]> git.ipfire.org Git - thirdparty/valgrind.git/log
thirdparty/valgrind.git
10 years agoMerge, from trunk, r2994
Julian Seward [Sun, 23 Nov 2014 17:32:07 +0000 (17:32 +0000)] 
Merge, from trunk, r2994
2994   fix stupid bug introduced in 2993

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3019

10 years agoMerge, from trunk, r2993
Julian Seward [Sun, 23 Nov 2014 17:31:10 +0000 (17:31 +0000)] 
Merge, from trunk, r2993
2993   arm64: implement ADDP etc

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3018

10 years agoMerge, from trunk, r2992.
Julian Seward [Sun, 23 Nov 2014 17:29:10 +0000 (17:29 +0000)] 
Merge, from trunk, r2992.
2992   arm64: Implement "fcvtpu w, s".  n-i-bz.

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3017

10 years agoMerge, from trunk, r2991.
Julian Seward [Sun, 23 Nov 2014 17:28:18 +0000 (17:28 +0000)] 
Merge, from trunk, r2991.
340856  disInstr(arm64): unhandled instruction 0x1E634C45 (fcsel)
2991

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3016

10 years agoMerge, from trunk, r2990.
Julian Seward [Sun, 23 Nov 2014 17:27:11 +0000 (17:27 +0000)] 
Merge, from trunk, r2990.
2990   Add detection of old ppc32 magic instructions from bug 278808.

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3015

10 years agoMerge, from trunk, r2988
Julian Seward [Sun, 23 Nov 2014 17:25:53 +0000 (17:25 +0000)] 
Merge, from trunk, r2988
340725  AVX2: Incorrect decoding of vpbroadcast{b,w} reg,reg forms
2988

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3014

10 years agoMerge, from trunk, r2987
Julian Seward [Sun, 23 Nov 2014 17:24:51 +0000 (17:24 +0000)] 
Merge, from trunk, r2987
340632  arm64: unhandled instruction fcvtas
2987

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3013

10 years agoMerge, from trunk, r2986
Julian Seward [Sun, 23 Nov 2014 17:23:24 +0000 (17:23 +0000)] 
Merge, from trunk, r2986
340033  arm64: unhandled instruction for dmb ishld and some other
        isb-dmb-dsb variants...
2986

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3012

10 years agoMerge, from trunk, r2985
Julian Seward [Sun, 23 Nov 2014 17:22:16 +0000 (17:22 +0000)] 
Merge, from trunk, r2985
335713  arm64: unhanded instruction: prfm (immediate)
2985

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3011

10 years agoMerge, from trunk, r2984
Julian Seward [Sun, 23 Nov 2014 17:21:12 +0000 (17:21 +0000)] 
Merge, from trunk, r2984
340509  arm64: unhandled instruction fcvtas
2984

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3010

10 years agoMerge, from trunk, r2983
Julian Seward [Sun, 23 Nov 2014 17:20:09 +0000 (17:20 +0000)] 
Merge, from trunk, r2983
339938  disInstr(arm64): unhandled instruction 0x4F8010A4 (fmla)
        == 339950
2983

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3009

10 years agoMerge, from trunk, r2982
Julian Seward [Sun, 23 Nov 2014 17:18:46 +0000 (17:18 +0000)] 
Merge, from trunk, r2982
339927  Unhandled instruction 0x9E7100C6 (fcvtmu) on aarch64
2982

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3008

10 years agoMerge, from trunk, r2981
Julian Seward [Sun, 23 Nov 2014 17:17:39 +0000 (17:17 +0000)] 
Merge, from trunk, r2981
339926  Unhandled instruction 0x1E674001 (frintx) on aarm64
2981

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3007

10 years agoMerge, from trunk, r2979
Julian Seward [Sun, 23 Nov 2014 12:52:05 +0000 (12:52 +0000)] 
Merge, from trunk, r2979
* add a missing extra m-reg check for some LD/ST vector cases

* implement
  LD1/ST1 (multiple 1-elem structs to/from 2 regs
  LD1/ST1 (multiple 1-elem structs to/from 3 regs
  LD1/ST1 (multiple 1-elem structs to/from 4 regs
  LD1R (single structure, replicate)
  LD2R (single structure, replicate)
  LD3R (single structure, replicate)
  LD4R (single structure, replicate)
  LD1/ST1 (single structure, to/from one lane)
  LD2/ST2 (single structure, to/from one lane)
  LD3/ST3 (single structure, to/from one lane)
  LD4/ST4 (single structure, to/from one lane)

I believe this completes the implementation of load and store
instructions for AArch64 ARMv8.

