1 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
3 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
6 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
8 * configure: Regenerate.
9 * include/Makefile.in: Regenerate.
10 * testsuite/Makefile.in: Regenerate.
12 2008-06-07 Joseph Myers <joseph@codesourcery.com>
14 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
15 powerpc-*-beos*): Remove.
16 * configure: Regenerate.
18 2008-05-09 Julian Brown <julian@codesourcery.com>
20 * Makefile.am (LTLDFLAGS): New.
21 (libffi_la_LDFLAGS): Use above.
22 * Makefile.in: Regenerate.
24 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
27 * aclocal.m4: Regenerate.
28 * configure: Regenerate.
30 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
32 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
33 * src/sh64/sysv.S: Likewise.
35 2008-03-26 Daniel Jacobowitz <dan@debian.org>
37 * src/arm/sysv.S: Fix ARM comment marker.
39 2008-03-26 Jakub Jelinek <jakub@redhat.com>
41 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
42 * src/s390/sysv.S: Likewise.
43 * src/powerpc/ppc_closure.S: Likewise.
44 * src/powerpc/sysv.S: Likewise.
45 * src/x86/unix64.S: Likewise.
46 * src/x86/sysv.S: Likewise.
47 * src/sparc/v8.S: Likewise.
48 * src/sparc/v9.S: Likewise.
49 * src/m68k/sysv.S: Likewise.
50 * src/arm/sysv.S: Likewise.
52 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
54 * aclocal.m4: Regenerate.
55 * configure: Likewise.
56 * Makefile.in: Likewise.
57 * include/Makefile.in: Likewise.
58 * testsuite/Makefile.in: Likewise.
60 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
61 Andreas Tobler <a.tobler@schweiz.org>
63 * configure.ac: Add amd64-*-freebsd* target.
64 * configure: Regenerate.
66 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
69 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
72 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
75 2008-01-24 David Edelsohn <edelsohn@gnu.org>
77 * configure: Regenerate.
79 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
81 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
83 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
86 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
87 signed/unsigned int8/16 for X86_DARWIN.
88 Updated copyright info.
89 Handle one and two byte structs with special cif->flags.
90 * src/x86/ffitarget.h: Add special types for one and two byte structs.
91 Updated copyright info.
92 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
94 Remove code to pop args from the stack after call.
95 Special-case signed/unsigned for int8/16, one and two byte structs.
96 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
97 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
99 Updated copyright info.
101 2007-12-08 David Daney <ddaney@avtrex.com>
103 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
104 SUBU, add with ADDU and use smaller code sequences.
106 2007-12-07 David Daney <ddaney@avtrex.com>
108 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
111 2007-12-06 David Daney <ddaney@avtrex.com>
113 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
115 (ffi_java_raw): New typedef.
116 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
117 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
119 (ffi_java_raw_closure) : Same.
120 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
122 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
124 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
125 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
126 sizeof(ffi_java_raw) for alignment calculations.
127 (ffi_java_ptrarray_to_raw): Same.
128 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
129 if FFI_SIZEOF_JAVA_RAW == 4.
130 (ffi_java_raw_to_rvalue): Same.
131 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
132 (ffi_java_translate_args): Same.
133 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
135 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
137 2007-12-06 David Daney <ddaney@avtrex.com>
139 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
142 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
145 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
146 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
147 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
148 set the NUM_FPR_ARG_REGISTERS according to.
149 Add support for potential soft-float support under hard-float
151 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
152 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
153 to the FFI_LINUX_SOFT_FLOAT ABI.
154 (ffi_prep_cif_machdep): Likewise.
155 (ffi_closure_helper_SYSV): Likewise.
156 * src/powerpc/ppc_closure.S: Make sure not to store float/double
157 on archs where __NO_FPRS__ is true.
158 Add FFI_TYPE_UINT128 support.
159 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
160 Adjust copyright notice.
162 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
164 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
165 * include/ffi_common.h: ... here.
168 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
170 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
171 * src/powerpc/linux64.S: Likewise.
172 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
173 * testsuite/libffi.call/return_ldl.c: New test.
175 2007-09-04 <aph@redhat.com>
177 * src/arm/sysv.S (UNWIND): New.
178 (Whole file): Conditionally compile unwinder directives.
179 * src/arm/sysv.S: Add unwinder directives.
181 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
182 Only treat r0 as a struct address if we're actually returning a
184 Only copy the bytes that are actually within a struct.
185 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
186 is returned in r0, not passed by address.
187 (ffi_call): Allocate a word-sized temporary for the case where
188 a composite is returned in r0.
189 (ffi_prep_incoming_args_SYSV): Align as necessary.
191 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
193 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
194 directly using the sys_cacheflush syscall.
196 2007-07-27 Andrew Haley <aph@redhat.com>
198 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
200 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
202 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
203 * configure.ac: Likewise.
204 * Makefile.in: Regenerate.
205 * include/Makefile.in: Likewise.
206 * testsuite/Makefile.in: Likewise.
207 * configure: Likewise.
209 2007-08-24 David Daney <ddaney@avtrex.com>
211 * testsuite/libffi.call/return_sl.c: New test.
213 2007-08-10 David Daney <ddaney@avtrex.com>
215 * testsuite/libffi.call/cls_multi_ushort.c,
216 testsuite/libffi.call/cls_align_uint16.c,
217 testsuite/libffi.call/nested_struct1.c,
218 testsuite/libffi.call/nested_struct3.c,
219 testsuite/libffi.call/cls_7_1_byte.c,
220 testsuite/libffi.call/nested_struct5.c,
221 testsuite/libffi.call/cls_double.c,
222 testsuite/libffi.call/nested_struct7.c,
223 testsuite/libffi.call/cls_sint.c,
224 testsuite/libffi.call/nested_struct9.c,
225 testsuite/libffi.call/cls_20byte1.c,
226 testsuite/libffi.call/cls_multi_sshortchar.c,
227 testsuite/libffi.call/cls_align_sint64.c,
228 testsuite/libffi.call/cls_3byte2.c,
229 testsuite/libffi.call/cls_multi_schar.c,
230 testsuite/libffi.call/cls_multi_uchar.c,
231 testsuite/libffi.call/cls_19byte.c,
232 testsuite/libffi.call/cls_9byte1.c,
233 testsuite/libffi.call/cls_align_float.c,
234 testsuite/libffi.call/closure_fn1.c,
235 testsuite/libffi.call/problem1.c,
236 testsuite/libffi.call/closure_fn3.c,
237 testsuite/libffi.call/cls_sshort.c,
238 testsuite/libffi.call/closure_fn5.c,
239 testsuite/libffi.call/cls_align_double.c,
240 testsuite/libffi.call/nested_struct.c,
241 testsuite/libffi.call/cls_2byte.c,
242 testsuite/libffi.call/nested_struct10.c,
243 testsuite/libffi.call/cls_4byte.c,
244 testsuite/libffi.call/cls_6byte.c,
245 testsuite/libffi.call/cls_8byte.c,
246 testsuite/libffi.call/cls_multi_sshort.c,
247 testsuite/libffi.call/cls_align_sint16.c,
248 testsuite/libffi.call/cls_align_uint32.c,
249 testsuite/libffi.call/cls_20byte.c,
250 testsuite/libffi.call/cls_float.c,
251 testsuite/libffi.call/nested_struct2.c,
252 testsuite/libffi.call/cls_5_1_byte.c,
253 testsuite/libffi.call/nested_struct4.c,
254 testsuite/libffi.call/cls_24byte.c,
255 testsuite/libffi.call/nested_struct6.c,
256 testsuite/libffi.call/cls_64byte.c,
257 testsuite/libffi.call/nested_struct8.c,
258 testsuite/libffi.call/cls_uint.c,
259 testsuite/libffi.call/cls_multi_ushortchar.c,
260 testsuite/libffi.call/cls_schar.c,
261 testsuite/libffi.call/cls_uchar.c,
262 testsuite/libffi.call/cls_align_uint64.c,
263 testsuite/libffi.call/cls_ulonglong.c,
264 testsuite/libffi.call/cls_align_longdouble.c,
265 testsuite/libffi.call/cls_1_1byte.c,
266 testsuite/libffi.call/cls_12byte.c,
267 testsuite/libffi.call/cls_3_1byte.c,
268 testsuite/libffi.call/cls_3byte1.c,
269 testsuite/libffi.call/cls_4_1byte.c,
270 testsuite/libffi.call/cls_6_1_byte.c,
271 testsuite/libffi.call/cls_16byte.c,
272 testsuite/libffi.call/cls_18byte.c,
273 testsuite/libffi.call/closure_fn0.c,
274 testsuite/libffi.call/cls_9byte2.c,
275 testsuite/libffi.call/closure_fn2.c,
276 testsuite/libffi.call/closure_fn4.c,
277 testsuite/libffi.call/cls_ushort.c,
278 testsuite/libffi.call/closure_fn6.c,
279 testsuite/libffi.call/cls_5byte.c,
280 testsuite/libffi.call/cls_align_pointer.c,
281 testsuite/libffi.call/cls_7byte.c,
282 testsuite/libffi.call/cls_align_sint32.c,
283 testsuite/libffi.special/unwindtest_ffi_call.cc,
284 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
286 2007-08-10 David Daney <ddaney@avtrex.com>
289 * configure.ac: Don't treat mips64 as a special case.
290 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
291 * configure: Regenerate
292 * Makefile.in: Ditto.
293 * fficonfig.h.in: Ditto.
294 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
295 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
296 (FFI_DEFAULT_ABI): Set for n64 case.
297 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
298 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
299 (ffi_closure_N32): New function.
300 (.eh_frame): New section
301 * src/mips/o32.S: Clean up comments.
302 (ffi_closure_O32): Pass ffi_closure parameter in $12.
303 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
304 _MIPS_SIM == _ABIN32 throughout.
305 (FFI_MIPS_STOP_HERE): New, use in place of
307 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
308 to support n32/n64 ABIs.
309 (calc_n32_struct_flags): Rewrite.
310 (calc_n32_return_struct_flags): Remove unused variable. Reverse
311 position of flag bits.
312 (ffi_prep_cif_machdep): Rewrite n32 portion.
313 (ffi_call): Enable for n64. Add special handling for small structure
315 (ffi_prep_closure_loc): Add n32 and n64 support.
316 (ffi_closure_mips_inner_O32): Add cast to silence warning.
317 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
319 2007-08-08 David Daney <ddaney@avtrex.com>
321 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
322 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
324 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
325 * testsuite/libffi.call/cls_sint.c (main): Ditto.
326 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
327 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
328 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
329 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
330 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
331 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
332 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
333 * testsuite/libffi.call/return_fl3.c (main): Ditto.
334 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
335 * testsuite/libffi.call/nested_struct.c (main): Ditto.
336 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
337 * testsuite/libffi.call/return_ll1.c (main): Ditto.
338 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
339 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
340 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
341 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
342 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
343 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
344 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
345 * testsuite/libffi.call/cls_uint.c (main): Ditto.
346 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
347 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
348 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
349 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
350 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
351 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
352 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
353 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
354 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
356 2007-08-07 Andrew Haley <aph@redhat.com>
358 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
361 2007-08-06 Andrew Haley <aph@redhat.com>
364 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
365 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
368 2007-08-02 David Daney <ddaney@avtrex.com>
370 * testsuite/libffi.call/return_ul.c (main): Define return type as
371 ffi_arg. Use proper printf conversion specifier.
373 2007-07-30 Andrew Haley <aph@redhat.com>
376 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
377 signed/unsigned int8/16.
378 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
380 Remove code to pop args from the stack after call.
381 Special-case signed/unsigned int8/16.
382 * testsuite/libffi.call/return_sc.c (main): Revert.
