1 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
3 * configure: Regenerate.
5 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
7 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
8 signed/unsigned int8/16 return values.
9 * src/sparc/v8.S (ffi_call_v8): Likewise.
10 (ffi_closure_v8): Likewise.
12 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
13 Steve Ellcey <sje@cup.hp.com>
15 * configure: Regenerate for new libtool.
17 * include/Makefile.in: Ditto.
20 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
22 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
23 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
24 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
25 Adjust copyright notice.
26 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
27 register or two register to use for FFI_SYSV structs.
28 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
29 (ffi_closure_helper_SYSV): Fix the return type for
30 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
31 Adjust copyright notice.
33 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
35 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
38 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
40 * configure: Regenerate.
41 * include/Makefile.in: Regenerate.
42 * testsuite/Makefile.in: Regenerate.
44 2008-06-07 Joseph Myers <joseph@codesourcery.com>
46 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
47 powerpc-*-beos*): Remove.
48 * configure: Regenerate.
50 2008-05-09 Julian Brown <julian@codesourcery.com>
52 * Makefile.am (LTLDFLAGS): New.
53 (libffi_la_LDFLAGS): Use above.
54 * Makefile.in: Regenerate.
56 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
59 * aclocal.m4: Regenerate.
60 * configure: Regenerate.
62 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
64 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
65 * src/sh64/sysv.S: Likewise.
67 2008-03-26 Daniel Jacobowitz <dan@debian.org>
69 * src/arm/sysv.S: Fix ARM comment marker.
71 2008-03-26 Jakub Jelinek <jakub@redhat.com>
73 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
74 * src/s390/sysv.S: Likewise.
75 * src/powerpc/ppc_closure.S: Likewise.
76 * src/powerpc/sysv.S: Likewise.
77 * src/x86/unix64.S: Likewise.
78 * src/x86/sysv.S: Likewise.
79 * src/sparc/v8.S: Likewise.
80 * src/sparc/v9.S: Likewise.
81 * src/m68k/sysv.S: Likewise.
82 * src/arm/sysv.S: Likewise.
84 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
86 * aclocal.m4: Regenerate.
87 * configure: Likewise.
88 * Makefile.in: Likewise.
89 * include/Makefile.in: Likewise.
90 * testsuite/Makefile.in: Likewise.
92 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
93 Andreas Tobler <a.tobler@schweiz.org>
95 * configure.ac: Add amd64-*-freebsd* target.
96 * configure: Regenerate.
98 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
101 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
104 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
107 2008-01-24 David Edelsohn <edelsohn@gnu.org>
109 * configure: Regenerate.
111 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
113 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
115 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
118 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
119 signed/unsigned int8/16 for X86_DARWIN.
120 Updated copyright info.
121 Handle one and two byte structs with special cif->flags.
122 * src/x86/ffitarget.h: Add special types for one and two byte structs.
123 Updated copyright info.
124 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
126 Remove code to pop args from the stack after call.
127 Special-case signed/unsigned for int8/16, one and two byte structs.
128 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
129 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
131 Updated copyright info.
133 2007-12-08 David Daney <ddaney@avtrex.com>
135 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
136 SUBU, add with ADDU and use smaller code sequences.
138 2007-12-07 David Daney <ddaney@avtrex.com>
140 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
143 2007-12-06 David Daney <ddaney@avtrex.com>
145 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
147 (ffi_java_raw): New typedef.
148 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
149 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
151 (ffi_java_raw_closure) : Same.
152 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
154 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
156 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
157 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
158 sizeof(ffi_java_raw) for alignment calculations.
159 (ffi_java_ptrarray_to_raw): Same.
160 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
161 if FFI_SIZEOF_JAVA_RAW == 4.
162 (ffi_java_raw_to_rvalue): Same.
163 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
164 (ffi_java_translate_args): Same.
165 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
167 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
169 2007-12-06 David Daney <ddaney@avtrex.com>
171 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
174 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
177 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
178 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
179 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
180 set the NUM_FPR_ARG_REGISTERS according to.
181 Add support for potential soft-float support under hard-float
183 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
184 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
185 to the FFI_LINUX_SOFT_FLOAT ABI.
186 (ffi_prep_cif_machdep): Likewise.
187 (ffi_closure_helper_SYSV): Likewise.
188 * src/powerpc/ppc_closure.S: Make sure not to store float/double
189 on archs where __NO_FPRS__ is true.
190 Add FFI_TYPE_UINT128 support.
191 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
192 Adjust copyright notice.
194 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
196 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
197 * include/ffi_common.h: ... here.
200 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
202 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
203 * src/powerpc/linux64.S: Likewise.
204 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
205 * testsuite/libffi.call/return_ldl.c: New test.
207 2007-09-04 <aph@redhat.com>
209 * src/arm/sysv.S (UNWIND): New.
210 (Whole file): Conditionally compile unwinder directives.
211 * src/arm/sysv.S: Add unwinder directives.
213 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
214 Only treat r0 as a struct address if we're actually returning a
216 Only copy the bytes that are actually within a struct.
217 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
218 is returned in r0, not passed by address.
219 (ffi_call): Allocate a word-sized temporary for the case where
220 a composite is returned in r0.
221 (ffi_prep_incoming_args_SYSV): Align as necessary.
223 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
225 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
226 directly using the sys_cacheflush syscall.
228 2007-07-27 Andrew Haley <aph@redhat.com>
230 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
232 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
234 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
235 * configure.ac: Likewise.
236 * Makefile.in: Regenerate.
237 * include/Makefile.in: Likewise.
238 * testsuite/Makefile.in: Likewise.
239 * configure: Likewise.
241 2007-08-24 David Daney <ddaney@avtrex.com>
243 * testsuite/libffi.call/return_sl.c: New test.
245 2007-08-10 David Daney <ddaney@avtrex.com>
247 * testsuite/libffi.call/cls_multi_ushort.c,
248 testsuite/libffi.call/cls_align_uint16.c,
249 testsuite/libffi.call/nested_struct1.c,
250 testsuite/libffi.call/nested_struct3.c,
251 testsuite/libffi.call/cls_7_1_byte.c,
252 testsuite/libffi.call/nested_struct5.c,
253 testsuite/libffi.call/cls_double.c,
254 testsuite/libffi.call/nested_struct7.c,
255 testsuite/libffi.call/cls_sint.c,
256 testsuite/libffi.call/nested_struct9.c,
257 testsuite/libffi.call/cls_20byte1.c,
258 testsuite/libffi.call/cls_multi_sshortchar.c,
259 testsuite/libffi.call/cls_align_sint64.c,
260 testsuite/libffi.call/cls_3byte2.c,
261 testsuite/libffi.call/cls_multi_schar.c,
262 testsuite/libffi.call/cls_multi_uchar.c,
263 testsuite/libffi.call/cls_19byte.c,
264 testsuite/libffi.call/cls_9byte1.c,
265 testsuite/libffi.call/cls_align_float.c,
266 testsuite/libffi.call/closure_fn1.c,
267 testsuite/libffi.call/problem1.c,
268 testsuite/libffi.call/closure_fn3.c,
269 testsuite/libffi.call/cls_sshort.c,
270 testsuite/libffi.call/closure_fn5.c,
271 testsuite/libffi.call/cls_align_double.c,
272 testsuite/libffi.call/nested_struct.c,
273 testsuite/libffi.call/cls_2byte.c,
274 testsuite/libffi.call/nested_struct10.c,
275 testsuite/libffi.call/cls_4byte.c,
276 testsuite/libffi.call/cls_6byte.c,
277 testsuite/libffi.call/cls_8byte.c,
278 testsuite/libffi.call/cls_multi_sshort.c,
279 testsuite/libffi.call/cls_align_sint16.c,
280 testsuite/libffi.call/cls_align_uint32.c,
281 testsuite/libffi.call/cls_20byte.c,
282 testsuite/libffi.call/cls_float.c,
283 testsuite/libffi.call/nested_struct2.c,
284 testsuite/libffi.call/cls_5_1_byte.c,
285 testsuite/libffi.call/nested_struct4.c,
286 testsuite/libffi.call/cls_24byte.c,
287 testsuite/libffi.call/nested_struct6.c,
288 testsuite/libffi.call/cls_64byte.c,
289 testsuite/libffi.call/nested_struct8.c,
290 testsuite/libffi.call/cls_uint.c,
291 testsuite/libffi.call/cls_multi_ushortchar.c,
292 testsuite/libffi.call/cls_schar.c,
293 testsuite/libffi.call/cls_uchar.c,
294 testsuite/libffi.call/cls_align_uint64.c,
295 testsuite/libffi.call/cls_ulonglong.c,
296 testsuite/libffi.call/cls_align_longdouble.c,
297 testsuite/libffi.call/cls_1_1byte.c,
298 testsuite/libffi.call/cls_12byte.c,
299 testsuite/libffi.call/cls_3_1byte.c,
300 testsuite/libffi.call/cls_3byte1.c,
301 testsuite/libffi.call/cls_4_1byte.c,
302 testsuite/libffi.call/cls_6_1_byte.c,
303 testsuite/libffi.call/cls_16byte.c,
304 testsuite/libffi.call/cls_18byte.c,
305 testsuite/libffi.call/closure_fn0.c,
306 testsuite/libffi.call/cls_9byte2.c,
307 testsuite/libffi.call/closure_fn2.c,
308 testsuite/libffi.call/closure_fn4.c,
309 testsuite/libffi.call/cls_ushort.c,
310 testsuite/libffi.call/closure_fn6.c,
311 testsuite/libffi.call/cls_5byte.c,
312 testsuite/libffi.call/cls_align_pointer.c,
313 testsuite/libffi.call/cls_7byte.c,
314 testsuite/libffi.call/cls_align_sint32.c,
315 testsuite/libffi.special/unwindtest_ffi_call.cc,
316 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
318 2007-08-10 David Daney <ddaney@avtrex.com>
321 * configure.ac: Don't treat mips64 as a special case.
322 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
323 * configure: Regenerate
324 * Makefile.in: Ditto.
325 * fficonfig.h.in: Ditto.
326 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
327 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
328 (FFI_DEFAULT_ABI): Set for n64 case.
329 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
330 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
331 (ffi_closure_N32): New function.
332 (.eh_frame): New section
333 * src/mips/o32.S: Clean up comments.
334 (ffi_closure_O32): Pass ffi_closure parameter in $12.
335 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
336 _MIPS_SIM == _ABIN32 throughout.
337 (FFI_MIPS_STOP_HERE): New, use in place of
339 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
340 to support n32/n64 ABIs.
341 (calc_n32_struct_flags): Rewrite.
342 (calc_n32_return_struct_flags): Remove unused variable. Reverse
343 position of flag bits.
344 (ffi_prep_cif_machdep): Rewrite n32 portion.
345 (ffi_call): Enable for n64. Add special handling for small structure
347 (ffi_prep_closure_loc): Add n32 and n64 support.
348 (ffi_closure_mips_inner_O32): Add cast to silence warning.
349 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
351 2007-08-08 David Daney <ddaney@avtrex.com>
353 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
354 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
356 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
357 * testsuite/libffi.call/cls_sint.c (main): Ditto.
358 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
359 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
360 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
361 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
362 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
363 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
364 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
365 * testsuite/libffi.call/return_fl3.c (main): Ditto.
366 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
367 * testsuite/libffi.call/nested_struct.c (main): Ditto.
368 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
369 * testsuite/libffi.call/return_ll1.c (main): Ditto.
370 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
371 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
372 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
373 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
374 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
375 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
376 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
377 * testsuite/libffi.call/cls_uint.c (main): Ditto.