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3006

10 years agoMerge, from trunk, r2976
Julian Seward [Sun, 23 Nov 2014 12:49:14 +0000 (12:49 +0000)] 
Merge, from trunk, r2976
Implement SIMD (de)interleaving loads/stores:
  LD1/ST1 (multiple 1-elem structs to/from 1 reg
  LD2/ST2 (multiple 2-elem structs to/from 2 regs
  LD3/ST3 (multiple 3-elem structs to/from 3 regs
  LD4/ST4 (multiple 4-elem structs to/from 4 regs
Also:
  LDNP, STNP  (load/store vector pair, non-temporal)

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3005

10 years agoMerge, from trunk, r2975
Julian Seward [Sun, 23 Nov 2014 12:16:11 +0000 (12:16 +0000)] 
Merge, from trunk, r2975
339858  arm64 dmb sy not implemented
2975 (subsequently overwritten by 2986)

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3004

10 years agoMerge, from trunk, 2962, 2966, 2967, 2973
Julian Seward [Sun, 23 Nov 2014 12:14:41 +0000 (12:14 +0000)] 
Merge, from trunk, 2962, 2966, 2967, 2973
339433  ppc64 lxvw4x instruction uses four 32-byte loads

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3003

10 years agoMerge, from trunk, r2960
Julian Seward [Sun, 23 Nov 2014 11:57:49 +0000 (11:57 +0000)] 
Merge, from trunk, r2960
339182  ppc64: AvSplat ought to load destination vector register with 16/16
        bytes stored prior

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3002

10 years agoMerge, from trunk, r2957:
Julian Seward [Sun, 23 Nov 2014 11:56:09 +0000 (11:56 +0000)] 
Merge, from trunk, r2957:
339336  PPC64 store quad instruction (stq) is not supposed to
        change the address register contents

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@3001

10 years agoCreate VEX_3_10_BRANCH as a copy of trunk r2949.
Julian Seward [Mon, 8 Sep 2014 11:50:47 +0000 (11:50 +0000)] 
Create VEX_3_10_BRANCH as a copy of trunk r2949.

git-svn-id: svn://svn.valgrind.org/vex/branches/VEX_3_10_BRANCH@2950

10 years agoChange how FXSAVE and FXRSTOR are done, so as to avoid pushing the XMM
Julian Seward [Sun, 7 Sep 2014 23:23:17 +0000 (23:23 +0000)] 
Change how FXSAVE and FXRSTOR are done, so as to avoid pushing the XMM
register contents themselves through the helper functions.  This
avoids the false positives reported in #291310.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2949

10 years agoMinor refactoring to avoid special handling of emulation
Florian Krohm [Sat, 6 Sep 2014 21:43:28 +0000 (21:43 +0000)] 
Minor refactoring to avoid special handling of emulation
failures in s390_irgen_PFPO.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2948

10 years agoHandle fcvtpu Xd,Sn. Fixes #335564.
Julian Seward [Sat, 6 Sep 2014 08:08:47 +0000 (08:08 +0000)] 
Handle fcvtpu Xd,Sn.  Fixes #335564.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2947

10 years agoUse const instead of a comment.
Florian Krohm [Fri, 5 Sep 2014 21:52:29 +0000 (21:52 +0000)] 
Use const instead of a comment.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2946

10 years agoiselStmt, case Ist_Dirty: remove pointless conditional. Spotted by
Julian Seward [Fri, 5 Sep 2014 20:38:34 +0000 (20:38 +0000)] 
iselStmt, case Ist_Dirty: remove pointless conditional.  Spotted by
Coverity and Florian Krohm.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2945

10 years agomips64: fix jmpKind for BLTZ and BGEZ
Petar Jovanovic [Thu, 4 Sep 2014 17:34:56 +0000 (17:34 +0000)] 
mips64: fix jmpKind for BLTZ and BGEZ

As spotted by Maran Pakkirisamy, jump kind for BLTZ and BGEZ should be
Ijk_Boring and not Ijk_Call.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2944

10 years agoarm64: enable support for: str bN, [reg, reg etc]
Julian Seward [Thu, 4 Sep 2014 11:44:03 +0000 (11:44 +0000)] 
arm64: enable support for: str bN, [reg, reg etc]
and str hN, [reg, reg etc].  Fixes #337762.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2943

10 years agomips64: implement Cavium BBIT0 and BBIT1 instructions
Petar Jovanovic [Wed, 3 Sep 2014 14:39:56 +0000 (14:39 +0000)] 
mips64: implement Cavium BBIT0 and BBIT1 instructions