384 2007-07-26 Richard Guenther <rguenther@suse.de>
387 * testsuite/libffi.call/return_sc.c (main): Verify call
388 result as signed char, not ffi_arg.
390 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
392 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
393 * configure: Regenerate.
395 2007-07-11 David Daney <ddaney@avtrex.com>
397 * src/mips/ffi.c: Don't include sys/cachectl.h.
398 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
401 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
403 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
404 from (ffi_prep_closure): ... this.
405 (FFI_INIT_TRAMPOLINE): Adjust.
407 2005-12-31 Phil Blundell <pb@reciva.com>
409 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
410 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
411 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
412 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
413 (FFI_CLOSURES): Enable closure support.
415 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
417 * testsuite/libffi.call/cls_multi_ushort.c,
418 testsuite/libffi.call/cls_align_uint16.c,
419 testsuite/libffi.call/nested_struct1.c,
420 testsuite/libffi.call/nested_struct3.c,
421 testsuite/libffi.call/cls_7_1_byte.c,
422 testsuite/libffi.call/cls_double.c,
423 testsuite/libffi.call/nested_struct5.c,
424 testsuite/libffi.call/nested_struct7.c,
425 testsuite/libffi.call/cls_sint.c,
426 testsuite/libffi.call/nested_struct9.c,
427 testsuite/libffi.call/cls_20byte1.c,
428 testsuite/libffi.call/cls_multi_sshortchar.c,
429 testsuite/libffi.call/cls_align_sint64.c,
430 testsuite/libffi.call/cls_3byte2.c,
431 testsuite/libffi.call/cls_multi_schar.c,
432 testsuite/libffi.call/cls_multi_uchar.c,
433 testsuite/libffi.call/cls_19byte.c,
434 testsuite/libffi.call/cls_9byte1.c,
435 testsuite/libffi.call/cls_align_float.c,
436 testsuite/libffi.call/closure_fn1.c,
437 testsuite/libffi.call/problem1.c,
438 testsuite/libffi.call/closure_fn3.c,
439 testsuite/libffi.call/cls_sshort.c,
440 testsuite/libffi.call/closure_fn5.c,
441 testsuite/libffi.call/cls_align_double.c,
442 testsuite/libffi.call/cls_2byte.c,
443 testsuite/libffi.call/nested_struct.c,
444 testsuite/libffi.call/nested_struct10.c,
445 testsuite/libffi.call/cls_4byte.c,
446 testsuite/libffi.call/cls_6byte.c,
447 testsuite/libffi.call/cls_8byte.c,
448 testsuite/libffi.call/cls_multi_sshort.c,
449 testsuite/libffi.call/cls_align_uint32.c,
450 testsuite/libffi.call/cls_align_sint16.c,
451 testsuite/libffi.call/cls_float.c,
452 testsuite/libffi.call/cls_20byte.c,
453 testsuite/libffi.call/cls_5_1_byte.c,
454 testsuite/libffi.call/nested_struct2.c,
455 testsuite/libffi.call/cls_24byte.c,
456 testsuite/libffi.call/nested_struct4.c,
457 testsuite/libffi.call/nested_struct6.c,
458 testsuite/libffi.call/cls_64byte.c,
459 testsuite/libffi.call/nested_struct8.c,
460 testsuite/libffi.call/cls_uint.c,
461 testsuite/libffi.call/cls_multi_ushortchar.c,
462 testsuite/libffi.call/cls_schar.c,
463 testsuite/libffi.call/cls_uchar.c,
464 testsuite/libffi.call/cls_align_uint64.c,
465 testsuite/libffi.call/cls_ulonglong.c,
466 testsuite/libffi.call/cls_align_longdouble.c,
467 testsuite/libffi.call/cls_1_1byte.c,
468 testsuite/libffi.call/cls_12byte.c,
469 testsuite/libffi.call/cls_3_1byte.c,
470 testsuite/libffi.call/cls_3byte1.c,
471 testsuite/libffi.call/cls_4_1byte.c,
472 testsuite/libffi.call/cls_6_1_byte.c,
473 testsuite/libffi.call/cls_16byte.c,
474 testsuite/libffi.call/cls_18byte.c,
475 testsuite/libffi.call/closure_fn0.c,
476 testsuite/libffi.call/cls_9byte2.c,
477 testsuite/libffi.call/closure_fn2.c,
478 testsuite/libffi.call/closure_fn4.c,
479 testsuite/libffi.call/cls_ushort.c,
480 testsuite/libffi.call/closure_fn6.c,
481 testsuite/libffi.call/cls_5byte.c,
482 testsuite/libffi.call/cls_align_pointer.c,
483 testsuite/libffi.call/cls_7byte.c,
484 testsuite/libffi.call/cls_align_sint32.c,
485 testsuite/libffi.special/unwindtest_ffi_call.cc,
486 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
488 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
490 * aclocal.m4: Regenerated.
492 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
494 * configure: Regenerate.
496 2007-05-23 Steve Ellcey <sje@cup.hp.com>
498 * Makefile.in: Regenerate.
499 * configure: Regenerate.
500 * aclocal.m4: Regenerate.
501 * include/Makefile.in: Regenerate.
502 * testsuite/Makefile.in: Regenerate.
504 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
506 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
507 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
508 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
509 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
510 (FFI_CLOSURES): Enable closure support.
512 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
514 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
515 * configure: Regenerate.
516 * fficonfig.h.in: Regenerate.
517 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
518 CFI_OFFSET,CFI_DEF_CFA): New macros.
519 (ffi_call_SYSV): Add callframe annotation.
521 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
523 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
524 numerous test suite failures.
525 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
527 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
529 * Makefile.am (EXTRA_DIST): Bring up to date.
530 * Makefile.in: Regenerate.
531 * src/frv/eabi.S: Remove RCS keyword.
533 2007-04-06 Richard Henderson <rth@redhat.com>
535 * configure.ac: Tidy target case.
536 (HAVE_LONG_DOUBLE): Allow the target to override.
537 * configure: Regenerate.
538 * include/ffi.h.in: Don't define ffi_type_foo if
539 LIBFFI_HIDE_BASIC_TYPES is defined.
540 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
542 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
543 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
544 (ffi_type_longdouble): Special case for Alpha. Don't define
545 if long double == double.
547 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
548 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
549 (ffi_call, ffi_closure_osf_inner): Likewise.
550 (ffi_closure_osf_inner): Likewise. Mark hidden.
551 (ffi_call_osf, ffi_closure_osf): Mark hidden.
552 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
553 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
554 (load_table): Handle 128-bit long double.
556 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
558 2007-04-06 Tom Tromey <tromey@redhat.com>
561 * README: Fixed bug in example.
563 2007-04-03 Jakub Jelinek <jakub@redhat.com>
565 * src/closures.c: Include sys/statfs.h.
566 (_GNU_SOURCE): Define on Linux.
567 (FFI_MMAP_EXEC_SELINUX): Define.
568 (selinux_enabled): New variable.
569 (selinux_enabled_check): New function.
570 (is_selinux_enabled): Define.
573 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
575 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
576 Use 'volatile float sum' to create sum of floats to avoid false
577 negative due to excess precision on ix86 targets.
580 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
582 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
584 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
586 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
588 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
589 (ffi_prep_closure_loc): New.
590 (ffi_prep_raw_closure_loc): New.
591 (ffi_prep_java_raw_closure_loc): New.
592 * src/closures.c: New file.
593 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
594 Replace sflags with exec_offset.
595 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
596 sub_segment_exec_offset): New macros.
597 (get_segment_flags, set_segment_flags, check_segment_merge): New
599 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
600 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
601 destroy_mspace): Use new macros.
602 (sys_alloc): Silence warning.
603 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
604 * Makefile.in: Rebuilt.
605 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
606 terms of ffi_prep_closure_loc.
607 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
609 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
611 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
613 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
615 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
616 (ffi_prep_closure): ... this.
617 * src/pa/ffi.c: Likewise.
618 * src/cris/ffi.c: Likewise. Adjust.
619 * src/frv/ffi.c: Likewise.
620 * src/ia64/ffi.c: Likewise.
621 * src/mips/ffi.c: Likewise.
622 * src/powerpc/ffi_darwin.c: Likewise.
623 * src/s390/ffi.c: Likewise.
624 * src/sh/ffi.c: Likewise.
625 * src/sh64/ffi.c: Likewise.
626 * src/sparc/ffi.c: Likewise.
627 * src/x86/ffi64.c: Likewise.
628 * src/x86/ffi.c: Likewise.
629 (FFI_INIT_TRAMPOLINE): Adjust.
630 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
631 (ffi_prep_raw_closure): ... this.
632 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
633 (ffi_prep_closure): ... this.
634 (flush_icache): Adjust.
636 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
638 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
641 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
643 * Makefile.am: Add dummy install-pdf target.
644 * Makefile.in: Regenerate
646 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
648 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
649 ffi_closure_helper_SYSV): Add long double handling.
651 2007-02-02 Jakub Jelinek <jakub@redhat.com>
653 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
654 immediately after bctrl instruction.
656 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
658 * Makefile.am (all-recursive, install-recursive,
659 mostlyclean-recursive, clean-recursive, distclean-recursive,
660 maintainer-clean-recursive): Add missing targets.
661 * Makefile.in: Rebuilt.
663 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
665 * configure.ac: Add TARGET for x86_64-*-darwin*.
666 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
668 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
669 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
670 * configure: Regenerate.
671 * Makefile.in: Regenerate.
672 * include/Makefile.in: Regenerate.
673 * testsuite/Makefile.in: Regenerate.
674 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
677 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
679 * aclocal.m4: Regenerate with aclocal -I .. as written in the
682 2006-10-31 Geoffrey Keating <geoffk@apple.com>
684 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
685 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
687 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
688 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
690 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
691 Sandro Tolaini <tolaini@libero.it>
693 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
695 * configure: Regenerated.
696 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
697 (EXTRA_DIST): Add src/x86/darwin.S.
698 * Makefile.in: Regenerated.
699 * include/Makefile.in: Regenerated.
700 * testsuite/Makefile.in: Regenerated.
702 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
703 X86_WIN32, and additionally align stack to 16 bytes.
704 * src/x86/darwin.S: New, based on sysv.S.
705 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
707 2006-09-12 David Daney <ddaney@avtrex.com>
710 * include/Makefile.am: Install both ffi.h and ffitarget.h in
711 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
712 * aclocal.m4: Regenerated for automake 1.9.6.
713 * Makefile.in: Regenerated.
714 * include/Makefile.in: Regenerated.
715 * testsuite/Makefile.in: Regenerated.
717 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
719 * include/ffi_common.h (struct): Revert accidental commit.
721 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
723 * include/ffi_common.h: Remove lint directives.
724 * include/ffi.h.in: Likewise.
726 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
728 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
729 for 32-bit architectures.
730 * testsuite/libffi.call/return_ul.c: New test case.
732 2006-07-19 David Daney <ddaney@avtrex.com>
734 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
735 xfail remains for mips64.
737 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
739 * Makefile.am: Add install-html target. Add install-html to .PHONY
740 * Makefile.in: Regenerate.
741 * aclocal.m4: Regenerate.
742 * include/Makefile.in: Regenerate.
743 * testsuite/Makefile.in: Regenerate.
745 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
747 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
750 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
752 * README: Remove notice about 'Crazy Comments'.
753 * src/debug.c: Remove lint directives. Cleanup white spaces.
754 * src/java_raw_api.c: Likewise.
755 * src/prep_cif.c: Likewise.
756 * src/raw_api.c: Likewise.
757 * src/ffitest.c: Delete. No longer needed, all test cases migrated
759 * src/arm/ffi.c: Remove lint directives.
