openkylin
/
linux
mirror of https://gitee.com/openkylin/linux.git
9
0
Fork 0
Code Issues Projects Releases Wiki Activity
linux/arch/mips/lib/memset.S

298 lines
6.5 KiB
ArmAsm
Raw Normal View History

Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
* This file is subject to the terms and conditions of the GNU General Public
* License. See the file "COPYING" in the main directory of this archive
* for more details.
*
* Copyright (C) 1998, 1999, 2000 by Ralf Baechle
* Copyright (C) 1999, 2000 Silicon Graphics, Inc.
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
* Copyright (C) 2007 by Maciej W. Rozycki
* Copyright (C) 2011, 2012 MIPS Technologies, Inc.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
#include <asm/asm.h>
kbuild: mips use generic asm-offsets.h support Removed obsolete stuff from arch makefile. mips had a special rule for generating asm-offsets.h so preserved it using an architecture specific hook in top-level Kbuild file. Renamed .h file to asm-offsets.h Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
2005-09-10 04:32:31 +08:00
#include <asm/asm-offsets.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#include <asm/regdef.h>
[MIPS] Unify memset.S The 32-bit version and 64-bit version are almost equal. Unify them. This makes further improvements (for example, supporting CDEX, etc.) easier. Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-12-17 23:07:40 +08:00
#if LONGSIZE == 4
#define LONG_S_L swl
#define LONG_S_R swr
#else
#define LONG_S_L sdl
#define LONG_S_R sdr
#endif
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
#ifdef CONFIG_CPU_MICROMIPS
#define STORSIZE (LONGSIZE * 2)
#define STORMASK (STORSIZE - 1)
#define FILL64RG t8
#define FILLPTRG t7
#undef LONG_S
#define LONG_S LONG_SP
#else
#define STORSIZE LONGSIZE
#define STORMASK LONGMASK
#define FILL64RG a1
#define FILLPTRG t0
#endif
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
#define LEGACY_MODE 1
#define EVA_MODE 2
MIPS: lib: memset: Add EVA support for the __bzero function. Build the __bzero function using the EVA load/store instructions when operating in the EVA mode. This function is only used when accessing user code so there is no need to build two distinct symbols for user and kernel operations respectively. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 18:11:45 +08:00
/*
* No need to protect it with EVA #ifdefery. The generated block of code
* will never be assembled if EVA is not enabled.
*/
#define __EVAFY(insn, reg, addr) __BUILD_EVA_INSN(insn##e, reg, addr)
#define ___BUILD_EVA_INSN(insn, reg, addr) __EVAFY(insn, reg, addr)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#define EX(insn,reg,addr,handler) \
MIPS: lib: memset: Add EVA support for the __bzero function. Build the __bzero function using the EVA load/store instructions when operating in the EVA mode. This function is only used when accessing user code so there is no need to build two distinct symbols for user and kernel operations respectively. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 18:11:45 +08:00
.if \mode == LEGACY_MODE; \
9: insn reg, addr; \
.else; \
9: ___BUILD_EVA_INSN(insn, reg, addr); \
.endif; \
MIPS: Whitespace cleanup. Having received another series of whitespace patches I decided to do this once and for all rather than dealing with this kind of patches trickling in forever. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2013-01-22 19:59:30 +08:00
.section __ex_table,"a"; \
PTR 9b, handler; \
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.previous
MIPS: lib: memset: Add EVA support for the __bzero function. Build the __bzero function using the EVA load/store instructions when operating in the EVA mode. This function is only used when accessing user code so there is no need to build two distinct symbols for user and kernel operations respectively. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 18:11:45 +08:00
.macro f_fill64 dst, offset, val, fixup, mode
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
EX(LONG_S, \val, (\offset + 0 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 1 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 2 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 3 * STORSIZE)(\dst), \fixup)
#if ((defined(CONFIG_CPU_MICROMIPS) && (LONGSIZE == 4)) || !defined(CONFIG_CPU_MICROMIPS))
EX(LONG_S, \val, (\offset + 4 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 5 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 6 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 7 * STORSIZE)(\dst), \fixup)
#endif
#if (!defined(CONFIG_CPU_MICROMIPS) && (LONGSIZE == 4))
EX(LONG_S, \val, (\offset + 8 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 9 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 10 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 11 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 12 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 13 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 14 * STORSIZE)(\dst), \fixup)
EX(LONG_S, \val, (\offset + 15 * STORSIZE)(\dst), \fixup)