Implement Cavium specific instructions:
- BBIT0 (Branch on bit clear)
- BBIT1 (Branch on bit set)

This should fix the hang issue reported in:
https://bugs.kde.org/show_bug.cgi?id=336139

git-svn-id: svn://svn.valgrind.org/vex/trunk@2942

10 years agoAdd a missing return statement. Spotted by the Coverity checker.
Florian Krohm [Tue, 2 Sep 2014 14:54:39 +0000 (14:54 +0000)] 
Add a missing return statement. Spotted by the Coverity checker.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2941

10 years agoarm64: route all whole-vector shift/rotate/slice operations
Julian Seward [Mon, 1 Sep 2014 14:13:15 +0000 (14:13 +0000)] 
arm64: route all whole-vector shift/rotate/slice operations
through Iop_SliceV128, so as to give it some testing.  Implement
Iop_SliceV128 in the back end.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2940

10 years agoRename Iop_Extract{64,V128} to Iop_Slice{64,V128}, improve their
Julian Seward [Mon, 1 Sep 2014 11:32:47 +0000 (11:32 +0000)] 
Rename Iop_Extract{64,V128} to Iop_Slice{64,V128}, improve their
documentation, and swap the sense of the first and second args
so as to be more in keeping with the rest of the ops here, so
that the more significant arg is arg1 rather than arg2.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2939

10 years agoAdd support for generating ProfInc sequences on ARM64, so as to
Julian Seward [Fri, 29 Aug 2014 21:58:03 +0000 (21:58 +0000)] 
Add support for generating ProfInc sequences on ARM64, so as to
make it possible to profile vex-generated ARM64 code.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2938

10 years agomips: remove unused macro
Petar Jovanovic [Fri, 29 Aug 2014 17:34:22 +0000 (17:34 +0000)] 
mips: remove unused macro

Remove unused instance of macro ASM_VOLATILE_CASE(rd, sel).

git-svn-id: svn://svn.valgrind.org/vex/trunk@2937

10 years agomips: fix typo (IRType/IRTemp)
Petar Jovanovic [Fri, 29 Aug 2014 17:05:40 +0000 (17:05 +0000)] 
mips: fix typo (IRType/IRTemp)

Fix typo (IRType used instead of IRTemp).

git-svn-id: svn://svn.valgrind.org/vex/trunk@2936

10 years agoAdd support for four IROps that Memcheck generates on arm64, that
Julian Seward [Tue, 26 Aug 2014 18:30:48 +0000 (18:30 +0000)] 
Add support for four IROps that Memcheck generates on arm64, that
the front end doesn't generate.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2935

10 years agoComment-only change.
Julian Seward [Tue, 26 Aug 2014 18:29:26 +0000 (18:29 +0000)] 
Comment-only change.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2934

10 years agoarm64: implement:
Julian Seward [Sun, 24 Aug 2014 20:36:14 +0000 (20:36 +0000)] 
arm64: implement:
  {zip,uzp,trn}{1,2} (vector)
  urecpe, ursqrte (vector)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2933

10 years agoRename IROps for reciprocal estimate, reciprocal step, reciprocal sqrt
Julian Seward [Sun, 24 Aug 2014 14:00:19 +0000 (14:00 +0000)] 
Rename IROps for reciprocal estimate, reciprocal step, reciprocal sqrt
estimate and reciprocal sqrt step, to be more consistent.  Remove
64FxWhatever versions of those ops since they are never used.  As a
side effect, observe that RSqrt32Fx4 and Rsqrte32Fx4 are the same and
hence fix the duplication, at the same time.  No functional change.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2932

10 years agoBug 330319 - vex amd64->IR: unhandled instruction bytes: 0xF 0x1 0xD5 (xend)
Mark Wielaard [Thu, 21 Aug 2014 17:46:00 +0000 (17:46 +0000)] 
Bug 330319 - vex amd64->IR: unhandled instruction bytes: 0xF 0x1 0xD5 (xend)

We are never in an transaction (xbegin immediately aborts).
So this just always generates a General Protection Fault.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2931

10 years agoputGST_masked: correctly handle the case where the mask is for
Julian Seward [Wed, 20 Aug 2014 08:54:06 +0000 (08:54 +0000)] 
putGST_masked: correctly handle the case where the mask is for
FPSCR.RN or FPSCR.DRN, but does not cover the entire field.  Then it
is important to update the exposed parts but leave the not-exposed
parts unchanged.  This is a regression relative to circa 5 years ago.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2929