378 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
379 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
380 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
381 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
382 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
383 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
384 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
385 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
386 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
388 2007-08-07 Andrew Haley <aph@redhat.com>
390 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
393 2007-08-06 Andrew Haley <aph@redhat.com>
396 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
397 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
400 2007-08-02 David Daney <ddaney@avtrex.com>
402 * testsuite/libffi.call/return_ul.c (main): Define return type as
403 ffi_arg. Use proper printf conversion specifier.
405 2007-07-30 Andrew Haley <aph@redhat.com>
408 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
409 signed/unsigned int8/16.
410 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
412 Remove code to pop args from the stack after call.
413 Special-case signed/unsigned int8/16.
414 * testsuite/libffi.call/return_sc.c (main): Revert.
416 2007-07-26 Richard Guenther <rguenther@suse.de>
419 * testsuite/libffi.call/return_sc.c (main): Verify call
420 result as signed char, not ffi_arg.
422 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
424 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
425 * configure: Regenerate.
427 2007-07-11 David Daney <ddaney@avtrex.com>
429 * src/mips/ffi.c: Don't include sys/cachectl.h.
430 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
433 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
435 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
436 from (ffi_prep_closure): ... this.
437 (FFI_INIT_TRAMPOLINE): Adjust.
439 2005-12-31 Phil Blundell <pb@reciva.com>
441 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
442 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
443 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
444 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
445 (FFI_CLOSURES): Enable closure support.
447 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
449 * testsuite/libffi.call/cls_multi_ushort.c,
450 testsuite/libffi.call/cls_align_uint16.c,
451 testsuite/libffi.call/nested_struct1.c,
452 testsuite/libffi.call/nested_struct3.c,
453 testsuite/libffi.call/cls_7_1_byte.c,
454 testsuite/libffi.call/cls_double.c,
455 testsuite/libffi.call/nested_struct5.c,
456 testsuite/libffi.call/nested_struct7.c,
457 testsuite/libffi.call/cls_sint.c,
458 testsuite/libffi.call/nested_struct9.c,
459 testsuite/libffi.call/cls_20byte1.c,
460 testsuite/libffi.call/cls_multi_sshortchar.c,
461 testsuite/libffi.call/cls_align_sint64.c,
462 testsuite/libffi.call/cls_3byte2.c,
463 testsuite/libffi.call/cls_multi_schar.c,
464 testsuite/libffi.call/cls_multi_uchar.c,
465 testsuite/libffi.call/cls_19byte.c,
466 testsuite/libffi.call/cls_9byte1.c,
467 testsuite/libffi.call/cls_align_float.c,
468 testsuite/libffi.call/closure_fn1.c,
469 testsuite/libffi.call/problem1.c,
470 testsuite/libffi.call/closure_fn3.c,
471 testsuite/libffi.call/cls_sshort.c,
472 testsuite/libffi.call/closure_fn5.c,
473 testsuite/libffi.call/cls_align_double.c,
474 testsuite/libffi.call/cls_2byte.c,
475 testsuite/libffi.call/nested_struct.c,
476 testsuite/libffi.call/nested_struct10.c,
477 testsuite/libffi.call/cls_4byte.c,
478 testsuite/libffi.call/cls_6byte.c,
479 testsuite/libffi.call/cls_8byte.c,
480 testsuite/libffi.call/cls_multi_sshort.c,
481 testsuite/libffi.call/cls_align_uint32.c,
482 testsuite/libffi.call/cls_align_sint16.c,
483 testsuite/libffi.call/cls_float.c,
484 testsuite/libffi.call/cls_20byte.c,
485 testsuite/libffi.call/cls_5_1_byte.c,
486 testsuite/libffi.call/nested_struct2.c,
487 testsuite/libffi.call/cls_24byte.c,
488 testsuite/libffi.call/nested_struct4.c,
489 testsuite/libffi.call/nested_struct6.c,
490 testsuite/libffi.call/cls_64byte.c,
491 testsuite/libffi.call/nested_struct8.c,
492 testsuite/libffi.call/cls_uint.c,
493 testsuite/libffi.call/cls_multi_ushortchar.c,
494 testsuite/libffi.call/cls_schar.c,
495 testsuite/libffi.call/cls_uchar.c,
496 testsuite/libffi.call/cls_align_uint64.c,
497 testsuite/libffi.call/cls_ulonglong.c,
498 testsuite/libffi.call/cls_align_longdouble.c,
499 testsuite/libffi.call/cls_1_1byte.c,
500 testsuite/libffi.call/cls_12byte.c,
501 testsuite/libffi.call/cls_3_1byte.c,
502 testsuite/libffi.call/cls_3byte1.c,
503 testsuite/libffi.call/cls_4_1byte.c,
504 testsuite/libffi.call/cls_6_1_byte.c,
505 testsuite/libffi.call/cls_16byte.c,
506 testsuite/libffi.call/cls_18byte.c,
507 testsuite/libffi.call/closure_fn0.c,
508 testsuite/libffi.call/cls_9byte2.c,
509 testsuite/libffi.call/closure_fn2.c,
510 testsuite/libffi.call/closure_fn4.c,
511 testsuite/libffi.call/cls_ushort.c,
512 testsuite/libffi.call/closure_fn6.c,
513 testsuite/libffi.call/cls_5byte.c,
514 testsuite/libffi.call/cls_align_pointer.c,
515 testsuite/libffi.call/cls_7byte.c,
516 testsuite/libffi.call/cls_align_sint32.c,
517 testsuite/libffi.special/unwindtest_ffi_call.cc,
518 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
520 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
522 * aclocal.m4: Regenerated.
524 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
526 * configure: Regenerate.
528 2007-05-23 Steve Ellcey <sje@cup.hp.com>
530 * Makefile.in: Regenerate.
531 * configure: Regenerate.
532 * aclocal.m4: Regenerate.
533 * include/Makefile.in: Regenerate.
534 * testsuite/Makefile.in: Regenerate.
536 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
538 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
539 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
540 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
541 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
542 (FFI_CLOSURES): Enable closure support.
544 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
546 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
547 * configure: Regenerate.
548 * fficonfig.h.in: Regenerate.
549 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
550 CFI_OFFSET,CFI_DEF_CFA): New macros.
551 (ffi_call_SYSV): Add callframe annotation.
553 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
555 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
556 numerous test suite failures.
557 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
559 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
561 * Makefile.am (EXTRA_DIST): Bring up to date.
562 * Makefile.in: Regenerate.
563 * src/frv/eabi.S: Remove RCS keyword.
565 2007-04-06 Richard Henderson <rth@redhat.com>
567 * configure.ac: Tidy target case.
568 (HAVE_LONG_DOUBLE): Allow the target to override.
569 * configure: Regenerate.
570 * include/ffi.h.in: Don't define ffi_type_foo if
571 LIBFFI_HIDE_BASIC_TYPES is defined.
572 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
574 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
575 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
576 (ffi_type_longdouble): Special case for Alpha. Don't define
577 if long double == double.
579 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
580 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
581 (ffi_call, ffi_closure_osf_inner): Likewise.
582 (ffi_closure_osf_inner): Likewise. Mark hidden.
583 (ffi_call_osf, ffi_closure_osf): Mark hidden.
584 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
585 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
586 (load_table): Handle 128-bit long double.
588 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
590 2007-04-06 Tom Tromey <tromey@redhat.com>
593 * README: Fixed bug in example.
595 2007-04-03 Jakub Jelinek <jakub@redhat.com>
597 * src/closures.c: Include sys/statfs.h.
598 (_GNU_SOURCE): Define on Linux.
599 (FFI_MMAP_EXEC_SELINUX): Define.
600 (selinux_enabled): New variable.
601 (selinux_enabled_check): New function.
602 (is_selinux_enabled): Define.
605 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
607 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
608 Use 'volatile float sum' to create sum of floats to avoid false
609 negative due to excess precision on ix86 targets.
612 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
614 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
616 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
618 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
620 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
621 (ffi_prep_closure_loc): New.
622 (ffi_prep_raw_closure_loc): New.
623 (ffi_prep_java_raw_closure_loc): New.
624 * src/closures.c: New file.
625 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
626 Replace sflags with exec_offset.
627 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
628 sub_segment_exec_offset): New macros.
629 (get_segment_flags, set_segment_flags, check_segment_merge): New
631 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
632 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
633 destroy_mspace): Use new macros.
634 (sys_alloc): Silence warning.
635 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
636 * Makefile.in: Rebuilt.
637 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
638 terms of ffi_prep_closure_loc.
639 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
641 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
643 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
645 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
647 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
648 (ffi_prep_closure): ... this.
649 * src/pa/ffi.c: Likewise.
650 * src/cris/ffi.c: Likewise. Adjust.
651 * src/frv/ffi.c: Likewise.
652 * src/ia64/ffi.c: Likewise.
653 * src/mips/ffi.c: Likewise.
654 * src/powerpc/ffi_darwin.c: Likewise.
655 * src/s390/ffi.c: Likewise.
656 * src/sh/ffi.c: Likewise.
657 * src/sh64/ffi.c: Likewise.
658 * src/sparc/ffi.c: Likewise.
659 * src/x86/ffi64.c: Likewise.
660 * src/x86/ffi.c: Likewise.
661 (FFI_INIT_TRAMPOLINE): Adjust.
662 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
663 (ffi_prep_raw_closure): ... this.
664 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
665 (ffi_prep_closure): ... this.
666 (flush_icache): Adjust.
668 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
670 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
673 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
675 * Makefile.am: Add dummy install-pdf target.
676 * Makefile.in: Regenerate
678 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
680 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
681 ffi_closure_helper_SYSV): Add long double handling.
683 2007-02-02 Jakub Jelinek <jakub@redhat.com>
685 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
686 immediately after bctrl instruction.
688 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
690 * Makefile.am (all-recursive, install-recursive,
691 mostlyclean-recursive, clean-recursive, distclean-recursive,
692 maintainer-clean-recursive): Add missing targets.
693 * Makefile.in: Rebuilt.
695 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
697 * configure.ac: Add TARGET for x86_64-*-darwin*.
698 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
700 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
701 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
702 * configure: Regenerate.
703 * Makefile.in: Regenerate.
704 * include/Makefile.in: Regenerate.
705 * testsuite/Makefile.in: Regenerate.
706 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
709 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
711 * aclocal.m4: Regenerate with aclocal -I .. as written in the
714 2006-10-31 Geoffrey Keating <geoffk@apple.com>
716 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
717 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
719 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
720 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
722 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
723 Sandro Tolaini <tolaini@libero.it>
725 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
727 * configure: Regenerated.
728 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
729 (EXTRA_DIST): Add src/x86/darwin.S.
730 * Makefile.in: Regenerated.
731 * include/Makefile.in: Regenerated.
732 * testsuite/Makefile.in: Regenerated.
734 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
735 X86_WIN32, and additionally align stack to 16 bytes.
736 * src/x86/darwin.S: New, based on sysv.S.
737 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
739 2006-09-12 David Daney <ddaney@avtrex.com>
742 * include/Makefile.am: Install both ffi.h and ffitarget.h in
743 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
744 * aclocal.m4: Regenerated for automake 1.9.6.
745 * Makefile.in: Regenerated.
746 * include/Makefile.in: Regenerated.
747 * testsuite/Makefile.in: Regenerated.
749 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
751 * include/ffi_common.h (struct): Revert accidental commit.
753 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
755 * include/ffi_common.h: Remove lint directives.
756 * include/ffi.h.in: Likewise.
758 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
760 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
761 for 32-bit architectures.
762 * testsuite/libffi.call/return_ul.c: New test case.
764 2006-07-19 David Daney <ddaney@avtrex.com>
766 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
767 xfail remains for mips64.
769 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
771 * Makefile.am: Add install-html target. Add install-html to .PHONY
772 * Makefile.in: Regenerate.
773 * aclocal.m4: Regenerate.
