1 2008-05-09 Julian Brown <julian@codesourcery.com>
3 * Makefile.am (LTLDFLAGS): New.
4 (libffi_la_LDFLAGS): Use above.
5 * Makefile.in: Regenerate.
7 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
10 * aclocal.m4: Regenerate.
11 * configure: Regenerate.
13 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
15 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
16 * src/sh64/sysv.S: Likewise.
18 2008-03-26 Daniel Jacobowitz <dan@debian.org>
20 * src/arm/sysv.S: Fix ARM comment marker.
22 2008-03-26 Jakub Jelinek <jakub@redhat.com>
24 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
25 * src/s390/sysv.S: Likewise.
26 * src/powerpc/ppc_closure.S: Likewise.
27 * src/powerpc/sysv.S: Likewise.
28 * src/x86/unix64.S: Likewise.
29 * src/x86/sysv.S: Likewise.
30 * src/sparc/v8.S: Likewise.
31 * src/sparc/v9.S: Likewise.
32 * src/m68k/sysv.S: Likewise.
33 * src/arm/sysv.S: Likewise.
35 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
37 * aclocal.m4: Regenerate.
38 * configure: Likewise.
39 * Makefile.in: Likewise.
40 * include/Makefile.in: Likewise.
41 * testsuite/Makefile.in: Likewise.
43 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
44 Andreas Tobler <a.tobler@schweiz.org>
46 * configure.ac: Add amd64-*-freebsd* target.
47 * configure: Regenerate.
49 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
52 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
55 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
58 2008-01-24 David Edelsohn <edelsohn@gnu.org>
60 * configure: Regenerate.
62 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
64 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
66 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
69 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
70 signed/unsigned int8/16 for X86_DARWIN.
71 Updated copyright info.
72 Handle one and two byte structs with special cif->flags.
73 * src/x86/ffitarget.h: Add special types for one and two byte structs.
74 Updated copyright info.
75 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
77 Remove code to pop args from the stack after call.
78 Special-case signed/unsigned for int8/16, one and two byte structs.
79 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
80 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
82 Updated copyright info.
84 2007-12-08 David Daney <ddaney@avtrex.com>
86 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
87 SUBU, add with ADDU and use smaller code sequences.
89 2007-12-07 David Daney <ddaney@avtrex.com>
91 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
94 2007-12-06 David Daney <ddaney@avtrex.com>
96 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
98 (ffi_java_raw): New typedef.
99 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
100 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
102 (ffi_java_raw_closure) : Same.
103 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
105 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
107 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
108 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
109 sizeof(ffi_java_raw) for alignment calculations.
110 (ffi_java_ptrarray_to_raw): Same.
111 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
112 if FFI_SIZEOF_JAVA_RAW == 4.
113 (ffi_java_raw_to_rvalue): Same.
114 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
115 (ffi_java_translate_args): Same.
116 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
118 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
120 2007-12-06 David Daney <ddaney@avtrex.com>
122 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
125 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
128 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
129 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
130 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
131 set the NUM_FPR_ARG_REGISTERS according to.
132 Add support for potential soft-float support under hard-float
134 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
135 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
136 to the FFI_LINUX_SOFT_FLOAT ABI.
137 (ffi_prep_cif_machdep): Likewise.
138 (ffi_closure_helper_SYSV): Likewise.
139 * src/powerpc/ppc_closure.S: Make sure not to store float/double
140 on archs where __NO_FPRS__ is true.
141 Add FFI_TYPE_UINT128 support.
142 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
143 Adjust copyright notice.
145 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
147 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
148 * include/ffi_common.h: ... here.
151 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
153 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
154 * src/powerpc/linux64.S: Likewise.
155 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
156 * testsuite/libffi.call/return_ldl.c: New test.
158 2007-09-04 <aph@redhat.com>
160 * src/arm/sysv.S (UNWIND): New.
161 (Whole file): Conditionally compile unwinder directives.
162 * src/arm/sysv.S: Add unwinder directives.
164 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
165 Only treat r0 as a struct address if we're actually returning a
167 Only copy the bytes that are actually within a struct.
168 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
169 is returned in r0, not passed by address.
170 (ffi_call): Allocate a word-sized temporary for the case where
171 a composite is returned in r0.
172 (ffi_prep_incoming_args_SYSV): Align as necessary.
174 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
176 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
177 directly using the sys_cacheflush syscall.
179 2007-07-27 Andrew Haley <aph@redhat.com>
181 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
183 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
185 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
186 * configure.ac: Likewise.
187 * Makefile.in: Regenerate.
188 * include/Makefile.in: Likewise.
189 * testsuite/Makefile.in: Likewise.
190 * configure: Likewise.
192 2007-08-24 David Daney <ddaney@avtrex.com>
194 * testsuite/libffi.call/return_sl.c: New test.