760 * src/m32r/ffi.c: Likewise.
761 * src/pa/ffi.c: Likewise.
762 * src/powerpc/ffi.c: Likewise.
763 * src/powerpc/ffi_darwin.c: Likewise.
764 * src/sh/ffi.c: Likewise.
765 * src/sh64/ffi.c: Likewise.
766 * src/x86/ffi.c: Likewise.
767 * testsuite/libffi.call/float2.c: Likewise.
768 * testsuite/libffi.call/promotion.c: Likewise.
769 * testsuite/libffi.call/struct1.c: Likewise.
771 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
773 * src/pa/hpux32.S: Correct unwind offset calculation for
775 * src/pa/linux.S: Likewise.
777 2006-04-12 James E Wilson <wilson@specifix.com>
780 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
781 (hfa_type_load): Call stf_spill.
782 (hfa_type_store): Call ldf_fill.
783 (ffi_call): Adjust calls to above routines. Add local temps for
786 2006-04-10 Matthias Klose <doko@debian.org>
788 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
789 directory names containing underscores.
791 2006-04-07 James E Wilson <wilson@specifix.com>
793 * testsuite/libffi.call/float4.c: New testcase.
795 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
796 Andreas Tobler <a.tobler@schweiz.ch>
798 * Makefile.am: Add PA_HPUX port.
799 * Makefile.in: Regenerate.
800 * include/Makefile.in: Likewise.
801 * testsuite/Makefile.in: Likewise.
802 * configure.ac: Add PA_HPUX rules.
803 * configure: Regenerate.
804 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
805 Add PA_HPUX and PA64_HPUX.
806 Rename FFI_LINUX ABI to FFI_PA32 ABI.
807 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
808 (FFI_TYPE_SMALL_STRUCT2): Define.
809 (FFI_TYPE_SMALL_STRUCT4): Likewise.
810 (FFI_TYPE_SMALL_STRUCT8): Likewise.
811 (FFI_TYPE_SMALL_STRUCT3): Redefine.
812 (FFI_TYPE_SMALL_STRUCT5): Likewise.
813 (FFI_TYPE_SMALL_STRUCT6): Likewise.
814 (FFI_TYPE_SMALL_STRUCT7): Likewise.
815 * src/pa/ffi.c (ROUND_DOWN): Delete.
816 (fldw, fstw, fldd, fstd): Use '__asm__'.
817 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
818 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
819 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
820 Simplify incrementing of stack slot variable. Change type of local
822 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
824 (ffi_prep_cif_machdep): Likewise.
825 (ffi_call): Likewise.
826 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
827 return type to ffi_status. Simplify incrementing of stack slot
828 variable. Only copy floating point argument registers when PA_LINUX
829 is true. Reformat debug statement.
830 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
831 FFI_TYPE_SMALL_STRUCT8.
832 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
834 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
835 Add nops to cache flush. Add trampoline for PA_HPUX.
836 * src/pa/hpux32.S: New file.
837 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
838 ffi_prep_args_LINUX to ffi_prep_args_pa32.
839 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
840 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
841 argument type checks so that common argument types appear first.
842 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
843 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
845 2006-03-24 Alan Modra <amodra@bigpond.net.au>
847 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
848 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
849 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
851 (ffi_prep_args64): Assert using IBM extended double.
852 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
853 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
854 (ffi_call): Handle FFI_LINUX.
855 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
856 gpr3 return pointer as for struct return. Handle FFI_LINUX
857 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
859 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
860 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
861 Don't use r6 as pointer to results, instead use sp offset. Don't
862 make a special call to load lr with case table address, instead
863 use offset from previous call.
864 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
865 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
868 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
870 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
871 passed with FP registers correctly.
872 (ffi_closure_helper_SYSV): Likewise.
873 * src/sh64/sysv.S: Likewise.
875 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
877 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
878 args and userdata unused.
879 (closure_test_fn1): Mark cif and userdata unused.
880 (main): Remove unused res.
882 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
884 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
885 -O2, -O3, -Os and the warning flags -W -Wall.
886 * testsuite/libffi.special/special.exp: Likewise.
887 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
888 unused parameter unused for gcc or else do nothing.
889 * testsuite/libffi.special/ffitestcxx.h: Likewise.
890 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
892 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
893 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
894 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
895 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
896 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
897 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
898 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
899 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
900 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
901 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
902 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
903 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
904 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
905 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
906 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
907 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
908 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
909 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
910 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
911 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
912 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
913 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
914 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
915 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
917 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
919 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
921 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
922 void* to avoid compiler warning.
924 (cls_struct_align_gn): Mark cif and userdata unused.
925 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
927 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
929 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
931 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
933 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
935 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
936 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
937 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
939 (main): Cast res_call to silence gcc.
940 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
942 (main): Cast res_call to silence gcc.
943 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
945 (main): Cast res_call to silence gcc.
946 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
948 (main): Cast res_call to silence gcc.
949 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
951 (main): Cast res_call to silence gcc.
952 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
954 (main): Cast res_call to silence gcc.
955 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
957 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
958 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
960 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
961 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
963 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
964 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
966 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
967 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
969 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
970 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
972 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
974 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
975 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
976 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
977 Cleanup white spaces.
978 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
979 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
980 cif and userdata unused.
981 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
983 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
984 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
985 formatters to silence gcc.
986 (B_gn): Mark cif and userdata unused.
987 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
989 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
990 (B_gn): Mark cif and userdata unused.
991 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
993 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
994 (B_gn): Mark cif and userdata unused.
995 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
997 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
998 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
999 * testsuite/libffi.call/problem1.c (stub): Likewise.
1000 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1002 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1003 in the last commit for this test case in the test case itself.
1004 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1006 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1007 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1008 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1009 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1010 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1011 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1013 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1015 * src/sh/sysv.S: Fix register numbers in the FDE for
1018 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1020 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1021 declaration to avoid a false negative on ix86. See PR323.
1023 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1025 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1026 and cast integer to void * if needed. Update the pointer to
1027 the FP register saved area correctly.
1029 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1031 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1033 * testsuite/libffi.call/nested_struct4.c: Likewise.
1035 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1037 * testsuite/libffi.call/return_dbl.c: New test case.
1038 * testsuite/libffi.call/return_dbl1.c: Likewise.
1039 * testsuite/libffi.call/return_dbl2.c: Likewise.
1040 * testsuite/libffi.call/return_fl.c: Likewise.
1041 * testsuite/libffi.call/return_fl1.c: Likewise.
1042 * testsuite/libffi.call/return_fl2.c: Likewise.
1043 * testsuite/libffi.call/return_fl3.c: Likewise.
1044 * testsuite/libffi.call/closure_fn6.c: Likewise.
1046 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1048 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1049 here to be used by other test cases too.
1051 * testsuite/libffi.call/nested_struct10.c: New test case.
1052 * testsuite/libffi.call/nested_struct9.c: Likewise.
1053 * testsuite/libffi.call/nested_struct8.c: Likewise.
1054 * testsuite/libffi.call/nested_struct7.c: Likewise.
1055 * testsuite/libffi.call/nested_struct6.c: Likewise.
1056 * testsuite/libffi.call/nested_struct5.c: Likewise.
1057 * testsuite/libffi.call/nested_struct4.c: Likewise.
1059 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1061 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1062 * configure: Rebuilt.
1064 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1066 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1067 instead do the shifting inline.
1068 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1069 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1070 and 8 byte structs, for the remaining struct sizes don't call
1071 __lshrdi3, instead do the shifting inline.
1073 2005-12-07 Thiemo Seufer <ths@networkno.de>
1075 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1076 missing parentheses.
1077 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1078 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1079 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1080 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1081 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1083 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1085 (ffi_prep_closure): Improve trampoline instruction scheduling.
1086 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1088 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1090 * src/powerpc/ffi.c: Formatting.
1091 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1092 (ffi_prep_args64): Likewise.
1094 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1096 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1097 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1099 2005-09-26 Tom Tromey <tromey@redhat.com>
1101 * testsuite/libffi.call/float1.c (value_type): New typedef.
1102 (CANARY): New define.
1103 (main): Check for result buffer overflow.
1104 * src/powerpc/linux64.S: Handle linux64 long double returns.
1105 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1106 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1108 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1111 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1113 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1115 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1117 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1118 (AH_BOTTOM): Add FFI_HIDDEN definition.
1119 * configure: Rebuilt.
1120 * fficonfig.h.in: Rebuilt.
1121 * src/powerpc/ffi.c (hidden): Remove.
1122 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1123 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1124 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1125 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1126 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1127 add FFI_HIDDEN to its prototype.
1128 (ffi_closure_SYSV_inner): New.
1129 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1130 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1132 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1135 * configure: Rebuilt.
1136 * configure.ac: Handle i*86-*-gnu*.
1138 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1140 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1141 DW_CFA_offset_extended_sf rather than
1142 DW_CFA_GNU_negative_offset_extended.
1143 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1145 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1147 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1149 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1150 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1151 partially on register.
1152 (ffi_closure_helper_SYSV): Likewise.
1153 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1155 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1157 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1159 * src/sh64/ffi.c (simple_type): Remove.
1160 (return_type): Handle small structures correctly.
1161 (ffi_prep_args): Likewise.
1162 (ffi_call): Likewise.
1163 (ffi_closure_helper_SYSV): Likewise.
1164 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1165 Emit position independent code if PIC and remove wrong datalabel
1166 prefixes from EH data.
1168 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1170 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1171 * Makefile.in: Regenerate.
1172 * include/Makefile.in: Likewise.
1173 * testsuite/Makefile.in: Likewise.
1174 * configure.ac: Add POWERPC_FREEBSD rules.
1175 * configure: Regenerate.
1176 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1177 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1178 * src/powerpc/ffi.c: Add flags to handle small structure returns
1180 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1182 (ffi_closure_helper_SYSV): Likewise.
1183 * src/powerpc/ppc_closure.S: Add return types for small structures.
1184 * src/powerpc/sysv.S: Add bits to handle small structures for
1187 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1189 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1190 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1191 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1193 2005-07-05 Randolph Chung <tausq@debian.org>
1195 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1196 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1197 structures. Kill compilation warnings.
1198 (ffi_closure_inner_LINUX): Print return values as hex in debug
1199 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1200 Properly handle 5-7 byte structure returns.
1201 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1202 (FFI_TYPE_SMALL_STRUCT2): Remove.
1203 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1204 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1205 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1206 (checksmst1, checksmst2): Remove.
1207 (checksmst3): Optimize handling of 3-byte struct returns.
1208 (checksmst567): Properly handle 5-7 byte struct returns.
1210 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1213 * src/mips/n32.S: Enforce PIC code.
1214 * src/mips/o32.S: Likewise.
1216 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1218 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1219 * configure: Regenerate.
1221 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1223 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1224 to call ffi_closure_helper_SYSV. Append @local instead.
1225 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1227 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1229 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1230 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1231 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1232 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1233 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1235 2005-05-09 Mike Stump <mrs@apple.com>
1237 * configure: Regenerate.
1239 2005-05-08 Richard Henderson <rth@redhat.com>
1242 * src/alpha/osf.S: Update unwind into to match code.
1244 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1245 Richard Henderson <rth@redhat.com>
1247 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1249 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1250 (ffi_prep_closure): Set carry bit if sse-used flag set.
1251 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1252 Only load sse registers if ssecount non-zero.
1253 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1255 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1257 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1258 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1259 * configure: Regenerate.
1261 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1263 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1264 have Tcl8.3-compatible intermediate variable.
1266 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1267 Hans-Peter Nilsson <hp@axis.com>
1269 * Makefile.am: Add CRIS support.