[MIPS] Unify memset.S The 32-bit version and 64-bit version are almost equal. Unify them. This makes further improvements (for example, supporting CDEX, etc.) easier. Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-12-17 23:07:40 +08:00
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.endm
.set noreorder
.align 5
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
/*
* Macro to generate the __bzero{,_user} symbol
* Arguments:
* mode: LEGACY_MODE or EVA_MODE
*/
.macro __BUILD_BZERO mode
/* Initialize __memset if this is the first time we call this macro */
.ifnotdef __memset
.set __memset, 1
.hidden __memset /* Make sure it does not leak */
.endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
sltiu t0, a2, STORSIZE /* very small region? */
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
bnez t0, .Lsmall_memset\@
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
andi t0, a0, STORMASK /* aligned? */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
#ifdef CONFIG_CPU_MICROMIPS
move t8, a1 /* used by 'swp' instruction */
move t9, a1
#endif
[MIPS] R4000/R4400 daddiu erratum workaround This complements the generic R4000/R4400 errata workaround code and adds bits for the daddiu problem. In most places it just modifies handwritten assembly code so that the assembler is allowed to use a temporary register as daddiu may now be treated as a macro that expands to a sequence of li and daddu. It is the AT register or, where AT is unavailable or used explicitly for another purpose, an explicitly-named register is selected, using the .set at=<reg> feature added recently to gas. This feature is only used if CONFIG_CPU_DADDI_WORKAROUNDS has been set, so if the workaround remains disabled, the required version of binutils stays unchanged. Similarly, daddiu instructions put in branch delay slots in noreorder fragments are now taken out of them and the assembler is allowed to reorder them itself as possible (which it does making the whole idea of scheduling them into delay slots manually questionable). Also in the very few places where such a simple conversion was not possible, a handcoded longer sequence is implemented. Other than that there are changes to code responsible for building the TLB fault and page clear/copy handlers to avoid daddiu as appropriate. These are only effective if the erratum is verified to be present at the run time. Finally there is a trivial update to __delay(), because it uses daddiu in a branch delay slot. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-10-23 19:43:25 +08:00
#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
beqz t0, 1f
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
PTR_SUBU t0, STORSIZE /* alignment in bytes */
[MIPS] R4000/R4400 daddiu erratum workaround This complements the generic R4000/R4400 errata workaround code and adds bits for the daddiu problem. In most places it just modifies handwritten assembly code so that the assembler is allowed to use a temporary register as daddiu may now be treated as a macro that expands to a sequence of li and daddu. It is the AT register or, where AT is unavailable or used explicitly for another purpose, an explicitly-named register is selected, using the .set at=<reg> feature added recently to gas. This feature is only used if CONFIG_CPU_DADDI_WORKAROUNDS has been set, so if the workaround remains disabled, the required version of binutils stays unchanged. Similarly, daddiu instructions put in branch delay slots in noreorder fragments are now taken out of them and the assembler is allowed to reorder them itself as possible (which it does making the whole idea of scheduling them into delay slots manually questionable). Also in the very few places where such a simple conversion was not possible, a handcoded longer sequence is implemented. Other than that there are changes to code responsible for building the TLB fault and page clear/copy handlers to avoid daddiu as appropriate. These are only effective if the erratum is verified to be present at the run time. Finally there is a trivial update to __delay(), because it uses daddiu in a branch delay slot. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-10-23 19:43:25 +08:00
#else
.set noat
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
li AT, STORSIZE
[MIPS] R4000/R4400 daddiu erratum workaround This complements the generic R4000/R4400 errata workaround code and adds bits for the daddiu problem. In most places it just modifies handwritten assembly code so that the assembler is allowed to use a temporary register as daddiu may now be treated as a macro that expands to a sequence of li and daddu. It is the AT register or, where AT is unavailable or used explicitly for another purpose, an explicitly-named register is selected, using the .set at=<reg> feature added recently to gas. This feature is only used if CONFIG_CPU_DADDI_WORKAROUNDS has been set, so if the workaround remains disabled, the required version of binutils stays unchanged. Similarly, daddiu instructions put in branch delay slots in noreorder fragments are now taken out of them and the assembler is allowed to reorder them itself as possible (which it does making the whole idea of scheduling them into delay slots manually questionable). Also in the very few places where such a simple conversion was not possible, a handcoded longer sequence is implemented. Other than that there are changes to code responsible for building the TLB fault and page clear/copy handlers to avoid daddiu as appropriate. These are only effective if the erratum is verified to be present at the run time. Finally there is a trivial update to __delay(), because it uses daddiu in a branch delay slot. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-10-23 19:43:25 +08:00