196 2007-08-10 David Daney <ddaney@avtrex.com>
198 * testsuite/libffi.call/cls_multi_ushort.c,
199 testsuite/libffi.call/cls_align_uint16.c,
200 testsuite/libffi.call/nested_struct1.c,
201 testsuite/libffi.call/nested_struct3.c,
202 testsuite/libffi.call/cls_7_1_byte.c,
203 testsuite/libffi.call/nested_struct5.c,
204 testsuite/libffi.call/cls_double.c,
205 testsuite/libffi.call/nested_struct7.c,
206 testsuite/libffi.call/cls_sint.c,
207 testsuite/libffi.call/nested_struct9.c,
208 testsuite/libffi.call/cls_20byte1.c,
209 testsuite/libffi.call/cls_multi_sshortchar.c,
210 testsuite/libffi.call/cls_align_sint64.c,
211 testsuite/libffi.call/cls_3byte2.c,
212 testsuite/libffi.call/cls_multi_schar.c,
213 testsuite/libffi.call/cls_multi_uchar.c,
214 testsuite/libffi.call/cls_19byte.c,
215 testsuite/libffi.call/cls_9byte1.c,
216 testsuite/libffi.call/cls_align_float.c,
217 testsuite/libffi.call/closure_fn1.c,
218 testsuite/libffi.call/problem1.c,
219 testsuite/libffi.call/closure_fn3.c,
220 testsuite/libffi.call/cls_sshort.c,
221 testsuite/libffi.call/closure_fn5.c,
222 testsuite/libffi.call/cls_align_double.c,
223 testsuite/libffi.call/nested_struct.c,
224 testsuite/libffi.call/cls_2byte.c,
225 testsuite/libffi.call/nested_struct10.c,
226 testsuite/libffi.call/cls_4byte.c,
227 testsuite/libffi.call/cls_6byte.c,
228 testsuite/libffi.call/cls_8byte.c,
229 testsuite/libffi.call/cls_multi_sshort.c,
230 testsuite/libffi.call/cls_align_sint16.c,
231 testsuite/libffi.call/cls_align_uint32.c,
232 testsuite/libffi.call/cls_20byte.c,
233 testsuite/libffi.call/cls_float.c,
234 testsuite/libffi.call/nested_struct2.c,
235 testsuite/libffi.call/cls_5_1_byte.c,
236 testsuite/libffi.call/nested_struct4.c,
237 testsuite/libffi.call/cls_24byte.c,
238 testsuite/libffi.call/nested_struct6.c,
239 testsuite/libffi.call/cls_64byte.c,
240 testsuite/libffi.call/nested_struct8.c,
241 testsuite/libffi.call/cls_uint.c,
242 testsuite/libffi.call/cls_multi_ushortchar.c,
243 testsuite/libffi.call/cls_schar.c,
244 testsuite/libffi.call/cls_uchar.c,
245 testsuite/libffi.call/cls_align_uint64.c,
246 testsuite/libffi.call/cls_ulonglong.c,
247 testsuite/libffi.call/cls_align_longdouble.c,
248 testsuite/libffi.call/cls_1_1byte.c,
249 testsuite/libffi.call/cls_12byte.c,
250 testsuite/libffi.call/cls_3_1byte.c,
251 testsuite/libffi.call/cls_3byte1.c,
252 testsuite/libffi.call/cls_4_1byte.c,
253 testsuite/libffi.call/cls_6_1_byte.c,
254 testsuite/libffi.call/cls_16byte.c,
255 testsuite/libffi.call/cls_18byte.c,
256 testsuite/libffi.call/closure_fn0.c,
257 testsuite/libffi.call/cls_9byte2.c,
258 testsuite/libffi.call/closure_fn2.c,
259 testsuite/libffi.call/closure_fn4.c,
260 testsuite/libffi.call/cls_ushort.c,
261 testsuite/libffi.call/closure_fn6.c,
262 testsuite/libffi.call/cls_5byte.c,
263 testsuite/libffi.call/cls_align_pointer.c,
264 testsuite/libffi.call/cls_7byte.c,
265 testsuite/libffi.call/cls_align_sint32.c,
266 testsuite/libffi.special/unwindtest_ffi_call.cc,
267 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
269 2007-08-10 David Daney <ddaney@avtrex.com>
272 * configure.ac: Don't treat mips64 as a special case.
273 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
274 * configure: Regenerate
275 * Makefile.in: Ditto.
276 * fficonfig.h.in: Ditto.
277 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
278 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
279 (FFI_DEFAULT_ABI): Set for n64 case.
280 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
281 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
282 (ffi_closure_N32): New function.
283 (.eh_frame): New section
284 * src/mips/o32.S: Clean up comments.
285 (ffi_closure_O32): Pass ffi_closure parameter in $12.
