[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] [PATCH 2/6] x86/xsaves: enable xsaves/xrstors in xen



This patch uses xsaves/xrstors instead of xsaveopt/xrstor
when perform task switch in xen if the feature is supported
in hardware.

Please note that xsaves/xrstors only use compact format.

Signed-off-by: Shuai Ruan <shuai.ruan@xxxxxxxxx>
---
 xen/arch/x86/xstate.c        | 83 ++++++++++++++++++++++++++++----------------
 xen/include/asm-x86/xstate.h |  3 +-
 2 files changed, 55 insertions(+), 31 deletions(-)

diff --git a/xen/arch/x86/xstate.c b/xen/arch/x86/xstate.c
index e34eda3..ff67986 100644
--- a/xen/arch/x86/xstate.c
+++ b/xen/arch/x86/xstate.c
@@ -14,6 +14,7 @@
 #include <asm/xstate.h>
 #include <asm/asm_defns.h>
 
+#define XSTATE_COMPACTION_ENABLED  (1ULL << 63)
 static bool_t __read_mostly cpu_has_xsaveopt;
 static bool_t __read_mostly cpu_has_xsavec;
 bool_t __read_mostly cpu_has_xgetbv1;
@@ -102,7 +103,9 @@ void xsave(struct vcpu *v, uint64_t mask)
         typeof(ptr->fpu_sse.fip.sel) fcs = ptr->fpu_sse.fip.sel;
         typeof(ptr->fpu_sse.fdp.sel) fds = ptr->fpu_sse.fdp.sel;
 
-        if ( cpu_has_xsaveopt )
+        if ( cpu_has_xsaves )
+            xsaves(lmask, hmask, ptr);
+        else if ( cpu_has_xsaveopt )
         {
             /*
              * xsaveopt may not write the FPU portion even when the respective
@@ -155,7 +158,9 @@ void xsave(struct vcpu *v, uint64_t mask)
     }
     else
     {
-        if ( cpu_has_xsaveopt )
+        if ( cpu_has_xsaves )
+            xsaves(lmask, hmask, ptr);
+        else if ( cpu_has_xsaveopt )
             asm volatile ( ".byte 0x0f,0xae,0x37"
                            : "=m" (*ptr)
                            : "a" (lmask), "d" (hmask), "D" (ptr) );
@@ -198,36 +203,54 @@ void xrstor(struct vcpu *v, uint64_t mask)
     switch ( __builtin_expect(ptr->fpu_sse.x[FPU_WORD_SIZE_OFFSET], 8) )
     {
     default:
-        asm volatile ( "1: .byte 0x48,0x0f,0xae,0x2f\n"
-                       ".section .fixup,\"ax\"      \n"
-                       "2: mov %5,%%ecx             \n"
-                       "   xor %1,%1                \n"
-                       "   rep stosb                \n"
-                       "   lea %2,%0                \n"
-                       "   mov %3,%1                \n"
-                       "   jmp 1b                   \n"
-                       ".previous                   \n"
-                       _ASM_EXTABLE(1b, 2b)
-                       : "+&D" (ptr), "+&a" (lmask)
-                       : "m" (*ptr), "g" (lmask), "d" (hmask),
-                         "m" (xsave_cntxt_size)
-                       : "ecx" );
+        if ( cpu_has_xsaves )
+        {
+            if ( !(v->arch.xsave_area->xsave_hdr.xcomp_bv &
+                 XSTATE_COMPACTION_ENABLED) )
+                v->arch.xsave_area->xsave_hdr.xcomp_bv = get_xcr0() |
+                                    XSTATE_COMPACTION_ENABLED;
+            xrstors(lmask, hmask, ptr);
+        }
+        else
+            asm volatile ( "1: .byte 0x48,0x0f,0xae,0x2f\n"
+                           ".section .fixup,\"ax\"      \n"
+                           "2: mov %5,%%ecx             \n"
+                           "   xor %1,%1                \n"
+                           "   rep stosb                \n"
+                           "   lea %2,%0                \n"
+                           "   mov %3,%1                \n"
+                           "   jmp 1b                   \n"
+                           ".previous                   \n"
+                           _ASM_EXTABLE(1b, 2b)
+                           : "+&D" (ptr), "+&a" (lmask)
+                           : "m" (*ptr), "g" (lmask), "d" (hmask),
+                             "m" (xsave_cntxt_size)
+                           : "ecx" );
         break;
     case 4: case 2:
-        asm volatile ( "1: .byte 0x0f,0xae,0x2f\n"
-                       ".section .fixup,\"ax\" \n"
-                       "2: mov %5,%%ecx        \n"
-                       "   xor %1,%1           \n"
-                       "   rep stosb           \n"
-                       "   lea %2,%0           \n"
-                       "   mov %3,%1           \n"
-                       "   jmp 1b              \n"
-                       ".previous              \n"
-                       _ASM_EXTABLE(1b, 2b)
-                       : "+&D" (ptr), "+&a" (lmask)
-                       : "m" (*ptr), "g" (lmask), "d" (hmask),
-                         "m" (xsave_cntxt_size)
-                       : "ecx" );
+        if ( cpu_has_xsaves )
+        {
+            if ( !(v->arch.xsave_area->xsave_hdr.xcomp_bv &
+                 XSTATE_COMPACTION_ENABLED) )
+                v->arch.xsave_area->xsave_hdr.xcomp_bv = get_xcr0() |
+                                    XSTATE_COMPACTION_ENABLED;
+            xrstors(lmask, hmask, ptr);
+        }
+        else
+            asm volatile ( "1: .byte 0x48,0x0f,0xae,0x2f\n"
+                           ".section .fixup,\"ax\"      \n"
+                           "2: mov %5,%%ecx             \n"
+                           "   xor %1,%1                \n"
+                           "   rep stosb                \n"
+                           "   lea %2,%0                \n"
+                           "   mov %3,%1                \n"
+                           "   jmp 1b                   \n"
+                           ".previous                   \n"
+                           _ASM_EXTABLE(1b, 2b)
+                           : "+&D" (ptr), "+&a" (lmask)
+                           : "m" (*ptr), "g" (lmask), "d" (hmask),
+                             "m" (xsave_cntxt_size)
+                           : "ecx" );
         break;
     }
 }
diff --git a/xen/include/asm-x86/xstate.h b/xen/include/asm-x86/xstate.h
index 59c7156..d03d824 100644
--- a/xen/include/asm-x86/xstate.h
+++ b/xen/include/asm-x86/xstate.h
@@ -72,7 +72,8 @@ struct __packed __attribute__((aligned (64))) xsave_struct
 
     struct {
         u64 xstate_bv;
-        u64 reserved[7];
+        u64 xcomp_bv;
+        u64 reserved[6];
     } xsave_hdr;                             /* The 64-byte header */
 
     struct { char x[XSTATE_YMM_SIZE]; } ymm; /* YMM */
-- 
1.9.1


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.