774 * include/Makefile.in: Regenerate.
775 * testsuite/Makefile.in: Regenerate.
777 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
779 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
782 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
784 * README: Remove notice about 'Crazy Comments'.
785 * src/debug.c: Remove lint directives. Cleanup white spaces.
786 * src/java_raw_api.c: Likewise.
787 * src/prep_cif.c: Likewise.
788 * src/raw_api.c: Likewise.
789 * src/ffitest.c: Delete. No longer needed, all test cases migrated
791 * src/arm/ffi.c: Remove lint directives.
792 * src/m32r/ffi.c: Likewise.
793 * src/pa/ffi.c: Likewise.
794 * src/powerpc/ffi.c: Likewise.
795 * src/powerpc/ffi_darwin.c: Likewise.
796 * src/sh/ffi.c: Likewise.
797 * src/sh64/ffi.c: Likewise.
798 * src/x86/ffi.c: Likewise.
799 * testsuite/libffi.call/float2.c: Likewise.
800 * testsuite/libffi.call/promotion.c: Likewise.
801 * testsuite/libffi.call/struct1.c: Likewise.
803 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
805 * src/pa/hpux32.S: Correct unwind offset calculation for
807 * src/pa/linux.S: Likewise.
809 2006-04-12 James E Wilson <wilson@specifix.com>
812 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
813 (hfa_type_load): Call stf_spill.
814 (hfa_type_store): Call ldf_fill.
815 (ffi_call): Adjust calls to above routines. Add local temps for
818 2006-04-10 Matthias Klose <doko@debian.org>
820 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
821 directory names containing underscores.
823 2006-04-07 James E Wilson <wilson@specifix.com>
825 * testsuite/libffi.call/float4.c: New testcase.
827 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
828 Andreas Tobler <a.tobler@schweiz.ch>
830 * Makefile.am: Add PA_HPUX port.
831 * Makefile.in: Regenerate.
832 * include/Makefile.in: Likewise.
833 * testsuite/Makefile.in: Likewise.
834 * configure.ac: Add PA_HPUX rules.
835 * configure: Regenerate.
836 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
837 Add PA_HPUX and PA64_HPUX.
838 Rename FFI_LINUX ABI to FFI_PA32 ABI.
839 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
840 (FFI_TYPE_SMALL_STRUCT2): Define.
841 (FFI_TYPE_SMALL_STRUCT4): Likewise.
842 (FFI_TYPE_SMALL_STRUCT8): Likewise.
843 (FFI_TYPE_SMALL_STRUCT3): Redefine.
844 (FFI_TYPE_SMALL_STRUCT5): Likewise.
845 (FFI_TYPE_SMALL_STRUCT6): Likewise.
846 (FFI_TYPE_SMALL_STRUCT7): Likewise.
847 * src/pa/ffi.c (ROUND_DOWN): Delete.
848 (fldw, fstw, fldd, fstd): Use '__asm__'.
849 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
850 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
851 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
852 Simplify incrementing of stack slot variable. Change type of local
854 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
856 (ffi_prep_cif_machdep): Likewise.
857 (ffi_call): Likewise.
858 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
859 return type to ffi_status. Simplify incrementing of stack slot
860 variable. Only copy floating point argument registers when PA_LINUX
861 is true. Reformat debug statement.
862 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
863 FFI_TYPE_SMALL_STRUCT8.
864 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
866 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
867 Add nops to cache flush. Add trampoline for PA_HPUX.
868 * src/pa/hpux32.S: New file.
869 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
870 ffi_prep_args_LINUX to ffi_prep_args_pa32.
871 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
872 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
873 argument type checks so that common argument types appear first.
874 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
875 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
877 2006-03-24 Alan Modra <amodra@bigpond.net.au>
879 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
880 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
881 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
883 (ffi_prep_args64): Assert using IBM extended double.
884 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
885 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
886 (ffi_call): Handle FFI_LINUX.
887 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
888 gpr3 return pointer as for struct return. Handle FFI_LINUX
889 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
891 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
892 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
893 Don't use r6 as pointer to results, instead use sp offset. Don't
894 make a special call to load lr with case table address, instead
895 use offset from previous call.
896 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
897 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
900 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
902 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
903 passed with FP registers correctly.
904 (ffi_closure_helper_SYSV): Likewise.
905 * src/sh64/sysv.S: Likewise.
907 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
909 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
910 args and userdata unused.
911 (closure_test_fn1): Mark cif and userdata unused.
912 (main): Remove unused res.
914 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
916 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
917 -O2, -O3, -Os and the warning flags -W -Wall.
918 * testsuite/libffi.special/special.exp: Likewise.
919 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
920 unused parameter unused for gcc or else do nothing.
921 * testsuite/libffi.special/ffitestcxx.h: Likewise.
922 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
924 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
925 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
926 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
927 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
928 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
929 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
930 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
931 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
932 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
933 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
934 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
935 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
936 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
937 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
938 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
939 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
940 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
941 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
942 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
943 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
944 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
945 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
946 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
947 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
949 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
951 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
953 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
954 void* to avoid compiler warning.
956 (cls_struct_align_gn): Mark cif and userdata unused.
957 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
959 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
961 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
963 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
965 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
967 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
968 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
969 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
971 (main): Cast res_call to silence gcc.
972 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
974 (main): Cast res_call to silence gcc.
975 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
977 (main): Cast res_call to silence gcc.
978 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
980 (main): Cast res_call to silence gcc.
981 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
983 (main): Cast res_call to silence gcc.
984 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
986 (main): Cast res_call to silence gcc.
987 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
989 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
990 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
992 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
993 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
995 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
996 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
998 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
999 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1001 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1002 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1003 and userdata unused.
1004 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1006 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1007 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1008 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1009 Cleanup white spaces.
1010 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1011 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1012 cif and userdata unused.
1013 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1015 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1016 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1017 formatters to silence gcc.
1018 (B_gn): Mark cif and userdata unused.
1019 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1021 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1022 (B_gn): Mark cif and userdata unused.
1023 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1025 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1026 (B_gn): Mark cif and userdata unused.
1027 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1029 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1030 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1031 * testsuite/libffi.call/problem1.c (stub): Likewise.
1032 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1034 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1035 in the last commit for this test case in the test case itself.
1036 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1038 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1039 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1040 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1041 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1042 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1043 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1045 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1047 * src/sh/sysv.S: Fix register numbers in the FDE for
1050 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1052 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1053 declaration to avoid a false negative on ix86. See PR323.
1055 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1057 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1058 and cast integer to void * if needed. Update the pointer to
1059 the FP register saved area correctly.
1061 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1063 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1065 * testsuite/libffi.call/nested_struct4.c: Likewise.
1067 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1069 * testsuite/libffi.call/return_dbl.c: New test case.
1070 * testsuite/libffi.call/return_dbl1.c: Likewise.
1071 * testsuite/libffi.call/return_dbl2.c: Likewise.
1072 * testsuite/libffi.call/return_fl.c: Likewise.
1073 * testsuite/libffi.call/return_fl1.c: Likewise.
1074 * testsuite/libffi.call/return_fl2.c: Likewise.
1075 * testsuite/libffi.call/return_fl3.c: Likewise.
1076 * testsuite/libffi.call/closure_fn6.c: Likewise.
1078 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1080 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1081 here to be used by other test cases too.
1083 * testsuite/libffi.call/nested_struct10.c: New test case.
1084 * testsuite/libffi.call/nested_struct9.c: Likewise.
1085 * testsuite/libffi.call/nested_struct8.c: Likewise.
1086 * testsuite/libffi.call/nested_struct7.c: Likewise.
1087 * testsuite/libffi.call/nested_struct6.c: Likewise.
1088 * testsuite/libffi.call/nested_struct5.c: Likewise.
1089 * testsuite/libffi.call/nested_struct4.c: Likewise.
1091 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1093 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1094 * configure: Rebuilt.
1096 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1098 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1099 instead do the shifting inline.
1100 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1101 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1102 and 8 byte structs, for the remaining struct sizes don't call
1103 __lshrdi3, instead do the shifting inline.
1105 2005-12-07 Thiemo Seufer <ths@networkno.de>
1107 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1108 missing parentheses.
1109 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1110 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1111 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1112 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1113 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1115 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1117 (ffi_prep_closure): Improve trampoline instruction scheduling.
1118 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1120 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1122 * src/powerpc/ffi.c: Formatting.
1123 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1124 (ffi_prep_args64): Likewise.
1126 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1128 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1129 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1131 2005-09-26 Tom Tromey <tromey@redhat.com>
1133 * testsuite/libffi.call/float1.c (value_type): New typedef.
1134 (CANARY): New define.
1135 (main): Check for result buffer overflow.
1136 * src/powerpc/linux64.S: Handle linux64 long double returns.
1137 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1138 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1140 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1143 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1145 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1147 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1149 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1150 (AH_BOTTOM): Add FFI_HIDDEN definition.
1151 * configure: Rebuilt.
1152 * fficonfig.h.in: Rebuilt.
1153 * src/powerpc/ffi.c (hidden): Remove.
1154 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1155 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1156 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1157 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1158 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1159 add FFI_HIDDEN to its prototype.
1160 (ffi_closure_SYSV_inner): New.
1161 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1162 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1164 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1167 * configure: Rebuilt.
1168 * configure.ac: Handle i*86-*-gnu*.
1170 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1172 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1173 DW_CFA_offset_extended_sf rather than
1174 DW_CFA_GNU_negative_offset_extended.
1175 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1177 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1179 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1181 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1182 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1183 partially on register.
1184 (ffi_closure_helper_SYSV): Likewise.
1185 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1187 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1189 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1191 * src/sh64/ffi.c (simple_type): Remove.
1192 (return_type): Handle small structures correctly.
1193 (ffi_prep_args): Likewise.
1194 (ffi_call): Likewise.
1195 (ffi_closure_helper_SYSV): Likewise.
1196 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1197 Emit position independent code if PIC and remove wrong datalabel
1198 prefixes from EH data.
1200 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1202 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1203 * Makefile.in: Regenerate.
1204 * include/Makefile.in: Likewise.
1205 * testsuite/Makefile.in: Likewise.
1206 * configure.ac: Add POWERPC_FREEBSD rules.
1207 * configure: Regenerate.
1208 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1209 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1210 * src/powerpc/ffi.c: Add flags to handle small structure returns
1212 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1214 (ffi_closure_helper_SYSV): Likewise.
1215 * src/powerpc/ppc_closure.S: Add return types for small structures.
1216 * src/powerpc/sysv.S: Add bits to handle small structures for
1219 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1221 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1222 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1223 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1225 2005-07-05 Randolph Chung <tausq@debian.org>
1227 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1228 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1229 structures. Kill compilation warnings.
1230 (ffi_closure_inner_LINUX): Print return values as hex in debug
1231 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1232 Properly handle 5-7 byte structure returns.
1233 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1234 (FFI_TYPE_SMALL_STRUCT2): Remove.
1235 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1236 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1237 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1238 (checksmst1, checksmst2): Remove.
1239 (checksmst3): Optimize handling of 3-byte struct returns.
1240 (checksmst567): Properly handle 5-7 byte struct returns.
1242 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1245 * src/mips/n32.S: Enforce PIC code.
1246 * src/mips/o32.S: Likewise.
1248 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1250 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1251 * configure: Regenerate.
1253 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1255 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1256 to call ffi_closure_helper_SYSV. Append @local instead.
1257 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1259 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1261 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1262 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1263 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1264 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1265 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1267 2005-05-09 Mike Stump <mrs@apple.com>
1269 * configure: Regenerate.
1271 2005-05-08 Richard Henderson <rth@redhat.com>
1274 * src/alpha/osf.S: Update unwind into to match code.