286 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
287 _MIPS_SIM == _ABIN32 throughout.
288 (FFI_MIPS_STOP_HERE): New, use in place of
290 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
291 to support n32/n64 ABIs.
292 (calc_n32_struct_flags): Rewrite.
293 (calc_n32_return_struct_flags): Remove unused variable. Reverse
294 position of flag bits.
295 (ffi_prep_cif_machdep): Rewrite n32 portion.
296 (ffi_call): Enable for n64. Add special handling for small structure
298 (ffi_prep_closure_loc): Add n32 and n64 support.
299 (ffi_closure_mips_inner_O32): Add cast to silence warning.
300 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
302 2007-08-08 David Daney <ddaney@avtrex.com>
304 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
305 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
307 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
308 * testsuite/libffi.call/cls_sint.c (main): Ditto.
309 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
310 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
311 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
312 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
313 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
314 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
315 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
316 * testsuite/libffi.call/return_fl3.c (main): Ditto.
317 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
318 * testsuite/libffi.call/nested_struct.c (main): Ditto.
319 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
320 * testsuite/libffi.call/return_ll1.c (main): Ditto.
321 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
322 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
323 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
324 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
325 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
326 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
327 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
328 * testsuite/libffi.call/cls_uint.c (main): Ditto.
329 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
330 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
331 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
332 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
333 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
334 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
335 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
336 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
337 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
339 2007-08-07 Andrew Haley <aph@redhat.com>
341 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
344 2007-08-06 Andrew Haley <aph@redhat.com>
347 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
348 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
351 2007-08-02 David Daney <ddaney@avtrex.com>
353 * testsuite/libffi.call/return_ul.c (main): Define return type as
354 ffi_arg. Use proper printf conversion specifier.
356 2007-07-30 Andrew Haley <aph@redhat.com>
359 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
360 signed/unsigned int8/16.
361 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
363 Remove code to pop args from the stack after call.
364 Special-case signed/unsigned int8/16.
365 * testsuite/libffi.call/return_sc.c (main): Revert.
367 2007-07-26 Richard Guenther <rguenther@suse.de>
370 * testsuite/libffi.call/return_sc.c (main): Verify call
371 result as signed char, not ffi_arg.
373 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
375 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
376 * configure: Regenerate.
378 2007-07-11 David Daney <ddaney@avtrex.com>
380 * src/mips/ffi.c: Don't include sys/cachectl.h.
381 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
384 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
386 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
387 from (ffi_prep_closure): ... this.
388 (FFI_INIT_TRAMPOLINE): Adjust.
390 2005-12-31 Phil Blundell <pb@reciva.com>
392 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
393 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
394 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
395 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
396 (FFI_CLOSURES): Enable closure support.
398 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
400 * testsuite/libffi.call/cls_multi_ushort.c,
401 testsuite/libffi.call/cls_align_uint16.c,
402 testsuite/libffi.call/nested_struct1.c,
403 testsuite/libffi.call/nested_struct3.c,
404 testsuite/libffi.call/cls_7_1_byte.c,
405 testsuite/libffi.call/cls_double.c,
406 testsuite/libffi.call/nested_struct5.c,
407 testsuite/libffi.call/nested_struct7.c,
408 testsuite/libffi.call/cls_sint.c,
409 testsuite/libffi.call/nested_struct9.c,
410 testsuite/libffi.call/cls_20byte1.c,
411 testsuite/libffi.call/cls_multi_sshortchar.c,
412 testsuite/libffi.call/cls_align_sint64.c,
413 testsuite/libffi.call/cls_3byte2.c,
414 testsuite/libffi.call/cls_multi_schar.c,
415 testsuite/libffi.call/cls_multi_uchar.c,
416 testsuite/libffi.call/cls_19byte.c,
417 testsuite/libffi.call/cls_9byte1.c,
418 testsuite/libffi.call/cls_align_float.c,
419 testsuite/libffi.call/closure_fn1.c,
420 testsuite/libffi.call/problem1.c,
421 testsuite/libffi.call/closure_fn3.c,
422 testsuite/libffi.call/cls_sshort.c,
423 testsuite/libffi.call/closure_fn5.c,
