This is the mail archive of the gdb-patches@sources.redhat.com mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: SH patch 2 (was Re: [RFA] SH: Deprecate deprecated functions, use new frame interface)


Corinna Vinschen writes:
 > On Fri, Sep 12, 2003 at 01:40:58PM -0400, Elena Zannoni wrote:
 > > Seems ok, but....could you split the renaming of insn to inst and
 > > removal of tdep into a separate patch, and repost the 2 patches?
 > > I.e. the renaming/tdep and the real cleanups after that. It's not easy
 > > to see what's going on with this diff.
 > > 
 > > thanks
 > > elena
 > 
 > ...and this is patch two, removing old deprecated functionality in
 > favor of shiny new frame stuff.
 > 
 > Corinna
 > 
 > ChangeLog:
 > 
 >  	* sh-tdep.c (struct frame_extra_info): Remove.
 >  	(struct sh_frame_cache): New structure.
 >  	(GET_SOURCE_REG): New macro extracting source register of an opcode.
 >  	(GET_TARGET_REG): Ditto but target register.
 >  	(GET_PUSHED_REG): Remove.
 >  	(IS_MOV_ARG_TO_REG): New macro.
 >  	(IS_MOV_ARG_TO_IND_R14): New macro.
 >  	(IS_MOV_ARG_TO_IND_R14_WITH_DISP): New macro.
 >  	(IS_MOVW_PCREL_TO_REG): New macro.
 >  	(IS_MOVL_PCREL_TO_REG): New macro.
 >  	(IS_SUB_REG_FROM_SP): New macro.
 >  	(IS_ARG_MOV): Remove.
 >  	(IS_MOV_TO_R14): Remove.
 >  	(IS_RESTORE_FP): New macro.
 >  	(IS_RTS): New macro.
 >  	(IS_LDS): New macro.
 >  	(IS_MOV_FP_SP): New macro.
 >  	(IS_ADD_REG_TO_FP): New macro.
 >  	(IS_ADD_IMM_FP): New macro.
 >  	(sh_skip_prologue_hard_way): Remove.
 >  	(sh_saved_pc_after_call): Remove.
 >  	(sh_frame_chain): Remove.
 >  	(sh_find_callers_reg): Remove.
 >  	(sh_nofp_frame_init_saved_regs): Remove.
 >  	(sh_fp_frame_init_saved_regs): Remove.
 >  	(sh_init_extra_frame_info): Remove.
 >  	(sh_analyze_prologue): New function.
 >  	(sh_skip_prologue): Remove deprecated code.  Rely on new function
 >  	sh_analyze_prologue when after_prologue fails.
 >  	(sh_frame_saved_pc): Remove.
 >  	(sh_alloc_frame_cache): New function.
 >  	(sh_frame_cache): Ditto.
 >  	(sh_frame_prev_register): Ditto.
 >  	(sh_frame_this_id): Ditto.
 >  	(sh_frame_unwind): New structure defining the heuristic frame
 >  	sniffer interface.
 >  	(sh_frame_sniffer): New function.
 >  	(sh_unwind_sp): Ditto.
 >  	(sh_unwind_pc): Ditto.
 >  	(sh_unwind_dummy_id): Ditto.
 >  	(sh_frame_base_address): Ditto.
 >  	(sh_frame_base): New structure defining new frame base code.
 >  	(sh_in_function_epilogue_p): New function.
 >  	(sh_gdbarch_init): Restructure and simplify to eliminate deprecated
 >  	code and to call all new code instead.  Initialize dwarf2 and
 >  	heuristic frame sniffer.


It looks ok, I assume that the sh64 prologue analysis is unaffected,
right?  I.e. the macros that were used for decomposing the
instructions in sh64 case are still there, etc etc.
How are the test results? Did they change?

elena



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]