1 /* PLT trampolines. x86-64 version.
2 Copyright (C) 2004-2016 Free Software Foundation, Inc.
3 This file is part of the GNU C Library.
5 The GNU C Library is free software; you can redistribute it and/or
6 modify it under the terms of the GNU Lesser General Public
7 License as published by the Free Software Foundation; either
8 version 2.1 of the License, or (at your option) any later version.
10 The GNU C Library is distributed in the hope that it will be useful,
11 but WITHOUT ANY WARRANTY; without even the implied warranty of
12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 Lesser General Public License for more details.
15 You should have received a copy of the GNU Lesser General Public
16 License along with the GNU C Library; if not, see
17 <http://www.gnu.org/licenses/>. */
21 #include <link-defines.h>
23 #ifndef DL_STACK_ALIGNMENT
26 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58066
28 __tls_get_addr may be called with 8-byte stack alignment. Although
29 this bug has been fixed in GCC 4.9.4, 5.3 and 6, we can't assume
30 that stack will be always aligned at 16 bytes. We use unaligned
31 16-byte move to load and store SSE registers, which has no penalty
32 on modern processors if stack is 16-byte aligned. */
33 # define DL_STACK_ALIGNMENT 8
36 #ifndef DL_RUNTIME_UNALIGNED_VEC_SIZE
37 /* The maximum size in bytes of unaligned vector load and store in the
38 dynamic linker. Since SSE optimized memory/string functions with
39 aligned SSE register load and store are used in the dynamic linker,
40 we must set this to 8 so that _dl_runtime_resolve_sse will align the
41 stack before calling _dl_fixup. */
42 # define DL_RUNTIME_UNALIGNED_VEC_SIZE 8
45 /* True if _dl_runtime_resolve should align stack to VEC_SIZE bytes. */
46 #define DL_RUNTIME_RESOLVE_REALIGN_STACK \
47 (VEC_SIZE > DL_STACK_ALIGNMENT \
48 && VEC_SIZE > DL_RUNTIME_UNALIGNED_VEC_SIZE)
50 /* Align vector register save area to 16 bytes. */
51 #define REGISTER_SAVE_VEC_OFF 0
53 /* Area on stack to save and restore registers used for parameter
54 passing when calling _dl_fixup. */
56 # define REGISTER_SAVE_RAX (REGISTER_SAVE_VEC_OFF + VEC_SIZE * 8)
57 # define PRESERVE_BND_REGS_PREFIX
59 /* Align bound register save area to 16 bytes. */
60 # define REGISTER_SAVE_BND0 (REGISTER_SAVE_VEC_OFF + VEC_SIZE * 8)
61 # define REGISTER_SAVE_BND1 (REGISTER_SAVE_BND0 + 16)
62 # define REGISTER_SAVE_BND2 (REGISTER_SAVE_BND1 + 16)
63 # define REGISTER_SAVE_BND3 (REGISTER_SAVE_BND2 + 16)
64 # define REGISTER_SAVE_RAX (REGISTER_SAVE_BND3 + 16)
65 # ifdef HAVE_MPX_SUPPORT
66 # define PRESERVE_BND_REGS_PREFIX bnd
68 # define PRESERVE_BND_REGS_PREFIX .byte 0xf2
71 #define REGISTER_SAVE_RCX (REGISTER_SAVE_RAX + 8)
72 #define REGISTER_SAVE_RDX (REGISTER_SAVE_RCX + 8)
73 #define REGISTER_SAVE_RSI (REGISTER_SAVE_RDX + 8)
74 #define REGISTER_SAVE_RDI (REGISTER_SAVE_RSI + 8)
75 #define REGISTER_SAVE_R8 (REGISTER_SAVE_RDI + 8)
76 #define REGISTER_SAVE_R9 (REGISTER_SAVE_R8 + 8)
80 #ifdef HAVE_AVX512_ASM_SUPPORT
82 # define VMOVA vmovdqa64
83 # if DL_RUNTIME_RESOLVE_REALIGN_STACK || VEC_SIZE <= DL_STACK_ALIGNMENT
84 # define VMOV vmovdqa64
86 # define VMOV vmovdqu64
88 # define VEC(i) zmm##i
89 # define _dl_runtime_resolve _dl_runtime_resolve_avx512
90 # define _dl_runtime_profile _dl_runtime_profile_avx512
91 # include "dl-trampoline.h"
92 # undef _dl_runtime_resolve
93 # undef _dl_runtime_profile
99 strong_alias (_dl_runtime_resolve_avx, _dl_runtime_resolve_avx512)
100 .hidden _dl_runtime_resolve_avx512
101 strong_alias (_dl_runtime_profile_avx, _dl_runtime_profile_avx512)
102 .hidden _dl_runtime_profile_avx512
106 #define VMOVA vmovdqa
107 #if DL_RUNTIME_RESOLVE_REALIGN_STACK || VEC_SIZE <= DL_STACK_ALIGNMENT
108 # define VMOV vmovdqa
110 # define VMOV vmovdqu
112 #define VEC(i) ymm##i
113 #define _dl_runtime_resolve _dl_runtime_resolve_avx
114 #define _dl_runtime_profile _dl_runtime_profile_avx
115 #include "dl-trampoline.h"
116 #undef _dl_runtime_resolve
117 #undef _dl_runtime_profile
123 /* movaps/movups is 1-byte shorter. */
126 #if DL_RUNTIME_RESOLVE_REALIGN_STACK || VEC_SIZE <= DL_STACK_ALIGNMENT
131 #define VEC(i) xmm##i
132 #define _dl_runtime_resolve _dl_runtime_resolve_sse
133 #define _dl_runtime_profile _dl_runtime_profile_sse
135 #include "dl-trampoline.h"