1276 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1277 Richard Henderson <rth@redhat.com>
1279 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1281 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1282 (ffi_prep_closure): Set carry bit if sse-used flag set.
1283 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1284 Only load sse registers if ssecount non-zero.
1285 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1287 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1289 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1290 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1291 * configure: Regenerate.
1293 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1295 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1296 have Tcl8.3-compatible intermediate variable.
1298 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1299 Hans-Peter Nilsson <hp@axis.com>
1301 * Makefile.am: Add CRIS support.
1302 * configure.ac: Likewise.
1303 * Makefile.in, configure, testsuite/Makefile.in,
1304 include/Makefile.in: Regenerate.
1305 * src/cris: New directory.
1306 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1307 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1309 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1310 \r?\n in output tests.
1312 2005-04-12 Mike Stump <mrs@apple.com>
1314 * configure: Regenerate.
1316 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1318 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1320 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1322 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1324 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1325 to 64 bits in ILP32 mode.
1326 Load 64 bits even for short data.
1328 2005-03-23 Mike Stump <mrs@apple.com>
1330 * src/powerpc/darwin.S: Update for -m64 multilib.
1331 * src/powerpc/darwin_closure.S: Likewise.
1333 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1335 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1336 Do not set tool_include_dir.
1337 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1339 * include/Makefile.am: Set gcc_version and toollibffidir.
1340 * include/Makefile.in: Regenerate.
1342 2005-02-22 Andrew Haley <aph@redhat.com>
1344 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1345 odd-numbered register pairs for 64-bit integer types.
1347 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1350 * testsuite/libffi.call/return_ll1.c: New test case.
1352 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1354 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1355 * testsuite/libffi.call/float.c: Ditto.
1356 * testsuite/libffi.call/float2.c: Ditto.
1357 * testsuite/libffi.call/float3.c: Ditto.
1359 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1361 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1363 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1365 * testsuite/libffi.special/special.exp (cxx_options): Add
1368 2004-12-31 Richard Henderson <rth@redhat.com>
1370 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1371 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1372 offset parameters with a type parameter; deduce size and structure
1373 alignment. Update all users.
1375 2004-12-31 Richard Henderson <rth@redhat.com>
1377 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1378 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1379 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1380 into ffi_prep_closure.
1381 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1384 2004-12-27 Richard Henderson <rth@redhat.com>
1386 * src/x86/unix64.S: Fix typo in unwind info.
1388 2004-12-25 Richard Henderson <rth@redhat.com>
1390 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1391 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1392 (merge_classes): Check for it.
1394 (classify_argument): Pass byte_offset by value; perform all updates
1396 (examine_argument): Add classes argument; handle
1397 X86_64_COMPLEX_X87_CLASS.
1398 (ffi_prep_args): Merge into ...
1399 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1400 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1401 (ffi_fill_return_value): Remove.
1402 (ffi_prep_closure): Remove dead assert.
1403 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1404 Rewrite to use struct register_args instead of va_list. Create
1405 flags for handling structure returns.
1406 * src/x86/unix64.S: Remove dead strings.
1407 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1408 stack frame with ffi_call. Handle structure returns properly.
1409 (float2sse, floatfloat2sse, double2sse): Remove.
1410 (sse2float, sse2double, sse2floatfloat): Remove.
1411 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1412 to handle structure returns properly.
1414 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1416 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1418 * Makefile.in: Regenerated.
1420 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1422 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1423 * configure, aclocal.m4, Makefile.in: Regenerate.
1424 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1426 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1428 * configure: Regenerate for libtool change.
1430 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1432 * configure: Regenerate for libtool reversion.
1434 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1436 * configure: Regenerate for libtool change.
1438 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1440 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1442 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1444 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1445 of jal. Use an absolute encoding for the frame information.
1447 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1449 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1450 * acinclude.m4: Delete logic for sincludes.
1451 * aclocal.m4, Makefile.in, configure: Regenerate.
1452 * include/Makefile: Likewise.
1453 * testsuite/Makefile: Likewise.
1455 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1457 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1458 on a 8-byte boundary.
1459 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1461 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1463 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1464 long long values. Round stack allocation to a multiple of 8 bytes
1465 for ATPCS compatibility.
1466 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1467 names. Handle returning long long types. Add Thumb and interworking
1468 support. Improve soft-float code.
1470 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1472 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1473 (libffi_exit): New function.
1474 (libffi_init): Build the testglue wrapper if needed.
1476 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1479 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1481 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1483 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1485 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1487 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1488 * testsuite/libffi.call/float3.c: New test case.
1490 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1492 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1493 the function returning a structure pointed with R2.
1494 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1495 the structure return value if T bit set. Emit position
1496 independent code and EH data if PIC.
1498 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1500 * Makefile.am: Add m32r support.
1501 * configure.ac: Likewise.
1502 * Makefile.in: Regenerate.
1503 * confiugre: Regenerate.
1504 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1505 (uint64, sint64, double, longdouble)
1506 * src/m32r: New directory.
1507 * src/m32r/ffi.c: New file.
1508 * src/m32r/sysv.S: Likewise.
1509 * src/m32r/ffitarget.h: Likewise.
1511 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1513 * testsuite/libffi.call/negint.c: New test case.
1515 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1518 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1519 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1520 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1523 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1525 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1526 * testsuite/libffi.call/promotion.c: Likewise.
1527 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1529 * testsuite/libffi.call/return_sc.c: Likewise.
1530 * testsuite/libffi.call/return_uc.c: Likewise.
1532 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1534 * src/powerpc/darwin.S: Fix comments and identation.
1535 * src/powerpc/darwin_closure.S: Likewise.
1537 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1539 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1540 (ffi_prep_args): Handle longdouble arguments.
1541 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1543 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1544 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1546 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1547 * src/types.c: Defined longdouble size and alignment for darwin.
1549 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1551 * src/powerpc/aix.S: Remove whitespaces.
1552 * src/powerpc/aix_closure.S: Likewise.
1553 * src/powerpc/asm.h: Likewise.
1554 * src/powerpc/ffi.c: Likewise.
1555 * src/powerpc/ffitarget.h: Likewise.
1556 * src/powerpc/linux64.S: Likewise.
1557 * src/powerpc/linux64_closure.S: Likewise.
1558 * src/powerpc/ppc_closure.S: Likewise.
1559 * src/powerpc/sysv.S: Likewise.
1561 2004-08-30 Anthony Green <green@redhat.com>
1563 * Makefile.am: Add frv support.
1564 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1565 * configure.ac: Read configure.host.
1566 * configure.in: Read configure.host.
1567 * configure.host: New file. frv-elf needs libgloss.
1568 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1569 alignment. This is needed to frv and shouldn't harm the others.
1570 * include/ffi_common.h (ALIGN_DOWN): New macro.
1571 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1573 2004-08-24 David Daney <daney@avtrex.com>
1575 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1576 * testsuite/libffi.call/closure_fn1.c: Likewise.
1577 * testsuite/libffi.call/closure_fn2.c Likewise.
1578 * testsuite/libffi.call/closure_fn3.c: Likewise.
1579 * testsuite/libffi.call/closure_fn4.c: Likewise.
1580 * testsuite/libffi.call/closure_fn5.c: Likewise.
1581 * testsuite/libffi.call/cls_18byte.c: Likewise.
1582 * testsuite/libffi.call/cls_19byte.c: Likewise.
1583 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1584 * testsuite/libffi.call/cls_20byte.c: Likewise.
1585 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1586 * testsuite/libffi.call/cls_24byte.c: Likewise.
1587 * testsuite/libffi.call/cls_2byte.c: Likewise.
1588 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1589 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1590 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1591 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1592 * testsuite/libffi.call/cls_4byte.c: Likewise.
1593 * testsuite/libffi.call/cls_64byte.c: Likewise.
1594 * testsuite/libffi.call/cls_6byte.c: Likewise.
1595 * testsuite/libffi.call/cls_7byte.c: Likewise.
1596 * testsuite/libffi.call/cls_8byte.c: Likewise.
1597 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1598 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1599 * testsuite/libffi.call/cls_align_double.c: Likewise.
1600 * testsuite/libffi.call/cls_align_float.c: Likewise.
1601 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1602 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1603 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1604 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1605 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1606 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1607 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1608 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1609 * testsuite/libffi.call/cls_double.c: Likewise.
1610 * testsuite/libffi.call/cls_float.c: Likewise.
1611 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1612 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1613 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1614 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1615 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1616 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1617 * testsuite/libffi.call/cls_schar.c: Likewise.
1618 * testsuite/libffi.call/cls_sint.c: Likewise.
1619 * testsuite/libffi.call/cls_sshort.c: Likewise.
1620 * testsuite/libffi.call/cls_uchar.c: Likewise.
1621 * testsuite/libffi.call/cls_uint.c: Likewise.
1622 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1623 * testsuite/libffi.call/cls_ushort.c: Likewise.
1624 * testsuite/libffi.call/nested_struct.c: Likewise.
1625 * testsuite/libffi.call/nested_struct1.c: Likewise.
1626 * testsuite/libffi.call/nested_struct2.c: Likewise.
1627 * testsuite/libffi.call/nested_struct3.c: Likewise.
1628 * testsuite/libffi.call/problem1.c: Likewise.
1629 * testsuite/libffi.special/unwindtest.cc: Likewise.
1630 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1632 * testsuite/libffi.call/cls_16byte.c: Likewise.
1633 * testsuite/libffi.call/cls_5byte.c: Likewise.
1635 2004-08-23 David Daney <daney@avtrex.com>
1638 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1639 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1640 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1641 parameters and return types.
1642 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1643 (ffi_prep_closure): Ditto.
1644 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1645 alignment calculations.
1646 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1647 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1649 2004-08-14 Casey Marshall <csm@gnu.org>
1651 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1652 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1653 integer (O32 ABI only).
1654 (ffi_prep_closure): new function.
1655 (ffi_closure_mips_inner_O32): new function.
1656 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1657 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1658 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1659 64 bit integers correctly.
1660 (ffi_closure_O32): new function.
1661 Added DWARF-2 unwind info for both functions.
1663 2004-08-10 Andrew Haley <aph@redhat.com>
1665 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1667 2004-08-01 Robert Millan <robertmh@gnu.org>
1669 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1670 * configure: Regenerate.
1672 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1674 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1675 and mmap() explicitly instead of relying on preset autoconf cache
1677 * aclocal.m4: Regenerate.
1678 * configure: Regenerate.
1680 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1682 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1683 (ffi_check_float_struct): Remove unused prototype.
1685 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1687 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1688 character on Darwin, use '\n\t' instead.
1690 2004-06-26 Matthias Klose <doko@debian.org>
1692 * libtool-version: Fix typo in revision/age.
1694 2004-06-17 Matthias Klose <doko@debian.org>
1696 * libtool-version: New.
1697 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1698 * Makefile.in: Regenerate.
1700 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1702 * Makefile.am: Remove useless multilib rules.
1703 * Makefile.in: Regenerate.
1704 * aclocal.m4: Regenerate with automake 1.8.5.
1705 * configure.ac: Remove useless multilib configury.
1706 * configure: Regenerate.
1708 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1710 * .cvsignore: New file.
1712 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1714 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1716 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1719 2004-06-06 Sean McNeil <sean@mcneil.com>
1721 * configure.ac: Add x86_64-*-freebsd* support.
1722 * configure: Regenerate.
1724 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1727 * configure.ac: Test for existence of mmap and sys/mman.h before
1728 checking blacklist. Fix suggested by Jim Wilson.
1729 * configure: Regenerate.
1731 2004-04-26 Matt Austern <austern@apple.com>
1733 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1735 * src/powerpc/darwin_closure.S: Likewise.
1737 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1739 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1740 error. Reported by Thomas Heller <theller@python.net>.