424 testsuite/libffi.call/cls_align_double.c,
425 testsuite/libffi.call/cls_2byte.c,
426 testsuite/libffi.call/nested_struct.c,
427 testsuite/libffi.call/nested_struct10.c,
428 testsuite/libffi.call/cls_4byte.c,
429 testsuite/libffi.call/cls_6byte.c,
430 testsuite/libffi.call/cls_8byte.c,
431 testsuite/libffi.call/cls_multi_sshort.c,
432 testsuite/libffi.call/cls_align_uint32.c,
433 testsuite/libffi.call/cls_align_sint16.c,
434 testsuite/libffi.call/cls_float.c,
435 testsuite/libffi.call/cls_20byte.c,
436 testsuite/libffi.call/cls_5_1_byte.c,
437 testsuite/libffi.call/nested_struct2.c,
438 testsuite/libffi.call/cls_24byte.c,
439 testsuite/libffi.call/nested_struct4.c,
440 testsuite/libffi.call/nested_struct6.c,
441 testsuite/libffi.call/cls_64byte.c,
442 testsuite/libffi.call/nested_struct8.c,
443 testsuite/libffi.call/cls_uint.c,
444 testsuite/libffi.call/cls_multi_ushortchar.c,
445 testsuite/libffi.call/cls_schar.c,
446 testsuite/libffi.call/cls_uchar.c,
447 testsuite/libffi.call/cls_align_uint64.c,
448 testsuite/libffi.call/cls_ulonglong.c,
449 testsuite/libffi.call/cls_align_longdouble.c,
450 testsuite/libffi.call/cls_1_1byte.c,
451 testsuite/libffi.call/cls_12byte.c,
452 testsuite/libffi.call/cls_3_1byte.c,
453 testsuite/libffi.call/cls_3byte1.c,
454 testsuite/libffi.call/cls_4_1byte.c,
455 testsuite/libffi.call/cls_6_1_byte.c,
456 testsuite/libffi.call/cls_16byte.c,
457 testsuite/libffi.call/cls_18byte.c,
458 testsuite/libffi.call/closure_fn0.c,
459 testsuite/libffi.call/cls_9byte2.c,
460 testsuite/libffi.call/closure_fn2.c,
461 testsuite/libffi.call/closure_fn4.c,
462 testsuite/libffi.call/cls_ushort.c,
463 testsuite/libffi.call/closure_fn6.c,
464 testsuite/libffi.call/cls_5byte.c,
465 testsuite/libffi.call/cls_align_pointer.c,
466 testsuite/libffi.call/cls_7byte.c,
467 testsuite/libffi.call/cls_align_sint32.c,
468 testsuite/libffi.special/unwindtest_ffi_call.cc,
469 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
471 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
473 * aclocal.m4: Regenerated.
475 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
477 * configure: Regenerate.
479 2007-05-23 Steve Ellcey <sje@cup.hp.com>
481 * Makefile.in: Regenerate.
482 * configure: Regenerate.
483 * aclocal.m4: Regenerate.
484 * include/Makefile.in: Regenerate.
485 * testsuite/Makefile.in: Regenerate.
487 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
489 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
490 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
491 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
492 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
493 (FFI_CLOSURES): Enable closure support.
495 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
497 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
498 * configure: Regenerate.
499 * fficonfig.h.in: Regenerate.
500 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
501 CFI_OFFSET,CFI_DEF_CFA): New macros.
502 (ffi_call_SYSV): Add callframe annotation.
504 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
506 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
507 numerous test suite failures.
508 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
510 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
512 * Makefile.am (EXTRA_DIST): Bring up to date.
513 * Makefile.in: Regenerate.
514 * src/frv/eabi.S: Remove RCS keyword.
516 2007-04-06 Richard Henderson <rth@redhat.com>
518 * configure.ac: Tidy target case.
519 (HAVE_LONG_DOUBLE): Allow the target to override.
520 * configure: Regenerate.
521 * include/ffi.h.in: Don't define ffi_type_foo if
522 LIBFFI_HIDE_BASIC_TYPES is defined.
523 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
525 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
526 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
527 (ffi_type_longdouble): Special case for Alpha. Don't define
528 if long double == double.
530 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
531 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
532 (ffi_call, ffi_closure_osf_inner): Likewise.
533 (ffi_closure_osf_inner): Likewise. Mark hidden.
534 (ffi_call_osf, ffi_closure_osf): Mark hidden.
535 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
536 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
537 (load_table): Handle 128-bit long double.
539 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
541 2007-04-06 Tom Tromey <tromey@redhat.com>
544 * README: Fixed bug in example.
546 2007-04-03 Jakub Jelinek <jakub@redhat.com>
548 * src/closures.c: Include sys/statfs.h.
549 (_GNU_SOURCE): Define on Linux.
550 (FFI_MMAP_EXEC_SELINUX): Define.
551 (selinux_enabled): New variable.
552 (selinux_enabled_check): New function.
553 (is_selinux_enabled): Define.
556 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
558 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
559 Use 'volatile float sum' to create sum of floats to avoid false
560 negative due to excess precision on ix86 targets.
563 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
565 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
567 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
569 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
571 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
572 (ffi_prep_closure_loc): New.
573 (ffi_prep_raw_closure_loc): New.
574 (ffi_prep_java_raw_closure_loc): New.
575 * src/closures.c: New file.
