_SVE_OFFSET + SVE_PT_SVE_SIZE(vq, SVE_PT_REGS_SVE);
 	write_buf = malloc(data_size);
 	if (!write_buf) {
-		ksft_test_result_fail("Error allocating %d byte buffer for %s VL %u\n",
+		ksft_test_result_fail("Error allocating %ld byte buffer for %s VL %u\n",
 				      data_size, type->name, vl);
 		return;
 	}
@@ -545,7 +547,7 @@ static void ptrace_set_fpsimd_get_sve_data(pid_t child,
 	read_sve = read_buf;
 
 	if (read_sve->vl != vl) {
-		ksft_test_result_fail("Child VL != expected VL %d\n",
+		ksft_test_result_fail("Child VL != expected VL: %u != %u\n",
 				      read_sve->vl, vl);
 		goto out;
 	}
@@ -555,7 +557,7 @@ static void ptrace_set_fpsimd_get_sve_data(pid_t child,
 	case SVE_PT_REGS_FPSIMD:
 		expected_size = SVE_PT_FPSIMD_SIZE(vq, SVE_PT_REGS_FPSIMD);
 		if (read_sve_size < expected_size) {
-			ksft_test_result_fail("Read %d bytes, expected %d\n",
+			ksft_test_result_fail("Read %ld bytes, expected %ld\n",
 					      read_sve_size, expected_size);
 			goto out;
 		}
@@ -571,7 +573,7 @@ static void ptrace_set_fpsimd_get_sve_data(pid_t child,
 	case SVE_PT_REGS_SVE:
 		expected_size = SVE_PT_SVE_SIZE(vq, SVE_PT_REGS_SVE);
 		if (read_sve_size < expected_size) {
-			ksft_test_result_fail("Read %d bytes, expected %d\n",
+			ksft_test_result_fail("Read %ld bytes, expected %ld\n",
 					      read_sve_size, expected_size);
 			goto out;
 		}
diff --git a/tools/testing/selftests/arm64/fp/za-ptrace.c b/tools/testing/selftests/arm64/fp/za-ptrace.c
index ac27d87396fc..08c777f87ea2 100644
--- a/tools/testing/selftests/arm64/fp/za-ptrace.c
+++ b/tools/testing/selftests/arm64/fp/za-ptrace.c
@@ -48,10 +48,12 @@ static void fill_buf(char *buf, size_t size)
 static int do_child(void)
 {
 	if (ptrace(PTRACE_TRACEME, -1, NULL, NULL))
-		ksft_exit_fail_msg("PTRACE_TRACEME", strerror(errno));
+		ksft_exit_fail_msg("ptrace(PTRACE_TRACEME) failed: %s (%d)",
+				   strerror(errno), errno);
 
 	if (raise(SIGSTOP))
-		ksft_exit_fail_msg("raise(SIGSTOP)", strerror(errno));
+		ksft_exit_fail_msg("raise(SIGSTOP) failed: %s (%d)\n",
+				   strerror(errno), errno);
 
 	return EXIT_SUCCESS;
 }
@@ -201,7 +203,7 @@ static void ptrace_set_get_data(pid_t child, unsigned int vl)
 	data_size = ZA_PT_SIZE(vq);
 	write_buf = malloc(data_size);
 	if (!write_buf) {
-		ksft_test_result_fail("Error allocating %d byte buffer for VL %u\n",
+		ksft_test_result_fail("Error allocating %ld byte buffer for VL %u\n",
 				      data_size, vl);
 		return;
 	}
diff --git a/tools/testing/selftests/arm64/fp/zt-ptrace.c b/tools/testing/selftests/arm64/fp/zt-ptrace.c
index 996d9614a131..584b8d59b7ea 100644
--- a/tools/testing/selftests/arm64/fp/zt-ptrace.c
+++ b/tools/testing/selftests/arm64/fp/zt-ptrace.c
@@ -43,10 +43,12 @@ static void fill_buf(char *buf, size_t size)
 static int do_child(void)
 {
 	if (ptrace(PTRACE_TRACEME, -1, NULL, NULL))
-		ksft_exit_fail_msg("PTRACE_TRACEME", strerror(errno));
+		ksft_exit_fail_msg("ptrace(PTRACE_TRACEME) failed: %s (%d)\n",
+				   strerror(errno), errno);
 
 	if (raise(SIGSTOP))
-		ksft_exit_fail_msg("raise(SIGSTOP)", strerror(errno));
+		ksft_exit_fail_msg("raise(SIGSTOP) failed: %s (%d)\n",
+				   strerror(errno), errno);
 
 	return EXIT_SUCCESS;
 }
@@ -231,7 +233,7 @@ static void ptrace_enable_za_via_zt(pid_t child)
 		/* Should have register data */
 		if (za_out->size < ZA_PT_SIZE(vq)) {
 			ksft_print_msg("ZA data less than expected: %u < %u\n",
-				       za_out->size, ZA_PT_SIZE(vq));
+				       za_out->size, (unsigned int)ZA_PT_SIZE(vq));
 			fail = true;
 			vq = 0;
 		}
-- 
2.50.1


From 0cc6b94a445c53ab152554b4cf60575e1396adf6 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 8 Nov 2024 13:49:18 +0000
Subject: [PATCH 10/16] kselftest/arm64: Fix printf() warning in the arm64 MTE
 prctl() test

While prctl() returns an 'int', the PR_MTE_TCF_MASK is defined as
unsigned long which results in the larger type following a bitwise 'and'
operation. Cast the printf() argument to 'int'.