1270 * configure.ac: Likewise.
1271 * Makefile.in, configure, testsuite/Makefile.in,
1272 include/Makefile.in: Regenerate.
1273 * src/cris: New directory.
1274 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1275 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1277 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1278 \r?\n in output tests.
1280 2005-04-12 Mike Stump <mrs@apple.com>
1282 * configure: Regenerate.
1284 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1286 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1288 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1290 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1292 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1293 to 64 bits in ILP32 mode.
1294 Load 64 bits even for short data.
1296 2005-03-23 Mike Stump <mrs@apple.com>
1298 * src/powerpc/darwin.S: Update for -m64 multilib.
1299 * src/powerpc/darwin_closure.S: Likewise.
1301 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1303 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1304 Do not set tool_include_dir.
1305 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1307 * include/Makefile.am: Set gcc_version and toollibffidir.
1308 * include/Makefile.in: Regenerate.
1310 2005-02-22 Andrew Haley <aph@redhat.com>
1312 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1313 odd-numbered register pairs for 64-bit integer types.
1315 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1318 * testsuite/libffi.call/return_ll1.c: New test case.
1320 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1322 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1323 * testsuite/libffi.call/float.c: Ditto.
1324 * testsuite/libffi.call/float2.c: Ditto.
1325 * testsuite/libffi.call/float3.c: Ditto.
1327 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1329 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1331 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1333 * testsuite/libffi.special/special.exp (cxx_options): Add
1336 2004-12-31 Richard Henderson <rth@redhat.com>
1338 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1339 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1340 offset parameters with a type parameter; deduce size and structure
1341 alignment. Update all users.
1343 2004-12-31 Richard Henderson <rth@redhat.com>
1345 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1346 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1347 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1348 into ffi_prep_closure.
1349 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1352 2004-12-27 Richard Henderson <rth@redhat.com>
1354 * src/x86/unix64.S: Fix typo in unwind info.
1356 2004-12-25 Richard Henderson <rth@redhat.com>
1358 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1359 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1360 (merge_classes): Check for it.
1362 (classify_argument): Pass byte_offset by value; perform all updates
1364 (examine_argument): Add classes argument; handle
1365 X86_64_COMPLEX_X87_CLASS.
1366 (ffi_prep_args): Merge into ...
1367 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1368 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1369 (ffi_fill_return_value): Remove.
1370 (ffi_prep_closure): Remove dead assert.
1371 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1372 Rewrite to use struct register_args instead of va_list. Create
1373 flags for handling structure returns.
1374 * src/x86/unix64.S: Remove dead strings.
1375 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1376 stack frame with ffi_call. Handle structure returns properly.
1377 (float2sse, floatfloat2sse, double2sse): Remove.
1378 (sse2float, sse2double, sse2floatfloat): Remove.
1379 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1380 to handle structure returns properly.
1382 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1384 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1386 * Makefile.in: Regenerated.
1388 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1390 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1391 * configure, aclocal.m4, Makefile.in: Regenerate.
1392 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1394 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1396 * configure: Regenerate for libtool change.
1398 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1400 * configure: Regenerate for libtool reversion.
1402 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1404 * configure: Regenerate for libtool change.
1406 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1408 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1410 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1412 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1413 of jal. Use an absolute encoding for the frame information.
1415 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1417 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1418 * acinclude.m4: Delete logic for sincludes.
1419 * aclocal.m4, Makefile.in, configure: Regenerate.
1420 * include/Makefile: Likewise.
1421 * testsuite/Makefile: Likewise.
1423 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1425 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1426 on a 8-byte boundary.
1427 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1429 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1431 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1432 long long values. Round stack allocation to a multiple of 8 bytes
1433 for ATPCS compatibility.
1434 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1435 names. Handle returning long long types. Add Thumb and interworking
1436 support. Improve soft-float code.
1438 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1440 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1441 (libffi_exit): New function.
1442 (libffi_init): Build the testglue wrapper if needed.
1444 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1447 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1449 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1451 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1453 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1455 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1456 * testsuite/libffi.call/float3.c: New test case.
1458 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1460 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1461 the function returning a structure pointed with R2.
1462 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1463 the structure return value if T bit set. Emit position
1464 independent code and EH data if PIC.
1466 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1468 * Makefile.am: Add m32r support.
1469 * configure.ac: Likewise.
1470 * Makefile.in: Regenerate.
1471 * confiugre: Regenerate.
1472 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1473 (uint64, sint64, double, longdouble)
1474 * src/m32r: New directory.
1475 * src/m32r/ffi.c: New file.
1476 * src/m32r/sysv.S: Likewise.
1477 * src/m32r/ffitarget.h: Likewise.
1479 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1481 * testsuite/libffi.call/negint.c: New test case.
1483 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1486 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1487 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1488 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1491 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1493 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1494 * testsuite/libffi.call/promotion.c: Likewise.
1495 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1497 * testsuite/libffi.call/return_sc.c: Likewise.
1498 * testsuite/libffi.call/return_uc.c: Likewise.
1500 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1502 * src/powerpc/darwin.S: Fix comments and identation.
1503 * src/powerpc/darwin_closure.S: Likewise.
1505 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1507 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1508 (ffi_prep_args): Handle longdouble arguments.
1509 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1511 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1512 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1514 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1515 * src/types.c: Defined longdouble size and alignment for darwin.
1517 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1519 * src/powerpc/aix.S: Remove whitespaces.
1520 * src/powerpc/aix_closure.S: Likewise.
1521 * src/powerpc/asm.h: Likewise.
1522 * src/powerpc/ffi.c: Likewise.
1523 * src/powerpc/ffitarget.h: Likewise.
1524 * src/powerpc/linux64.S: Likewise.
1525 * src/powerpc/linux64_closure.S: Likewise.
1526 * src/powerpc/ppc_closure.S: Likewise.
1527 * src/powerpc/sysv.S: Likewise.
1529 2004-08-30 Anthony Green <green@redhat.com>
1531 * Makefile.am: Add frv support.
1532 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1533 * configure.ac: Read configure.host.
1534 * configure.in: Read configure.host.
1535 * configure.host: New file. frv-elf needs libgloss.
1536 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1537 alignment. This is needed to frv and shouldn't harm the others.
1538 * include/ffi_common.h (ALIGN_DOWN): New macro.
1539 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1541 2004-08-24 David Daney <daney@avtrex.com>
1543 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1544 * testsuite/libffi.call/closure_fn1.c: Likewise.
1545 * testsuite/libffi.call/closure_fn2.c Likewise.
1546 * testsuite/libffi.call/closure_fn3.c: Likewise.
1547 * testsuite/libffi.call/closure_fn4.c: Likewise.
1548 * testsuite/libffi.call/closure_fn5.c: Likewise.
1549 * testsuite/libffi.call/cls_18byte.c: Likewise.
1550 * testsuite/libffi.call/cls_19byte.c: Likewise.
1551 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1552 * testsuite/libffi.call/cls_20byte.c: Likewise.
1553 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1554 * testsuite/libffi.call/cls_24byte.c: Likewise.
1555 * testsuite/libffi.call/cls_2byte.c: Likewise.
1556 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1557 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1558 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1559 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1560 * testsuite/libffi.call/cls_4byte.c: Likewise.
1561 * testsuite/libffi.call/cls_64byte.c: Likewise.
1562 * testsuite/libffi.call/cls_6byte.c: Likewise.
1563 * testsuite/libffi.call/cls_7byte.c: Likewise.
1564 * testsuite/libffi.call/cls_8byte.c: Likewise.
1565 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1566 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1567 * testsuite/libffi.call/cls_align_double.c: Likewise.
1568 * testsuite/libffi.call/cls_align_float.c: Likewise.
1569 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1570 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1571 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1572 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1573 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1574 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1575 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1576 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1577 * testsuite/libffi.call/cls_double.c: Likewise.
1578 * testsuite/libffi.call/cls_float.c: Likewise.
1579 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1580 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1581 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1582 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1583 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1584 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1585 * testsuite/libffi.call/cls_schar.c: Likewise.
1586 * testsuite/libffi.call/cls_sint.c: Likewise.
1587 * testsuite/libffi.call/cls_sshort.c: Likewise.
1588 * testsuite/libffi.call/cls_uchar.c: Likewise.
1589 * testsuite/libffi.call/cls_uint.c: Likewise.
1590 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1591 * testsuite/libffi.call/cls_ushort.c: Likewise.
1592 * testsuite/libffi.call/nested_struct.c: Likewise.
1593 * testsuite/libffi.call/nested_struct1.c: Likewise.
1594 * testsuite/libffi.call/nested_struct2.c: Likewise.
1595 * testsuite/libffi.call/nested_struct3.c: Likewise.
1596 * testsuite/libffi.call/problem1.c: Likewise.
1597 * testsuite/libffi.special/unwindtest.cc: Likewise.
1598 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1600 * testsuite/libffi.call/cls_16byte.c: Likewise.
1601 * testsuite/libffi.call/cls_5byte.c: Likewise.
1603 2004-08-23 David Daney <daney@avtrex.com>
1606 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1607 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1608 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1609 parameters and return types.
1610 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1611 (ffi_prep_closure): Ditto.
1612 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1613 alignment calculations.
1614 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1615 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1617 2004-08-14 Casey Marshall <csm@gnu.org>
1619 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1620 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1621 integer (O32 ABI only).
1622 (ffi_prep_closure): new function.
1623 (ffi_closure_mips_inner_O32): new function.
1624 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1625 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1626 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1627 64 bit integers correctly.
1628 (ffi_closure_O32): new function.
1629 Added DWARF-2 unwind info for both functions.
1631 2004-08-10 Andrew Haley <aph@redhat.com>
1633 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1635 2004-08-01 Robert Millan <robertmh@gnu.org>
1637 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1638 * configure: Regenerate.
1640 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1642 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1643 and mmap() explicitly instead of relying on preset autoconf cache
1645 * aclocal.m4: Regenerate.
1646 * configure: Regenerate.
1648 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1650 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1651 (ffi_check_float_struct): Remove unused prototype.
1653 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1655 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1656 character on Darwin, use '\n\t' instead.
1658 2004-06-26 Matthias Klose <doko@debian.org>
1660 * libtool-version: Fix typo in revision/age.
1662 2004-06-17 Matthias Klose <doko@debian.org>
1664 * libtool-version: New.
1665 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1666 * Makefile.in: Regenerate.
1668 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1670 * Makefile.am: Remove useless multilib rules.
1671 * Makefile.in: Regenerate.
1672 * aclocal.m4: Regenerate with automake 1.8.5.
1673 * configure.ac: Remove useless multilib configury.
1674 * configure: Regenerate.
1676 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1678 * .cvsignore: New file.
1680 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1682 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1684 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1687 2004-06-06 Sean McNeil <sean@mcneil.com>
1689 * configure.ac: Add x86_64-*-freebsd* support.
1690 * configure: Regenerate.
1692 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1695 * configure.ac: Test for existence of mmap and sys/mman.h before
1696 checking blacklist. Fix suggested by Jim Wilson.
1697 * configure: Regenerate.
1699 2004-04-26 Matt Austern <austern@apple.com>
1701 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1703 * src/powerpc/darwin_closure.S: Likewise.
1705 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1707 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1708 error. Reported by Thomas Heller <theller@python.net>.
1709 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1710 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1712 2004-03-20 Matthias Klose <doko@debian.org>
1714 * src/pa/linux.S: Fix typo.
1716 2004-03-19 Matthias Klose <doko@debian.org>
1718 * Makefile.am: Update.
1719 * Makefile.in: Regenerate.
1720 * src/pa/ffi.h.in: Remove.
1721 * src/pa/ffitarget.h: New file.
