ftmemsim-valgrind/coregrind/m_dispatch/dispatch-x86-linux.S
Mark Wielaard 461cc5c003 Cleanup GPL header address notices by using http://www.gnu.org/licenses/
Sync VEX/LICENSE.GPL with top-level COPYING file. We used 3 different
addresses for writing to the FSF to receive a copy of the GPL. Replace
all different variants with an URL <http://www.gnu.org/licenses/>.

The following files might still have some slightly different (L)GPL
copyright notice because they were derived from other programs:

- files under coregrind/m_demangle which come from libiberty:
  cplus-dem.c, d-demangle.c, demangle.h, rust-demangle.c,
  safe-ctype.c and safe-ctype.h
- coregrind/m_demangle/dyn-string.[hc] derived from GCC.
- coregrind/m_demangle/ansidecl.h derived from glibc.
- VEX files for FMA detived from glibc:
  host_generic_maddf.h and host_generic_maddf.c
- files under coregrin/m_debuginfo derived from LZO:
  lzoconf.h, lzodefs.h, minilzo-inl.c and minilzo.h
- files under coregrind/m_gdbserver detived from GDB:
  gdb/signals.h, inferiors.c, regcache.c, regcache.h,
  regdef.h, remote-utils.c, server.c, server.h, signals.c,
  target.c, target.h and utils.c

Plus the following test files:

- none/tests/ppc32/testVMX.c derived from testVMX.
- ppc tests derived from QEMU: jm-insns.c, ppc64_helpers.h
  and test_isa_3_0.c
- tests derived from bzip2 (with embedded GPL text in code):
  hackedbz2.c, origin5-bz2.c, varinfo6.c
- tests detived from glibc: str_tester.c, pth_atfork1.c
- test detived from GCC libgomp: tc17_sembar.c
- performance tests derived from bzip2 or tinycc (with embedded GPL
  text in code): bz2.c, test_input_for_tinycc.c and tinycc.c
2019-05-26 20:07:51 +02:00

