Properly handle psrldq when optimizing for Atom.

gcc/

2010-06-21  H.J. Lu  <hongjiu.lu@intel.com>

	PR target/44615
	* config/i386/atom.md (atom_sseishft_2): Also check sseishft1.

	* config/i386/i386.md (type): Add sseishft1

	* config/i386/ppro_insn (ppro_insn): Also check sseishft1.
	(ppro_insn_load): Likewise.
	(ppro_insn_store): Likewise.
	(ppro_insn_both): Likewise.

	* config/i386/sse.md (sse2_lshrv1ti3): Add atom_unit.
	(*vec_extractv2di_1_rex64_avx): Replace sseishft with sseishft1
	for type.
	(*vec_extractv2di_1_avx): Likewise.
	(*vec_extractv2di_1_rex64): Replace sseishft with sseishft1 for
	type.  Remove atom_unit.
	(*vec_extractv2di_1_sse2): Likewise.

gcc/testsuite/

2010-06-21  H.J. Lu  <hongjiu.lu@intel.com>

	PR target/44615
	* gcc.target/i386/sse2-vec-2a.c: New.

git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@161118 138bc75d-0d04-0410-961f-82ee72b054a4

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 85041a1..1359a3b 100644 (file)
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,23 @@
+2010-06-21  H.J. Lu  <hongjiu.lu@intel.com>
+
+       PR target/44615
+       * config/i386/atom.md (atom_sseishft_2): Also check sseishft1.
+
+       * config/i386/i386.md (type): Add sseishft1
+
+       * config/i386/ppro_insn (ppro_insn): Also check sseishft1.
+       (ppro_insn_load): Likewise.
+       (ppro_insn_store): Likewise.
+       (ppro_insn_both): Likewise.
+
+       * config/i386/sse.md (sse2_lshrv1ti3): Add atom_unit.
+       (*vec_extractv2di_1_rex64_avx): Replace sseishft with sseishft1
+       for type.
+       (*vec_extractv2di_1_avx): Likewise.
+       (*vec_extractv2di_1_rex64): Replace sseishft with sseishft1 for
+       type.  Remove atom_unit.
+       (*vec_extractv2di_1_sse2): Likewise.
+
 2010-06-21  DJ Delorie  <dj@redhat.com>
 
        * diagnostic.h (diagnostic_classification_change_t): New.

diff --git a/gcc/config/i386/atom.md b/gcc/config/i386/atom.md
index a9c4c5d..1664269 100644 (file)
--- a/gcc/config/i386/atom.md
+++ b/gcc/config/i386/atom.md
@@ -501,9 +501,10 @@
 ;; if palignr or psrldq
 (define_insn_reservation  "atom_sseishft_2" 1
   (and (eq_attr "cpu" "atom")
-       (and (eq_attr "type" "sseishft")
-            (and (eq_attr "atom_unit" "sishuf")
-                 (match_operand 2 "immediate_operand"))))
+       (ior (eq_attr "type" "sseishft1")
+           (and (eq_attr "type" "sseishft")
+                (and (eq_attr "atom_unit" "sishuf")
+                     (match_operand 2 "immediate_operand")))))
   "atom-simple-0")
 
 ;; if reg/mem op

diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md
index 777f8e7..4894669 100644 (file)
--- a/gcc/config/i386/i386.md
+++ b/gcc/config/i386/i386.md
@@ -357,7 +357,7 @@
    push,pop,call,callv,leave,
    str,bitmanip,
    fmov,fop,fsgn,fmul,fdiv,fpspc,fcmov,fcmp,fxch,fistp,fisttp,frndint,
-   sselog,sselog1,sseiadd,sseiadd1,sseishft,sseimul,
+   sselog,sselog1,sseiadd,sseiadd1,sseishft,sseishft1,sseimul,
    sse,ssemov,sseadd,ssemul,ssecmp,ssecomi,ssecvt,ssecvt1,sseicvt,ssediv,sseins,
    ssemuladd,sse4arg,lwp,
    mmx,mmxmov,mmxadd,mmxmul,mmxcmp,mmxcvt,mmxshft"

diff --git a/gcc/config/i386/ppro.md b/gcc/config/i386/ppro.md
index 5e163d8..20f457a 100644 (file)
--- a/gcc/config/i386/ppro.md
+++ b/gcc/config/i386/ppro.md
@@ -731,7 +731,7 @@
 (define_insn_reservation "ppro_insn" 1
                         (and (eq_attr "cpu" "pentiumpro")
                              (and (eq_attr "memory" "none,unknown")
-                                  (eq_attr "type" "alu,alu1,negnot,incdec,icmp,test,setcc,icmov,push,pop,fxch,sseiadd,sseishft,sseimul,mmx,mmxadd,mmxcmp")))
+                                  (eq_attr "type" "alu,alu1,negnot,incdec,icmp,test,setcc,icmov,push,pop,fxch,sseiadd,sseishft,sseishft1,sseimul,mmx,mmxadd,mmxcmp")))
                         "decodern,(p0|p1)")
 