Cc: Shuah Khan <skhan@linuxfoundation.org>
Cc: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241108134920.1233992-3-catalin.marinas@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/mte/check_prctl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/arm64/mte/check_prctl.c b/tools/testing/selftests/arm64/mte/check_prctl.c
index f139a33a43ef..4c89e9538ca0 100644
--- a/tools/testing/selftests/arm64/mte/check_prctl.c
+++ b/tools/testing/selftests/arm64/mte/check_prctl.c
@@ -85,7 +85,7 @@ void set_mode_test(const char *name, int hwcap2, int mask)
 		ksft_test_result_pass("%s\n", name);
 	} else {
 		ksft_print_msg("Got %x, expected %x\n",
-			       (ret & PR_MTE_TCF_MASK), mask);
+			       (ret & (int)PR_MTE_TCF_MASK), mask);
 		ksft_test_result_fail("%s\n", name);
 	}
 }
-- 
2.50.1


From 694e2803fece8d066bd85ce8607c630ce2b69859 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 8 Nov 2024 13:49:19 +0000
Subject: [PATCH 11/16] kselftest/arm64: Fix printf() compiler warnings in the
 arm64 syscall-abi.c tests

Fix the incorrect length modifiers in arm64/abi/syscall-abi.c.

Cc: Shuah Khan <skhan@linuxfoundation.org>
Cc: Mark Brown <broonie@kernel.org>
Reviewed-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241108134920.1233992-4-catalin.marinas@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/abi/syscall-abi.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/arm64/abi/syscall-abi.c b/tools/testing/selftests/arm64/abi/syscall-abi.c
index d704511a0955..5ec9a18ec802 100644
--- a/tools/testing/selftests/arm64/abi/syscall-abi.c
+++ b/tools/testing/selftests/arm64/abi/syscall-abi.c
@@ -81,7 +81,7 @@ static int check_gpr(struct syscall_cfg *cfg, int sve_vl, int sme_vl, uint64_t s
 	 */
 	for (i = 9; i < ARRAY_SIZE(gpr_in); i++) {
 		if (gpr_in[i] != gpr_out[i]) {
-			ksft_print_msg("%s SVE VL %d mismatch in GPR %d: %llx != %llx\n",
+			ksft_print_msg("%s SVE VL %d mismatch in GPR %d: %lx != %lx\n",
 				       cfg->name, sve_vl, i,
 				       gpr_in[i], gpr_out[i]);
 			errors++;
@@ -112,7 +112,7 @@ static int check_fpr(struct syscall_cfg *cfg, int sve_vl, int sme_vl,
 	if (!sve_vl && !(svcr & SVCR_SM_MASK)) {
 		for (i = 0; i < ARRAY_SIZE(fpr_in); i++) {
 			if (fpr_in[i] != fpr_out[i]) {
-				ksft_print_msg("%s Q%d/%d mismatch %llx != %llx\n",
+				ksft_print_msg("%s Q%d/%d mismatch %lx != %lx\n",
 					       cfg->name,
 					       i / 2, i % 2,
 					       fpr_in[i], fpr_out[i]);
@@ -294,13 +294,13 @@ static int check_svcr(struct syscall_cfg *cfg, int sve_vl, int sme_vl,
 	int errors = 0;
 
 	if (svcr_out & SVCR_SM_MASK) {
-		ksft_print_msg("%s Still in SM, SVCR %llx\n",
+		ksft_print_msg("%s Still in SM, SVCR %lx\n",
 			       cfg->name, svcr_out);
 		errors++;
 	}
 
 	if ((svcr_in & SVCR_ZA_MASK) != (svcr_out & SVCR_ZA_MASK)) {
-		ksft_print_msg("%s PSTATE.ZA changed, SVCR %llx != %llx\n",
+		ksft_print_msg("%s PSTATE.ZA changed, SVCR %lx != %lx\n",
 			       cfg->name, svcr_in, svcr_out);
 		errors++;
 	}
-- 
2.50.1


From 116e50d6474e82579086c0397d2fa3999815f29e Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Wed, 6 Nov 2024 17:07:51 +0000
Subject: [PATCH 12/16] kselftest/arm64: Check that SVCR is 0 in signal
 handlers

We don't currently validate that we exit streaming mode and clear ZA when
we enter a signal handler. Add simple checks for this in the SSVE and ZA
tests.

Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241106-arm64-fpmr-signal-test-v1-1-31fa34ce58fe@kernel.org
[catalin.marinas@arm.com: Use %lx in fprintf() as uint64_t seems to be unsigned long in glibc]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/signal/sve_helpers.h  | 13 +++++++++++++
 .../selftests/arm64/signal/testcases/ssve_regs.c    |  5 +++++
 .../selftests/arm64/signal/testcases/za_regs.c      |  5 +++++
 3 files changed, 23 insertions(+)

diff --git a/tools/testing/selftests/arm64/signal/sve_helpers.h b/tools/testing/selftests/arm64/signal/sve_helpers.h
index 50948ce471cc..ca133b93375f 100644
--- a/tools/testing/selftests/arm64/signal/sve_helpers.h
+++ b/tools/testing/selftests/arm64/signal/sve_helpers.h
@@ -18,4 +18,17 @@ extern unsigned int nvls;
 
 int sve_fill_vls(bool use_sme, int min_vls);
 
+static inline uint64_t get_svcr(void)
+{
+	uint64_t val;
+
+	asm volatile (
+		"mrs	%0, S3_3_C4_C2_2\n"
+		: "=r"(val)
+		:
+		: "cc");
+
+	return val;
+}
+
 #endif
diff --git a/tools/testing/selftests/arm64/signal/testcases/ssve_regs.c b/tools/testing/selftests/arm64/signal/testcases/ssve_regs.c
index 6dbe48cf8b09..1dbca9afb13c 100644
--- a/tools/testing/selftests/arm64/signal/testcases/ssve_regs.c
+++ b/tools/testing/selftests/arm64/signal/testcases/ssve_regs.c
@@ -85,6 +85,11 @@ static int do_one_sme_vl(struct tdescr *td, siginfo_t *si, ucontext_t *uc,
 	fprintf(stderr, "Got expected size %u and VL %d\n",
 		head->size, ssve->vl);
 
+	if (get_svcr() != 0) {
+		fprintf(stderr, "Unexpected SVCR %lx\n", get_svcr());
+		return 1;
+	}
+
 	return 0;
 }
 
diff --git a/tools/testing/selftests/arm64/signal/testcases/za_regs.c b/tools/testing/selftests/arm64/signal/testcases/za_regs.c
index b9e13f27f1f9..badaead5326a 100644
--- a/tools/testing/selftests/arm64/signal/testcases/za_regs.c
+++ b/tools/testing/selftests/arm64/signal/testcases/za_regs.c
@@ -91,6 +91,11 @@ static int do_one_sme_vl(struct tdescr *td, siginfo_t *si, ucontext_t *uc,
 		return 1;
 	}
 
+	if (get_svcr() != 0) {
+		fprintf(stderr, "Unexpected SVCR %lx\n", get_svcr());
+		return 1;
+	}
+
 	return 0;
 }
 
-- 
2.50.1


From c297aa7d3fb6755890b78b483e82c9cf07370d50 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Mon, 11 Nov 2024 18:32:58 +0000
Subject: [PATCH 13/16] kselftest/arm64: Enable build of PAC tests with LLVM=1

Currently we don't build the PAC selftests when building with LLVM=1 since
we attempt to test for PAC support in the toolchain before we've set up the
build system to point at LLVM in lib.mk, which has to be one of the last
things in the Makefile.

Since all versions of LLVM supported for use with the kernel have PAC
support we can just sidestep the issue by just assuming PAC is there when
doing a LLVM=1 build.

Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241111-arm64-selftest-pac-clang-v1-1-08599ceee418@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/pauth/Makefile | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tools/testing/selftests/arm64/pauth/Makefile b/tools/testing/selftests/arm64/pauth/Makefile
index 72e290b0b10c..b5a1c80e0ead 100644
--- a/tools/testing/selftests/arm64/pauth/Makefile
+++ b/tools/testing/selftests/arm64/pauth/Makefile
@@ -7,8 +7,14 @@ CC := $(CROSS_COMPILE)gcc
 endif
 
 CFLAGS += -mbranch-protection=pac-ret
+
+# All supported LLVMs have PAC, test for GCC
+ifeq ($(LLVM),1)
+pauth_cc_support := 1
+else
 # check if the compiler supports ARMv8.3 and branch protection with PAuth
 pauth_cc_support := $(shell if ($(CC) $(CFLAGS) -march=armv8.3-a -E -x c /dev/null -o /dev/null 2>&1) then echo "1"; fi)
+endif
 
 ifeq ($(pauth_cc_support),1)
 TEST_GEN_PROGS := pac
-- 
2.50.1


From c0350076c13eac4f1d7f7ab6acd43bb252baef7a Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Tue, 12 Nov 2024 13:08:14 +0000
Subject: [PATCH 14/16] kselftets/arm64: Use flag bits for features in
 fp-ptrace assembler code

The assembler portions of fp-ptrace are passed feature flags by the C code
indicating which architectural features are supported. Currently these use
an entire register for each flag which is wasteful and gets cumbersome as
new flags are added. Switch to using flag bits in a single register to make
things easier to maintain.