10 years agoarm64: implement:
Julian Seward [Mon, 18 Aug 2014 12:28:02 +0000 (12:28 +0000)] 
arm64: implement:
suqadd, usqadd (scalar)
suqadd, usqadd (vector)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2928

10 years agoarm64: implement srhadd, urhadd (vector)
Julian Seward [Sun, 17 Aug 2014 19:59:09 +0000 (19:59 +0000)] 
arm64: implement srhadd, urhadd (vector)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2927

10 years agoarm64: implement
Julian Seward [Sun, 17 Aug 2014 18:32:14 +0000 (18:32 +0000)] 
arm64: implement
  sshr, ushr, ssra, usra (scalar, imm)
  srshr, urshr, srsra, ursra (scalar, imm)
  srshr, urshr, srsra, ursra (vector, imm)
  sshl, srshl, ushl, urshl (scalar, imm)
  sshl, srshl, ushl, urshl (vector, vector)
  ssra, usra (vector, imm)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2926

10 years agoNo functional change. Remove commented out code copied from the
Julian Seward [Fri, 15 Aug 2014 09:29:36 +0000 (09:29 +0000)] 
No functional change.  Remove commented out code copied from the
arm32 port, which is never going to get used.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2925

10 years agoRename Iop_QSalN*, Iop_QShlN* and Iop_QShlN*S so as to more accurately
Julian Seward [Fri, 15 Aug 2014 09:11:08 +0000 (09:11 +0000)] 
Rename Iop_QSalN*, Iop_QShlN* and Iop_QShlN*S so as to more accurately
reflect what they actually do, which is a zero-fill shift left followed
by one of three flavours of saturation (S->S, U->U or S->U).

git-svn-id: svn://svn.valgrind.org/vex/trunk@2924

10 years agoarm64: implement: {uqshl, sqshl, sqshlu} (scalar, imm) and fix two
Julian Seward [Fri, 15 Aug 2014 05:35:35 +0000 (05:35 +0000)] 
arm64: implement: {uqshl, sqshl, sqshlu} (scalar, imm) and fix two
bugs in the implementation of the (vector, imm) variants.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2923

10 years agoarm64: implement: {uqshl, sqshl, sqshlu} (vector, imm).
Julian Seward [Thu, 14 Aug 2014 22:26:52 +0000 (22:26 +0000)] 
arm64: implement: {uqshl, sqshl, sqshlu} (vector, imm).

git-svn-id: svn://svn.valgrind.org/vex/trunk@2922

10 years agoAdd a simple folding rule for Iop_ZeroHI64ofV128.
Julian Seward [Thu, 14 Aug 2014 22:25:31 +0000 (22:25 +0000)] 
Add a simple folding rule for Iop_ZeroHI64ofV128.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2921

10 years agoarm64: implement: uqshrn{2}, sqrshrun{2}, sqshrun{2} (scalar, imm)
Julian Seward [Wed, 13 Aug 2014 13:10:47 +0000 (13:10 +0000)] 
arm64: implement: uqshrn{2}, sqrshrun{2}, sqshrun{2} (scalar, imm)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2920

10 years agoSmall cleanups in VEX:
Philippe Waroquiers [Mon, 11 Aug 2014 22:45:47 +0000 (22:45 +0000)] 
Small cleanups in VEX:
  * rm unused arm64 function
  * ijk_nodecode: always set the 4 components of the result
    (avoid a compiler warning that a part is not initialised)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2919

10 years agoarm64: implement:
Julian Seward [Mon, 11 Aug 2014 14:02:47 +0000 (14:02 +0000)] 
arm64: implement:
uqshrn{2}, sqrshrun{2}, sqshrun{2} (vector, imm)
sqxtn{2}, uqxtn{2}, sqxtun{2} (vector and scalar)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2918

10 years agoAdd a new folding rule:
Julian Seward [Mon, 11 Aug 2014 14:01:00 +0000 (14:01 +0000)] 
Add a new folding rule:
ZeroHI64ofV128( ZeroHI64ofV128(x) ) --> ZeroHI64ofV128(x)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2917

11 years agoUnbreak the build
Philippe Waroquiers [Sat, 9 Aug 2014 08:13:19 +0000 (08:13 +0000)] 
Unbreak the build
priv/guest_ppc_toIR.c: In function disInstr_PPC:
priv/guest_ppc_toIR.c:20160:7: error: dis undeclared (first use in this function)
       dis.continueAt   = 0;
       ^

git-svn-id: svn://svn.valgrind.org/vex/trunk@2916

11 years agoThis commit is for Bugzilla 334834.
Carl Love [Fri, 8 Aug 2014 22:26:31 +0000 (22:26 +0000)] 
This commit is for Bugzilla 334834.