1723 2004-02-10 Randolph Chung <tausq@debian.org>
1725 * Makefile.am: Add PA support.
1726 * Makefile.in: Regenerate.
1727 * include/Makefile.in: Regenerate.
1728 * configure.ac: Add PA target.
1729 * configure: Regenerate.
1730 * src/pa/ffi.c: New file.
1731 * src/pa/ffi.h.in: Add PA support.
1732 * src/pa/linux.S: New file.
1733 * prep_cif.c: Add PA support.
1735 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1737 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1739 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1740 with ecif->cif->flags.
1741 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1743 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1744 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1745 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1746 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1748 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1750 * configure.in: Rename file to ...
1751 * configure.ac: ... this.
1752 * fficonfig.h.in: Regenerate.
1753 * Makefile.in: Regenerate.
1754 * include/Makefile.in: Regenerate.
1755 * testsuite/Makefile.in: Regenerate.
1757 2004-03-12 Matt Austern <austern@apple.com>
1759 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1760 changes in EH format resulting from addition of linkonce support.
1761 * src/powerpc/darwin_closure.S: Likewise.
1763 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1764 Paolo Bonzini <bonzini@gnu.org>
1766 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1767 Remove VPATH. Remove rules for object files. Remove multilib support.
1768 (AM_CCASFLAGS): Add.
1769 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1770 (AC_PREREQ): Bump version to 2.59.
1771 (AC_INIT): Fill with version info and bug address.
1772 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1773 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1774 De-precious CC so that the right flags are passed down to multilibs.
1775 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1776 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1777 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1778 * configure: Rebuilt.
1779 * aclocal.m4: Likewise.
1780 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1781 * fficonfig.h.in: Likewise.
1783 2004-03-11 Andreas Schwab <schwab@suse.de>
1785 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1786 arguments from fp registers only for the first 8 parameter slots.
1787 Don't convert a float parameter when passed in memory.
1789 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1791 * configure: Regenerate for config/accross.m4 correction.
1793 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1795 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1796 ecif->cif->bytes to bytes.
1797 (ffi_prep_cif_machdep): Add braces around nested if statement.
1799 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1801 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1803 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1804 (ffi_closure_helper_LINUX64): Fix typo.
1805 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1807 * testsuite/libffi.call/float.c: Likewise.
1808 * testsuite/libffi.call/float2.c: Likewise.
1810 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1812 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1813 long double function return and long double arg handling.
1814 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1815 Use "end_pfr" instead of "nf". Correct long double handling.
1817 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1819 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1820 space for long double return value. Adjust stack frame and offsets.
1821 Load f2 long double return.
1823 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1825 * src/types.c: Use 16 byte long double for POWERPC64.
1827 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1829 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1830 when the structure return address is passed in %o0.
1831 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1832 (ffi_v9_layout_struct): Align the field following a nested structure
1833 on a word boundary. Use memmove instead of memcpy.
1834 (ffi_call): Update call to ffi_V9_return_struct.
1835 (ffi_prep_closure): Define 'ctx' only for V8.
1836 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1837 and ffi_closure_sparc_inner_v9.
1838 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1839 Always skip the structure return address. For structures and long
1840 doubles, copy the argument directly.
1841 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1842 if required. Shift the maximum floating-point slot accordingly. For
1843 big structures, copy the argument directly; otherwise, left-justify the
1844 argument and call ffi_v9_layout_struct to lay out the structure on
1846 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1847 (ffi_closure_v8): Pass the structure return address. Update call to
1848 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1849 Skip the 'unimp' insn when returning long doubles and structures.
1850 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1851 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1852 FFI_TYPE_INT handling. Load structures both in integers and
1853 floating-point registers on return.
1854 * README: Update status of the SPARC port.
1856 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1858 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1860 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1862 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1864 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1865 value as of type ffi_arg, not unsigned int.
1867 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1869 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1872 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1874 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1877 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1879 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1882 2003-12-10 Richard Henderson <rth@redhat.com>
1884 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1885 size_t instead of int.
1887 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1889 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1890 * testsuite/libffi.call/many_win32.c (main): Replace variable
1891 int i with unsigned long ul.
1893 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1894 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1895 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1896 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1897 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1898 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1899 * testsuite/libffi.call/cls_align_float.c: Likewise.
1900 * testsuite/libffi.call/cls_align_double.c: Likewise.
1901 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1902 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1904 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1907 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1908 Align arguments to 32 bits.
1910 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1913 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1914 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1915 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1916 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1917 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1918 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1920 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1922 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1924 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1925 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1927 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1929 * Makefile.in: Rebuilt.
1930 * configure: Likewise.
1931 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1934 2003-11-21 Andreas Jaeger <aj@suse.de>
1935 Andreas Tobler <a.tobler@schweiz.ch>
1937 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1938 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1939 * Makefile.in: Rebuilt.
1940 * aclocal.m4: Likewise.
1941 * configure: Likewise.
1942 * fficonfig.h.in: Likewise.
1943 * testsuite/lib/libffi-dg.exp: Add include dir.
1944 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1945 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1946 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1947 for ffi_closure if available.
1948 * testsuite/libffi.call/closure_fn1.c: Likewise.
1949 * testsuite/libffi.call/closure_fn2.c: Likewise.
1950 * testsuite/libffi.call/closure_fn3.c: Likewise.
1951 * testsuite/libffi.call/closure_fn4.c: Likewise.
1952 * testsuite/libffi.call/closure_fn5.c: Likewise.
1953 * testsuite/libffi.call/cls_12byte.c: Likewise.
1954 * testsuite/libffi.call/cls_16byte.c: Likewise.
1955 * testsuite/libffi.call/cls_18byte.c: Likewise.
1956 * testsuite/libffi.call/cls_19byte.c: Likewise.
1957 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1958 * testsuite/libffi.call/cls_20byte.c: Likewise.
1959 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1960 * testsuite/libffi.call/cls_24byte.c: Likewise.
1961 * testsuite/libffi.call/cls_2byte.c: Likewise.
1962 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1963 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1964 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1965 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1966 * testsuite/libffi.call/cls_4byte.c: Likewise.
1967 * testsuite/libffi.call/cls_5byte.c: Likewise.
1968 * testsuite/libffi.call/cls_64byte.c: Likewise.
1969 * testsuite/libffi.call/cls_6byte.c: Likewise.
1970 * testsuite/libffi.call/cls_7byte.c: Likewise.
1971 * testsuite/libffi.call/cls_8byte.c: Likewise.
1972 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1973 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1974 * testsuite/libffi.call/cls_double.c: Likewise.
1975 * testsuite/libffi.call/cls_float.c: Likewise.
1976 * testsuite/libffi.call/cls_schar.c: Likewise.
1977 * testsuite/libffi.call/cls_sint.c: Likewise.
1978 * testsuite/libffi.call/cls_sshort.c: Likewise.
1979 * testsuite/libffi.call/cls_uchar.c: Likewise.
1980 * testsuite/libffi.call/cls_uint.c: Likewise.
1981 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1982 * testsuite/libffi.call/cls_ushort.c: Likewise.
1983 * testsuite/libffi.call/nested_struct.c: Likewise.
1984 * testsuite/libffi.call/nested_struct1.c: Likewise.
1985 * testsuite/libffi.call/nested_struct2.c: Likewise.
1986 * testsuite/libffi.call/nested_struct3.c: Likewise.
1987 * testsuite/libffi.call/problem1.c: Likewise.
1988 * testsuite/libffi.special/unwindtest.cc: Likewise.
1990 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
1992 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
1994 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
1996 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
1997 Add -lgcc_s to additional flags.
1999 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2001 * configure.in, include/Makefile.am: PR libgcj/11147, install
2002 the ffitarget.h header file in a gcc versioned and target
2004 * configure: Regenerated.
2005 * Makefile.in, include/Makefile.in: Likewise.
2006 * testsuite/Makefile.in: Likewise.
2008 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2010 * testsuite/libffi.call/closure_fn0.c: Print result and check
2011 with dg-output to make debugging easier.
2012 * testsuite/libffi.call/closure_fn1.c: Likewise.
2013 * testsuite/libffi.call/closure_fn2.c: Likewise.
2014 * testsuite/libffi.call/closure_fn3.c: Likewise.
2015 * testsuite/libffi.call/closure_fn4.c: Likewise.
2016 * testsuite/libffi.call/closure_fn5.c: Likewise.
2017 * testsuite/libffi.call/cls_12byte.c: Likewise.
2018 * testsuite/libffi.call/cls_16byte.c: Likewise.
2019 * testsuite/libffi.call/cls_18byte.c: Likewise.
2020 * testsuite/libffi.call/cls_19byte.c: Likewise.
2021 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2022 * testsuite/libffi.call/cls_20byte.c: Likewise.
2023 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2024 * testsuite/libffi.call/cls_24byte.c: Likewise.
2025 * testsuite/libffi.call/cls_2byte.c: Likewise.
2026 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2027 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2028 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2029 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2030 * testsuite/libffi.call/cls_4byte.c: Likewise.
2031 * testsuite/libffi.call/cls_5byte.c: Likewise.
2032 * testsuite/libffi.call/cls_64byte.c: Likewise.
2033 * testsuite/libffi.call/cls_6byte.c: Likewise.
2034 * testsuite/libffi.call/cls_7byte.c: Likewise.
2035 * testsuite/libffi.call/cls_8byte.c: Likewise.
2036 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2037 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2038 * testsuite/libffi.call/cls_double.c: Likewise.
2039 * testsuite/libffi.call/cls_float.c: Likewise.
2040 * testsuite/libffi.call/cls_schar.c: Likewise.
2041 * testsuite/libffi.call/cls_sint.c: Likewise.
2042 * testsuite/libffi.call/cls_sshort.c: Likewise.
2043 * testsuite/libffi.call/cls_uchar.c: Likewise.
2044 * testsuite/libffi.call/cls_uint.c: Likewise.
2045 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2046 * testsuite/libffi.call/cls_ushort.c: Likewise.
2047 * testsuite/libffi.call/problem1.c: Likewise.
2049 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2052 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2054 * testsuite/libffi.call/cls_9byte2.c: New test case.
2055 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2056 * testsuite/libffi.call/cls_64byte.c: Likewise.
2057 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2058 * testsuite/libffi.call/cls_19byte.c: Likewise.
2059 * testsuite/libffi.call/cls_18byte.c: Likewise.
2060 * testsuite/libffi.call/closure_fn4.c: Likewise.
2061 * testsuite/libffi.call/closure_fn5.c: Likewise.
2062 * testsuite/libffi.call/cls_schar.c: Likewise.
2063 * testsuite/libffi.call/cls_sint.c: Likewise.
2064 * testsuite/libffi.call/cls_sshort.c: Likewise.
2065 * testsuite/libffi.call/nested_struct2.c: Likewise.
2066 * testsuite/libffi.call/nested_struct3.c: Likewise.
2068 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2070 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2071 * testsuite/libffi.call/cls_uchar.c: Likewise.
2072 * testsuite/libffi.call/cls_uint.c: Likewise.
2073 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2074 * testsuite/libffi.call/cls_ushort.c: Likewise.
2075 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2077 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2079 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2082 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2084 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2085 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2087 2003-10-22 David Daney <ddaney@avtrex.com>
2089 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2090 __attribute__((__mode__(__SI__))) and friends.
2092 2003-10-22 Andreas Schwab <schwab@suse.de>
2094 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2096 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2098 * configure.in: AC_LINK_FILES(ffitarget.h).
2099 * configure: Regenerate.
2100 * Makefile.in: Likewise.
2101 * include/Makefile.in: Likewise.
2102 * testsuite/Makefile.in: Likewise.