No functional change.

Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241112-arm64-fp-ptrace-fpmr-v2-1-250b57c61254@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 .../selftests/arm64/fp/fp-ptrace-asm.S        | 32 +++++++++++--------
 tools/testing/selftests/arm64/fp/fp-ptrace.c  | 17 +++++++---
 tools/testing/selftests/arm64/fp/fp-ptrace.h  | 10 ++++++
 3 files changed, 41 insertions(+), 18 deletions(-)

diff --git a/tools/testing/selftests/arm64/fp/fp-ptrace-asm.S b/tools/testing/selftests/arm64/fp/fp-ptrace-asm.S
index 7ad59d92d02b..5e7e9c878f2c 100644
--- a/tools/testing/selftests/arm64/fp/fp-ptrace-asm.S
+++ b/tools/testing/selftests/arm64/fp/fp-ptrace-asm.S
@@ -15,10 +15,7 @@
 
 // Load and save register values with pauses for ptrace
 //
-// x0 - SVE in use
-// x1 - SME in use
-// x2 - SME2 in use
-// x3 - FA64 supported
+// x0 - HAVE_ flags indicating which features are in use
 
 .globl load_and_save
 load_and_save:
@@ -44,7 +41,7 @@ load_and_save:
 	ldp	q30, q31, [x7, #16 * 30]
 
 	// SME?
-	cbz	x1, check_sve_in
+	tbz	x0, #HAVE_SME_SHIFT, check_sve_in
 
 	adrp	x7, svcr_in
 	ldr	x7, [x7, :lo12:svcr_in]
@@ -64,7 +61,7 @@ load_and_save:
 	bne	1b
 
 	// ZT?
-	cbz	x2, check_sm_in
+	tbz	x0, #HAVE_SME2_SHIFT, check_sm_in
 	adrp	x6, zt_in
 	add	x6, x6, :lo12:zt_in
 	_ldr_zt 6
@@ -72,12 +69,16 @@ load_and_save:
 	// In streaming mode?
 check_sm_in:
 	tbz	x7, #SVCR_SM_SHIFT, check_sve_in
-	mov	x4, x3		// Load FFR if we have FA64
+
+	// Load FFR if we have FA64
+	mov	x4, #0
+	tbz	x0, #HAVE_FA64_SHIFT, load_sve
+	mov	x4, #1
 	b	load_sve
 
 	// SVE?
 check_sve_in:
-	cbz	x0, wait_for_writes
+	tbz	x0, #HAVE_SVE_SHIFT, wait_for_writes
 	mov	x4, #1
 
 load_sve:
@@ -165,8 +166,7 @@ wait_for_writes:
 	stp	q28, q29, [x7, #16 * 28]
 	stp	q30, q31, [x7, #16 * 30]
 
-	// SME?
-	cbz	x1, check_sve_out
+	tbz	x0, #HAVE_SME_SHIFT, check_sve_out
 
 	rdsvl	11, 1
 	adrp	x6, sme_vl_out
@@ -187,7 +187,7 @@ wait_for_writes:
 	bne	1b
 
 	// ZT?
-	cbz	x2, check_sm_out
+	tbz	x0, #HAVE_SME2_SHIFT, check_sm_out
 	adrp	x6, zt_out
 	add	x6, x6, :lo12:zt_out
 	_str_zt 6
@@ -195,12 +195,16 @@ wait_for_writes:
 	// In streaming mode?
 check_sm_out:
 	tbz	x7, #SVCR_SM_SHIFT, check_sve_out
-	mov	x4, x3				// FFR?
+
+	// Do we have FA64 and FFR?
+	mov	x4, #0
+	tbz	x0, #HAVE_FA64_SHIFT, read_sve
+	mov	x4, #1
 	b	read_sve
 
 	// SVE?
 check_sve_out:
-	cbz	x0, wait_for_reads
+	tbz	x0, #HAVE_SVE_SHIFT, wait_for_reads
 	mov	x4, #1
 
 	rdvl	x7, #1
@@ -271,7 +275,7 @@ wait_for_reads:
 	brk #0
 
 	// Ensure we don't leave ourselves in streaming mode
-	cbz	x1, out
+	tbz	x0, #HAVE_SME_SHIFT, out
 	msr	S3_3_C4_C2_2, xzr
 
 out:
diff --git a/tools/testing/selftests/arm64/fp/fp-ptrace.c b/tools/testing/selftests/arm64/fp/fp-ptrace.c
index c7ceafe5f471..d96af27487fa 100644
--- a/tools/testing/selftests/arm64/fp/fp-ptrace.c
+++ b/tools/testing/selftests/arm64/fp/fp-ptrace.c
@@ -82,7 +82,7 @@ uint64_t sve_vl_out;
 uint64_t sme_vl_out;
 uint64_t svcr_in, svcr_expected, svcr_out;
 