576 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
577 Replace sflags with exec_offset.
578 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
579 sub_segment_exec_offset): New macros.
580 (get_segment_flags, set_segment_flags, check_segment_merge): New
582 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
583 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
584 destroy_mspace): Use new macros.
585 (sys_alloc): Silence warning.
586 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
587 * Makefile.in: Rebuilt.
588 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
589 terms of ffi_prep_closure_loc.
590 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
592 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
594 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
596 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
598 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
599 (ffi_prep_closure): ... this.
600 * src/pa/ffi.c: Likewise.
601 * src/cris/ffi.c: Likewise. Adjust.
602 * src/frv/ffi.c: Likewise.
603 * src/ia64/ffi.c: Likewise.
604 * src/mips/ffi.c: Likewise.
605 * src/powerpc/ffi_darwin.c: Likewise.
606 * src/s390/ffi.c: Likewise.
607 * src/sh/ffi.c: Likewise.
608 * src/sh64/ffi.c: Likewise.
609 * src/sparc/ffi.c: Likewise.
610 * src/x86/ffi64.c: Likewise.
611 * src/x86/ffi.c: Likewise.
612 (FFI_INIT_TRAMPOLINE): Adjust.
613 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
614 (ffi_prep_raw_closure): ... this.
615 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
616 (ffi_prep_closure): ... this.
617 (flush_icache): Adjust.
619 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
621 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
624 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
626 * Makefile.am: Add dummy install-pdf target.
627 * Makefile.in: Regenerate
629 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
631 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
632 ffi_closure_helper_SYSV): Add long double handling.
634 2007-02-02 Jakub Jelinek <jakub@redhat.com>
636 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
637 immediately after bctrl instruction.
639 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
641 * Makefile.am (all-recursive, install-recursive,
642 mostlyclean-recursive, clean-recursive, distclean-recursive,
643 maintainer-clean-recursive): Add missing targets.
644 * Makefile.in: Rebuilt.
646 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
648 * configure.ac: Add TARGET for x86_64-*-darwin*.
649 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
651 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
652 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
653 * configure: Regenerate.
654 * Makefile.in: Regenerate.
655 * include/Makefile.in: Regenerate.
656 * testsuite/Makefile.in: Regenerate.
657 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
660 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
662 * aclocal.m4: Regenerate with aclocal -I .. as written in the
665 2006-10-31 Geoffrey Keating <geoffk@apple.com>
667 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
668 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
670 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
671 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
673 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
674 Sandro Tolaini <tolaini@libero.it>
676 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
678 * configure: Regenerated.
679 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
680 (EXTRA_DIST): Add src/x86/darwin.S.
681 * Makefile.in: Regenerated.
682 * include/Makefile.in: Regenerated.
683 * testsuite/Makefile.in: Regenerated.
685 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
686 X86_WIN32, and additionally align stack to 16 bytes.
687 * src/x86/darwin.S: New, based on sysv.S.
688 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
690 2006-09-12 David Daney <ddaney@avtrex.com>
693 * include/Makefile.am: Install both ffi.h and ffitarget.h in
694 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
695 * aclocal.m4: Regenerated for automake 1.9.6.
696 * Makefile.in: Regenerated.
697 * include/Makefile.in: Regenerated.
698 * testsuite/Makefile.in: Regenerated.
700 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
702 * include/ffi_common.h (struct): Revert accidental commit.
704 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
706 * include/ffi_common.h: Remove lint directives.
707 * include/ffi.h.in: Likewise.
709 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
711 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
712 for 32-bit architectures.
713 * testsuite/libffi.call/return_ul.c: New test case.
715 2006-07-19 David Daney <ddaney@avtrex.com>
717 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
718 xfail remains for mips64.
720 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
722 * Makefile.am: Add install-html target. Add install-html to .PHONY
723 * Makefile.in: Regenerate.
724 * aclocal.m4: Regenerate.
725 * include/Makefile.in: Regenerate.
726 * testsuite/Makefile.in: Regenerate.
728 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
730 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
733 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
735 * README: Remove notice about 'Crazy Comments'.
736 * src/debug.c: Remove lint directives. Cleanup white spaces.
737 * src/java_raw_api.c: Likewise.
738 * src/prep_cif.c: Likewise.
739 * src/raw_api.c: Likewise.
740 * src/ffitest.c: Delete. No longer needed, all test cases migrated
742 * src/arm/ffi.c: Remove lint directives.
743 * src/m32r/ffi.c: Likewise.
744 * src/pa/ffi.c: Likewise.
745 * src/powerpc/ffi.c: Likewise.
746 * src/powerpc/ffi_darwin.c: Likewise.
747 * src/sh/ffi.c: Likewise.
748 * src/sh64/ffi.c: Likewise.
749 * src/x86/ffi.c: Likewise.