The compiler is generating a message
"dres.continueAt’ may be used uninitialized in this function".

Setting dres.continueAT to zero since this is an error case
and we should not be continuing.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2915

11 years agoThis commit is for Bugzilla 334834. The Bugzilla contains patch 2 of 3
Carl Love [Thu, 7 Aug 2014 23:25:23 +0000 (23:25 +0000)] 
This commit is for Bugzilla 334834.  The Bugzilla contains patch 2 of 3
to add PPC64 LE support.  The other two patches can be found in Bugzillas
334384 and 334836.

POWER PC, add the functional Little Endian support, patch 2 VEX part

The IBM POWER processor now supports both Big Endian and Little Endian.
The ABI for Little Endian also changes.  Specifically, the function
descriptor is not used, the stack size changed, accessing the TOC
changed.  Functions now have a local and a global entry point.  Register
r2 contains the TOC for local calls and register r12 contains the TOC
for global calls.  This patch makes the functional changes to the
Valgrind tool.  The patch makes the changes needed for the
none/tests/ppc32 and none/tests/ppc64 Makefile.am.  A number of the
ppc specific tests have Endian dependencies that are not fixed in
this patch.  They are fixed in the next patch.

Per Julian's comments renamed coregrind/m_dispatch/dispatch-ppc64-linux.S
to coregrind/m_dispatch/dispatch-ppc64be-linux.S  Created new file for LE
coregrind/m_dispatch/dispatch-ppc64le-linux.S.  The same was done for
coregrind/m_syswrap/syscall-ppc-linux.S.

Signed-off-by: Carl Love <carll@us.ibm.com>
git-svn-id: svn://svn.valgrind.org/vex/trunk@2914

11 years agoarm64: add support for: sqshl, uqshl, sqrshl, uqrshl (reg) (vector and scalar)
Julian Seward [Mon, 4 Aug 2014 08:09:47 +0000 (08:09 +0000)] 
arm64: add support for: sqshl, uqshl, sqrshl, uqrshl (reg) (vector and scalar)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2913

11 years agoAdd a folding rule: XorV128(t,0) ==> t.
Julian Seward [Mon, 4 Aug 2014 08:09:23 +0000 (08:09 +0000)] 
Add a folding rule: XorV128(t,0) ==> t.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2912

11 years agoarm64: implement remaining SQDMULH and SQRDMULH cases.
Julian Seward [Sun, 3 Aug 2014 12:45:19 +0000 (12:45 +0000)] 
arm64: implement remaining SQDMULH and SQRDMULH cases.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2911

11 years agoImprove infrastructure for dealing with endianness in VEX. This patch
Julian Seward [Thu, 24 Jul 2014 12:42:03 +0000 (12:42 +0000)] 
Improve infrastructure for dealing with endianness in VEX.  This patch
removes all decisions about endianness from VEX.  Instead, it requires
that the LibVEX_* calls pass in information about the guest or host
endianness (depending on context) and in turn it passes that info
through to all the places that need it:

* the front ends (xx_toIR.c)
* the back ends (xx_isel.c)
* the patcher functions (Chain, UnChain, PatchProfInc)

Mostly it is boring and ugly plumbing.  As far as types go, there is a
new type "VexEndness" that carries the endianness.  This also makes it
possible to stop using Bools to indicate endianness.  VexArchInfo has
a new field of type VexEndness.  Apart from that, no other changes in
types.

Followups: MIPS front and back ends have not yet been fixed up to use
the passed-in endianness information.  Currently they assume that the
endianness of both host and guest is the same as the endianness of the
target for which VEX is being compiled.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2910

11 years agoarm64: implement:
Julian Seward [Tue, 22 Jul 2014 09:27:49 +0000 (09:27 +0000)] 
arm64: implement:
{sqdmlal,sqdmlsl,sqdmull}{d_s_s[],s_h_h[]}
{sqdmlal,sqdmlsl,sqdmull}{d_s_s,s_h_h}
{sqdmlal,sqdmlsl,sqdmull}{2d_(2s_2s)/(4s_4s), 4s_(4h_4h)/(8h_8h)}
sqrdmulh 4s,2s,8h,4h (vector)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2909

11 years agoComment-only change.
Julian Seward [Tue, 22 Jul 2014 09:26:36 +0000 (09:26 +0000)] 
Comment-only change.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2908

