[Patch,AVR] PR54222: Add fixed point support

[Georg-Johann Lay]

This patch adds fixed point support to the avr target.

It's based on the work of Sean, see

 http://lists.gnu.org/archive/html/avr-gcc-list/2012-07/msg00030.html

This patch has several changes compared to Sean's patch:

* Additions and subtractions are merged with the existing INT_MODE
  operations by means of mode iterator.  The output routines are
  generic enough to handle fixed-point, too.  The changes were minimal,
  e.g. some new fixed-point constraints.

* Similar for shifts and comparisons.

* The patch neither implements NEG nor ABS.  The standard requires
  saturation, e.g. -0x80 must not become 0x80 again.  This is work
  still to be done, but just a matter of optimization.

* avr-modes.def adjusts TAmode and UTAmode to be 64-bit modes.
  The GCC default with 128 bits is too extreme for AVR.
  Besides that, the libgcc machinery won't generate TA/UTA functions
  because it thinks the mode it too big (is does not code for
  ADJUST_BYTESIZE) and the respective functions are empty.
  TA/UTA have 48 fractional bits so that the user can pick
  2 different resolutions of 64-bit Accum types.

* To make TA/UTA work, avr-lib.h needs some hand-made defines.

* There are no middle-end changes.  The original patch changed:
  rtl.h, cse.c, fold-const.c, varasm.c.

The patch works out fine.  However, because of PR53923 which
shreds the AVR port, currently no reasonable testing is possible.

Work to be done is better testing after PR53923 is fixed and the
AVR port works properly again.  And there are many possible tweaks,
e.g. the above mentioned NEG implementation etc, but that can be
done later.

Ok for trunk?

Johann


libgcc/
	PR target/54222
	* config/avr/lib1funcs-fixed.S: New file.
	* config/avr/lib1funcs.S: Include it.  Undefine some divmodsi
	after they are used.
	(neg2, neg4): New macros.
	* config/avr/avr-lib.h (TA, UTA): Adjust according to gcc's
	avr-modes.def.
	* config/avr/t-avr (LIB1ASMFUNCS): Add: _fractqqsf, _fractuqqsf,
	_fracthqsf, _fractuhqsf, _fracthasf, _fractuhasf, _fractsasf,
	_fractusasf, _fractsfqq, _fractsfuqq, _fractsfhq, _fractsfuhq,
	_fractsfha, _fractsfsa, _mulqq3, _muluqq3, _mulhq3, _muluhq3,
	_mulha3, _muluha3, _mulsa3, _mulusa3, _divqq3, _udivuqq3, _divhq3,
	_udivuhq3, _divha3, _udivuha3, _divsa3, _udivusa3.

gcc/
	PR target/54222
	* avr-modes.def (HA, SA, DA, TA, UTA): Adjust modes.
	* avr/avr-fixed.md: New file.
	* avr/avr.md: Include it.
	(cc): Add: minus.
	(adjust_len): Add: minus, minus64, ufract, sfract.
	(ALL1, ALL2, ALL4, ORDERED234): New mode iterators.
	(MOVMODE): Add: QQ, UQQ, HQ, UHQ, HA, UHA, SQ, USQ, SA, USA.
	(MPUSH): Add: HQ, UHQ, HA, UHA, SQ, USQ, SA, USA.
	(pushqi1, xload8_A, xload_8, movqi_insn, *reload_inqi, addqi3,
	subqi3, ashlqi3, *ashlqi3, ashrqi3, lshrqi3, *lshrqi3, *cmpqi,
	cbranchqi4, *cpse.eq): Generalize to handle all 8-bit modes in ALL1.
	(*movhi, reload_inhi, addhi3, *addhi3, addhi3_clobber, subhi3,
	ashlhi3, *ashlhi3_const, ashrhi3, *ashirhi3_const, lshrhi3,
	*lshrhi3_const, *cmphi, cbranchhi4): Generalize to handle all
	16-bit modes in ALL2.
	(subhi3, casesi, strlenhi): Add clobber when expanding minus:HI.
	(*movsi, *reload_insi, addsi3, subsi3, ashlsi3, *ashlsi3_const,
	ashrsi3, *ashrhi3_const, *ashrsi3_const, lshrsi3, *lshrsi3_const,
	*reversed_tstsi, *cmpsi, cbranchsi4): Generalize to handle all
	32-bit modes in ALL4.
	* avr-dimode.md (ALL8): New mode iterator.
	(adddi3, adddi3_insn, adddi3_const_insn, subdi3, subdi3_insn,
	subdi3_const_insn, cbranchdi4, compare_di2,
	compare_const_di2, ashrdi3, lshrdi3, rotldi3, ashldi3_insn,
	ashrdi3_insn, lshrdi3_insn, rotldi3_insn): Generalize to handle
	all 64-bit modes in ALL8.
	* config/avr/avr-protos.h (avr_to_int_mode): New prototype.
	(avr_out_fract, avr_out_minus, avr_out_minus64): New prototypes.
	* config/avr/avr.c (TARGET_FIXED_POINT_SUPPORTED_P): Return true.
	(avr_scalar_mode_supported_p): Allow if ALL_FIXED_POINT_MODE_P.
	(avr_builtin_setjmp_frame_value): Use gen_subhi3 and return new
	pseudo instead of gen_rtx_MINUS.
	(avr_print_operand, avr_operand_rtx_cost): Handle: CONST_FIXED.
	(notice_update_cc): Handle: CC_MINUS.
	(output_movqi): Generalize to handle respective fixed-point modes.
	(output_movhi, output_movsisf, avr_2word_insn_p): Ditto.
	(avr_out_compare, avr_out_plus_1): Also handle fixed-point modes.
	(avr_assemble_integer): Ditto.
	(output_reload_in_const, output_reload_insisf): Ditto.
	(avr_out_fract, avr_out_minus, avr_out_minus64): New functions.
	(avr_to_int_mode): New function.
	(adjust_insn_length): Handle: ADJUST_LEN_SFRACT,
	ADJUST_LEN_UFRACT, ADJUST_LEN_MINUS, ADJUST_LEN_MINUS64.
	* config/avr/predicates.md (const0_operand): Allow const_fixed.
	(const_operand, const_or_immediate_operand): New.
	(nonmemory_or_const_operand): New.
	* config/avr/constraints.md (Ynn, Y00, Y01, Y02, Ym1, Ym2, YIJ):
	New constraints.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: fixed-48.diff
Type: text/x-patch
Size: 147887 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/gcc-patches/attachments/20120810/0462a772/attachment.bin>