1741 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1742 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1744 2004-03-20 Matthias Klose <doko@debian.org>
1746 * src/pa/linux.S: Fix typo.
1748 2004-03-19 Matthias Klose <doko@debian.org>
1750 * Makefile.am: Update.
1751 * Makefile.in: Regenerate.
1752 * src/pa/ffi.h.in: Remove.
1753 * src/pa/ffitarget.h: New file.
1755 2004-02-10 Randolph Chung <tausq@debian.org>
1757 * Makefile.am: Add PA support.
1758 * Makefile.in: Regenerate.
1759 * include/Makefile.in: Regenerate.
1760 * configure.ac: Add PA target.
1761 * configure: Regenerate.
1762 * src/pa/ffi.c: New file.
1763 * src/pa/ffi.h.in: Add PA support.
1764 * src/pa/linux.S: New file.
1765 * prep_cif.c: Add PA support.
1767 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1769 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1771 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1772 with ecif->cif->flags.
1773 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1775 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1776 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1777 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1778 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1780 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1782 * configure.in: Rename file to ...
1783 * configure.ac: ... this.
1784 * fficonfig.h.in: Regenerate.
1785 * Makefile.in: Regenerate.
1786 * include/Makefile.in: Regenerate.
1787 * testsuite/Makefile.in: Regenerate.
1789 2004-03-12 Matt Austern <austern@apple.com>
1791 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1792 changes in EH format resulting from addition of linkonce support.
1793 * src/powerpc/darwin_closure.S: Likewise.
1795 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1796 Paolo Bonzini <bonzini@gnu.org>
1798 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1799 Remove VPATH. Remove rules for object files. Remove multilib support.
1800 (AM_CCASFLAGS): Add.
1801 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1802 (AC_PREREQ): Bump version to 2.59.
1803 (AC_INIT): Fill with version info and bug address.
1804 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1805 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1806 De-precious CC so that the right flags are passed down to multilibs.
1807 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1808 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1809 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1810 * configure: Rebuilt.
1811 * aclocal.m4: Likewise.
1812 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1813 * fficonfig.h.in: Likewise.
1815 2004-03-11 Andreas Schwab <schwab@suse.de>
1817 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1818 arguments from fp registers only for the first 8 parameter slots.
1819 Don't convert a float parameter when passed in memory.
1821 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1823 * configure: Regenerate for config/accross.m4 correction.
1825 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1827 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1828 ecif->cif->bytes to bytes.
1829 (ffi_prep_cif_machdep): Add braces around nested if statement.
1831 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1833 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1835 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1836 (ffi_closure_helper_LINUX64): Fix typo.
1837 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1839 * testsuite/libffi.call/float.c: Likewise.
1840 * testsuite/libffi.call/float2.c: Likewise.
1842 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1844 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1845 long double function return and long double arg handling.
1846 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1847 Use "end_pfr" instead of "nf". Correct long double handling.
1849 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1851 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1852 space for long double return value. Adjust stack frame and offsets.
1853 Load f2 long double return.
1855 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1857 * src/types.c: Use 16 byte long double for POWERPC64.
1859 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1861 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1862 when the structure return address is passed in %o0.
1863 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1864 (ffi_v9_layout_struct): Align the field following a nested structure
1865 on a word boundary. Use memmove instead of memcpy.
1866 (ffi_call): Update call to ffi_V9_return_struct.
1867 (ffi_prep_closure): Define 'ctx' only for V8.
1868 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1869 and ffi_closure_sparc_inner_v9.
1870 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1871 Always skip the structure return address. For structures and long
1872 doubles, copy the argument directly.
1873 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1874 if required. Shift the maximum floating-point slot accordingly. For
1875 big structures, copy the argument directly; otherwise, left-justify the
1876 argument and call ffi_v9_layout_struct to lay out the structure on
1878 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1879 (ffi_closure_v8): Pass the structure return address. Update call to
1880 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1881 Skip the 'unimp' insn when returning long doubles and structures.
1882 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1883 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1884 FFI_TYPE_INT handling. Load structures both in integers and
1885 floating-point registers on return.
1886 * README: Update status of the SPARC port.
1888 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1890 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1892 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1894 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1896 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1897 value as of type ffi_arg, not unsigned int.
1899 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1901 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1904 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1906 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1909 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1911 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1914 2003-12-10 Richard Henderson <rth@redhat.com>
1916 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1917 size_t instead of int.
1919 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1921 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1922 * testsuite/libffi.call/many_win32.c (main): Replace variable
1923 int i with unsigned long ul.
1925 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1926 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1927 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1928 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1929 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1930 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1931 * testsuite/libffi.call/cls_align_float.c: Likewise.
1932 * testsuite/libffi.call/cls_align_double.c: Likewise.
1933 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1934 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1936 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1939 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1940 Align arguments to 32 bits.
1942 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1945 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1946 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1947 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1948 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1949 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1950 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1952 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1954 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1956 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1957 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1959 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1961 * Makefile.in: Rebuilt.
1962 * configure: Likewise.
1963 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1966 2003-11-21 Andreas Jaeger <aj@suse.de>
1967 Andreas Tobler <a.tobler@schweiz.ch>
1969 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1970 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1971 * Makefile.in: Rebuilt.
1972 * aclocal.m4: Likewise.
1973 * configure: Likewise.
1974 * fficonfig.h.in: Likewise.
1975 * testsuite/lib/libffi-dg.exp: Add include dir.
1976 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1977 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1978 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1979 for ffi_closure if available.
1980 * testsuite/libffi.call/closure_fn1.c: Likewise.
1981 * testsuite/libffi.call/closure_fn2.c: Likewise.
1982 * testsuite/libffi.call/closure_fn3.c: Likewise.
1983 * testsuite/libffi.call/closure_fn4.c: Likewise.
1984 * testsuite/libffi.call/closure_fn5.c: Likewise.
1985 * testsuite/libffi.call/cls_12byte.c: Likewise.
1986 * testsuite/libffi.call/cls_16byte.c: Likewise.
1987 * testsuite/libffi.call/cls_18byte.c: Likewise.
1988 * testsuite/libffi.call/cls_19byte.c: Likewise.
1989 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1990 * testsuite/libffi.call/cls_20byte.c: Likewise.
1991 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1992 * testsuite/libffi.call/cls_24byte.c: Likewise.
1993 * testsuite/libffi.call/cls_2byte.c: Likewise.
1994 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1995 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1996 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1997 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1998 * testsuite/libffi.call/cls_4byte.c: Likewise.
1999 * testsuite/libffi.call/cls_5byte.c: Likewise.
2000 * testsuite/libffi.call/cls_64byte.c: Likewise.
2001 * testsuite/libffi.call/cls_6byte.c: Likewise.
2002 * testsuite/libffi.call/cls_7byte.c: Likewise.
2003 * testsuite/libffi.call/cls_8byte.c: Likewise.
2004 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2005 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2006 * testsuite/libffi.call/cls_double.c: Likewise.
2007 * testsuite/libffi.call/cls_float.c: Likewise.
2008 * testsuite/libffi.call/cls_schar.c: Likewise.
2009 * testsuite/libffi.call/cls_sint.c: Likewise.
2010 * testsuite/libffi.call/cls_sshort.c: Likewise.
2011 * testsuite/libffi.call/cls_uchar.c: Likewise.
2012 * testsuite/libffi.call/cls_uint.c: Likewise.
2013 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2014 * testsuite/libffi.call/cls_ushort.c: Likewise.
2015 * testsuite/libffi.call/nested_struct.c: Likewise.
2016 * testsuite/libffi.call/nested_struct1.c: Likewise.
2017 * testsuite/libffi.call/nested_struct2.c: Likewise.
2018 * testsuite/libffi.call/nested_struct3.c: Likewise.
2019 * testsuite/libffi.call/problem1.c: Likewise.
2020 * testsuite/libffi.special/unwindtest.cc: Likewise.
2022 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2024 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2026 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2028 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2029 Add -lgcc_s to additional flags.
2031 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2033 * configure.in, include/Makefile.am: PR libgcj/11147, install
2034 the ffitarget.h header file in a gcc versioned and target
2036 * configure: Regenerated.
2037 * Makefile.in, include/Makefile.in: Likewise.
2038 * testsuite/Makefile.in: Likewise.
2040 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2042 * testsuite/libffi.call/closure_fn0.c: Print result and check
2043 with dg-output to make debugging easier.
2044 * testsuite/libffi.call/closure_fn1.c: Likewise.
2045 * testsuite/libffi.call/closure_fn2.c: Likewise.
2046 * testsuite/libffi.call/closure_fn3.c: Likewise.
2047 * testsuite/libffi.call/closure_fn4.c: Likewise.
2048 * testsuite/libffi.call/closure_fn5.c: Likewise.
2049 * testsuite/libffi.call/cls_12byte.c: Likewise.
2050 * testsuite/libffi.call/cls_16byte.c: Likewise.
2051 * testsuite/libffi.call/cls_18byte.c: Likewise.
2052 * testsuite/libffi.call/cls_19byte.c: Likewise.
2053 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2054 * testsuite/libffi.call/cls_20byte.c: Likewise.
2055 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2056 * testsuite/libffi.call/cls_24byte.c: Likewise.
2057 * testsuite/libffi.call/cls_2byte.c: Likewise.
2058 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2059 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2060 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2061 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2062 * testsuite/libffi.call/cls_4byte.c: Likewise.
2063 * testsuite/libffi.call/cls_5byte.c: Likewise.
2064 * testsuite/libffi.call/cls_64byte.c: Likewise.
2065 * testsuite/libffi.call/cls_6byte.c: Likewise.
2066 * testsuite/libffi.call/cls_7byte.c: Likewise.
2067 * testsuite/libffi.call/cls_8byte.c: Likewise.
2068 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2069 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2070 * testsuite/libffi.call/cls_double.c: Likewise.
2071 * testsuite/libffi.call/cls_float.c: Likewise.
2072 * testsuite/libffi.call/cls_schar.c: Likewise.
2073 * testsuite/libffi.call/cls_sint.c: Likewise.
2074 * testsuite/libffi.call/cls_sshort.c: Likewise.
2075 * testsuite/libffi.call/cls_uchar.c: Likewise.
2076 * testsuite/libffi.call/cls_uint.c: Likewise.
2077 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2078 * testsuite/libffi.call/cls_ushort.c: Likewise.
2079 * testsuite/libffi.call/problem1.c: Likewise.
2081 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2084 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2086 * testsuite/libffi.call/cls_9byte2.c: New test case.
2087 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2088 * testsuite/libffi.call/cls_64byte.c: Likewise.
2089 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2090 * testsuite/libffi.call/cls_19byte.c: Likewise.
2091 * testsuite/libffi.call/cls_18byte.c: Likewise.
2092 * testsuite/libffi.call/closure_fn4.c: Likewise.
2093 * testsuite/libffi.call/closure_fn5.c: Likewise.
2094 * testsuite/libffi.call/cls_schar.c: Likewise.
2095 * testsuite/libffi.call/cls_sint.c: Likewise.
2096 * testsuite/libffi.call/cls_sshort.c: Likewise.
2097 * testsuite/libffi.call/nested_struct2.c: Likewise.
2098 * testsuite/libffi.call/nested_struct3.c: Likewise.
2100 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2102 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2103 * testsuite/libffi.call/cls_uchar.c: Likewise.
2104 * testsuite/libffi.call/cls_uint.c: Likewise.
2105 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2106 * testsuite/libffi.call/cls_ushort.c: Likewise.
2107 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2109 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2111 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2114 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2116 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2117 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2119 2003-10-22 David Daney <ddaney@avtrex.com>
2121 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2122 __attribute__((__mode__(__SI__))) and friends.