beqz t0, 1f
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
PTR_SUBU t0, AT /* alignment in bytes */
[MIPS] R4000/R4400 daddiu erratum workaround This complements the generic R4000/R4400 errata workaround code and adds bits for the daddiu problem. In most places it just modifies handwritten assembly code so that the assembler is allowed to use a temporary register as daddiu may now be treated as a macro that expands to a sequence of li and daddu. It is the AT register or, where AT is unavailable or used explicitly for another purpose, an explicitly-named register is selected, using the .set at=<reg> feature added recently to gas. This feature is only used if CONFIG_CPU_DADDI_WORKAROUNDS has been set, so if the workaround remains disabled, the required version of binutils stays unchanged. Similarly, daddiu instructions put in branch delay slots in noreorder fragments are now taken out of them and the assembler is allowed to reorder them itself as possible (which it does making the whole idea of scheduling them into delay slots manually questionable). Also in the very few places where such a simple conversion was not possible, a handcoded longer sequence is implemented. Other than that there are changes to code responsible for building the TLB fault and page clear/copy handlers to avoid daddiu as appropriate. These are only effective if the erratum is verified to be present at the run time. Finally there is a trivial update to __delay(), because it uses daddiu in a branch delay slot. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-10-23 19:43:25 +08:00
.set at
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
MIPS: lib: memset: Add MIPS R6 support MIPS R6 dropped the unaligned load and store instructions so we need to re-write this part of the code for R6 to store one byte at a time. Signed-off-by: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com> Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-11-18 17:04:34 +08:00
#ifndef CONFIG_CPU_MIPSR6
[MIPS] IP28: added cache barrier to assembly routines IP28 needs special treatment to avoid speculative accesses. gcc takes care for .c code, but for assembly code we need to do it manually. This is taken from Peter Fuersts IP28 patches. Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-11-25 18:47:56 +08:00
R10KCBARRIER(0(ra))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#ifdef __MIPSEB__
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
EX(LONG_S_L, a1, (a0), .Lfirst_fixup\@) /* make word/dword aligned */
MIPS: lib: memset: Clean up some MIPS{EL,EB} ifdefery The toolchain defines exactly one of __MIPSEB__ and __MIPSEL__. As a result, simplify the ifdefery a little bit. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/8522/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2014-11-19 16:58:10 +08:00
#else
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
EX(LONG_S_R, a1, (a0), .Lfirst_fixup\@) /* make word/dword aligned */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#endif
PTR_SUBU a0, t0 /* long align ptr */
PTR_ADDU a2, t0 /* correct size */
MIPS: lib: memset: Add MIPS R6 support MIPS R6 dropped the unaligned load and store instructions so we need to re-write this part of the code for R6 to store one byte at a time. Signed-off-by: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com> Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-11-18 17:04:34 +08:00
#else /* CONFIG_CPU_MIPSR6 */
#define STORE_BYTE(N) \
EX(sb, a1, N(a0), .Lbyte_fixup\@); \
beqz t0, 0f; \
PTR_ADDU t0, 1;
PTR_ADDU a2, t0 /* correct size */
PTR_ADDU t0, 1
STORE_BYTE(0)
STORE_BYTE(1)
#if LONGSIZE == 4
EX(sb, a1, 2(a0), .Lbyte_fixup\@)
#else
STORE_BYTE(2)
STORE_BYTE(3)
STORE_BYTE(4)
STORE_BYTE(5)
EX(sb, a1, 6(a0), .Lbyte_fixup\@)
#endif
0:
ori a0, STORMASK
xori a0, STORMASK
PTR_ADDIU a0, STORSIZE
#endif /* CONFIG_CPU_MIPSR6 */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
1: ori t1, a2, 0x3f /* # of full blocks */
xori t1, 0x3f
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
beqz t1, .Lmemset_partial\@ /* no block to fill */
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
andi t0, a2, 0x40-STORSIZE
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
PTR_ADDU t1, a0 /* end address */
.set reorder
1: PTR_ADDIU a0, 64