-void load_and_save(int sve, int sme, int sme2, int fa64);
+void load_and_save(int flags);
 
 static bool got_alarm;
 
@@ -198,7 +198,7 @@ static int vl_expected(struct test_config *config)
 
 static void run_child(struct test_config *config)
 {
-	int ret;
+	int ret, flags;
 
 	/* Let the parent attach to us */
 	ret = ptrace(PTRACE_TRACEME, 0, 0, 0);
@@ -224,8 +224,17 @@ static void run_child(struct test_config *config)
 	}
 
 	/* Load values and wait for the parent */
-	load_and_save(sve_supported(), sme_supported(),
-		      sme2_supported(), fa64_supported());
+	flags = 0;
+	if (sve_supported())
+		flags |= HAVE_SVE;
+	if (sme_supported())
+		flags |= HAVE_SME;
+	if (sme2_supported())
+		flags |= HAVE_SME2;
+	if (fa64_supported())
+		flags |= HAVE_FA64;
+
+	load_and_save(flags);
 
 	exit(0);
 }
diff --git a/tools/testing/selftests/arm64/fp/fp-ptrace.h b/tools/testing/selftests/arm64/fp/fp-ptrace.h
index db4f2c4d750c..36ca627e1980 100644
--- a/tools/testing/selftests/arm64/fp/fp-ptrace.h
+++ b/tools/testing/selftests/arm64/fp/fp-ptrace.h
@@ -10,4 +10,14 @@
 #define SVCR_SM (1 << SVCR_SM_SHIFT)
 #define SVCR_ZA (1 << SVCR_ZA_SHIFT)
 
+#define HAVE_SVE_SHIFT		0
+#define HAVE_SME_SHIFT		1
+#define HAVE_SME2_SHIFT		2
+#define HAVE_FA64_SHIFT		3
+
+#define HAVE_SVE	(1 << HAVE_SVE_SHIFT)
+#define HAVE_SME	(1 << HAVE_SME_SHIFT)
+#define HAVE_SME2	(1 << HAVE_SME2_SHIFT)
+#define HAVE_FA64	(1 << HAVE_FA64_SHIFT)
+
 #endif
-- 
2.50.1


From 7e9c5b00009a625cc304c865192978c01c0cc077 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Tue, 12 Nov 2024 13:08:15 +0000
Subject: [PATCH 15/16] kselftest/arm64: Expand the set of ZA writes fp-ptrace
 does

Currently our test for implementable ZA writes is written in a bit of a
convoluted fashion which excludes all changes where we clear SVCR.SM even
though we can actually support that since changing the vector length resets
SVCR. Make the logic more direct, enabling us to actually run these cases.

Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241112-arm64-fp-ptrace-fpmr-v2-2-250b57c61254@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 tools/testing/selftests/arm64/fp/fp-ptrace.c | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

diff --git a/tools/testing/selftests/arm64/fp/fp-ptrace.c b/tools/testing/selftests/arm64/fp/fp-ptrace.c
index d96af27487fa..56cf6e02c535 100644
--- a/tools/testing/selftests/arm64/fp/fp-ptrace.c
+++ b/tools/testing/selftests/arm64/fp/fp-ptrace.c
@@ -1078,21 +1078,19 @@ static void sve_write(pid_t child, struct test_config *config)
 
 static bool za_write_supported(struct test_config *config)
 {
-	if (config->svcr_expected & SVCR_SM) {
-		if (!(config->svcr_in & SVCR_SM))
+	if (config->sme_vl_in != config->sme_vl_expected) {
+		/* Changing the SME VL exits streaming mode. */
+		if (config->svcr_expected & SVCR_SM) {
 			return false;
-
-		/* Changing the SME VL exits streaming mode */
-		if (config->sme_vl_in != config->sme_vl_expected) {
+		}
+	} else {
+		/* Otherwise we can't change streaming mode */
+		if ((config->svcr_in & SVCR_SM) !=
+		    (config->svcr_expected & SVCR_SM)) {
 			return false;
 		}
 	}
 
-	/* Can't disable SM outside a VL change */
-	if ((config->svcr_in & SVCR_SM) &&
-	    !(config->svcr_expected & SVCR_SM))
-		return false;
-
 	return true;
 }
 
-- 
2.50.1


From 7dbd26d0b22d69d36ab3e76ee7f152482a19cbed Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@kernel.org>
Date: Tue, 12 Nov 2024 13:08:16 +0000
Subject: [PATCH 16/16] kselftest/arm64: Add FPMR coverage to fp-ptrace

Add coverage for FPMR to fp-ptrace. FPMR can be available independently of
SVE and SME, if SME is supported then FPMR is cleared by entering and
exiting streaming mode. As with other registers we generate random values
to load into the register, we restrict these to bitfields which are always
defined. We also leave bitfields where the valid values are affected by
the set of supported FP8 formats zero to reduce complexity, it is unlikely
that specific bitfields will be affected by ptrace issues.

Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20241112-arm64-fp-ptrace-fpmr-v2-3-250b57c61254@kernel.org
[catalin.marinas@arm.com: use REG_FPMR instead of FPMR]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 .../selftests/arm64/fp/fp-ptrace-asm.S        |  23 +++-
 tools/testing/selftests/arm64/fp/fp-ptrace.c  | 126 ++++++++++++++++++
 tools/testing/selftests/arm64/fp/fp-ptrace.h  |   2 +
 tools/testing/selftests/arm64/fp/sme-inst.h   |   2 +
 4 files changed, 146 insertions(+), 7 deletions(-)

diff --git a/tools/testing/selftests/arm64/fp/fp-ptrace-asm.S b/tools/testing/selftests/arm64/fp/fp-ptrace-asm.S
index 5e7e9c878f2c..82c3ab70e1cf 100644
--- a/tools/testing/selftests/arm64/fp/fp-ptrace-asm.S
+++ b/tools/testing/selftests/arm64/fp/fp-ptrace-asm.S
@@ -71,14 +71,12 @@ check_sm_in:
 	tbz	x7, #SVCR_SM_SHIFT, check_sve_in
 
 	// Load FFR if we have FA64
-	mov	x4, #0
-	tbz	x0, #HAVE_FA64_SHIFT, load_sve
-	mov	x4, #1
+	ubfx	x4, x0, #HAVE_FA64_SHIFT, #1
 	b	load_sve
 
 	// SVE?
 check_sve_in:
-	tbz	x0, #HAVE_SVE_SHIFT, wait_for_writes
+	tbz	x0, #HAVE_SVE_SHIFT, check_fpmr_in
 	mov	x4, #1
 
 load_sve:
@@ -143,6 +141,13 @@ load_sve:
 	ldr	p14, [x7, #14, MUL VL]
 	ldr	p15, [x7, #15, MUL VL]
 
+	// This has to come after we set PSTATE.SM
+check_fpmr_in:
+	tbz	x0, #HAVE_FPMR_SHIFT, wait_for_writes
+	adrp	x7, fpmr_in
+	ldr	x7, [x7, :lo12:fpmr_in]
+	msr	REG_FPMR, x7
+
 wait_for_writes:
 	// Wait for the parent
 	brk #0
@@ -166,6 +171,12 @@ wait_for_writes:
 	stp	q28, q29, [x7, #16 * 28]
 	stp	q30, q31, [x7, #16 * 30]
 
+	tbz	x0, #HAVE_FPMR_SHIFT, check_sme_out
+	mrs	x7, REG_FPMR
+	adrp	x6, fpmr_out
+	str	x7, [x6, :lo12:fpmr_out]
+
+check_sme_out:
 	tbz	x0, #HAVE_SME_SHIFT, check_sve_out
 
 	rdsvl	11, 1
@@ -197,9 +208,7 @@ check_sm_out:
 	tbz	x7, #SVCR_SM_SHIFT, check_sve_out
 
 	// Do we have FA64 and FFR?
-	mov	x4, #0
-	tbz	x0, #HAVE_FA64_SHIFT, read_sve
-	mov	x4, #1
+	ubfx	x4, x0, #HAVE_FA64_SHIFT, #1
 	b	read_sve
 
 	// SVE?
diff --git a/tools/testing/selftests/arm64/fp/fp-ptrace.c b/tools/testing/selftests/arm64/fp/fp-ptrace.c
index 56cf6e02c535..4930e03a7b99 100644
--- a/tools/testing/selftests/arm64/fp/fp-ptrace.c
+++ b/tools/testing/selftests/arm64/fp/fp-ptrace.c
@@ -31,6 +31,14 @@
 
 #include "fp-ptrace.h"
 
+#include <linux/bits.h>
+
+#define FPMR_LSCALE2_MASK                               GENMASK(37, 32)
+#define FPMR_NSCALE_MASK                                GENMASK(31, 24)
+#define FPMR_LSCALE_MASK                                GENMASK(22, 16)
+#define FPMR_OSC_MASK                                   GENMASK(15, 15)
+#define FPMR_OSM_MASK                                   GENMASK(14, 14)
+
 /* <linux/elf.h> and <sys/auxv.h> don't like each other, so: */
 #ifndef NT_ARM_SVE
 #define NT_ARM_SVE 0x405
@@ -48,11 +56,22 @@
 #define NT_ARM_ZT 0x40d
 #endif
 