2103 * fficonfig.h.in: Likewise.
2105 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2106 Richard Henderson <rth@redhat.com>
2108 Avoid that ffi.h includes fficonfig.h.
2110 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2111 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2112 (TARGET_SRC_MIPS_SGI): Removed.
2113 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2114 (MIPS_SGI): Removed.
2115 (CLEANFILES): Removed.
2116 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2118 * acconfig.h: Removed.
2119 * configure.in: Compute sizeofs only for double and long double.
2120 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2121 into AC_DEFINE instead of using acconfig.h. Create
2122 include/ffitarget.h instead of include/fficonfig.h. Rename
2123 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2124 AC_DEFINE EH_FRAME_FLAGS.
2125 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2126 (hack_DATA): Add ffitarget.h.
2127 * include/ffi.h.in: Remove all system specific definitions.
2128 Declare raw API even if it is not installed, why bother?
2129 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2130 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2131 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2132 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2133 * include/ffi_common.h (bool): Do not define.
2134 (ffi_assert): Accept failed assertion.
2135 (ffi_type_test): Return void and accept file/line.
2136 (FFI_ASSERT): Pass stringized failed assertion.
2137 (FFI_ASSERT_AT): New macro.
2138 (FFI_ASSERT_VALID_TYPE): New macro.
2139 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2140 UINT64, SINT64): Define here with gcc's __attribute__ macro
2142 (FLOAT32, ALIGN): Define here instead of in ffi.h
2143 * include/ffi-mips.h: Removed. Its content moved to
2144 src/mips/ffitarget.h after separating assembly and C sections.
2145 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2146 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2147 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2148 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2149 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2150 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2151 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2152 (ffi_type_test): Rewritten.
2153 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2154 FFI_ASSERT_VALID_TYPE.
2155 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2156 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2157 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2158 src/s390/ffitarget.h, src/sh/ffitarget.h,
2159 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2160 src/x86/ffitarget.h: New files.
2161 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2162 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2163 src/powerpc/aix.S, src/powerpc/darwin.S,
2164 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2165 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2166 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2167 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2168 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2171 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2173 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2174 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2176 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2178 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2179 Used when FFI_DEBUG = 1.
2181 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2183 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2186 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2188 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2189 FFI_MIPS_O32 for O32 ABI.
2191 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2193 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2194 SPARC64. Cleanup whitespaces.
2196 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2198 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2199 strongarm, xscale. Cleanup whitespaces.
2200 * testsuite/libffi.call/closure_fn1.c: Likewise.
2201 * testsuite/libffi.call/closure_fn2.c: Likewise.
2202 * testsuite/libffi.call/closure_fn3.c: Likewise.
2203 * testsuite/libffi.call/cls_12byte.c: Likewise.
2204 * testsuite/libffi.call/cls_16byte.c: Likewise.
2205 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2206 * testsuite/libffi.call/cls_20byte.c: Likewise.
2207 * testsuite/libffi.call/cls_24byte.c: Likewise.
2208 * testsuite/libffi.call/cls_2byte.c: Likewise.
2209 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2210 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2211 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2212 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2213 * testsuite/libffi.call/cls_4byte.c: Likewise.
2214 * testsuite/libffi.call/cls_5byte.c: Likewise.
2215 * testsuite/libffi.call/cls_6byte.c: Likewise.
2216 * testsuite/libffi.call/cls_7byte.c: Likewise.
2217 * testsuite/libffi.call/cls_8byte.c: Likewise.
2218 * testsuite/libffi.call/cls_double.c: Likewise.
2219 * testsuite/libffi.call/cls_float.c: Likewise.
2220 * testsuite/libffi.call/cls_uchar.c: Likewise.
2221 * testsuite/libffi.call/cls_uint.c: Likewise.
2222 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2223 * testsuite/libffi.call/cls_ushort.c: Likewise.
2224 * testsuite/libffi.call/nested_struct.c: Likewise.
2225 * testsuite/libffi.call/nested_struct1.c: Likewise.
2226 * testsuite/libffi.call/problem1.c: Likewise.
2227 * testsuite/libffi.special/unwindtest.cc: Likewise.
2228 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2230 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2232 * src/powerpc/aix.S: Cleanup whitespaces.
2233 * src/powerpc/aix_closure.S: Likewise.
2235 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2237 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2238 * src/powerpc/darwin_closure.S: Likewise.
2239 * src/powerpc/ffi_darwin.c: Likewise.
2241 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2242 David Edelsohn <edelsohn@gnu.org>
2244 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2245 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2247 * src/powerpc/darwin_closure.S: Likewise.
2248 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2249 according to the Darwin/AIX ABI.
2250 (ffi_prep_cif_machdep): Likewise.
2251 (ffi_closure_helper_DARWIN): Likewise.
2252 Remove the outgoing parameter stack logic. Simplify the evaluation
2253 of the different CASE types.
2254 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2255 statement in the trampoline code.
2257 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2259 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2260 for the register size.
2261 (ffi_closure_helper_SYSV): Handle the structure return value
2263 (ffi_closure_helper_SYSV): Return the appropriate type when
2264 the registers are used for the structure return value.
2265 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2266 the 64-bit return value. Update copyright years.
2268 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2270 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2271 srcdir for ffi_mips.h.
2273 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2275 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2277 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2278 placement of float result.
2279 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2280 cast of "resp" for big-endian 64 bit machines.
2282 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2284 * src/types.c (double, longdouble): Merge identical SH and ARM
2285 typedefs, and add POWERPC64.
2286 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2287 struct split over gpr and rest.
2288 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2289 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2291 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2293 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2296 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2298 * configure: Regenerate.
2300 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2302 * Makefile.am: Remove build rules for ffitest.
2303 * Makefile.in: Rebuilt.
2305 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2307 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2308 about implicit declaration of abort().
2310 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2312 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2313 * Makefile.in: Rebuilt.
2314 * configure.in: Add dejagnu test framework.
2315 * configure: Rebuilt.
2317 * testsuite/Makefile.am: New file.
2318 * testsuite/Makefile.in: Built
2319 * testsuite/lib/libffi-dg.exp: New file.
2320 * testsuite/config/default.exp: Likewise.
2321 * testsuite/libffi.call/call.exp: Likewise.
2322 * testsuite/libffi.call/ffitest.h: Likewise.
2323 * testsuite/libffi.call/closure_fn0.c: Likewise.
2324 * testsuite/libffi.call/closure_fn1.c: Likewise.
2325 * testsuite/libffi.call/closure_fn2.c: Likewise.
2326 * testsuite/libffi.call/closure_fn3.c: Likewise.
2327 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2328 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2329 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2330 * testsuite/libffi.call/cls_2byte.c: Likewise.
2331 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2332 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2333 * testsuite/libffi.call/cls_4byte.c: Likewise.
2334 * testsuite/libffi.call/cls_5byte.c: Likewise.
2335 * testsuite/libffi.call/cls_6byte.c: Likewise.
2336 * testsuite/libffi.call/cls_7byte.c: Likewise.
2337 * testsuite/libffi.call/cls_8byte.c: Likewise.
2338 * testsuite/libffi.call/cls_12byte.c: Likewise.
2339 * testsuite/libffi.call/cls_16byte.c: Likewise.
2340 * testsuite/libffi.call/cls_20byte.c: Likewise.
2341 * testsuite/libffi.call/cls_24byte.c: Likewise.
2342 * testsuite/libffi.call/cls_double.c: Likewise.
2343 * testsuite/libffi.call/cls_float.c: Likewise.
2344 * testsuite/libffi.call/cls_uchar.c: Likewise.
2345 * testsuite/libffi.call/cls_uint.c: Likewise.
2346 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2347 * testsuite/libffi.call/cls_ushort.c: Likewise.
2348 * testsuite/libffi.call/float.c: Likewise.
2349 * testsuite/libffi.call/float1.c: Likewise.
2350 * testsuite/libffi.call/float2.c: Likewise.
2351 * testsuite/libffi.call/many.c: Likewise.
2352 * testsuite/libffi.call/many_win32.c: Likewise.
2353 * testsuite/libffi.call/nested_struct.c: Likewise.
2354 * testsuite/libffi.call/nested_struct1.c: Likewise.
2355 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2356 * testsuite/libffi.call/problem1.c: Likewise.
2357 * testsuite/libffi.call/promotion.c: Likewise.
2358 * testsuite/libffi.call/return_ll.c: Likewise.
2359 * testsuite/libffi.call/return_sc.c: Likewise.
2360 * testsuite/libffi.call/return_uc.c: Likewise.
2361 * testsuite/libffi.call/strlen.c: Likewise.
2362 * testsuite/libffi.call/strlen_win32.c: Likewise.
2363 * testsuite/libffi.call/struct1.c: Likewise.
2364 * testsuite/libffi.call/struct2.c: Likewise.
2365 * testsuite/libffi.call/struct3.c: Likewise.
2366 * testsuite/libffi.call/struct4.c: Likewise.
2367 * testsuite/libffi.call/struct5.c: Likewise.
2368 * testsuite/libffi.call/struct6.c: Likewise.
2369 * testsuite/libffi.call/struct7.c: Likewise.
2370 * testsuite/libffi.call/struct8.c: Likewise.
2371 * testsuite/libffi.call/struct9.c: Likewise.
2372 * testsuite/libffi.special/special.exp: New file.
2373 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2374 * testsuite/libffi.special/unwindtest.cc: Likewise.
2377 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2379 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2382 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2384 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2386 (ffi_closure_helper_LINUX64): Likewise.
2387 * src/powerpc/linux64.S: Remove code writing to parm save area.
2388 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2389 address in lr from ffi_closure_helper_LINUX64 call to calculate
2390 table address. Optimize function tail.
2392 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2394 * src/sparc/ffi.c: Handle all floating point registers.
2395 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2397 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2399 * README: Note that libffi is not part of GCC. Update the project
2402 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2404 * src/powerpc/ppc_closure.S: Include ffi.h.
2406 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2408 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2409 Use C style comments.
2411 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2413 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2414 * Makefile.in: Regenerate.
2415 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2416 * configure: Regenerate.
2417 * include/ffi.h.in: Add SHmedia support.
2418 * src/sh64/ffi.c: New file.
2419 * src/sh64/sysv.S: New file.
2421 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2423 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2424 should be read-only.
2425 * configure: Rebuilt.
2426 * fficonfig.h.in: Rebuilt.
2427 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2428 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2429 * src/powerpc/linux64.S: Likewise.
2430 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2431 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2432 if -fpic/-fPIC/-mrelocatable.
2433 * src/powerpc/powerpc_closure.S: Likewise.
2434 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2435 #write in .eh_frame flags.
2436 * src/sparc/v9.S: Likewise.
2437 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2438 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2439 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2441 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2443 Fixes PR bootstrap/10656
2444 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2445 support for .register pseudo-op.
2446 * src/sparc/v8.S: Use it.
2447 * fficonfig.h.in: Rebuilt.
2448 * configure: Rebuilt.
2450 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2452 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2453 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2454 Make it the default on POWERPC64.
2455 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2456 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2457 * configure: Rebuilt.
2458 * src/powerpc/ffi.c (hidden): Define.
2459 (ffi_prep_args_SYSV): Renamed from
2460 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2461 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2462 ASM_NEEDS_REGISTERS64): New.
2463 (ffi_prep_args64): New function.
2464 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2465 (ffi_call): Likewise.
2466 (ffi_prep_closure): Likewise.
2467 (flush_icache): Surround by #ifndef POWERPC64.
2468 (ffi_dblfl): New union type.
2469 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2470 (ffi_closure_helper_LINUX64): New function.
2471 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2473 * src/powerpc/sysv.S: Likewise.