11 years agoarm64: implement: sqneg, {u,s}q{add,sub} (scalar),
Julian Seward [Mon, 21 Jul 2014 09:19:50 +0000 (09:19 +0000)] 
arm64: implement: sqneg, {u,s}q{add,sub} (scalar),
{sqdmlal,sqdmlsl,sqdmull} (vector x element)

As part of this, rename Iop_QDMulLong* to Iop_QDMull* so as to be
consistent with their non-saturating equivalents.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2907

11 years agoInitialise a couple of scalars that gcc -Og thinks might be
Julian Seward [Mon, 21 Jul 2014 07:55:45 +0000 (07:55 +0000)] 
Initialise a couple of scalars that gcc -Og thinks might be
uninitialised, presumably because at -Og it doesn't do enough
block straightening-outening or whatever to see that they are
always assigned before use.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2906

11 years agoAdd a few more algebraic optimisations for Iop_And8/16.
Florian Krohm [Fri, 18 Jul 2014 21:23:46 +0000 (21:23 +0000)] 
Add a few more algebraic optimisations for Iop_And8/16.
Observed on s390.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2905

11 years agoRemove fields from VexAbiInfo that only had relevance to the old AIX5
Julian Seward [Wed, 16 Jul 2014 23:14:33 +0000 (23:14 +0000)] 
Remove fields from VexAbiInfo that only had relevance to the old AIX5
port: guest_ppc_sc_continues_at_LR and host_ppc32_regalign_int64_args.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2904

11 years agoComment out an unsed function to avoid a compiler warning.
Florian Krohm [Wed, 16 Jul 2014 20:29:38 +0000 (20:29 +0000)] 
Comment out an unsed function to avoid a compiler warning.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2903

11 years agoFix algebraic simplification for Iop_AndV256.
Florian Krohm [Wed, 16 Jul 2014 20:17:49 +0000 (20:17 +0000)] 
Fix algebraic simplification for Iop_AndV256.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2902

11 years agoarm64: implement: LD1/ST1 (multi 1-elem structs, 2 regs, post index)
Julian Seward [Tue, 15 Jul 2014 11:08:42 +0000 (11:08 +0000)] 
arm64: implement: LD1/ST1 (multi 1-elem structs, 2 regs, post index)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2901

11 years agoarm64: implement "mrs Xt, cntvct_el0" by pass-through to the host.
Julian Seward [Mon, 14 Jul 2014 20:39:23 +0000 (20:39 +0000)] 
arm64: implement "mrs Xt, cntvct_el0" by pass-through to the host.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2900

11 years agoarm64: implement: {sli,sri} (vector & scalar), sqabs (vector & scalar)
Julian Seward [Fri, 11 Jul 2014 12:05:47 +0000 (12:05 +0000)] 
arm64: implement: {sli,sri} (vector & scalar), sqabs (vector & scalar)

Fix instruction decoding bug in dis_AdvSIMD_vector_x_indexed_elem
introduced in r2874 but not exposed until recently.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2899

11 years agoarm64: implement: shll #imm, shrn #imm, rshrn #imm,
Julian Seward [Thu, 10 Jul 2014 14:22:45 +0000 (14:22 +0000)] 
arm64: implement: shll #imm, shrn #imm, rshrn #imm,
{smlal,umlal,smlsl,umlsl,smull,umull} (elem)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2898

11 years agoarm32: support (ARM) PLDW [reg, reg]. The non-W variant was already
Julian Seward [Thu, 3 Jul 2014 12:27:21 +0000 (12:27 +0000)] 
arm32: support (ARM) PLDW [reg, reg].  The non-W variant was already
accepted.  Fixes #323178.  (vasily.golubev@gmail.com)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2897

11 years agoarm32: support (ARM) PLDW [reg, #imm]. The non-W variant was already
Julian Seward [Thu, 3 Jul 2014 11:01:38 +0000 (11:01 +0000)] 
arm32: support (ARM) PLDW [reg, #imm].  The non-W variant was already
accepted.  Fixes #323179.  (vasily.golubev@gmail.com)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2896