312 lines
10 KiB
ArmAsm

/*--------------------------------------------------------------------*/
/*--- The core dispatch loop, for jumping to a code address. ---*/
/*--- dispatch-x86-linux.S ---*/
/*--------------------------------------------------------------------*/
/*
This file is part of Valgrind, a dynamic binary instrumentation
framework.
Copyright (C) 2000-2017 Julian Seward
jseward@acm.org
This program is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License as
published by the Free Software Foundation; either version 2 of the
License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, see <http://www.gnu.org/licenses/>.
The GNU General Public License is contained in the file COPYING.
*/
#include "pub_core_basics_asm.h"
#if defined(VGP_x86_linux)
#include "pub_core_dispatch_asm.h"
#include "pub_core_transtab_asm.h"
#include "libvex_guest_offsets.h" /* for OFFSET_x86_EIP */
/*------------------------------------------------------------*/
/*--- ---*/
/*--- The dispatch loop. VG_(disp_run_translations) is ---*/
/*--- used to run all translations, ---*/
/*--- including no-redir ones. ---*/
/*--- ---*/
/*------------------------------------------------------------*/
/*----------------------------------------------------*/
/*--- Entry and preamble (set everything up) ---*/
/*----------------------------------------------------*/
/* signature:
void VG_(disp_run_translations)( UWord* two_words,
void* guest_state,
Addr host_addr );
*/
.text
.globl VG_(disp_run_translations)
.type VG_(disp_run_translations), @function
VG_(disp_run_translations):
/* 0(%esp) holds our return address. */
/* 4(%esp) holds two_words */
/* 8(%esp) holds guest_state */
/* 12(%esp) holds host_addr */
/* The preamble */
/* Save integer registers, since this is a pseudo-function. */
pushl %eax
pushl %ebx
pushl %ecx
pushl %edx
pushl %esi
pushl %edi
pushl %ebp
/* 28+4(%esp) holds two_words */
/* 28+8(%esp) holds guest_state */
/* 28+12(%esp) holds host_addr */
/* Get the host CPU in the state expected by generated code. */
/* set host FPU control word to the default mode expected
by VEX-generated code. See comments in libvex.h for
more info. */
finit
pushl $0x027F
fldcw (%esp)
addl $4, %esp
/* set host SSE control word to the default mode expected
by VEX-generated code. */
cmpl $0, VG_(machine_x86_have_mxcsr)
jz L1
pushl $0x1F80
ldmxcsr (%esp)
addl $4, %esp
L1:
/* set dir flag to known value */
cld
/* Set up the guest state pointer */
movl 28+8(%esp), %ebp
/* and jump into the code cache. Chained translations in
the code cache run, until for whatever reason, they can't
continue. When that happens, the translation in question
will jump (or call) to one of the continuation points
VG_(cp_...) below. */
jmpl *28+12(%esp)
/*NOTREACHED*/
/*----------------------------------------------------*/
/*--- Postamble and exit. ---*/
/*----------------------------------------------------*/
postamble:
/* At this point, %eax and %edx contain two
words to be returned to the caller. %eax
holds a TRC value, and %edx optionally may
hold another word (for CHAIN_ME exits, the
address of the place to patch.) */
/* We're leaving. Check that nobody messed with %mxcsr
or %fpucw. We can't mess with %eax or %edx here as they
holds the tentative return value, but any others are OK. */
#if !defined(ENABLE_INNER)
/* This check fails for self-hosting, so skip in that case */
pushl $0
fstcw (%esp)
cmpl $0x027F, (%esp)
popl %esi /* get rid of the word without trashing %eflags */
jnz invariant_violation
#endif
# cmpl $0, VG_(machine_x86_have_mxcsr)
jz L2
pushl $0
stmxcsr (%esp)
andl $0xFFFFFFC0, (%esp) /* mask out status flags */
cmpl $0x1F80, (%esp)
popl %esi
jnz invariant_violation
L2: /* otherwise we're OK */
jmp remove_frame
invariant_violation:
movl $VG_TRC_INVARIANT_FAILED, %eax
movl $0, %edx
remove_frame:
/* Stash return values */
movl 28+4(%esp), %edi /* two_words */
movl %eax, 0(%edi)
movl %edx, 4(%edi)
/* Restore int regs and return. */
popl %ebp
popl %edi
popl %esi
popl %edx
popl %ecx
popl %ebx
popl %eax
ret
/*----------------------------------------------------*/