750 * testsuite/libffi.call/float2.c: Likewise.
751 * testsuite/libffi.call/promotion.c: Likewise.
752 * testsuite/libffi.call/struct1.c: Likewise.
754 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
756 * src/pa/hpux32.S: Correct unwind offset calculation for
758 * src/pa/linux.S: Likewise.
760 2006-04-12 James E Wilson <wilson@specifix.com>
763 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
764 (hfa_type_load): Call stf_spill.
765 (hfa_type_store): Call ldf_fill.
766 (ffi_call): Adjust calls to above routines. Add local temps for
769 2006-04-10 Matthias Klose <doko@debian.org>
771 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
772 directory names containing underscores.
774 2006-04-07 James E Wilson <wilson@specifix.com>
776 * testsuite/libffi.call/float4.c: New testcase.
778 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
779 Andreas Tobler <a.tobler@schweiz.ch>
781 * Makefile.am: Add PA_HPUX port.
782 * Makefile.in: Regenerate.
783 * include/Makefile.in: Likewise.
784 * testsuite/Makefile.in: Likewise.
785 * configure.ac: Add PA_HPUX rules.
786 * configure: Regenerate.
787 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
788 Add PA_HPUX and PA64_HPUX.
789 Rename FFI_LINUX ABI to FFI_PA32 ABI.
790 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
791 (FFI_TYPE_SMALL_STRUCT2): Define.
792 (FFI_TYPE_SMALL_STRUCT4): Likewise.
793 (FFI_TYPE_SMALL_STRUCT8): Likewise.
794 (FFI_TYPE_SMALL_STRUCT3): Redefine.
795 (FFI_TYPE_SMALL_STRUCT5): Likewise.
796 (FFI_TYPE_SMALL_STRUCT6): Likewise.
797 (FFI_TYPE_SMALL_STRUCT7): Likewise.
798 * src/pa/ffi.c (ROUND_DOWN): Delete.
799 (fldw, fstw, fldd, fstd): Use '__asm__'.
800 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
801 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
802 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
803 Simplify incrementing of stack slot variable. Change type of local
805 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
807 (ffi_prep_cif_machdep): Likewise.
808 (ffi_call): Likewise.
809 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
810 return type to ffi_status. Simplify incrementing of stack slot
811 variable. Only copy floating point argument registers when PA_LINUX
812 is true. Reformat debug statement.
813 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
814 FFI_TYPE_SMALL_STRUCT8.
815 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
817 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
818 Add nops to cache flush. Add trampoline for PA_HPUX.
819 * src/pa/hpux32.S: New file.
820 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
821 ffi_prep_args_LINUX to ffi_prep_args_pa32.
822 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
823 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
824 argument type checks so that common argument types appear first.
825 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
826 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
828 2006-03-24 Alan Modra <amodra@bigpond.net.au>
830 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
831 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
832 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
834 (ffi_prep_args64): Assert using IBM extended double.
835 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
836 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
837 (ffi_call): Handle FFI_LINUX.
838 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
839 gpr3 return pointer as for struct return. Handle FFI_LINUX
840 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
842 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
843 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
844 Don't use r6 as pointer to results, instead use sp offset. Don't
845 make a special call to load lr with case table address, instead
846 use offset from previous call.
847 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
848 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
851 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
853 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
854 passed with FP registers correctly.
855 (ffi_closure_helper_SYSV): Likewise.
856 * src/sh64/sysv.S: Likewise.
858 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
860 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
861 args and userdata unused.
862 (closure_test_fn1): Mark cif and userdata unused.
863 (main): Remove unused res.
865 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
867 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
868 -O2, -O3, -Os and the warning flags -W -Wall.
869 * testsuite/libffi.special/special.exp: Likewise.
870 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
871 unused parameter unused for gcc or else do nothing.
872 * testsuite/libffi.special/ffitestcxx.h: Likewise.
873 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
875 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
876 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
877 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
878 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
879 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
880 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
881 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
882 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
883 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
884 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
885 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
886 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
887 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
888 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
889 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
890 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
891 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
892 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
893 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
894 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
895 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
896 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
897 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
898 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
900 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
902 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
904 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
905 void* to avoid compiler warning.
907 (cls_struct_align_gn): Mark cif and userdata unused.
908 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
910 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
912 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
914 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
916 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
918 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
919 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
920 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
922 (main): Cast res_call to silence gcc.
923 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
925 (main): Cast res_call to silence gcc.
926 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
928 (main): Cast res_call to silence gcc.
929 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
931 (main): Cast res_call to silence gcc.
932 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
934 (main): Cast res_call to silence gcc.
935 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
937 (main): Cast res_call to silence gcc.
938 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
940 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
941 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
943 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
944 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
946 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
947 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
949 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
950 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
952 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
953 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
955 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
957 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
958 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
959 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
960 Cleanup white spaces.
961 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
962 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
963 cif and userdata unused.
