[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-arm] [PATCH v3 11/15] target/arm/kvm64: Add kvm_arch_get/put_sve
From: |
Andrew Jones |
Subject: |
[Qemu-arm] [PATCH v3 11/15] target/arm/kvm64: Add kvm_arch_get/put_sve |
Date: |
Fri, 2 Aug 2019 14:25:36 +0200 |
These are the SVE equivalents to kvm_arch_get/put_fpsimd. Note, the
swabbing is different than it is for fpsmid because the vector format
is a little-endian stream of words.
Signed-off-by: Andrew Jones <address@hidden>
---
target/arm/kvm64.c | 150 +++++++++++++++++++++++++++++++++++++++++++--
1 file changed, 146 insertions(+), 4 deletions(-)
diff --git a/target/arm/kvm64.c b/target/arm/kvm64.c
index 0b004d5d3050..f5c99984f25f 100644
--- a/target/arm/kvm64.c
+++ b/target/arm/kvm64.c
@@ -671,11 +671,12 @@ int kvm_arch_destroy_vcpu(CPUState *cs)
bool kvm_arm_reg_syncs_via_cpreg_list(uint64_t regidx)
{
/* Return true if the regidx is a register we should synchronize
- * via the cpreg_tuples array (ie is not a core reg we sync by
- * hand in kvm_arch_get/put_registers())
+ * via the cpreg_tuples array (ie is not a core or sve reg that
+ * we sync by hand in kvm_arch_get/put_registers())
*/
switch (regidx & KVM_REG_ARM_COPROC_MASK) {
case KVM_REG_ARM_CORE:
+ case KVM_REG_ARM64_SVE:
return false;
default:
return true;
@@ -761,6 +762,85 @@ static int kvm_arch_put_fpsimd(CPUState *cs)
return 0;
}
+/*
+ * SVE registers are encoded in KVM's memory in an endianness-invariant format.
+ * The byte at offset i from the start of the in-memory representation contains
+ * the bits [(7 + 8 * i) : (8 * i)] of the register value. As this means the
+ * lowest offsets are stored in the lowest memory addresses, then that nearly
+ * matches QEMU's representation, which is to use an array of host-endian
+ * uint64_t's, where the lower offsets are at the lower indices. To complete
+ * the translation we just need to byte swap the uint64_t's on big-endian
hosts.
+ */
+#ifdef HOST_WORDS_BIGENDIAN
+static uint64_t *sve_bswap64(uint64_t *dst, uint64_t *src, int nr)
+{
+ int i;
+
+ for (i = 0; i < nr; ++i) {
+ dst[i] = bswap64(src[i]);
+ }
+
+ return dst;
+}
+#endif
+
+/*
+ * KVM SVE registers come in slices where ZREGs have a slice size of 2048 bits
+ * and PREGS and the FFR have a slice size of 256 bits. However we simply hard
+ * code the slice index to zero for now as it's unlikely we'll need more than
+ * one slice for quite some time.
+ */
+static int kvm_arch_put_sve(CPUState *cs)
+{
+ ARMCPU *cpu = ARM_CPU(cs);
+ CPUARMState *env = &cpu->env;
+#ifdef HOST_WORDS_BIGENDIAN
+ uint64_t tmp[ARM_MAX_VQ * 2];
+#endif
+ uint64_t *r;
+ struct kvm_one_reg reg;
+ int n, ret;
+
+ for (n = 0; n < KVM_ARM64_SVE_NUM_ZREGS; ++n) {
+ r = &env->vfp.zregs[n].d[0];
+#ifdef HOST_WORDS_BIGENDIAN
+ r = sve_bswap64(tmp, r, cpu->sve_max_vq * 2);
+#endif
+ reg.addr = (uintptr_t)r;
+ reg.id = KVM_REG_ARM64_SVE_ZREG(n, 0);
+ ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, ®);
+ if (ret) {
+ return ret;
+ }
+ }
+
+ for (n = 0; n < KVM_ARM64_SVE_NUM_PREGS; ++n) {
+ r = &env->vfp.pregs[n].p[0];
+#ifdef HOST_WORDS_BIGENDIAN
+ r = sve_bswap64(tmp, r, DIV_ROUND_UP(cpu->sve_max_vq, 8));
+#endif
+ reg.addr = (uintptr_t)r;
+ reg.id = KVM_REG_ARM64_SVE_PREG(n, 0);
+ ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, ®);
+ if (ret) {
+ return ret;
+ }
+ }
+
+ r = &env->vfp.pregs[FFR_PRED_NUM].p[0];
+#ifdef HOST_WORDS_BIGENDIAN
+ r = sve_bswap64(tmp, r, DIV_ROUND_UP(cpu->sve_max_vq, 8));
+#endif
+ reg.addr = (uintptr_t)r;
+ reg.id = KVM_REG_ARM64_SVE_FFR(0);
+ ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, ®);
+ if (ret) {
+ return ret;
+ }
+
+ return 0;
+}
+
int kvm_arch_put_registers(CPUState *cs, int level)
{
struct kvm_one_reg reg;
@@ -855,7 +935,11 @@ int kvm_arch_put_registers(CPUState *cs, int level)
}
}
- ret = kvm_arch_put_fpsimd(cs);
+ if (cpu_isar_feature(aa64_sve, cpu)) {
+ ret = kvm_arch_put_sve(cs);
+ } else {
+ ret = kvm_arch_put_fpsimd(cs);
+ }
if (ret) {
return ret;
}
@@ -918,6 +1002,60 @@ static int kvm_arch_get_fpsimd(CPUState *cs)
return 0;
}
+/*
+ * KVM SVE registers come in slices where ZREGs have a slice size of 2048 bits
+ * and PREGS and the FFR have a slice size of 256 bits. However we simply hard
+ * code the slice index to zero for now as it's unlikely we'll need more than
+ * one slice for quite some time.