2124 2003-10-22 Andreas Schwab <schwab@suse.de>
2126 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2128 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2130 * configure.in: AC_LINK_FILES(ffitarget.h).
2131 * configure: Regenerate.
2132 * Makefile.in: Likewise.
2133 * include/Makefile.in: Likewise.
2134 * testsuite/Makefile.in: Likewise.
2135 * fficonfig.h.in: Likewise.
2137 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2138 Richard Henderson <rth@redhat.com>
2140 Avoid that ffi.h includes fficonfig.h.
2142 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2143 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2144 (TARGET_SRC_MIPS_SGI): Removed.
2145 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2146 (MIPS_SGI): Removed.
2147 (CLEANFILES): Removed.
2148 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2150 * acconfig.h: Removed.
2151 * configure.in: Compute sizeofs only for double and long double.
2152 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2153 into AC_DEFINE instead of using acconfig.h. Create
2154 include/ffitarget.h instead of include/fficonfig.h. Rename
2155 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2156 AC_DEFINE EH_FRAME_FLAGS.
2157 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2158 (hack_DATA): Add ffitarget.h.
2159 * include/ffi.h.in: Remove all system specific definitions.
2160 Declare raw API even if it is not installed, why bother?
2161 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2162 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2163 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2164 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2165 * include/ffi_common.h (bool): Do not define.
2166 (ffi_assert): Accept failed assertion.
2167 (ffi_type_test): Return void and accept file/line.
2168 (FFI_ASSERT): Pass stringized failed assertion.
2169 (FFI_ASSERT_AT): New macro.
2170 (FFI_ASSERT_VALID_TYPE): New macro.
2171 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2172 UINT64, SINT64): Define here with gcc's __attribute__ macro
2174 (FLOAT32, ALIGN): Define here instead of in ffi.h
2175 * include/ffi-mips.h: Removed. Its content moved to
2176 src/mips/ffitarget.h after separating assembly and C sections.
2177 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2178 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2179 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2180 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2181 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2182 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2183 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2184 (ffi_type_test): Rewritten.
2185 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2186 FFI_ASSERT_VALID_TYPE.
2187 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2188 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2189 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2190 src/s390/ffitarget.h, src/sh/ffitarget.h,
2191 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2192 src/x86/ffitarget.h: New files.
2193 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2194 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2195 src/powerpc/aix.S, src/powerpc/darwin.S,
2196 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2197 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2198 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2199 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2200 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2203 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2205 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2206 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2208 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2210 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2211 Used when FFI_DEBUG = 1.
2213 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2215 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2218 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2220 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2221 FFI_MIPS_O32 for O32 ABI.
2223 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2225 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2226 SPARC64. Cleanup whitespaces.
2228 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2230 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2231 strongarm, xscale. Cleanup whitespaces.
2232 * testsuite/libffi.call/closure_fn1.c: Likewise.
2233 * testsuite/libffi.call/closure_fn2.c: Likewise.
2234 * testsuite/libffi.call/closure_fn3.c: Likewise.
2235 * testsuite/libffi.call/cls_12byte.c: Likewise.
2236 * testsuite/libffi.call/cls_16byte.c: Likewise.
2237 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2238 * testsuite/libffi.call/cls_20byte.c: Likewise.
2239 * testsuite/libffi.call/cls_24byte.c: Likewise.
2240 * testsuite/libffi.call/cls_2byte.c: Likewise.
2241 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2242 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2243 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2244 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2245 * testsuite/libffi.call/cls_4byte.c: Likewise.
2246 * testsuite/libffi.call/cls_5byte.c: Likewise.
2247 * testsuite/libffi.call/cls_6byte.c: Likewise.
2248 * testsuite/libffi.call/cls_7byte.c: Likewise.
2249 * testsuite/libffi.call/cls_8byte.c: Likewise.
2250 * testsuite/libffi.call/cls_double.c: Likewise.
2251 * testsuite/libffi.call/cls_float.c: Likewise.
2252 * testsuite/libffi.call/cls_uchar.c: Likewise.
2253 * testsuite/libffi.call/cls_uint.c: Likewise.
2254 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2255 * testsuite/libffi.call/cls_ushort.c: Likewise.
2256 * testsuite/libffi.call/nested_struct.c: Likewise.
2257 * testsuite/libffi.call/nested_struct1.c: Likewise.
2258 * testsuite/libffi.call/problem1.c: Likewise.
2259 * testsuite/libffi.special/unwindtest.cc: Likewise.
2260 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2262 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2264 * src/powerpc/aix.S: Cleanup whitespaces.
2265 * src/powerpc/aix_closure.S: Likewise.
2267 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2269 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2270 * src/powerpc/darwin_closure.S: Likewise.
2271 * src/powerpc/ffi_darwin.c: Likewise.
2273 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2274 David Edelsohn <edelsohn@gnu.org>
2276 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2277 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2279 * src/powerpc/darwin_closure.S: Likewise.
2280 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2281 according to the Darwin/AIX ABI.
2282 (ffi_prep_cif_machdep): Likewise.
2283 (ffi_closure_helper_DARWIN): Likewise.
2284 Remove the outgoing parameter stack logic. Simplify the evaluation
2285 of the different CASE types.
2286 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2287 statement in the trampoline code.
2289 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2291 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2292 for the register size.
2293 (ffi_closure_helper_SYSV): Handle the structure return value
2295 (ffi_closure_helper_SYSV): Return the appropriate type when
2296 the registers are used for the structure return value.
2297 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2298 the 64-bit return value. Update copyright years.
2300 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2302 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2303 srcdir for ffi_mips.h.
2305 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2307 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2309 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2310 placement of float result.
2311 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2312 cast of "resp" for big-endian 64 bit machines.
2314 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2316 * src/types.c (double, longdouble): Merge identical SH and ARM
2317 typedefs, and add POWERPC64.
2318 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2319 struct split over gpr and rest.
2320 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2321 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2323 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2325 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2328 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2330 * configure: Regenerate.
2332 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2334 * Makefile.am: Remove build rules for ffitest.
2335 * Makefile.in: Rebuilt.
2337 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2339 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2340 about implicit declaration of abort().
2342 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2344 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2345 * Makefile.in: Rebuilt.
2346 * configure.in: Add dejagnu test framework.
2347 * configure: Rebuilt.
2349 * testsuite/Makefile.am: New file.
2350 * testsuite/Makefile.in: Built
2351 * testsuite/lib/libffi-dg.exp: New file.
2352 * testsuite/config/default.exp: Likewise.
2353 * testsuite/libffi.call/call.exp: Likewise.
2354 * testsuite/libffi.call/ffitest.h: Likewise.
2355 * testsuite/libffi.call/closure_fn0.c: Likewise.
2356 * testsuite/libffi.call/closure_fn1.c: Likewise.
2357 * testsuite/libffi.call/closure_fn2.c: Likewise.
2358 * testsuite/libffi.call/closure_fn3.c: Likewise.
2359 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2360 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2361 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2362 * testsuite/libffi.call/cls_2byte.c: Likewise.
2363 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2364 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2365 * testsuite/libffi.call/cls_4byte.c: Likewise.
2366 * testsuite/libffi.call/cls_5byte.c: Likewise.
2367 * testsuite/libffi.call/cls_6byte.c: Likewise.
2368 * testsuite/libffi.call/cls_7byte.c: Likewise.
2369 * testsuite/libffi.call/cls_8byte.c: Likewise.
2370 * testsuite/libffi.call/cls_12byte.c: Likewise.
2371 * testsuite/libffi.call/cls_16byte.c: Likewise.
2372 * testsuite/libffi.call/cls_20byte.c: Likewise.
2373 * testsuite/libffi.call/cls_24byte.c: Likewise.
2374 * testsuite/libffi.call/cls_double.c: Likewise.
2375 * testsuite/libffi.call/cls_float.c: Likewise.
2376 * testsuite/libffi.call/cls_uchar.c: Likewise.
2377 * testsuite/libffi.call/cls_uint.c: Likewise.
2378 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2379 * testsuite/libffi.call/cls_ushort.c: Likewise.
2380 * testsuite/libffi.call/float.c: Likewise.
2381 * testsuite/libffi.call/float1.c: Likewise.
2382 * testsuite/libffi.call/float2.c: Likewise.
2383 * testsuite/libffi.call/many.c: Likewise.
2384 * testsuite/libffi.call/many_win32.c: Likewise.
2385 * testsuite/libffi.call/nested_struct.c: Likewise.
2386 * testsuite/libffi.call/nested_struct1.c: Likewise.
2387 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2388 * testsuite/libffi.call/problem1.c: Likewise.
2389 * testsuite/libffi.call/promotion.c: Likewise.
2390 * testsuite/libffi.call/return_ll.c: Likewise.
2391 * testsuite/libffi.call/return_sc.c: Likewise.
2392 * testsuite/libffi.call/return_uc.c: Likewise.
2393 * testsuite/libffi.call/strlen.c: Likewise.
2394 * testsuite/libffi.call/strlen_win32.c: Likewise.
2395 * testsuite/libffi.call/struct1.c: Likewise.
2396 * testsuite/libffi.call/struct2.c: Likewise.
2397 * testsuite/libffi.call/struct3.c: Likewise.
2398 * testsuite/libffi.call/struct4.c: Likewise.
2399 * testsuite/libffi.call/struct5.c: Likewise.
2400 * testsuite/libffi.call/struct6.c: Likewise.
2401 * testsuite/libffi.call/struct7.c: Likewise.
2402 * testsuite/libffi.call/struct8.c: Likewise.
2403 * testsuite/libffi.call/struct9.c: Likewise.
2404 * testsuite/libffi.special/special.exp: New file.
2405 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2406 * testsuite/libffi.special/unwindtest.cc: Likewise.
2409 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2411 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2414 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2416 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2418 (ffi_closure_helper_LINUX64): Likewise.
2419 * src/powerpc/linux64.S: Remove code writing to parm save area.
2420 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2421 address in lr from ffi_closure_helper_LINUX64 call to calculate
2422 table address. Optimize function tail.
2424 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2426 * src/sparc/ffi.c: Handle all floating point registers.
2427 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2429 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2431 * README: Note that libffi is not part of GCC. Update the project
2434 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2436 * src/powerpc/ppc_closure.S: Include ffi.h.
2438 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2440 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2441 Use C style comments.
2443 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2445 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2446 * Makefile.in: Regenerate.
2447 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2448 * configure: Regenerate.
2449 * include/ffi.h.in: Add SHmedia support.
2450 * src/sh64/ffi.c: New file.
2451 * src/sh64/sysv.S: New file.
2453 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2455 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2456 should be read-only.
2457 * configure: Rebuilt.
2458 * fficonfig.h.in: Rebuilt.
2459 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2460 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2461 * src/powerpc/linux64.S: Likewise.
2462 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2463 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2464 if -fpic/-fPIC/-mrelocatable.
2465 * src/powerpc/powerpc_closure.S: Likewise.
2466 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2467 #write in .eh_frame flags.
2468 * src/sparc/v9.S: Likewise.
2469 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2470 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2471 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2473 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2475 Fixes PR bootstrap/10656
2476 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2477 support for .register pseudo-op.
2478 * src/sparc/v8.S: Use it.
2479 * fficonfig.h.in: Rebuilt.
2480 * configure: Rebuilt.
2482 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2484 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2485 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2486 Make it the default on POWERPC64.
2487 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2488 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2489 * configure: Rebuilt.
2490 * src/powerpc/ffi.c (hidden): Define.
2491 (ffi_prep_args_SYSV): Renamed from
2492 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2493 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2494 ASM_NEEDS_REGISTERS64): New.
2495 (ffi_prep_args64): New function.
2496 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2497 (ffi_call): Likewise.
2498 (ffi_prep_closure): Likewise.
2499 (flush_icache): Surround by #ifndef POWERPC64.
