Re: [patch] MIPS32 updated

[Nick Clifton <nickc at redhat dot com>]

Hi Anders,

: here's an updated patch for MIPS32 support.

Sorry about the delay in replying to this email.

: Changes from the previous post are:
: * Patch binutils to use defines instead of hardcoded mips processor
: numbers.
: * The readelf utility is patched to print the EF_MIPS_MACH
: field of the e_flags field.
: * Adds a few instructions.
: 
: Ok to apply?

Approved and applied.

Cheers
	Nick

: binutils/ChangeLog
: Fri Sep  1 14:06:19 2000  Anders Norlander  <anorland@acc.umu.se>
: 
: 	* readelf.c (get_machine_flags): Also print EF_MIPS_MACH field.
: 
: include/opcode/ChangeLog
: Fri Sep  1 11:41:03 2000  Anders Norlander  <anorland@acc.umu.se>
: 
: 	Use defines instead of hard-coded processor numbers.
: 	* mips.h (CPU_R2000, CPU_R3000, CPU_R3900, CPU_R4000, CPU_R4010,
: 	CPU_VR4100, CPU_R4111, CPU_R4300, CPU_R4400, CPU_R4600, CPU_R4650,
: 	CPU_R5000, CPU_R6000, CPU_R8000, CPU_R10000, CPU_MIPS32, CPU_4K,
: 	CPU_4KC, CPU_4KM, CPU_4KP): New defines.
: 	* mips.h (OPCODE_IS_MEMBER): Use.
: 
: bfd/ChangeLog
: Thu Aug 31 10:02:37 2000  Anders Norlander  <anorland@acc.umu.se>
: 	
:  	* cpu-mips.c (arch_info_struct): Add mips:4K
:  	* bfd-in2.h (bfd_mach_mips4K): New define.
:  	* archures.c: Add bfd_mach_mips4K to comment.
:  	* elf32-mips.c (_bfd_mips_elf_final_write_processing): Return E_MIPS_ARCH_2 for
:  	bfd_mach_mips4K.
: 
: 
: gas/ChangeLog
: Thu Aug 31 10:02:37 2000  Anders Norlander  <anorland@acc.umu.se>
: 
:  	* config/tc-mips.c (md_begin): Recognize 4Kc, 4Km and 4Kp processors.
:  	(md_parse_option): Ditto.
:  	(md_longopts): Add -mips32 option.
:  	(md_show_usage): Document new options.
:  	(mips_ip): Assemble sdbbp 20 bit 'm' args for MIPS32.
:  	(mips_ip): Assemble mfc0 with a sub-selection code.
:  	(validate_mips_insn): Handle 'H' (OP_*_SEL) and 'm' (OP_*_CODE20).
: 	(mips_cpu_to_str): New function.
: 	(mips_ip): Use mips_cpu_to_str instead of printing numeric cpu value.
: 
: 	* config/tc-mips.c: Use CPU_* defines instead of hardcoded numbers.
: 
:  	* doc/as.texinfo: Document new options.
:  	* doc/c-mips.texi: Ditto.
: 
: 
: include/elf/ChangeLog
: Thu Aug 31 10:02:37 2000  Anders Norlander  <anorland@acc.umu.se>
: 
:  	* mips.h (E_MIPS_MACH_4K): New define.
: 
: 
: include/opcode/ChangeLog
: Thu Aug 31 10:02:37 2000  Anders Norlander  <anorland@acc.umu.se>
: 
:  	* mips.h (OP_MASK_SEL, OP_SH_SEL): Define.
: 	(OP_MASK_CODE20, OP_SH_CODE20): Define.
:  	* mips.h: Add 'P' to used characters.
:  	* mips.h: Use 'H' for coprocessor select field.
: 	* mips.h: Use 'm' for 20 bit breakpoint code.
:  	* mips.h: Document new arg characters and add to used characters.
:  	* mips.h (INSN_MIPS32): New define for MIPS32 extensions.
:  	(OPCODE_IS_MEMBER): Recognize MIPS32 instructions.
: 
: opcodes/ChangeLog
: Thu Aug 31 10:02:37 2000  Anders Norlander  <anorland@acc.umu.se>
:  
:  	* mips-opc.c (mips_builtin_opcodes): Support cache instruction on 4K cores.
:  	(mips_builtin_opcodes): Add mfc0 and mtc0 with sub-selection values.
:   
: 	* mips-opc.c (mips_builtin_opcodes): Add clo and clz opcodes.
:  	(P4): New define.
:  	(mips_builtin_opcodes): Add msub and msubu instructions for MIPS32.
:  	(mips_builtin_opcodes): Add madd/maddu aliases for mad/madu for MIPS32.
:  	(mips_builtin_opcodes): Support wait, deret, eret, movn, pref for MIPS32.
: 	(mips_builtin_opcodes): Support tlbp, tlbr, tlbwi, tlbwr.
:  
:  	* mips-dis.c (print_insn_arg): Print sdbbp 'm' args.
:  	(print_insn_arg): Handle 'H' args.
:  	(set_mips_isa_type): Recognize 4K.
: 	* mips-dis.c: Use CPU_* defines instead of hardcoded numbers.