/*--- Continuation points ---*/
/*----------------------------------------------------*/
/* ------ Chain me to slow entry point ------ */
.global VG_(disp_cp_chain_me_to_slowEP)
VG_(disp_cp_chain_me_to_slowEP):
/* We got called. The return address indicates
where the patching needs to happen. Collect
the return address and, exit back to C land,
handing the caller the pair (Chain_me_S, RA) */
movl $VG_TRC_CHAIN_ME_TO_SLOW_EP, %eax
popl %edx
/* 5 = movl $VG_(disp_chain_me_to_slowEP), %edx;
2 = call *%edx */
subl $5+2, %edx
jmp postamble
/* ------ Chain me to fast entry point ------ */
.global VG_(disp_cp_chain_me_to_fastEP)
VG_(disp_cp_chain_me_to_fastEP):
/* We got called. The return address indicates
where the patching needs to happen. Collect
the return address and, exit back to C land,
handing the caller the pair (Chain_me_F, RA) */
movl $VG_TRC_CHAIN_ME_TO_FAST_EP, %eax
popl %edx
/* 5 = movl $VG_(disp_chain_me_to_fastEP), %edx;
2 = call *%edx */
subl $5+2, %edx
jmp postamble
/* ------ Indirect but boring jump ------ */
.global VG_(disp_cp_xindir)
VG_(disp_cp_xindir):
/* Where are we going? */
movl OFFSET_x86_EIP(%ebp), %eax // "guest"
/* stats only */
addl $1, VG_(stats__n_xIndirs_32)
// LIVE: %ebp (guest state ptr), %eax (guest address to go to).
// We use 4 temporaries:
// %esi (to point at the relevant FastCacheSet),
// %ebx, %ecx and %edx (scratch).
/* Try a fast lookup in the translation cache. This is pretty much
a handcoded version of VG_(lookupInFastCache). */
// Compute %esi = VG_TT_FAST_HASH(guest)
movl %eax, %esi // guest
shrl $VG_TT_FAST_BITS, %esi // (guest >> VG_TT_FAST_BITS)
xorl %eax, %esi // (guest >> VG_TT_FAST_BITS) ^ guest
andl $VG_TT_FAST_MASK, %esi // setNo
// Compute %esi = &VG_(tt_fast)[%esi]
shll $VG_FAST_CACHE_SET_BITS, %esi // setNo * sizeof(FastCacheSet)
leal VG_(tt_fast)(%esi), %esi // &VG_(tt_fast)[setNo]
// LIVE: %ebp (guest state ptr), %eax (guest addr), %esi (cache set)
// try way 0
cmpl %eax, FCS_g0(%esi) // cmp against .guest0
jnz 1f
// hit at way 0
jmp *FCS_h0(%esi) // goto .host0
ud2
1: // try way 1
cmpl %eax, FCS_g1(%esi) // cmp against .guest1
jnz 2f
// hit at way 1; swap upwards
/* stats only */
addl $1, VG_(stats__n_xIndir_hits1_32)
movl FCS_g0(%esi), %ebx // ebx = old .guest0
movl FCS_h0(%esi), %ecx // ecx = old .host0
movl FCS_h1(%esi), %edx // edx = old .host1
movl %eax, FCS_g0(%esi) // new .guest0 = guest
movl %edx, FCS_h0(%esi) // new .host0 = old .host1
movl %ebx, FCS_g1(%esi) // new .guest1 = old .guest0
movl %ecx, FCS_h1(%esi) // new .host1 = old .host0
jmp *%edx // goto old .host1 a.k.a. new .host0
ud2
2: // try way 2
cmpl %eax, FCS_g2(%esi) // cmp against .guest2
jnz 3f
// hit at way 2; swap upwards
/* stats only */
addl $1, VG_(stats__n_xIndir_hits2_32)
movl FCS_g1(%esi), %ebx
movl FCS_h1(%esi), %ecx
movl FCS_h2(%esi), %edx
movl %eax, FCS_g1(%esi)
movl %edx, FCS_h1(%esi)
movl %ebx, FCS_g2(%esi)
movl %ecx, FCS_h2(%esi)
jmp *%edx
ud2
3: // try way 3
cmpl %eax, FCS_g3(%esi) // cmp against .guest3
jnz 4f
// hit at way 3; swap upwards
/* stats only */
addl $1, VG_(stats__n_xIndir_hits3_32)
movl FCS_g2(%esi), %ebx
movl FCS_h2(%esi), %ecx
movl FCS_h3(%esi), %edx
movl %eax, FCS_g2(%esi)
movl %edx, FCS_h2(%esi)
movl %ebx, FCS_g3(%esi)
movl %ecx, FCS_h3(%esi)
jmp *%edx
ud2
4: // fast lookup failed
/* stats only */
addl $1, VG_(stats__n_xIndir_misses_32)
movl $VG_TRC_INNER_FASTMISS, %eax
movl $0, %edx
jmp postamble
/* ------ Assisted jump ------ */
.global VG_(disp_cp_xassisted)
VG_(disp_cp_xassisted):
/* %ebp contains the TRC */
movl %ebp, %eax
movl $0, %edx
jmp postamble
/* ------ Event check failed ------ */
.global VG_(disp_cp_evcheck_fail)
VG_(disp_cp_evcheck_fail):
movl $VG_TRC_INNER_COUNTERZERO, %eax
movl $0, %edx
jmp postamble
.size VG_(disp_run_translations), .-VG_(disp_run_translations)
#endif // defined(VGP_x86_linux)
/* Let the linker know we don't need an executable stack */
MARK_STACK_NO_EXEC
/*--------------------------------------------------------------------*/
/*--- end ---*/
/*--------------------------------------------------------------------*/