[MIPS] IP28: added cache barrier to assembly routines IP28 needs special treatment to avoid speculative accesses. gcc takes care for .c code, but for assembly code we need to do it manually. This is taken from Peter Fuersts IP28 patches. Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-11-25 18:47:56 +08:00
R10KCBARRIER(0(ra))
MIPS: lib: memset: Add EVA support for the __bzero function. Build the __bzero function using the EVA load/store instructions when operating in the EVA mode. This function is only used when accessing user code so there is no need to build two distinct symbols for user and kernel operations respectively. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 18:11:45 +08:00
f_fill64 a0, -64, FILL64RG, .Lfwd_fixup\@, \mode
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
bne t1, a0, 1b
.set noreorder
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
.Lmemset_partial\@:
[MIPS] IP28: added cache barrier to assembly routines IP28 needs special treatment to avoid speculative accesses. gcc takes care for .c code, but for assembly code we need to do it manually. This is taken from Peter Fuersts IP28 patches. Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-11-25 18:47:56 +08:00
R10KCBARRIER(0(ra))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
PTR_LA t1, 2f /* where to start */
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
#ifdef CONFIG_CPU_MICROMIPS
LONG_SRL t7, t0, 1
#endif
[MIPS] Unify memset.S The 32-bit version and 64-bit version are almost equal. Unify them. This makes further improvements (for example, supporting CDEX, etc.) easier. Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-12-17 23:07:40 +08:00
#if LONGSIZE == 4
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
PTR_SUBU t1, FILLPTRG
[MIPS] Unify memset.S The 32-bit version and 64-bit version are almost equal. Unify them. This makes further improvements (for example, supporting CDEX, etc.) easier. Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-12-17 23:07:40 +08:00
#else
.set noat
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
LONG_SRL AT, FILLPTRG, 1
[MIPS] Unify memset.S The 32-bit version and 64-bit version are almost equal. Unify them. This makes further improvements (for example, supporting CDEX, etc.) easier. Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-12-17 23:07:40 +08:00
PTR_SUBU t1, AT
[MIPS] R4000/R4400 daddiu erratum workaround This complements the generic R4000/R4400 errata workaround code and adds bits for the daddiu problem. In most places it just modifies handwritten assembly code so that the assembler is allowed to use a temporary register as daddiu may now be treated as a macro that expands to a sequence of li and daddu. It is the AT register or, where AT is unavailable or used explicitly for another purpose, an explicitly-named register is selected, using the .set at=<reg> feature added recently to gas. This feature is only used if CONFIG_CPU_DADDI_WORKAROUNDS has been set, so if the workaround remains disabled, the required version of binutils stays unchanged. Similarly, daddiu instructions put in branch delay slots in noreorder fragments are now taken out of them and the assembler is allowed to reorder them itself as possible (which it does making the whole idea of scheduling them into delay slots manually questionable). Also in the very few places where such a simple conversion was not possible, a handcoded longer sequence is implemented. Other than that there are changes to code responsible for building the TLB fault and page clear/copy handlers to avoid daddiu as appropriate. These are only effective if the erratum is verified to be present at the run time. Finally there is a trivial update to __delay(), because it uses daddiu in a branch delay slot. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-10-23 19:43:25 +08:00
.set at
[MIPS] Unify memset.S The 32-bit version and 64-bit version are almost equal. Unify them. This makes further improvements (for example, supporting CDEX, etc.) easier. Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2006-12-17 23:07:40 +08:00
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
jr t1
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
PTR_ADDU a0, t0 /* dest ptr */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.set push
.set noreorder
.set nomacro
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
/* ... but first do longs ... */
MIPS: lib: memset: Add EVA support for the __bzero function. Build the __bzero function using the EVA load/store instructions when operating in the EVA mode. This function is only used when accessing user code so there is no need to build two distinct symbols for user and kernel operations respectively. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 18:11:45 +08:00
f_fill64 a0, -64, FILL64RG, .Lpartial_fixup\@, \mode
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
2: .set pop
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
andi a2, STORMASK /* At most one long to go */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
beqz a2, 1f