+#ifndef NT_ARM_FPMR
+#define NT_ARM_FPMR 0x40e
+#endif
+
 #define ARCH_VQ_MAX 256
 
 /* VL 128..2048 in powers of 2 */
 #define MAX_NUM_VLS 5
 
+/*
+ * FPMR bits we can set without doing feature checks to see if values
+ * are valid.
+ */
+#define FPMR_SAFE_BITS (FPMR_LSCALE2_MASK | FPMR_NSCALE_MASK | \
+			FPMR_LSCALE_MASK | FPMR_OSC_MASK | FPMR_OSM_MASK)
+
 #define NUM_FPR 32
 __uint128_t v_in[NUM_FPR];
 __uint128_t v_expected[NUM_FPR];
@@ -78,6 +97,8 @@ char zt_in[ZT_SIG_REG_BYTES];
 char zt_expected[ZT_SIG_REG_BYTES];
 char zt_out[ZT_SIG_REG_BYTES];
 
+uint64_t fpmr_in, fpmr_expected, fpmr_out;
+
 uint64_t sve_vl_out;
 uint64_t sme_vl_out;
 uint64_t svcr_in, svcr_expected, svcr_out;
@@ -128,6 +149,11 @@ static bool fa64_supported(void)
 	return getauxval(AT_HWCAP2) & HWCAP2_SME_FA64;
 }
 
+static bool fpmr_supported(void)
+{
+	return getauxval(AT_HWCAP2) & HWCAP2_FPMR;
+}
+
 static bool compare_buffer(const char *name, void *out,
 			   void *expected, size_t size)
 {
@@ -233,6 +259,8 @@ static void run_child(struct test_config *config)
 		flags |= HAVE_SME2;
 	if (fa64_supported())
 		flags |= HAVE_FA64;
+	if (fpmr_supported())
+		flags |= HAVE_FPMR;
 
 	load_and_save(flags);
 
@@ -321,6 +349,14 @@ static void read_child_regs(pid_t child)
 		iov_child.iov_len = sizeof(zt_out);
 		read_one_child_regs(child, "ZT", &iov_parent, &iov_child);
 	}
+
+	if (fpmr_supported()) {
+		iov_parent.iov_base = &fpmr_out;
+		iov_parent.iov_len = sizeof(fpmr_out);
+		iov_child.iov_base = &fpmr_out;
+		iov_child.iov_len = sizeof(fpmr_out);
+		read_one_child_regs(child, "FPMR", &iov_parent, &iov_child);
+	}
 }
 
 static bool continue_breakpoint(pid_t child,
@@ -595,6 +631,26 @@ static bool check_ptrace_values_zt(pid_t child, struct test_config *config)
 	return compare_buffer("initial ZT", buf, zt_in, ZT_SIG_REG_BYTES);
 }
 
+static bool check_ptrace_values_fpmr(pid_t child, struct test_config *config)
+{
+	uint64_t val;
+	struct iovec iov;
+	int ret;
+
+	if (!fpmr_supported())
+		return true;
+
+	iov.iov_base = &val;
+	iov.iov_len = sizeof(val);
+	ret = ptrace(PTRACE_GETREGSET, child, NT_ARM_FPMR, &iov);
+	if (ret != 0) {
+		ksft_print_msg("Failed to read initial FPMR: %s (%d)\n",
+			       strerror(errno), errno);
+		return false;
+	}
+
+	return compare_buffer("initial FPMR", &val, &fpmr_in, sizeof(val));
+}
 
 static bool check_ptrace_values(pid_t child, struct test_config *config)
 {
@@ -629,6 +685,9 @@ static bool check_ptrace_values(pid_t child, struct test_config *config)
 	if (!check_ptrace_values_zt(child, config))
 		pass = false;
 
+	if (!check_ptrace_values_fpmr(child, config))
+		pass = false;
+
 	return pass;
 }
 
@@ -832,11 +891,18 @@ static void set_initial_values(struct test_config *config)
 {
 	int vq = __sve_vq_from_vl(vl_in(config));
 	int sme_vq = __sve_vq_from_vl(config->sme_vl_in);
+	bool sm_change;
 
 	svcr_in = config->svcr_in;
 	svcr_expected = config->svcr_expected;
 	svcr_out = 0;
 