2500 (ffi_dblfl): New union type.
2501 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2502 (ffi_closure_helper_LINUX64): New function.
2503 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2505 * src/powerpc/sysv.S: Likewise.
2506 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2507 * src/powerpc/linux64.S: New file.
2508 * src/powerpc/linux64_closure.S: New file.
2509 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2510 src/powerpc/linux64_closure.S.
2511 (TARGET_SRC_POWERPC): Likewise.
2513 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2514 closure_test_fn3): Fix result printing on big-endian 64-bit
2516 (main): Print tst2_arg instead of uninitialized tst2_result.
2518 * src/ffitest.c (main): Hide what closure pointer really points to
2521 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2523 * configure.in (arm-*-netbsdelf*): Add configuration.
2524 (configure): Regenerated.
2526 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2528 * include/Makefile.in: Regenerate.
2530 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2532 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2534 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2535 Receive closure pointer through parameter, read args using
2536 __builtin_dwarf_cfa.
2537 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2539 2003-03-12 Andreas Schwab <schwab@suse.de>
2541 * configure.in: Avoid trailing /. in toolexeclibdir.
2542 * configure: Rebuilt.
2544 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2546 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2548 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2550 * libffi/src/powerpc/darwin_closure.S:
2551 Fix alignement bug, allocate 8 bytes for the result.
2552 * libffi/src/powerpc/aix_closure.S:
2554 * libffi/src/powerpc/ffi_darwin.c:
2555 Update stackframe description for aix/darwin_closure.S.
2557 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2559 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2562 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2563 Andreas Schwab <schwab@suse.de>
2565 * configure.in: Adjust command to source config-ml.in to account
2566 for changes to the libffi_basedir definition.
2567 (libffi_basedir): Remove ${srcdir} from value and include trailing
2570 * configure: Regenerate.
2572 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2574 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2576 2003-01-28 Andrew Haley <aph@redhat.com>
2578 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2579 * src/x86/ffi64.c (ffi_prep_closure): New.
2580 (ffi_closure_UNIX64_inner): New.
2581 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2583 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2585 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2586 Remove USE_LIBDIR conditional.
2587 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2588 * Makefile.in, configure: Rebuilt.
2590 2003-01027 David Edelsohn <edelsohn@gnu.org>
2592 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2593 * Makefile.in: Regenerate.
2595 2003-01-22 Andrew Haley <aph@redhat.com>
2597 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2600 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2602 * src/powerpc/darwin.S: Add unwind info.
2603 * src/powerpc/darwin_closure.S: Likewise.
2605 2003-01-14 Andrew Haley <aph@redhat.com>
2607 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2608 (ffi_prep_cif_machdep): Likewise.
2609 * src/x86/unix64.S: Add unwind info.
2611 2003-01-14 Andreas Jaeger <aj@suse.de>
2613 * src/ffitest.c (main): Only use ffi_closures if those are
2616 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2618 * libffi/src/ffitest.c
2619 add closure testcases
2621 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2623 * libffi/src/powerpc/ffi.c
2624 fix alignment bug for float (4 byte aligned iso 8 byte)
2626 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2628 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2629 * src/powerpc/darwin.S: Remove RCS version string.
2631 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2633 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2634 * src/ffitest.c (main): Use static storage for closure.
2635 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2636 * src/sparc/v8.S (ffi_closure_v8): New.
2637 * src/sparc/v9.S (ffi_closure_v9): New.
2639 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2641 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2642 enumeration for X86_WIN32.
2643 * src/x86/win32.S: Added ffi_call_STDCALL function
2645 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2646 switch cases for recognising FFI_STDCALL and
2647 calling ffi_call_STDCALL if target is X86_WIN32.
2648 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2649 stdcall versions of the "my_strlen" and "many"
2650 test functions (for X86_WIN32).
2651 Added test cases to test stdcall invocation using
2654 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2656 * src/sh/sysv.S: Add DWARF2 unwind info.
2658 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2660 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2662 2002-11-26 Jim Wilson <wilson@redhat.com>
2664 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2666 2002-11-23 H.J. Lu <hjl@gnu.org>
2668 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2669 Include ../config/accross.m4.
2670 * aclocal.m4; Rebuild.
2671 * configure: Likewise.
2673 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2675 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2677 2002-11-11 DJ Delorie <dj@redhat.com>
2679 * configure.in: Look for common files in the right place.
2681 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2683 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2684 raw data as _Jv_word values, not ffi_raw.
2685 (ffi_java_ptrarray_to_raw): Likewise.
2686 (ffi_java_rvalue_to_raw): New function.
2687 (ffi_java_raw_call): Call it.
2688 (ffi_java_raw_to_rvalue): New function.
2689 (ffi_java_translate_args): Call it.
2690 * src/ffitest.c (closure_test_fn): Interpret return value
2691 as ffi_arg, not int.
2692 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2693 FFI_TYPE_POINTER case.
2694 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2695 values extended to word size.
2697 2002-10-02 Andreas Jaeger <aj@suse.de>
2699 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2701 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2703 * include/ffi.h.in: Fix i386 win32 compilation.
2705 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2707 * configure.in: Add s390x-*-linux-* target.
2708 * configure: Regenerate.
2709 * include/ffi.h.in: Define S390X for s390x targets.
2710 (FFI_CLOSURES): Define for s390/s390x.
2711 (FFI_TRAMPOLINE_SIZE): Likewise.
2712 (FFI_NATIVE_RAW_API): Likewise.
2713 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2714 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2715 * src/s390/ffi.c: Major rework of existing code. Add support for
2716 s390x targets. Add closure support.
2717 * src/s390/sysv.S: Likewise.
2719 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2721 * src/arm/sysv.S: Fix typo.
2723 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2725 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2726 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2727 (ffi_call_SYSV): Handle soft-float.
2729 2002-09-27 Bo Thorsen <bo@suse.de>
2731 * include/ffi.h.in: Fix multilib x86-64 support.
2733 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2735 * Makefile.am (all-multi): Fix multilib parallel build.
2737 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2739 * configure.in (sh[34]*-*-linux*): Add brackets.
2740 * configure: Regenerate.
2742 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2744 * Makefile.am: Add SH support.
2745 * Makefile.in: Regenerate.
2746 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2747 * configure: Regenerate.
2748 * include/ffi.h.in: Add SH support.
2749 * src/sh/ffi.c: New file.
2750 * src/sh/sysv.S: New file.
2751 * src/types.c: Add SH support.
2753 2002-07-16 Bo Thorsen <bo@suse.de>
2755 * src/x86/ffi64.c: New file that adds x86-64 support.
2756 * src/x86/unix64.S: New file that handles argument setup for
2758 * src/x86/sysv.S: Don't use this on x86-64.
2759 * src/x86/ffi.c: Don't use this on x86-64.
2761 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2763 * src/ffitest.c (struct6): New test that tests a special case in
2765 (struct7): Likewise.
2766 (struct8): Likewise.
2767 (struct9): Likewise.
2768 (closure_test_fn): Silence warning about this when it's not used.
2769 (main): Add the new tests.
2770 (main): Fix a couple of wrong casts and silence some compiler warnings.
2771 * include/ffi.h.in: Add x86-64 ABI definition.
2772 * fficonfig.h.in: Regenerate.
2773 * Makefile.am: Add x86-64 support.
2774 * configure.in: Likewise.
2775 * Makefile.in: Regenerate.
2776 * configure: Likewise.
2778 2002-06-24 Bo Thorsen <bo@suse.de>
2780 * src/types.c: Merge settings for similar architectures.
2781 Add x86-64 sizes and alignments.
2783 2002-06-23 Bo Thorsen <bo@suse.de>
2785 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2786 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2787 * src/mips/ffi.c (ffi_prep_args): Likewise.
2788 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2790 2002-07-18 H.J. Lu (hjl@gnu.org)
2792 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2793 (libffi_la_SOURCES): Support MIPS_LINUX.
2794 (libffi_convenience_la_SOURCES): Likewise.
2795 * Makefile.in: Regenerated.
2797 * configure.in (mips64*-*): Skip.
2798 (mips*-*-linux*): New.
2799 * configure: Regenerated.
2801 * src/mips/ffi.c: Include <sgidefs.h>.
2803 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2805 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2807 2002-05-27 Roger Sayle <roger@eyesopen.com>
2809 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2811 2002-05-27 Bo Thorsen <bo@suse.de>
2813 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2816 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2818 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2819 beginning of function (for older apple cc).
2821 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2823 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2824 script entry, and set LD to it when configuring multilibs.
2825 * configure: Rebuilt.
2827 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2829 * configure.in (sparc64-*-netbsd*): Add target.
2830 (sparc-*-netbsdelf*): Likewise.
2831 * configure: Regenerate.
2833 2002-04-28 David S. Miller <davem@redhat.com>
2835 * configure.in, configure: Fix SPARC test in previous change.
2837 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2839 * Makefile.am: Add Linux for S/390 support.
2840 * Makefile.in: Regenerate.
2841 * configure.in: Add Linux for S/390 support.
2842 * configure: Regenerate.
2843 * include/ffi.h.in: Add Linux for S/390 support.
2844 * src/s390/ffi.c: New file from libffi CVS tree.
2845 * src/s390/sysv.S: New file from libffi CVS tree.
2847 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2849 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2851 * src/sparc/v8.S: Use it.
2852 * src/sparc/v9.S: Likewise.
2853 * fficonfig.h.in: Rebuilt.
2854 * configure: Rebuilt.
2856 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2858 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2860 * src/ia64/unix.S: Add unwind information. Fix comments.
2861 Save sp in a way that's compatible with unwind info.
2862 (ffi_call_unix): Correctly restore sp in all cases.
2863 * src/ia64/ffi.c: Add, fix comments.
2865 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2867 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2869 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2871 * configure.in (alpha*-*-netbsd*): Add target.
2872 * configure: Regenerate.
2874 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2876 * src/sparc/v8.S: Add unwind info.
2877 * src/sparc/v9.S: Likewise.
2879 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2881 * configure.in: Enable i*86-*-netbsdelf*.
2882 * configure: Rebuilt.
2884 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2887 * src/mips/n32.s: Delete
2888 * src/mips/o32.s: Delete
2890 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2892 * configure.in: Enable alpha*-*-freebsd*.
2893 * configure: Rebuilt.
2895 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2897 * Makefile.am: libfficonvenience -> libffi_convenience.
2898 * Makefile.in: Rebuilt.
2900 * Makefile.am: Define ffitest_OBJECTS.
2901 * Makefile.in: Rebuilt.
2903 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2904 David Edelsohn <edelsohn@gnu.org>
2906 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2907 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2908 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2909 * Makefile.in: Regenerate.
2910 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2911 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2912 (flush_icache, flush_range): New functions.
2913 (ffi_closure_helper_DARWIN): New function.
2914 * src/powerpc/aix_closure.S: New file.
2915 * src/powerpc/darwin_closure.S: New file.
2917 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2919 * include/ffi.h.in: Add typedef for ffi_arg.
2920 * src/ffitest.c (main): Declare rint with ffi_arg.
2922 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2924 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2925 number of GPRs for floating-point arguments.
2927 2002-01-31 Anthony Green <green@redhat.com>
2929 * configure: Rebuilt.
2930 * configure.in: Replace CHECK_SIZEOF and endian tests with
2931 cross-compiler friendly macros.
2932 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2935 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2937 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2938 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2940 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2942 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2943 (TARGET_SRC_POWERPC_AIX): New.
2944 (POWERPC_AIX): New stanza.
2945 * Makefile.in: Regenerate.
2946 * configure.in: Add AIX case.
2947 * configure: Regenerate.
2948 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2949 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2950 size. Fix "long double" support.
2951 (ffi_call): Add FFI_AIX case.