2474 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2475 * src/powerpc/linux64.S: New file.
2476 * src/powerpc/linux64_closure.S: New file.
2477 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2478 src/powerpc/linux64_closure.S.
2479 (TARGET_SRC_POWERPC): Likewise.
2481 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2482 closure_test_fn3): Fix result printing on big-endian 64-bit
2484 (main): Print tst2_arg instead of uninitialized tst2_result.
2486 * src/ffitest.c (main): Hide what closure pointer really points to
2489 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2491 * configure.in (arm-*-netbsdelf*): Add configuration.
2492 (configure): Regenerated.
2494 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2496 * include/Makefile.in: Regenerate.
2498 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2500 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2502 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2503 Receive closure pointer through parameter, read args using
2504 __builtin_dwarf_cfa.
2505 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2507 2003-03-12 Andreas Schwab <schwab@suse.de>
2509 * configure.in: Avoid trailing /. in toolexeclibdir.
2510 * configure: Rebuilt.
2512 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2514 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2516 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2518 * libffi/src/powerpc/darwin_closure.S:
2519 Fix alignement bug, allocate 8 bytes for the result.
2520 * libffi/src/powerpc/aix_closure.S:
2522 * libffi/src/powerpc/ffi_darwin.c:
2523 Update stackframe description for aix/darwin_closure.S.
2525 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2527 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2530 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2531 Andreas Schwab <schwab@suse.de>
2533 * configure.in: Adjust command to source config-ml.in to account
2534 for changes to the libffi_basedir definition.
2535 (libffi_basedir): Remove ${srcdir} from value and include trailing
2538 * configure: Regenerate.
2540 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2542 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2544 2003-01-28 Andrew Haley <aph@redhat.com>
2546 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2547 * src/x86/ffi64.c (ffi_prep_closure): New.
2548 (ffi_closure_UNIX64_inner): New.
2549 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2551 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2553 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2554 Remove USE_LIBDIR conditional.
2555 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2556 * Makefile.in, configure: Rebuilt.
2558 2003-01027 David Edelsohn <edelsohn@gnu.org>
2560 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2561 * Makefile.in: Regenerate.
2563 2003-01-22 Andrew Haley <aph@redhat.com>
2565 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2568 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2570 * src/powerpc/darwin.S: Add unwind info.
2571 * src/powerpc/darwin_closure.S: Likewise.
2573 2003-01-14 Andrew Haley <aph@redhat.com>
2575 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2576 (ffi_prep_cif_machdep): Likewise.
2577 * src/x86/unix64.S: Add unwind info.
2579 2003-01-14 Andreas Jaeger <aj@suse.de>
2581 * src/ffitest.c (main): Only use ffi_closures if those are
2584 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2586 * libffi/src/ffitest.c
2587 add closure testcases
2589 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2591 * libffi/src/powerpc/ffi.c
2592 fix alignment bug for float (4 byte aligned iso 8 byte)
2594 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2596 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2597 * src/powerpc/darwin.S: Remove RCS version string.
2599 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2601 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2602 * src/ffitest.c (main): Use static storage for closure.
2603 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2604 * src/sparc/v8.S (ffi_closure_v8): New.
2605 * src/sparc/v9.S (ffi_closure_v9): New.
2607 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2609 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2610 enumeration for X86_WIN32.
2611 * src/x86/win32.S: Added ffi_call_STDCALL function
2613 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2614 switch cases for recognising FFI_STDCALL and
2615 calling ffi_call_STDCALL if target is X86_WIN32.
2616 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2617 stdcall versions of the "my_strlen" and "many"
2618 test functions (for X86_WIN32).
2619 Added test cases to test stdcall invocation using
2622 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2624 * src/sh/sysv.S: Add DWARF2 unwind info.
2626 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2628 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2630 2002-11-26 Jim Wilson <wilson@redhat.com>
2632 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2634 2002-11-23 H.J. Lu <hjl@gnu.org>
2636 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2637 Include ../config/accross.m4.
2638 * aclocal.m4; Rebuild.
2639 * configure: Likewise.
2641 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2643 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2645 2002-11-11 DJ Delorie <dj@redhat.com>
2647 * configure.in: Look for common files in the right place.
2649 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2651 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2652 raw data as _Jv_word values, not ffi_raw.
2653 (ffi_java_ptrarray_to_raw): Likewise.
2654 (ffi_java_rvalue_to_raw): New function.
2655 (ffi_java_raw_call): Call it.
2656 (ffi_java_raw_to_rvalue): New function.
2657 (ffi_java_translate_args): Call it.
2658 * src/ffitest.c (closure_test_fn): Interpret return value
2659 as ffi_arg, not int.
2660 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2661 FFI_TYPE_POINTER case.
2662 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2663 values extended to word size.
2665 2002-10-02 Andreas Jaeger <aj@suse.de>
2667 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2669 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2671 * include/ffi.h.in: Fix i386 win32 compilation.
2673 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2675 * configure.in: Add s390x-*-linux-* target.
2676 * configure: Regenerate.
2677 * include/ffi.h.in: Define S390X for s390x targets.
2678 (FFI_CLOSURES): Define for s390/s390x.
2679 (FFI_TRAMPOLINE_SIZE): Likewise.
2680 (FFI_NATIVE_RAW_API): Likewise.
2681 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2682 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2683 * src/s390/ffi.c: Major rework of existing code. Add support for
2684 s390x targets. Add closure support.
2685 * src/s390/sysv.S: Likewise.
2687 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2689 * src/arm/sysv.S: Fix typo.
2691 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2693 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2694 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2695 (ffi_call_SYSV): Handle soft-float.
2697 2002-09-27 Bo Thorsen <bo@suse.de>
2699 * include/ffi.h.in: Fix multilib x86-64 support.
2701 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2703 * Makefile.am (all-multi): Fix multilib parallel build.
2705 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2707 * configure.in (sh[34]*-*-linux*): Add brackets.
2708 * configure: Regenerate.
2710 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2712 * Makefile.am: Add SH support.
2713 * Makefile.in: Regenerate.
2714 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2715 * configure: Regenerate.
2716 * include/ffi.h.in: Add SH support.
2717 * src/sh/ffi.c: New file.
2718 * src/sh/sysv.S: New file.
2719 * src/types.c: Add SH support.
2721 2002-07-16 Bo Thorsen <bo@suse.de>
2723 * src/x86/ffi64.c: New file that adds x86-64 support.
2724 * src/x86/unix64.S: New file that handles argument setup for
2726 * src/x86/sysv.S: Don't use this on x86-64.
2727 * src/x86/ffi.c: Don't use this on x86-64.
2729 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2731 * src/ffitest.c (struct6): New test that tests a special case in
2733 (struct7): Likewise.
2734 (struct8): Likewise.
2735 (struct9): Likewise.
2736 (closure_test_fn): Silence warning about this when it's not used.
2737 (main): Add the new tests.
2738 (main): Fix a couple of wrong casts and silence some compiler warnings.
2739 * include/ffi.h.in: Add x86-64 ABI definition.
2740 * fficonfig.h.in: Regenerate.
2741 * Makefile.am: Add x86-64 support.
2742 * configure.in: Likewise.
2743 * Makefile.in: Regenerate.
2744 * configure: Likewise.
2746 2002-06-24 Bo Thorsen <bo@suse.de>
2748 * src/types.c: Merge settings for similar architectures.
2749 Add x86-64 sizes and alignments.
2751 2002-06-23 Bo Thorsen <bo@suse.de>
2753 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2754 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2755 * src/mips/ffi.c (ffi_prep_args): Likewise.
2756 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2758 2002-07-18 H.J. Lu (hjl@gnu.org)
2760 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2761 (libffi_la_SOURCES): Support MIPS_LINUX.
2762 (libffi_convenience_la_SOURCES): Likewise.
2763 * Makefile.in: Regenerated.
2765 * configure.in (mips64*-*): Skip.
2766 (mips*-*-linux*): New.
2767 * configure: Regenerated.
2769 * src/mips/ffi.c: Include <sgidefs.h>.
2771 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2773 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2775 2002-05-27 Roger Sayle <roger@eyesopen.com>
2777 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2779 2002-05-27 Bo Thorsen <bo@suse.de>
2781 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2784 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2786 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2787 beginning of function (for older apple cc).
2789 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2791 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2792 script entry, and set LD to it when configuring multilibs.
2793 * configure: Rebuilt.
2795 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2797 * configure.in (sparc64-*-netbsd*): Add target.
2798 (sparc-*-netbsdelf*): Likewise.
2799 * configure: Regenerate.
2801 2002-04-28 David S. Miller <davem@redhat.com>
2803 * configure.in, configure: Fix SPARC test in previous change.
2805 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2807 * Makefile.am: Add Linux for S/390 support.
2808 * Makefile.in: Regenerate.
2809 * configure.in: Add Linux for S/390 support.
2810 * configure: Regenerate.
2811 * include/ffi.h.in: Add Linux for S/390 support.
2812 * src/s390/ffi.c: New file from libffi CVS tree.
2813 * src/s390/sysv.S: New file from libffi CVS tree.
2815 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2817 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2819 * src/sparc/v8.S: Use it.
2820 * src/sparc/v9.S: Likewise.
2821 * fficonfig.h.in: Rebuilt.
2822 * configure: Rebuilt.
2824 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2826 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2828 * src/ia64/unix.S: Add unwind information. Fix comments.
2829 Save sp in a way that's compatible with unwind info.
2830 (ffi_call_unix): Correctly restore sp in all cases.
2831 * src/ia64/ffi.c: Add, fix comments.
2833 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2835 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2837 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2839 * configure.in (alpha*-*-netbsd*): Add target.
2840 * configure: Regenerate.
2842 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2844 * src/sparc/v8.S: Add unwind info.
2845 * src/sparc/v9.S: Likewise.
2847 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2849 * configure.in: Enable i*86-*-netbsdelf*.
2850 * configure: Rebuilt.
2852 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2855 * src/mips/n32.s: Delete
2856 * src/mips/o32.s: Delete
2858 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2860 * configure.in: Enable alpha*-*-freebsd*.
2861 * configure: Rebuilt.
2863 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2865 * Makefile.am: libfficonvenience -> libffi_convenience.
2866 * Makefile.in: Rebuilt.
2868 * Makefile.am: Define ffitest_OBJECTS.
2869 * Makefile.in: Rebuilt.
2871 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2872 David Edelsohn <edelsohn@gnu.org>
2874 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2875 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2876 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2877 * Makefile.in: Regenerate.
2878 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2879 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2880 (flush_icache, flush_range): New functions.
2881 (ffi_closure_helper_DARWIN): New function.
2882 * src/powerpc/aix_closure.S: New file.
2883 * src/powerpc/darwin_closure.S: New file.
2885 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2887 * include/ffi.h.in: Add typedef for ffi_arg.
2888 * src/ffitest.c (main): Declare rint with ffi_arg.
2890 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2892 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2893 number of GPRs for floating-point arguments.
2895 2002-01-31 Anthony Green <green@redhat.com>
2897 * configure: Rebuilt.
2898 * configure.in: Replace CHECK_SIZEOF and endian tests with
2899 cross-compiler friendly macros.
2900 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2903 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2905 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2906 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2908 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2910 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2911 (TARGET_SRC_POWERPC_AIX): New.
2912 (POWERPC_AIX): New stanza.
2913 * Makefile.in: Regenerate.
2914 * configure.in: Add AIX case.
2915 * configure: Regenerate.
2916 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2917 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2918 size. Fix "long double" support.
2919 (ffi_call): Add FFI_AIX case.
2920 * src/powerpc/aix.S: New.
2922 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2924 Implement Darwin PowerPC ABI.
