Wed, 03 Apr 2013 11:12:57 -0700
8011102: Clear AVX registers after return from JNI call
Summary: Execute vzeroupper instruction after JNI call and on exits in jit compiled code which use 256bit vectors.
Reviewed-by: roland
duke@435 | 1 | // |
kvn@3577 | 2 | // Copyright (c) 2003, 2012, Oracle and/or its affiliates. All rights reserved. |
duke@435 | 3 | // DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. |
duke@435 | 4 | // |
duke@435 | 5 | // This code is free software; you can redistribute it and/or modify it |
duke@435 | 6 | // under the terms of the GNU General Public License version 2 only, as |
duke@435 | 7 | // published by the Free Software Foundation. |
duke@435 | 8 | // |
duke@435 | 9 | // This code is distributed in the hope that it will be useful, but WITHOUT |
duke@435 | 10 | // ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or |
duke@435 | 11 | // FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
duke@435 | 12 | // version 2 for more details (a copy is included in the LICENSE file that |
duke@435 | 13 | // accompanied this code). |
duke@435 | 14 | // |
duke@435 | 15 | // You should have received a copy of the GNU General Public License version |
duke@435 | 16 | // 2 along with this work; if not, write to the Free Software Foundation, |
duke@435 | 17 | // Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. |
duke@435 | 18 | // |
trims@1907 | 19 | // Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA |
trims@1907 | 20 | // or visit www.oracle.com if you need additional information or have any |
trims@1907 | 21 | // questions. |
duke@435 | 22 | // |
duke@435 | 23 | // |
duke@435 | 24 | |
duke@435 | 25 | // AMD64 Linux Architecture Description File |
duke@435 | 26 | |
duke@435 | 27 | //----------OS-DEPENDENT ENCODING BLOCK---------------------------------------- |
duke@435 | 28 | // This block specifies the encoding classes used by the compiler to |
duke@435 | 29 | // output byte streams. Encoding classes generate functions which are |
duke@435 | 30 | // called by Machine Instruction Nodes in order to generate the bit |
duke@435 | 31 | // encoding of the instruction. Operands specify their base encoding |
duke@435 | 32 | // interface with the interface keyword. There are currently |
duke@435 | 33 | // supported four interfaces, REG_INTER, CONST_INTER, MEMORY_INTER, & |
duke@435 | 34 | // COND_INTER. REG_INTER causes an operand to generate a function |
duke@435 | 35 | // which returns its register number when queried. CONST_INTER causes |
duke@435 | 36 | // an operand to generate a function which returns the value of the |
duke@435 | 37 | // constant when queried. MEMORY_INTER causes an operand to generate |
duke@435 | 38 | // four functions which return the Base Register, the Index Register, |
duke@435 | 39 | // the Scale Value, and the Offset Value of the operand when queried. |
duke@435 | 40 | // COND_INTER causes an operand to generate six functions which return |
duke@435 | 41 | // the encoding code (ie - encoding bits for the instruction) |
duke@435 | 42 | // associated with each basic boolean condition for a conditional |
duke@435 | 43 | // instruction. Instructions specify two basic values for encoding. |
duke@435 | 44 | // They use the ins_encode keyword to specify their encoding class |
duke@435 | 45 | // (which must be one of the class names specified in the encoding |
duke@435 | 46 | // block), and they use the opcode keyword to specify, in order, their |
duke@435 | 47 | // primary, secondary, and tertiary opcode. Only the opcode sections |
duke@435 | 48 | // which a particular instruction needs for encoding need to be |
duke@435 | 49 | // specified. |
duke@435 | 50 | encode %{ |
duke@435 | 51 | // Build emit functions for each basic byte or larger field in the intel |
duke@435 | 52 | // encoding scheme (opcode, rm, sib, immediate), and call them from C++ |
duke@435 | 53 | // code in the enc_class source block. Emit functions will live in the |
duke@435 | 54 | // main source block for now. In future, we can generalize this by |
duke@435 | 55 | // adding a syntax that specifies the sizes of fields in an order, |
duke@435 | 56 | // so that the adlc can build the emit functions automagically |
duke@435 | 57 | |
duke@435 | 58 | %} |
duke@435 | 59 | |
duke@435 | 60 | |
duke@435 | 61 | // Platform dependent source |
duke@435 | 62 | |
kvn@3577 | 63 | source %{ |
duke@435 | 64 | |
duke@435 | 65 | %} |