On some cpus the overhead for virtualization instructions is in the same
range as a system call. Having to call multiple ioctls to get set registers
will make certain userspace handled exits more expensive than necessary.
Lets provide two sections in kvm_run to have a shared save area for
guest registers.
1. the first section is read-only, to handle registers that have side-effects
2. the second section is read/write, e.g. for general purpose registers.
We also provide two 64bit flags fields (architecture specific), that will
specify which parts of these fields are valid. Each bit will define that
a group of registers (like general purpose) or a single register is valid.
In that way we can extend and shrink the interface. (The structure definition
itself can only grow of course).
Signed-off-by: Christian Borntraeger <borntrae...@de.ibm.com>
---
Documentation/virtual/kvm/api.txt | 28 ++++++++++++++++++++++++++++
arch/ia64/include/asm/kvm.h | 7 +++++++
arch/powerpc/include/asm/kvm.h | 7 +++++++
arch/s390/include/asm/kvm.h | 6 ++++++
arch/x86/include/asm/kvm.h | 7 +++++++
include/linux/kvm.h | 20 ++++++++++++++++++++
6 files changed, 75 insertions(+), 0 deletions(-)
diff --git a/Documentation/virtual/kvm/api.txt
b/Documentation/virtual/kvm/api.txt
index da1f8fd..a149e22 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -1693,6 +1693,34 @@ developer registration required to access it).
/* Fix the size of the union. */
char padding[256];
};
+
+ /*
+ * Here are two fields that allow to access often used registers
+ * directly, to avoid the overhead of the ioctl system call. Each
+ * register can be extended or reduced by having flag bits specifying
+ * if a group of registers is valid.
+ */
+ __u64 kvm_valid_sync_ro_regs;
+ __u64 kvm_valid_sync_rw_regs;
+ union {
+ /* registers which can be only read */
+ struct kvm_sync_ro_regs sync_ro;
+ char padding[1024];
+ };
+ union {
+ /* read/write guest registers */
+ struct kvm_sync_rw_regs sync_rw;
+ char padding[1024];
+ };
+
+If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access
+certain guest registers without having to call SET/GET_*REGS. Thus we can
+avoid some system call overhead if userspace has to handle the exit.
+Userspace can query the validity of the structure by checking
+kvm_valid_sync_r[wo]_regs for specific bits. These bits are architecture
+specific and usually define the validity of a groups of registers. (e.g.
+one bit for general purpose registers)
+
};
6. Capabilities that can be enabled
diff --git a/arch/ia64/include/asm/kvm.h b/arch/ia64/include/asm/kvm.h
index bc90c75..bae1f1e 100644
--- a/arch/ia64/include/asm/kvm.h
+++ b/arch/ia64/include/asm/kvm.h
@@ -261,4 +261,11 @@ struct kvm_debug_exit_arch {
struct kvm_guest_debug_arch {
};
+/* definition of registers in kvm_run */
+struct kvm_sync_ro_regs {
+};
+
+struct kvm_sync_rw_regs {
+};
+
#endif
diff --git a/arch/powerpc/include/asm/kvm.h b/arch/powerpc/include/asm/kvm.h
index f7727d9..14b766b 100644
--- a/arch/powerpc/include/asm/kvm.h
+++ b/arch/powerpc/include/asm/kvm.h
@@ -265,6 +265,13 @@ struct kvm_debug_exit_arch {
struct kvm_guest_debug_arch {
};
+/* definition of registers in kvm_run */
+struct kvm_sync_ro_regs {
+};
+
+struct kvm_sync_rw_regs {
+};
+
#define KVM_REG_MASK 0x001f
#define KVM_REG_EXT_MASK 0xffe0
#define KVM_REG_GPR 0x0000
diff --git a/arch/s390/include/asm/kvm.h b/arch/s390/include/asm/kvm.h
index 82b32a1..dda27a0 100644
--- a/arch/s390/include/asm/kvm.h
+++ b/arch/s390/include/asm/kvm.h
@@ -41,4 +41,10 @@ struct kvm_debug_exit_arch {
struct kvm_guest_debug_arch {
};
+/* definition of registers in kvm_run */
+struct kvm_sync_ro_regs {
+};
+
+struct kvm_sync_rw_regs {
+};
#endif
diff --git a/arch/x86/include/asm/kvm.h b/arch/x86/include/asm/kvm.h
index 4d8dcbd..fb32847 100644
--- a/arch/x86/include/asm/kvm.h
+++ b/arch/x86/include/asm/kvm.h
@@ -321,4 +321,11 @@ struct kvm_xcrs {
__u64 padding[16];
};
+/* definition of registers in kvm_run */
+struct kvm_sync_ro_regs {
+};
+
+struct kvm_sync_rw_regs {
+};
+
#endif /* _ASM_X86_KVM_H */
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 68e67e5..9398672 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -273,6 +273,25 @@ struct kvm_run {
/* Fix the size of the union. */
char padding[256];
};
+
+ /*
+ * Here are two fields that allow to access often used registers
+ * directly, to avoid the overhead of the ioctl system call. Each
+ * register can be extended or reduced by having flag bits specifying
+ * if a group of registers is valid.
+ */
+ __u64 kvm_valid_sync_ro_regs;
+ __u64 kvm_valid_sync_rw_regs;
+ union {
+ /* registers which can be only read */
+ struct kvm_sync_ro_regs sync_ro;
+ char padding[1024];
+ };
+ union {
+ /* read/write guest registers */
+ struct kvm_sync_rw_regs sync_rw;
+ char padding[1024];
+ };
};
/* for KVM_REGISTER_COALESCED_MMIO / KVM_UNREGISTER_COALESCED_MMIO */
@@ -558,6 +577,7 @@ struct kvm_ppc_pvinfo {
#define KVM_CAP_PPC_PAPR 68
#define KVM_CAP_S390_GMAP 71
#define KVM_CAP_TSC_DEADLINE_TIMER 72
+#define KVM_CAP_SYNC_REGS 73
#ifdef KVM_CAP_IRQ_ROUTING
--
1.7.8.2
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