MIPS: lib: memset: Add MIPS R6 support MIPS R6 dropped the unaligned load and store instructions so we need to re-write this part of the code for R6 to store one byte at a time. Signed-off-by: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com> Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-11-18 17:04:34 +08:00
#ifndef CONFIG_CPU_MIPSR6
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
PTR_ADDU a0, a2 /* What's left */
[MIPS] IP28: added cache barrier to assembly routines IP28 needs special treatment to avoid speculative accesses. gcc takes care for .c code, but for assembly code we need to do it manually. This is taken from Peter Fuersts IP28 patches. Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-11-25 18:47:56 +08:00
R10KCBARRIER(0(ra))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#ifdef __MIPSEB__
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
EX(LONG_S_R, a1, -1(a0), .Llast_fixup\@)
MIPS: lib: memset: Clean up some MIPS{EL,EB} ifdefery The toolchain defines exactly one of __MIPSEB__ and __MIPSEL__. As a result, simplify the ifdefery a little bit. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/8522/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2014-11-19 16:58:10 +08:00
#else
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
EX(LONG_S_L, a1, -1(a0), .Llast_fixup\@)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#endif
MIPS: lib: memset: Add MIPS R6 support MIPS R6 dropped the unaligned load and store instructions so we need to re-write this part of the code for R6 to store one byte at a time. Signed-off-by: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com> Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-11-18 17:04:34 +08:00
#else
PTR_SUBU t0, $0, a2
PTR_ADDIU t0, 1
STORE_BYTE(0)
STORE_BYTE(1)
#if LONGSIZE == 4
EX(sb, a1, 2(a0), .Lbyte_fixup\@)
#else
STORE_BYTE(2)
STORE_BYTE(3)
STORE_BYTE(4)
STORE_BYTE(5)
EX(sb, a1, 6(a0), .Lbyte_fixup\@)
#endif
0:
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
1: jr ra
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
move a2, zero
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
.Lsmall_memset\@:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
beqz a2, 2f
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
PTR_ADDU t1, a0, a2
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
1: PTR_ADDIU a0, 1 /* fill bytewise */
[MIPS] IP28: added cache barrier to assembly routines IP28 needs special treatment to avoid speculative accesses. gcc takes care for .c code, but for assembly code we need to do it manually. This is taken from Peter Fuersts IP28 patches. Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-11-25 18:47:56 +08:00
R10KCBARRIER(0(ra))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
bne t1, a0, 1b
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
sb a1, -1(a0)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
2: jr ra /* done */
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
move a2, zero
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
.if __memset == 1
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
END(memset)
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
.set __memset, 0
.hidden __memset
.endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
MIPS: memset.S: Disable code unused with non-R6 MIPS configs This complements commit 8c56208aff77 ("MIPS: lib: memset: Add MIPS R6 support"). Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/12452/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2016-02-07 19:05:58 +08:00
#ifdef CONFIG_CPU_MIPSR6
MIPS: lib: memset: Add MIPS R6 support MIPS R6 dropped the unaligned load and store instructions so we need to re-write this part of the code for R6 to store one byte at a time. Signed-off-by: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com> Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-11-18 17:04:34 +08:00
.Lbyte_fixup\@:
PTR_SUBU a2, $0, t0
jr ra
PTR_ADDIU a2, 1
MIPS: memset.S: Disable code unused with non-R6 MIPS configs This complements commit 8c56208aff77 ("MIPS: lib: memset: Add MIPS R6 support"). Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/12452/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2016-02-07 19:05:58 +08:00
#endif /* CONFIG_CPU_MIPSR6 */
MIPS: lib: memset: Add MIPS R6 support MIPS R6 dropped the unaligned load and store instructions so we need to re-write this part of the code for R6 to store one byte at a time. Signed-off-by: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com> Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-11-18 17:04:34 +08:00
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
.Lfirst_fixup\@:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
jr ra
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
.Lfwd_fixup\@:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
PTR_L t0, TI_TASK($28)
andi a2, 0x3f