+	if (sme_supported() &&
+	    (svcr_in & SVCR_SM) != (svcr_expected & SVCR_SM))
+		sm_change = true;
+	else
+		sm_change = false;
+
 	fill_random(&v_in, sizeof(v_in));
 	memcpy(v_expected, v_in, sizeof(v_in));
 	memset(v_out, 0, sizeof(v_out));
@@ -883,6 +949,21 @@ static void set_initial_values(struct test_config *config)
 			memset(zt_expected, 0, ZT_SIG_REG_BYTES);
 		memset(zt_out, 0, sizeof(zt_out));
 	}
+
+	if (fpmr_supported()) {
+		fill_random(&fpmr_in, sizeof(fpmr_in));
+		fpmr_in &= FPMR_SAFE_BITS;
+
+		/* Entering or exiting streaming mode clears FPMR */
+		if (sm_change)
+			fpmr_expected = 0;
+		else
+			fpmr_expected = fpmr_in;
+	} else {
+		fpmr_in = 0;
+		fpmr_expected = 0;
+		fpmr_out = 0;
+	}
 }
 
 static bool check_memory_values(struct test_config *config)
@@ -933,6 +1014,12 @@ static bool check_memory_values(struct test_config *config)
 	if (!compare_buffer("saved ZT", zt_out, zt_expected, ZT_SIG_REG_BYTES))
 		pass = false;
 
+	if (fpmr_out != fpmr_expected) {
+		ksft_print_msg("Mismatch in saved FPMR: %lx != %lx\n",
+			       fpmr_out, fpmr_expected);
+		pass = false;
+	}
+
 	return pass;
 }
 
@@ -1010,6 +1097,36 @@ static void fpsimd_write(pid_t child, struct test_config *test_config)
 			       strerror(errno), errno);
 }
 
+static bool fpmr_write_supported(struct test_config *config)
+{
+	if (!fpmr_supported())
+		return false;
+
+	if (!sve_sme_same(config))
+		return false;
+
+	return true;
+}
+
+static void fpmr_write_expected(struct test_config *config)
+{
+	fill_random(&fpmr_expected, sizeof(fpmr_expected));
+	fpmr_expected &= FPMR_SAFE_BITS;
+}
+
+static void fpmr_write(pid_t child, struct test_config *config)
+{
+	struct iovec iov;
+	int ret;
+
+	iov.iov_len = sizeof(fpmr_expected);
+	iov.iov_base = &fpmr_expected;
+	ret = ptrace(PTRACE_SETREGSET, child, NT_ARM_FPMR, &iov);
+	if (ret != 0)
+		ksft_print_msg("Failed to write FPMR: %s (%d)\n",
+			       strerror(errno), errno);
+}
+
 static void sve_write_expected(struct test_config *config)
 {
 	int vl = vl_expected(config);
@@ -1266,6 +1383,12 @@ static struct test_definition base_test_defs[] = {
 		.set_expected_values = fpsimd_write_expected,
 		.modify_values = fpsimd_write,
 	},
+	{
+		.name = "FPMR write",
+		.supported = fpmr_write_supported,
+		.set_expected_values = fpmr_write_expected,
+		.modify_values = fpmr_write,
+	},
 };
 
 static struct test_definition sve_test_defs[] = {
@@ -1475,6 +1598,9 @@ int main(void)
 	if (fa64_supported())
 		ksft_print_msg("FA64 supported\n");
 
+	if (fpmr_supported())
+		ksft_print_msg("FPMR supported\n");
+
 	ksft_set_plan(tests);
 
 	/* Get signal handers ready before we start any children */
diff --git a/tools/testing/selftests/arm64/fp/fp-ptrace.h b/tools/testing/selftests/arm64/fp/fp-ptrace.h
index 36ca627e1980..c06919aaf1f7 100644
--- a/tools/testing/selftests/arm64/fp/fp-ptrace.h
+++ b/tools/testing/selftests/arm64/fp/fp-ptrace.h
@@ -14,10 +14,12 @@
 #define HAVE_SME_SHIFT		1
 #define HAVE_SME2_SHIFT		2
 #define HAVE_FA64_SHIFT		3
+#define HAVE_FPMR_SHIFT		4
 
 #define HAVE_SVE	(1 << HAVE_SVE_SHIFT)
 #define HAVE_SME	(1 << HAVE_SME_SHIFT)
 #define HAVE_SME2	(1 << HAVE_SME2_SHIFT)
 #define HAVE_FA64	(1 << HAVE_FA64_SHIFT)
+#define HAVE_FPMR	(1 << HAVE_FPMR_SHIFT)
 
 #endif
diff --git a/tools/testing/selftests/arm64/fp/sme-inst.h b/tools/testing/selftests/arm64/fp/sme-inst.h
index 9292bba5400b..85b9184e0835 100644
--- a/tools/testing/selftests/arm64/fp/sme-inst.h
+++ b/tools/testing/selftests/arm64/fp/sme-inst.h
@@ -5,6 +5,8 @@
 #ifndef SME_INST_H
 #define SME_INST_H
 
+#define REG_FPMR                                        S3_3_C4_C4_2
+
 /*
  * RDSVL X\nx, #\imm
  */
-- 
2.50.1