11 years agoarm64: implement: sadalp uadalp saddlp uaddlp saddlv uaddlv saddw{2}
Julian Seward [Mon, 30 Jun 2014 07:33:56 +0000 (07:33 +0000)] 
arm64: implement: sadalp uadalp saddlp uaddlp saddlv uaddlv saddw{2}
uaddw{2} ssubw{2} usubw{2} shadd uhadd shsub uhsub sqadd uqadd sqsub
uqsub smaxp umaxp sminp uminp

git-svn-id: svn://svn.valgrind.org/vex/trunk@2895

11 years agoarm64: change the representation of FPSR.QC so that it can be
Julian Seward [Sat, 28 Jun 2014 22:11:16 +0000 (22:11 +0000)] 
arm64: change the representation of FPSR.QC so that it can be
used efficiently to record SIMD saturation, and remove support
for all other bits of FPSR, since we don't model them anyway.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2894

11 years agoarm64: implement: sabal uabal sabdl uabdl saddl uaddl ssubl usubl
Julian Seward [Sat, 28 Jun 2014 12:21:37 +0000 (12:21 +0000)] 
arm64: implement: sabal uabal sabdl uabdl saddl uaddl ssubl usubl
smlal umlal smlsl umlsl smull umull

git-svn-id: svn://svn.valgrind.org/vex/trunk@2893

11 years agoarm64:
Julian Seward [Fri, 27 Jun 2014 10:43:22 +0000 (10:43 +0000)] 
arm64:
* implement: rev32, rev64, saba, uaba, sabd, uabd.
* factor out a large number of duplicated expressions of the form
  bitQ == 0 ? unop(Iop_ZeroHI64ofV128, mkexpr(t)) : mkexpr(t)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2892

11 years agoarm64: implement: rbit 16b,8b, rev16 16b,8b
Julian Seward [Thu, 26 Jun 2014 12:39:05 +0000 (12:39 +0000)] 
arm64: implement: rbit 16b,8b, rev16 16b,8b

git-svn-id: svn://svn.valgrind.org/vex/trunk@2891

11 years agoRename the vector subparts-of-lanes-reversal IROps to names
Julian Seward [Thu, 26 Jun 2014 10:49:33 +0000 (10:49 +0000)] 
Rename the vector subparts-of-lanes-reversal IROps to names
that are easier to understand.  No functional change.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2890

11 years agoThe vector versions of the count leading zeros/sign bits primops
Julian Seward [Thu, 26 Jun 2014 08:18:08 +0000 (08:18 +0000)] 
The vector versions of the count leading zeros/sign bits primops
(Iop_Cls* and Iop_Clz*) misleadingly imply a signedness in the
incoming lanes.  Rename them to fix this.  Fixes #326026.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2889

11 years agoarm64: implement pmull{2}.
Julian Seward [Thu, 26 Jun 2014 07:41:14 +0000 (07:41 +0000)] 
arm64: implement pmull{2}.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2888

11 years agoarm64: implement:
Julian Seward [Wed, 25 Jun 2014 13:05:23 +0000 (13:05 +0000)] 
arm64: implement:
LD3/ST3 (multi 3-elem structs, 3 regs, post index) (2d variants only)
pmul 16b_16b_16b, 8b_8b_8b

git-svn-id: svn://svn.valgrind.org/vex/trunk@2887

11 years agoRemove commented out junk which is never going to get used.
Julian Seward [Wed, 25 Jun 2014 12:19:02 +0000 (12:19 +0000)] 
Remove commented out junk which is never going to get used.
No functional change.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2886

11 years agoFix bogus-looking assertion.
Julian Seward [Wed, 25 Jun 2014 11:59:24 +0000 (11:59 +0000)] 
Fix bogus-looking assertion.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2885

11 years agoImplement LD1/ST1 {3 regs . 16b}, [ea] (no offset)
Julian Seward [Tue, 24 Jun 2014 10:26:52 +0000 (10:26 +0000)] 
Implement LD1/ST1 {3 regs . 16b}, [ea]  (no offset)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2884

11 years agoarm64: more SIMD instructions:
Julian Seward [Mon, 23 Jun 2014 09:09:41 +0000 (09:09 +0000)] 
arm64: more SIMD instructions:
ins (vec[], vec[])
mla, mls, mul (vec, vec, vec[])
various more movi/mvni cases
not 16b/8b

git-svn-id: svn://svn.valgrind.org/vex/trunk@2883

11 years agoFix an enum type confusion, PPCAvFpOp vs PPCAvOp, as excellently
Julian Seward [Fri, 20 Jun 2014 14:27:27 +0000 (14:27 +0000)] 
Fix an enum type confusion, PPCAvFpOp vs PPCAvOp, as excellently
detected by Clang.  Gcc, are you paying attention?

git-svn-id: svn://svn.valgrind.org/vex/trunk@2882

11 years agoIncrease the number of vector registers available for allocation from
Julian Seward [Fri, 20 Jun 2014 08:30:21 +0000 (08:30 +0000)] 
Increase the number of vector registers available for allocation from
3 to 5.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2881