964 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
966 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
967 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
968 formatters to silence gcc.
969 (B_gn): Mark cif and userdata unused.
970 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
972 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
973 (B_gn): Mark cif and userdata unused.
974 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
976 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
977 (B_gn): Mark cif and userdata unused.
978 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
980 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
981 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
982 * testsuite/libffi.call/problem1.c (stub): Likewise.
983 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
985 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
986 in the last commit for this test case in the test case itself.
987 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
989 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
990 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
991 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
992 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
993 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
994 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
996 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
998 * src/sh/sysv.S: Fix register numbers in the FDE for
1001 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1003 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1004 declaration to avoid a false negative on ix86. See PR323.
1006 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1008 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1009 and cast integer to void * if needed. Update the pointer to
1010 the FP register saved area correctly.
1012 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1014 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1016 * testsuite/libffi.call/nested_struct4.c: Likewise.
1018 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1020 * testsuite/libffi.call/return_dbl.c: New test case.
1021 * testsuite/libffi.call/return_dbl1.c: Likewise.
1022 * testsuite/libffi.call/return_dbl2.c: Likewise.
1023 * testsuite/libffi.call/return_fl.c: Likewise.
1024 * testsuite/libffi.call/return_fl1.c: Likewise.
1025 * testsuite/libffi.call/return_fl2.c: Likewise.
1026 * testsuite/libffi.call/return_fl3.c: Likewise.
1027 * testsuite/libffi.call/closure_fn6.c: Likewise.
1029 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1031 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1032 here to be used by other test cases too.
1034 * testsuite/libffi.call/nested_struct10.c: New test case.
1035 * testsuite/libffi.call/nested_struct9.c: Likewise.
1036 * testsuite/libffi.call/nested_struct8.c: Likewise.
1037 * testsuite/libffi.call/nested_struct7.c: Likewise.
1038 * testsuite/libffi.call/nested_struct6.c: Likewise.
1039 * testsuite/libffi.call/nested_struct5.c: Likewise.
1040 * testsuite/libffi.call/nested_struct4.c: Likewise.
1042 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1044 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1045 * configure: Rebuilt.
1047 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1049 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1050 instead do the shifting inline.
1051 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1052 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1053 and 8 byte structs, for the remaining struct sizes don't call
1054 __lshrdi3, instead do the shifting inline.
1056 2005-12-07 Thiemo Seufer <ths@networkno.de>
1058 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1059 missing parentheses.
1060 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1061 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1062 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1063 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1064 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1066 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1068 (ffi_prep_closure): Improve trampoline instruction scheduling.
1069 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1071 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1073 * src/powerpc/ffi.c: Formatting.
1074 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1075 (ffi_prep_args64): Likewise.
1077 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1079 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1080 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1082 2005-09-26 Tom Tromey <tromey@redhat.com>
1084 * testsuite/libffi.call/float1.c (value_type): New typedef.
1085 (CANARY): New define.
1086 (main): Check for result buffer overflow.
1087 * src/powerpc/linux64.S: Handle linux64 long double returns.
1088 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1089 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1091 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1094 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1096 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1098 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1100 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1101 (AH_BOTTOM): Add FFI_HIDDEN definition.
1102 * configure: Rebuilt.
1103 * fficonfig.h.in: Rebuilt.
1104 * src/powerpc/ffi.c (hidden): Remove.
1105 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1106 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1107 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1108 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1109 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1110 add FFI_HIDDEN to its prototype.
1111 (ffi_closure_SYSV_inner): New.
1112 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1113 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1115 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1118 * configure: Rebuilt.
1119 * configure.ac: Handle i*86-*-gnu*.
1121 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1123 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1124 DW_CFA_offset_extended_sf rather than
1125 DW_CFA_GNU_negative_offset_extended.
1126 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1128 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1130 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1132 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1133 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1134 partially on register.
1135 (ffi_closure_helper_SYSV): Likewise.
1136 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1138 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1140 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1142 * src/sh64/ffi.c (simple_type): Remove.
1143 (return_type): Handle small structures correctly.
1144 (ffi_prep_args): Likewise.
1145 (ffi_call): Likewise.
1146 (ffi_closure_helper_SYSV): Likewise.
1147 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1148 Emit position independent code if PIC and remove wrong datalabel
1149 prefixes from EH data.
1151 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1153 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1154 * Makefile.in: Regenerate.
1155 * include/Makefile.in: Likewise.
1156 * testsuite/Makefile.in: Likewise.
1157 * configure.ac: Add POWERPC_FREEBSD rules.
1158 * configure: Regenerate.
1159 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1160 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1161 * src/powerpc/ffi.c: Add flags to handle small structure returns
1163 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1165 (ffi_closure_helper_SYSV): Likewise.