2952 * src/powerpc/aix.S: New.
2954 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2956 Implement Darwin PowerPC ABI.
2957 * configure.in: Handle powerpc-*-darwin*.
2958 * Makefile.am: Set source files for POWERPC_DARWIN.
2959 * configure: Rebuilt.
2960 * Makefile.in: Rebuilt.
2961 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2963 * src/powerpc/darwin.S: New file.
2964 * src/powerpc/ffi_darwin.c: New file.
2966 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2968 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2970 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2972 * src/x86/sysv.S: Avoid gas-only .balign directive.
2973 Use C style comments.
2975 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2977 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2978 Fixes PR bootstrap/3563.
2980 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2982 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2984 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2986 * configure.in: Recognize sparc*-sun-* host.
2987 * configure: Regenerate.
2989 2001-06-06 Andrew Haley <aph@redhat.com>
2991 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
2993 2001-06-03 Andrew Haley <aph@redhat.com>
2995 * src/alpha/osf.S: Add unwind info.
2996 * src/powerpc/sysv.S: Add unwind info.
2997 * src/powerpc/ppc_closure.S: Likewise.
2999 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3001 * configure.in: Fix AC_ARG_ENABLE usage.
3002 * configure: Rebuilt.
3004 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3006 * configure.in: Remove warning about beta code.
3007 * configure: Rebuilt.
3009 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3011 * src/ia64/unix.S: Restore stack pointer when returning from
3013 * src/ia64/ffi.c: Fix typo in comment.
3015 2001-04-18 Jim Wilson <wilson@redhat.com>
3017 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3018 to eliminate RAW DV.
3020 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3022 * Makefile.am: Make a libtool convenience library.
3023 * Makefile.in: Rebuilt.
3025 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3027 * configure.in: Use different syntax for subdirectory creation.
3028 * configure: Rebuilt.
3030 2001-03-27 Jon Beniston <jon@beniston.com>
3032 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3033 * configure: Rebuilt.
3034 * Makefile.am: Added X86_WIN32 target support.
3035 * Makefile.in: Rebuilt.
3037 * include/ffi.h.in: Added X86_WIN32 target support.
3039 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3040 * src/types.c: Added X86_WIN32 target support.
3042 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3043 stuff removed and made to work with CygWin's gas.
3045 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3047 * configure.in: Make target subdirectory in build dir.
3048 * Makefile.am: Override suffix based rules to specify correct output
3050 * Makefile.in: Rebuilt.
3051 * configure: Rebuilt.
3053 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3055 * src/powerpc/ppc_closure.S: New file.
3056 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3057 involving long long and register pairs.
3058 (ffi_prep_closure): New function.
3059 (flush_icache): Likewise.
3060 (ffi_closure_helper_SYSV): Likewise.
3061 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3062 (FFI_TRAMPOLINE_SIZE): Likewise.
3063 (FFI_NATIVE_RAW_API): Likewise.
3064 * Makefile.in: Rebuilt.
3065 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3066 (TARGET_SRC_POWERPC): Likewise.
3068 2001-03-19 Tom Tromey <tromey@redhat.com>
3070 * Makefile.in: Rebuilt.
3071 * Makefile.am (ffitest_LDFLAGS): New macro.
3073 2001-03-02 Nick Clifton <nickc@redhat.com>
3075 * include/ffi.h.in: Remove RCS ident string.
3076 * include/ffi_mips.h: Remove RCS ident string.
3077 * src/debug.c: Remove RCS ident string.
3078 * src/ffitest.c: Remove RCS ident string.
3079 * src/prep_cif.c: Remove RCS ident string.
3080 * src/types.c: Remove RCS ident string.
3081 * src/alpha/ffi.c: Remove RCS ident string.
3082 * src/alpha/osf.S: Remove RCS ident string.
3083 * src/arm/ffi.c: Remove RCS ident string.
3084 * src/arm/sysv.S: Remove RCS ident string.
3085 * src/mips/ffi.c: Remove RCS ident string.
3086 * src/mips/n32.S: Remove RCS ident string.
3087 * src/mips/o32.S: Remove RCS ident string.
3088 * src/sparc/ffi.c: Remove RCS ident string.
3089 * src/sparc/v8.S: Remove RCS ident string.
3090 * src/sparc/v9.S: Remove RCS ident string.
3091 * src/x86/ffi.c: Remove RCS ident string.
3092 * src/x86/sysv.S: Remove RCS ident string.
3094 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3096 * include/ffi.h.in: Change sourceware.cygnus.com references to
3099 2000-12-09 Richard Henderson <rth@redhat.com>
3101 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3102 (ffi_closure_osf_inner): Index rather than increment avalue
3103 and arg_types. Give ffi_closure_osf the raw return value type.
3104 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3107 2000-12-07 Richard Henderson <rth@redhat.com>
3109 * src/raw_api.c (ffi_translate_args): Fix typo.
3110 (ffi_prep_closure): Likewise.
3112 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3113 FFI_TRAMPOLINE_SIZE.
3114 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3115 cif->bytes for new ffi_call_osf implementation.
3116 (ffi_prep_args): Absorb into ...
3117 (ffi_call): ... here. Do all stack allocation here and
3118 avoid a callback function.
3119 (ffi_prep_closure, ffi_closure_osf_inner): New.
3120 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3121 (ffi_closure_osf): New.
3123 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3125 * config.guess, config.sub, install-sh: Removed.
3126 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3127 * Makefile.in: Rebuilt.
3129 * acinclude.m4: Include libtool macros from the top level.
3130 * aclocal.m4, configure: Rebuilt.
3132 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3134 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3135 * configure: Rebuilt.
3137 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3139 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3140 memory correctly. Use conditional instructions, not branches where
3143 2000-05-04 Tom Tromey <tromey@cygnus.com>
3145 * configure: Rebuilt.
3146 * configure.in: Match `arm*-*-linux-*'.
3147 From Chris Dornan <cdornan@arm.com>.
3149 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3151 * Makefile.am (SUBDIRS): Define.
3152 (AM_MAKEFLAGS): Likewise.
3153 (Multilib support.): Add section.
3154 * Makefile.in: Rebuilt.
3155 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3156 New variables. Set for gcc using -print-multi-lib. Export them
3158 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3159 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3161 (extra_compiler_flags_value, extra_compiler_flags): Check these
3162 for extra compiler options which need to be passed down in
3165 2000-04-16 Anthony Green <green@redhat.com>
3167 * configure: Rebuilt.
3168 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3170 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3172 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3173 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3174 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3175 Replace all void * sizeofs with sizeof(int).
3176 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3177 different than DOUBLE.
3178 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3179 (ffi_prep_args_v9): New function.
3180 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3181 (ffi_V9_return_struct): New function.
3182 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3183 32bit code (not yet cross-arch calls).
3184 * src/sparc/v8.S: Add struct return delay nop.
3186 * src/sparc/v9.S: New file.
3187 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3188 is used on sparc64 only for structures larger than 32 bytes.
3189 Pass by reference for structures is done for structure arguments
3190 larger than 16 bytes.
3191 * src/ffitest.c (main): Use 64bit rint on sparc64.
3192 Run long long tests on sparc.
3193 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3195 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3197 * configure.in (sparc-*-linux*): New supported target.
3198 (sparc64-*-linux*): Likewise.
3199 * configure: Rebuilt.
3200 * Makefile.am: Add v9.S to SPARC files.
3201 * Makefile.in: Likewise.
3202 (LINK): Surround $(CCLD) into double quotes, so that multilib
3203 compiles work correctly.
3205 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3207 * configure: Rebuilt.
3208 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3209 proposed by Bryce McKinlay.
3211 2000-03-20 Tom Tromey <tromey@cygnus.com>
3213 * Makefile.in: Hand edit for java_raw_api.lo.
3215 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3217 * config.guess, config.sub: Update from the gcc tree.
3218 Fix for PR libgcj/168.
3220 2000-03-03 Tom Tromey <tromey@cygnus.com>
3222 * Makefile.in: Fixed ia64 by hand.
3224 * configure: Rebuilt.
3225 * configure.in (--enable-multilib): New option.
3226 (libffi_basedir): New subst.
3227 (AC_OUTPUT): Added multilib code.
3229 2000-03-02 Tom Tromey <tromey@cygnus.com>
3231 * Makefile.in: Rebuilt.
3232 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3235 2000-02-25 Hans Boehm <boehm@acm.org>
3237 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3239 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3241 (ffi_prep_raw_closure): Use ffi_translate_args, not
3242 ffi_closure_translate.
3243 * src/java_raw_api.c: New file.
3244 * src/ffitest.c (closure_test_fn): New function.
3245 (main): Define `rint' as long long on IA64. Added new test when
3246 FFI_CLOSURES is defined.
3247 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3248 (ffi_abi): Recognize IA64.
3249 (ffi_raw): Added `flt' field.
3250 Added "Java raw API" code.
3251 * configure.in: Recognize ia64.
3252 * Makefile.am (TARGET_SRC_IA64): New macro.
3253 (libffi_la_common_SOURCES): Added java_raw_api.c.
3254 (libffi_la_SOURCES): Define in IA64 case.
3256 2000-01-04 Tom Tromey <tromey@cygnus.com>
3258 * Makefile.in: Rebuilt with newer automake.
3260 1999-12-31 Tom Tromey <tromey@cygnus.com>
3262 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3264 1999-09-01 Tom Tromey <tromey@cygnus.com>
3266 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3268 * fficonfig.h.in: Rebuilt.
3269 * configure: Rebuilt.
3270 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3271 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3272 * acconfig.h: Don't #undef PACKAGE or VERSION.
3274 1999-08-09 Anthony Green <green@cygnus.com>
3276 * include/ffi.h.in: Try to work around messy header problem
3277 with PACKAGE and VERSION.
3279 * configure: Rebuilt.
3280 * configure.in: Change version to 2.00-beta.
3282 * fficonfig.h.in: Rebuilt.
3283 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3285 * src/x86/ffi.c (ffi_raw_call): Rename.
3287 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3289 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3290 (ffi_prep_incoming_args_SYSV): Ditto.
3291 (ffi_prep_closure): Ditto.
3292 (ffi_closure_raw_SYSV): Ditto.
3293 (ffi_prep_raw_closure): More ditto.
3294 (ffi_call_raw): Final ditto.
3296 * include/ffi.h.in: Add definitions for closure and raw API.
3298 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3301 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3303 * src/raw_api.c: New file.
3305 * include/ffi.h.in (ffi_raw): New type.
3306 (UINT_ARG, SINT_ARG): New defines.
3307 (ffi_closure, ffi_raw_closure): New types.
3308 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3310 * configure.in: Add check for endianness and sizeof void*.
3312 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3313 instead of directly.
3315 * configure: Rebuilt.
3317 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3319 * configure.in: Add x86 and powerpc BeOS configurations.
3320 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3322 1999-05-09 Anthony Green <green@cygnus.com>
3324 * configure.in: Add warning about this being beta code.
3325 Remove src/Makefile.am from the picture.
3326 * configure: Rebuilt.
3328 * Makefile.am: Move logic from src/Makefile.am. Add changes
3329 to support libffi as a target library.
3330 * Makefile.in: Rebuilt.
3332 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3333 Upgraded to new autoconf, automake, libtool.
3337 * LICENSE: Update copyright date.
3339 * src/Makefile.am, src/Makefile.in: Removed.
3341 1998-11-29 Anthony Green <green@cygnus.com>
3343 * include/ChangeLog: Removed.
3344 * src/ChangeLog: Removed.
3345 * src/mips/ChangeLog: Removed.
3346 * src/sparc/ChangeLog: Remboved.
3347 * src/x86/ChangeLog: Removed.
3349 * ChangeLog.v1: Created.