2925 * configure.in: Handle powerpc-*-darwin*.
2926 * Makefile.am: Set source files for POWERPC_DARWIN.
2927 * configure: Rebuilt.
2928 * Makefile.in: Rebuilt.
2929 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2931 * src/powerpc/darwin.S: New file.
2932 * src/powerpc/ffi_darwin.c: New file.
2934 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2936 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2938 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2940 * src/x86/sysv.S: Avoid gas-only .balign directive.
2941 Use C style comments.
2943 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2945 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2946 Fixes PR bootstrap/3563.
2948 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2950 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2952 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2954 * configure.in: Recognize sparc*-sun-* host.
2955 * configure: Regenerate.
2957 2001-06-06 Andrew Haley <aph@redhat.com>
2959 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
2961 2001-06-03 Andrew Haley <aph@redhat.com>
2963 * src/alpha/osf.S: Add unwind info.
2964 * src/powerpc/sysv.S: Add unwind info.
2965 * src/powerpc/ppc_closure.S: Likewise.
2967 2000-05-31 Jeff Sturm <jsturm@one-point.com>
2969 * configure.in: Fix AC_ARG_ENABLE usage.
2970 * configure: Rebuilt.
2972 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2974 * configure.in: Remove warning about beta code.
2975 * configure: Rebuilt.
2977 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
2979 * src/ia64/unix.S: Restore stack pointer when returning from
2981 * src/ia64/ffi.c: Fix typo in comment.
2983 2001-04-18 Jim Wilson <wilson@redhat.com>
2985 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
2986 to eliminate RAW DV.
2988 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
2990 * Makefile.am: Make a libtool convenience library.
2991 * Makefile.in: Rebuilt.
2993 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
2995 * configure.in: Use different syntax for subdirectory creation.
2996 * configure: Rebuilt.
2998 2001-03-27 Jon Beniston <jon@beniston.com>
3000 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3001 * configure: Rebuilt.
3002 * Makefile.am: Added X86_WIN32 target support.
3003 * Makefile.in: Rebuilt.
3005 * include/ffi.h.in: Added X86_WIN32 target support.
3007 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3008 * src/types.c: Added X86_WIN32 target support.
3010 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3011 stuff removed and made to work with CygWin's gas.
3013 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3015 * configure.in: Make target subdirectory in build dir.
3016 * Makefile.am: Override suffix based rules to specify correct output
3018 * Makefile.in: Rebuilt.
3019 * configure: Rebuilt.
3021 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3023 * src/powerpc/ppc_closure.S: New file.
3024 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3025 involving long long and register pairs.
3026 (ffi_prep_closure): New function.
3027 (flush_icache): Likewise.
3028 (ffi_closure_helper_SYSV): Likewise.
3029 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3030 (FFI_TRAMPOLINE_SIZE): Likewise.
3031 (FFI_NATIVE_RAW_API): Likewise.
3032 * Makefile.in: Rebuilt.
3033 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3034 (TARGET_SRC_POWERPC): Likewise.
3036 2001-03-19 Tom Tromey <tromey@redhat.com>
3038 * Makefile.in: Rebuilt.
3039 * Makefile.am (ffitest_LDFLAGS): New macro.
3041 2001-03-02 Nick Clifton <nickc@redhat.com>
3043 * include/ffi.h.in: Remove RCS ident string.
3044 * include/ffi_mips.h: Remove RCS ident string.
3045 * src/debug.c: Remove RCS ident string.
3046 * src/ffitest.c: Remove RCS ident string.
3047 * src/prep_cif.c: Remove RCS ident string.
3048 * src/types.c: Remove RCS ident string.
3049 * src/alpha/ffi.c: Remove RCS ident string.
3050 * src/alpha/osf.S: Remove RCS ident string.
3051 * src/arm/ffi.c: Remove RCS ident string.
3052 * src/arm/sysv.S: Remove RCS ident string.
3053 * src/mips/ffi.c: Remove RCS ident string.
3054 * src/mips/n32.S: Remove RCS ident string.
3055 * src/mips/o32.S: Remove RCS ident string.
3056 * src/sparc/ffi.c: Remove RCS ident string.
3057 * src/sparc/v8.S: Remove RCS ident string.
3058 * src/sparc/v9.S: Remove RCS ident string.
3059 * src/x86/ffi.c: Remove RCS ident string.
3060 * src/x86/sysv.S: Remove RCS ident string.
3062 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3064 * include/ffi.h.in: Change sourceware.cygnus.com references to
3067 2000-12-09 Richard Henderson <rth@redhat.com>
3069 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3070 (ffi_closure_osf_inner): Index rather than increment avalue
3071 and arg_types. Give ffi_closure_osf the raw return value type.
3072 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3075 2000-12-07 Richard Henderson <rth@redhat.com>
3077 * src/raw_api.c (ffi_translate_args): Fix typo.
3078 (ffi_prep_closure): Likewise.
3080 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3081 FFI_TRAMPOLINE_SIZE.
3082 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3083 cif->bytes for new ffi_call_osf implementation.
3084 (ffi_prep_args): Absorb into ...
3085 (ffi_call): ... here. Do all stack allocation here and
3086 avoid a callback function.
3087 (ffi_prep_closure, ffi_closure_osf_inner): New.
3088 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3089 (ffi_closure_osf): New.
3091 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3093 * config.guess, config.sub, install-sh: Removed.
3094 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3095 * Makefile.in: Rebuilt.
3097 * acinclude.m4: Include libtool macros from the top level.
3098 * aclocal.m4, configure: Rebuilt.
3100 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3102 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3103 * configure: Rebuilt.
3105 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3107 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3108 memory correctly. Use conditional instructions, not branches where
3111 2000-05-04 Tom Tromey <tromey@cygnus.com>
3113 * configure: Rebuilt.
3114 * configure.in: Match `arm*-*-linux-*'.
3115 From Chris Dornan <cdornan@arm.com>.
3117 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3119 * Makefile.am (SUBDIRS): Define.
3120 (AM_MAKEFLAGS): Likewise.
3121 (Multilib support.): Add section.
3122 * Makefile.in: Rebuilt.
3123 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3124 New variables. Set for gcc using -print-multi-lib. Export them
3126 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3127 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3129 (extra_compiler_flags_value, extra_compiler_flags): Check these
3130 for extra compiler options which need to be passed down in
3133 2000-04-16 Anthony Green <green@redhat.com>
3135 * configure: Rebuilt.
3136 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3138 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3140 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3141 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3142 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3143 Replace all void * sizeofs with sizeof(int).
3144 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3145 different than DOUBLE.
3146 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3147 (ffi_prep_args_v9): New function.
3148 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3149 (ffi_V9_return_struct): New function.
3150 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3151 32bit code (not yet cross-arch calls).
3152 * src/sparc/v8.S: Add struct return delay nop.
3154 * src/sparc/v9.S: New file.
3155 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3156 is used on sparc64 only for structures larger than 32 bytes.
3157 Pass by reference for structures is done for structure arguments
3158 larger than 16 bytes.
3159 * src/ffitest.c (main): Use 64bit rint on sparc64.
3160 Run long long tests on sparc.
3161 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3163 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3165 * configure.in (sparc-*-linux*): New supported target.
3166 (sparc64-*-linux*): Likewise.
3167 * configure: Rebuilt.
3168 * Makefile.am: Add v9.S to SPARC files.
3169 * Makefile.in: Likewise.
3170 (LINK): Surround $(CCLD) into double quotes, so that multilib
3171 compiles work correctly.
3173 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3175 * configure: Rebuilt.
3176 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3177 proposed by Bryce McKinlay.
3179 2000-03-20 Tom Tromey <tromey@cygnus.com>
3181 * Makefile.in: Hand edit for java_raw_api.lo.
3183 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3185 * config.guess, config.sub: Update from the gcc tree.
3186 Fix for PR libgcj/168.
3188 2000-03-03 Tom Tromey <tromey@cygnus.com>
3190 * Makefile.in: Fixed ia64 by hand.
3192 * configure: Rebuilt.
3193 * configure.in (--enable-multilib): New option.
3194 (libffi_basedir): New subst.
3195 (AC_OUTPUT): Added multilib code.
3197 2000-03-02 Tom Tromey <tromey@cygnus.com>
3199 * Makefile.in: Rebuilt.
3200 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3203 2000-02-25 Hans Boehm <boehm@acm.org>
3205 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3207 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3209 (ffi_prep_raw_closure): Use ffi_translate_args, not
3210 ffi_closure_translate.
3211 * src/java_raw_api.c: New file.
3212 * src/ffitest.c (closure_test_fn): New function.
3213 (main): Define `rint' as long long on IA64. Added new test when
3214 FFI_CLOSURES is defined.
3215 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3216 (ffi_abi): Recognize IA64.
3217 (ffi_raw): Added `flt' field.
3218 Added "Java raw API" code.
3219 * configure.in: Recognize ia64.
3220 * Makefile.am (TARGET_SRC_IA64): New macro.
3221 (libffi_la_common_SOURCES): Added java_raw_api.c.
3222 (libffi_la_SOURCES): Define in IA64 case.
3224 2000-01-04 Tom Tromey <tromey@cygnus.com>
3226 * Makefile.in: Rebuilt with newer automake.
3228 1999-12-31 Tom Tromey <tromey@cygnus.com>
3230 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3232 1999-09-01 Tom Tromey <tromey@cygnus.com>
3234 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3236 * fficonfig.h.in: Rebuilt.
3237 * configure: Rebuilt.
3238 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3239 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3240 * acconfig.h: Don't #undef PACKAGE or VERSION.
3242 1999-08-09 Anthony Green <green@cygnus.com>
3244 * include/ffi.h.in: Try to work around messy header problem
3245 with PACKAGE and VERSION.
3247 * configure: Rebuilt.
3248 * configure.in: Change version to 2.00-beta.
3250 * fficonfig.h.in: Rebuilt.
3251 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3253 * src/x86/ffi.c (ffi_raw_call): Rename.
3255 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3257 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3258 (ffi_prep_incoming_args_SYSV): Ditto.
3259 (ffi_prep_closure): Ditto.
3260 (ffi_closure_raw_SYSV): Ditto.
3261 (ffi_prep_raw_closure): More ditto.
3262 (ffi_call_raw): Final ditto.
3264 * include/ffi.h.in: Add definitions for closure and raw API.
3266 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3269 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3271 * src/raw_api.c: New file.
3273 * include/ffi.h.in (ffi_raw): New type.
3274 (UINT_ARG, SINT_ARG): New defines.
3275 (ffi_closure, ffi_raw_closure): New types.
3276 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3278 * configure.in: Add check for endianness and sizeof void*.
3280 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3281 instead of directly.
3283 * configure: Rebuilt.
3285 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3287 * configure.in: Add x86 and powerpc BeOS configurations.
3288 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3290 1999-05-09 Anthony Green <green@cygnus.com>
3292 * configure.in: Add warning about this being beta code.
3293 Remove src/Makefile.am from the picture.
3294 * configure: Rebuilt.
3296 * Makefile.am: Move logic from src/Makefile.am. Add changes
3297 to support libffi as a target library.
3298 * Makefile.in: Rebuilt.
3300 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3301 Upgraded to new autoconf, automake, libtool.
3305 * LICENSE: Update copyright date.
3307 * src/Makefile.am, src/Makefile.in: Removed.
3309 1998-11-29 Anthony Green <green@cygnus.com>
3311 * include/ChangeLog: Removed.
3312 * src/ChangeLog: Removed.
3313 * src/mips/ChangeLog: Removed.
3314 * src/sparc/ChangeLog: Remboved.
3315 * src/x86/ChangeLog: Removed.
3317 * ChangeLog.v1: Created.