MIPS: Separate two consecutive loads in memset.S partial_fixup is used in noreorder block. Separating two consecutive loads can save one cycle on processors with GPR intrelock and can fix load-use on processors that need a load delay slot. Also do so for fwd_fixup. [Ralf: Only R2000/R3000 class processors are lacking the the load-user interlock and even some of those got it retrofitted. With R2000/R3000 being fairly uncommon these days the impact of this bug should be minor.] Signed-off-by: Tony Wu <tung7970@gmail.com> To: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/1768/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-11-10 21:48:15 +08:00
LONG_L t0, THREAD_BUADDR(t0)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
LONG_ADDU a2, t1
jr ra
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
LONG_SUBU a2, t0
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
.Lpartial_fixup\@:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
PTR_L t0, TI_TASK($28)
MIPS: microMIPS: Optimise 'memset' core library function. Optimise 'memset' to use microMIPS instructions and/or optimisations for binary size reduction. When the microMIPS ISA is not being used, the library function compiles to the original binary code. Signed-off-by: Steven J. Hill <Steven.Hill@imgtec.com>
2013-03-26 02:40:49 +08:00
andi a2, STORMASK
MIPS: Separate two consecutive loads in memset.S partial_fixup is used in noreorder block. Separating two consecutive loads can save one cycle on processors with GPR intrelock and can fix load-use on processors that need a load delay slot. Also do so for fwd_fixup. [Ralf: Only R2000/R3000 class processors are lacking the the load-user interlock and even some of those got it retrofitted. With R2000/R3000 being fairly uncommon these days the impact of this bug should be minor.] Signed-off-by: Tony Wu <tung7970@gmail.com> To: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/1768/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-11-10 21:48:15 +08:00
LONG_L t0, THREAD_BUADDR(t0)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
LONG_ADDU a2, t1
jr ra
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
LONG_SUBU a2, t0
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
.Llast_fixup\@:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
jr ra
MIPS: lib: memset: Whitespace fixes Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:21:20 +08:00
andi v1, a2, STORMASK
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
.endm
/*
* memset(void *s, int c, size_t n)
*
* a0: start of area to clear
* a1: char to fill with
* a2: size of area to clear
*/
LEAF(memset)
beqz a1, 1f
move v0, a0 /* result */
andi a1, 0xff /* spread fillword */
LONG_SLL t1, a1, 8
or a1, t1
LONG_SLL t1, a1, 16
#if LONGSIZE == 8
or a1, t1
LONG_SLL t1, a1, 32
#endif
or a1, t1
1:
MIPS: lib: memset: Add EVA support for the __bzero function. Build the __bzero function using the EVA load/store instructions when operating in the EVA mode. This function is only used when accessing user code so there is no need to build two distinct symbols for user and kernel operations respectively. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 18:11:45 +08:00
#ifndef CONFIG_EVA
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
FEXPORT(__bzero)
MIPS: uaccess: Take EVA into account in [__]clear_user __clear_user() (and clear_user() which uses it), always access the user mode address space, which results in EVA store instructions when EVA is enabled even if the current user address limit is KERNEL_DS. Fix this by adding a new symbol __bzero_kernel for the normal kernel address space bzero in EVA mode, and call that from __clear_user() if eva_kernel_access(). Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Markos Chandras <markos.chandras@imgtec.com> Cc: Paul Burton <paul.burton@imgtec.com> Cc: Leonid Yegoshin <leonid.yegoshin@imgtec.com> Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/10844/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-08-05 23:41:39 +08:00
#else
FEXPORT(__bzero_kernel)
MIPS: lib: memset: Add EVA support for the __bzero function. Build the __bzero function using the EVA load/store instructions when operating in the EVA mode. This function is only used when accessing user code so there is no need to build two distinct symbols for user and kernel operations respectively. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 18:11:45 +08:00
#endif
MIPS: lib: memset: Use macro to build the __bzero symbol Build the __bzero symbol using a macor. In EVA mode we will need to use similar code to do the userspace load operations so it is better if we use a macro to avoid code duplications. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 17:23:16 +08:00
__BUILD_BZERO LEGACY_MODE
MIPS: lib: memset: Add EVA support for the __bzero function. Build the __bzero function using the EVA load/store instructions when operating in the EVA mode. This function is only used when accessing user code so there is no need to build two distinct symbols for user and kernel operations respectively. Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
2014-01-03 18:11:45 +08:00
#ifdef CONFIG_EVA
LEAF(__bzero)
__BUILD_BZERO EVA_MODE
END(__bzero)
#endif