+ */
+static int kvm_arch_get_sve(CPUState *cs)
+{
+ ARMCPU *cpu = ARM_CPU(cs);
+ CPUARMState *env = &cpu->env;
+ struct kvm_one_reg reg;
+ uint64_t *r;
+ int n, ret;
+
+ for (n = 0; n < KVM_ARM64_SVE_NUM_ZREGS; ++n) {
+ r = &env->vfp.zregs[n].d[0];
+ reg.addr = (uintptr_t)r;
+ reg.id = KVM_REG_ARM64_SVE_ZREG(n, 0);
+ ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, ®);
+ if (ret) {
+ return ret;
+ }
+#ifdef HOST_WORDS_BIGENDIAN
+ sve_bswap64(r, r, cpu->sve_max_vq * 2);
+#endif
+ }
+
+ for (n = 0; n < KVM_ARM64_SVE_NUM_PREGS; ++n) {
+ r = &env->vfp.pregs[n].p[0];
+ reg.addr = (uintptr_t)r;
+ reg.id = KVM_REG_ARM64_SVE_PREG(n, 0);
+ ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, ®);
+ if (ret) {
+ return ret;
+ }
+#ifdef HOST_WORDS_BIGENDIAN
+ sve_bswap64(r, r, DIV_ROUND_UP(cpu->sve_max_vq, 8));
+#endif
+ }
+
+ r = &env->vfp.pregs[FFR_PRED_NUM].p[0];
+ reg.addr = (uintptr_t)r;
+ reg.id = KVM_REG_ARM64_SVE_FFR(0);
+ ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, ®);
+ if (ret) {
+ return ret;
+ }
+#ifdef HOST_WORDS_BIGENDIAN
+ sve_bswap64(r, r, DIV_ROUND_UP(cpu->sve_max_vq, 8));
+#endif
+
+ return 0;
+}
+
int kvm_arch_get_registers(CPUState *cs)
{
struct kvm_one_reg reg;
@@ -1012,7 +1150,11 @@ int kvm_arch_get_registers(CPUState *cs)
env->spsr = env->banked_spsr[i];
}
- ret = kvm_arch_get_fpsimd(cs);
+ if (cpu_isar_feature(aa64_sve, cpu)) {
+ ret = kvm_arch_get_sve(cs);
+ } else {
+ ret = kvm_arch_get_fpsimd(cs);
+ }
if (ret) {
return ret;
}
--
2.20.1
- Re: [Qemu-arm] [Qemu-devel] [PATCH v3 03/15] target/arm/monitor: Introduce qmp_query_cpu_model_expansion, (continued)
[Qemu-arm] [PATCH v3 04/15] tests: arm: Introduce cpu feature tests, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH v3 05/15] target/arm/helper: zcr: Add build bug next to value range assumption, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH v3 06/15] target/arm/cpu: Use div-round-up to determine predicate register array size, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH v3 07/15] target/arm: Allow SVE to be disabled via a CPU property, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH v3 08/15] target/arm/cpu64: max cpu: Introduce sve<vl-bits> properties, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH v3 09/15] target/arm/kvm64: Fix error returns, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH v3 11/15] target/arm/kvm64: Add kvm_arch_get/put_sve,
Andrew Jones <=
[Qemu-arm] [PATCH v3 10/15] target/arm/kvm64: Move the get/put of fpsimd registers out, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH v3 12/15] target/arm/kvm64: max cpu: Enable SVE when available, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH v3 13/15] target/arm/kvm: scratch vcpu: Preserve input kvm_vcpu_init features, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH v3 15/15] target/arm/kvm: host cpu: Add support for sve<vl-bits> properties, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH v3 14/15] target/arm/cpu64: max cpu: Support sve properties with KVM, Andrew Jones, 2019/08/02
[Qemu-arm] [PATCH] HACK: Centralize sve property checks, Richard Henderson, 2019/08/09
Re: [Qemu-arm] [PATCH v3 00/15] target/arm/kvm: enable SVE in guests, Peter Maydell, 2019/08/15