Skip to contentSkip to conversation

ARM: advertise availability of v8 Crypto instructions

PATCH New Ard Biesheuvel 2 participants 2 messages Mar 10, 2015 – Mar 18, 2015 Arm (32-bit)

lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 ARM: advertise availability of v8 Crypto instructions

When running the 32-bit ARM kernel on ARMv8 capable bare metal (e.g., 32-bit Android userland and kernel on a Cortex-A53), or as a KVM guest on a 64-bit host, we should advertise the availability of the Crypto instructions, so that userland libraries such as OpenSSL may use them. (Support for the v8 Crypto instructions in the 32-bit build was added to OpenSSL more than six months ago)

This adds the ID feature bit detection, and sets elf_hwcap2 accordingly.

---
v3:
- reuse logic from arm64 kernel: the feature blocks are signed quantities that represent incremental functionality for non-negative values.
v2:
- drop redundant architecture check -> accessing ID_ISAR5 should be safe even on v7-M, and even if we don't expect to find crypto features there

 arch/arm/kernel/setup.c | 33 ++++++++++++++++++++++++++++++++-
 1 file changed, 32 insertions(+), 1 deletion(-)
arch/arm/kernel/setup.c +32 -1
index e55408e96559..fe07efca2f12 100644
@@ -375,7 +375,7 @@ void __init early_print(const char *str, ...)
 
 static void __init cpuid_init_hwcaps(void)
 {
-	unsigned int divide_instrs, vmsa;
+	unsigned int divide_instrs, vmsa, features, block;
 
 	if (cpu_architecture() < CPU_ARCH_ARMv7)
 		return;
@@ -393,6 +393,37 @@ static void __init cpuid_init_hwcaps(void)
 	vmsa = (read_cpuid_ext(CPUID_EXT_MMFR0) & 0xf) >> 0;
 	if (vmsa >= 5)
 		elf_hwcap |= HWCAP_LPAE;
+
+	/*
+	 * ID_ISAR5 contains 4-bit wide signed feature blocks.
+	 * The blocks we test below represent incremental functionality
+	 * for non-negative values. Negative values are reserved.
+	 */
+	features = read_cpuid_ext(ID_ISAR5);
+	block = (features >> 4) & 0xf;
+	if (!(block & 0x8)) {
+		switch (block) {
+		default:
+		case 2:
+			elf_hwcap2 |= HWCAP2_PMULL;
+		case 1:
+			elf_hwcap2 |= HWCAP2_AES;
+		case 0:
+			break;
+		}
+	}
+
+	block = (features >> 8) & 0xf;
+	if (block && !(block & 0x8))
+		elf_hwcap2 |= HWCAP2_SHA1;
+
+	block = (features >> 12) & 0xf;
+	if (block && !(block & 0x8))
+		elf_hwcap2 |= HWCAP2_SHA2;
+
+	block = (features >> 16) & 0xf;
+	if (block && !(block & 0x8))
+		elf_hwcap2 |= HWCAP2_CRC32;
 }
 
 static void __init elf_hwcap_fixup(void)
Signature · 1 line
1.8.3.2
#1 lore raw

PATCH v3 1/1 ARM: advertise availability of v8 Crypto instructions

arch/arm/kernel/setup.c static void __init cpuid_init_hwcaps(void) · line 403

Quoted 12 lines · Ard Biesheuvel
arch/arm/kernel/setup.c +8 -0
@@ -393,6 +393,37 @@ static void __init cpuid_init_hwcaps(void)
 	vmsa = (read_cpuid_ext(CPUID_EXT_MMFR0) & 0xf) >> 0;
 	if (vmsa >= 5)
 		elf_hwcap |= HWCAP_LPAE;
+
+	/*
+	 * ID_ISAR5 contains 4-bit wide signed feature blocks.
+	 * The blocks we test below represent incremental functionality
+	 * for non-negative values. Negative values are reserved.
+	 */
+	features = read_cpuid_ext(ID_ISAR5);
+	block = (features >> 4) & 0xf;

Okay, what I'd like to see is something like:

	block = cpuid_feature_extract_field(features, 4);

with this in cputype.h:

static inline int cpuid_feature_extract_field(u32 features, unsigned int field)

{
	int feature = (features >> field) & 15;

	/* feature registers are signed values */
	if (feature > 8)
		feature -= 16;

	return feature;
}

We can then use proper signed tests in the code - and convert all tests to use this helper.

Signature · 2 lines
FTTC broadband for 0.8mile line: currently at 10.5Mbps down 400kbps up
according to speedtest.net.