 ;; read-modify and register-memory instructions have 2 or three uops,
@@ -739,13 +739,13 @@
 (define_insn_reservation "ppro_insn_load" 3
                         (and (eq_attr "cpu" "pentiumpro")
                              (and (eq_attr "memory" "load")
-                                  (eq_attr "type" "alu,alu1,negnot,incdec,icmp,test,setcc,icmov,push,pop,fxch,sseiadd,sseishft,sseimul,mmx,mmxadd,mmxcmp")))
+                                  (eq_attr "type" "alu,alu1,negnot,incdec,icmp,test,setcc,icmov,push,pop,fxch,sseiadd,sseishft,sseishft1,sseimul,mmx,mmxadd,mmxcmp")))
                         "decoder0,p2+(p0|p1)")
 
 (define_insn_reservation "ppro_insn_store" 1
                         (and (eq_attr "cpu" "pentiumpro")
                              (and (eq_attr "memory" "store")
-                                  (eq_attr "type" "alu,alu1,negnot,incdec,icmp,test,setcc,icmov,push,pop,fxch,sseiadd,sseishft,sseimul,mmx,mmxadd,mmxcmp")))
+                                  (eq_attr "type" "alu,alu1,negnot,incdec,icmp,test,setcc,icmov,push,pop,fxch,sseiadd,sseishft,sseishft1,sseimul,mmx,mmxadd,mmxcmp")))
                         "decoder0,(p0|p1),p4+p3")
 
 ;; read-modify-store instructions produce 4 uops so they have to be
@@ -753,6 +753,6 @@
 (define_insn_reservation "ppro_insn_both" 4
                         (and (eq_attr "cpu" "pentiumpro")
                              (and (eq_attr "memory" "both")
-                                  (eq_attr "type" "alu,alu1,negnot,incdec,icmp,test,setcc,icmov,push,pop,fxch,sseiadd,sseishft,sseimul,mmx,mmxadd,mmxcmp")))
+                                  (eq_attr "type" "alu,alu1,negnot,incdec,icmp,test,setcc,icmov,push,pop,fxch,sseiadd,sseishft,sseishft1,sseimul,mmx,mmxadd,mmxcmp")))
                         "decoder0,p2+(p0|p1),p4+p3")
 

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index b4046b8..1f8770f 100644 (file)
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -5967,6 +5967,7 @@
   [(set_attr "type" "sseishft")
    (set_attr "prefix_data16" "1")
    (set_attr "length_immediate" "1")
+   (set_attr "atom_unit" "sishuf")
    (set_attr "mode" "TI")])
 
 (define_insn "lshr<mode>3"
@@ -7397,7 +7398,7 @@
    vpsrldq\t{$8, %1, %0|%0, %1, 8}
    vmovq\t{%H1, %0|%0, %H1}
    vmov{q}\t{%H1, %0|%0, %H1}"
-  [(set_attr "type" "ssemov,sseishft,ssemov,imov")
+  [(set_attr "type" "ssemov,sseishft1,ssemov,imov")
    (set_attr "length_immediate" "*,1,*,*")
    (set_attr "memory" "*,none,*,*")
    (set_attr "prefix" "vex")
@@ -7414,9 +7415,8 @@
    psrldq\t{$8, %0|%0, 8}
    movq\t{%H1, %0|%0, %H1}
    mov{q}\t{%H1, %0|%0, %H1}"
-  [(set_attr "type" "ssemov,sseishft,ssemov,imov")
+  [(set_attr "type" "ssemov,sseishft1,ssemov,imov")
    (set_attr "length_immediate" "*,1,*,*")
-   (set_attr "atom_unit" "*,sishuf,*,*")
    (set_attr "memory" "*,none,*,*")
    (set_attr "mode" "V2SF,TI,TI,DI")])
 
@@ -7432,7 +7432,7 @@
    vmovhps\t{%1, %0|%0, %1}
    vpsrldq\t{$8, %1, %0|%0, %1, 8}
    vmovq\t{%H1, %0|%0, %H1}"
-  [(set_attr "type" "ssemov,sseishft,ssemov")
+  [(set_attr "type" "ssemov,sseishft1,ssemov")
    (set_attr "length_immediate" "*,1,*")
    (set_attr "memory" "*,none,*")
    (set_attr "prefix" "vex")
@@ -7449,9 +7449,8 @@
    movhps\t{%1, %0|%0, %1}
    psrldq\t{$8, %0|%0, 8}
    movq\t{%H1, %0|%0, %H1}"
-  [(set_attr "type" "ssemov,sseishft,ssemov")
+  [(set_attr "type" "ssemov,sseishft1,ssemov")
    (set_attr "length_immediate" "*,1,*")
-   (set_attr "atom_unit" "*,sishuf,*")
    (set_attr "memory" "*,none,*")
    (set_attr "mode" "V2SF,TI,TI")])
 

diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog
index 77228c1..f8f0372 100644 (file)
--- a/gcc/testsuite/ChangeLog
+++ b/gcc/testsuite/ChangeLog
@@ -1,3 +1,8 @@
+2010-06-21  H.J. Lu  <hongjiu.lu@intel.com>
+
+       PR target/44615
+       * gcc.target/i386/sse2-vec-2a.c: New.
+
 2010-06-21  DJ Delorie  <dj@redhat.com>
 
        * gcc.dg/pragma-diag-1.c: New.

diff --git a/gcc/testsuite/gcc.target/i386/sse2-vec-2a.c b/gcc/testsuite/gcc.target/i386/sse2-vec-2a.c
new file mode 100644 (file)
index 0000000..f230f27
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-vec-2a.c
@@ -0,0 +1,5 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -msse2 -mtune=atom" } */
+/* { dg-require-effective-target sse2 } */
+
+#include "sse2-vec-2.c"