11 years agoImplement: dup_{d_d[], s_s[], h_h[], b_b[]}, ext
Julian Seward [Thu, 19 Jun 2014 22:20:47 +0000 (22:20 +0000)] 
Implement: dup_{d_d[], s_s[], h_h[], b_b[]}, ext

git-svn-id: svn://svn.valgrind.org/vex/trunk@2880

11 years agoImplement: orr_{8h,4h}_imm8_shifted, orr_{4s,2s}_imm8_shifted,
Julian Seward [Thu, 19 Jun 2014 14:21:37 +0000 (14:21 +0000)] 
Implement: orr_{8h,4h}_imm8_shifted, orr_{4s,2s}_imm8_shifted,
bic_{8h,4h}_imm8_shifted, bic_{4s,2s}_imm8_shifted, cls_std6_std6,
cm{eq,ge,gt,hi,hs,tst}_d_d_d, cm{ge,gt,le,lt}_d_d_zero,
cnt_{16,8}b_{16,8}b

git-svn-id: svn://svn.valgrind.org/vex/trunk@2879

11 years agoarm64: implement: addp std7_std7_std7, addv vector, addp d_2d
Julian Seward [Sun, 15 Jun 2014 21:55:33 +0000 (21:55 +0000)] 
arm64: implement: addp std7_std7_std7, addv vector, addp d_2d

git-svn-id: svn://svn.valgrind.org/vex/trunk@2878

11 years agoarm64: implement: abs d_d, neg d_d, abs std7_std7, addhn, subhn, raddhn, rsubhn
Julian Seward [Sun, 15 Jun 2014 19:36:29 +0000 (19:36 +0000)] 
arm64: implement: abs d_d, neg d_d, abs std7_std7, addhn, subhn, raddhn, rsubhn

git-svn-id: svn://svn.valgrind.org/vex/trunk@2877

11 years agoRemove temporary front end scaffolding for Cat{Even,Odd}Lanes
Julian Seward [Sun, 15 Jun 2014 08:17:35 +0000 (08:17 +0000)] 
Remove temporary front end scaffolding for Cat{Even,Odd}Lanes
and Interleave{LO,HI} operations, and instead generate real
UZP1/UZP2/ZIP1/ZIP2 instructions in the back end.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2876

11 years agoImplement LD1R (single structure, replicate).
Julian Seward [Sat, 14 Jun 2014 18:05:30 +0000 (18:05 +0000)] 
Implement LD1R (single structure, replicate).

git-svn-id: svn://svn.valgrind.org/vex/trunk@2875

11 years agoImplement FMUL 2d_2d_d[], 4s_4s_s[], 2s_2s_s[].
Julian Seward [Thu, 12 Jun 2014 13:16:01 +0000 (13:16 +0000)] 
Implement FMUL 2d_2d_d[], 4s_4s_s[], 2s_2s_s[].

git-svn-id: svn://svn.valgrind.org/vex/trunk@2874

11 years agoRemove the old SIMD decoder entirely.
Julian Seward [Thu, 12 Jun 2014 10:15:46 +0000 (10:15 +0000)] 
Remove the old SIMD decoder entirely.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2873

11 years agoMove remaining implemented SIMD instructions into the new SIMD/FP
Julian Seward [Wed, 11 Jun 2014 20:57:23 +0000 (20:57 +0000)] 
Move remaining implemented SIMD instructions into the new SIMD/FP
decoding framework.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2872

11 years agoReimplement the SIMD and FP instruction decoder, so as to avoid huge
Julian Seward [Tue, 10 Jun 2014 22:52:05 +0000 (22:52 +0000)] 
Reimplement the SIMD and FP instruction decoder, so as to avoid huge
amounts of duplicated decode, and to follow the documentation more
closely.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2871

11 years agomips: Fix non mips compiler warning.
Dejan Jevtic [Mon, 9 Jun 2014 10:54:49 +0000 (10:54 +0000)] 
mips: Fix non mips compiler warning.

git-svn-id: svn://svn.valgrind.org/vex/trunk@2870

11 years agoSupport ADC/ADCS/SBC/SBCS. Fixes #335496. (dimitry@google.com)
Julian Seward [Wed, 4 Jun 2014 13:09:44 +0000 (13:09 +0000)] 
Support ADC/ADCS/SBC/SBCS.  Fixes #335496.  (dimitry@google.com)

git-svn-id: svn://svn.valgrind.org/vex/trunk@2868