1166 * src/powerpc/ppc_closure.S: Add return types for small structures.
1167 * src/powerpc/sysv.S: Add bits to handle small structures for
1170 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1172 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1173 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1174 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1176 2005-07-05 Randolph Chung <tausq@debian.org>
1178 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1179 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1180 structures. Kill compilation warnings.
1181 (ffi_closure_inner_LINUX): Print return values as hex in debug
1182 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1183 Properly handle 5-7 byte structure returns.
1184 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1185 (FFI_TYPE_SMALL_STRUCT2): Remove.
1186 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1187 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1188 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1189 (checksmst1, checksmst2): Remove.
1190 (checksmst3): Optimize handling of 3-byte struct returns.
1191 (checksmst567): Properly handle 5-7 byte struct returns.
1193 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1196 * src/mips/n32.S: Enforce PIC code.
1197 * src/mips/o32.S: Likewise.
1199 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1201 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1202 * configure: Regenerate.
1204 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1206 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1207 to call ffi_closure_helper_SYSV. Append @local instead.
1208 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1210 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1212 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1213 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1214 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1215 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1216 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1218 2005-05-09 Mike Stump <mrs@apple.com>
1220 * configure: Regenerate.
1222 2005-05-08 Richard Henderson <rth@redhat.com>
1225 * src/alpha/osf.S: Update unwind into to match code.
1227 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1228 Richard Henderson <rth@redhat.com>
1230 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1232 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1233 (ffi_prep_closure): Set carry bit if sse-used flag set.
1234 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1235 Only load sse registers if ssecount non-zero.
1236 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1238 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1240 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1241 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1242 * configure: Regenerate.
1244 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1246 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1247 have Tcl8.3-compatible intermediate variable.
1249 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1250 Hans-Peter Nilsson <hp@axis.com>
1252 * Makefile.am: Add CRIS support.
1253 * configure.ac: Likewise.
1254 * Makefile.in, configure, testsuite/Makefile.in,
1255 include/Makefile.in: Regenerate.
1256 * src/cris: New directory.
1257 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1258 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1260 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1261 \r?\n in output tests.
1263 2005-04-12 Mike Stump <mrs@apple.com>
1265 * configure: Regenerate.
1267 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1269 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1271 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1273 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1275 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1276 to 64 bits in ILP32 mode.
1277 Load 64 bits even for short data.
1279 2005-03-23 Mike Stump <mrs@apple.com>
1281 * src/powerpc/darwin.S: Update for -m64 multilib.
1282 * src/powerpc/darwin_closure.S: Likewise.
1284 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1286 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1287 Do not set tool_include_dir.
1288 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1290 * include/Makefile.am: Set gcc_version and toollibffidir.
1291 * include/Makefile.in: Regenerate.
1293 2005-02-22 Andrew Haley <aph@redhat.com>
1295 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1296 odd-numbered register pairs for 64-bit integer types.
1298 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1301 * testsuite/libffi.call/return_ll1.c: New test case.
1303 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1305 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1306 * testsuite/libffi.call/float.c: Ditto.
1307 * testsuite/libffi.call/float2.c: Ditto.
1308 * testsuite/libffi.call/float3.c: Ditto.
1310 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1312 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1314 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1316 * testsuite/libffi.special/special.exp (cxx_options): Add
1319 2004-12-31 Richard Henderson <rth@redhat.com>
1321 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1322 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1323 offset parameters with a type parameter; deduce size and structure
1324 alignment. Update all users.
1326 2004-12-31 Richard Henderson <rth@redhat.com>
1328 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1329 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1330 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1331 into ffi_prep_closure.
1332 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1335 2004-12-27 Richard Henderson <rth@redhat.com>
1337 * src/x86/unix64.S: Fix typo in unwind info.
1339 2004-12-25 Richard Henderson <rth@redhat.com>
1341 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1342 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1343 (merge_classes): Check for it.
1345 (classify_argument): Pass byte_offset by value; perform all updates
1347 (examine_argument): Add classes argument; handle
1348 X86_64_COMPLEX_X87_CLASS.
1349 (ffi_prep_args): Merge into ...
1350 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1351 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1352 (ffi_fill_return_value): Remove.
1353 (ffi_prep_closure): Remove dead assert.
1354 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1355 Rewrite to use struct register_args instead of va_list. Create
1356 flags for handling structure returns.
1357 * src/x86/unix64.S: Remove dead strings.
1358 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1359 stack frame with ffi_call. Handle structure returns properly.
1360 (float2sse, floatfloat2sse, double2sse): Remove.
1361 (sse2float, sse2double, sse2floatfloat): Remove.
1362 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1363 to handle structure returns properly.
1365 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1367 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1369 * Makefile.in: Regenerated.
1371 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1373 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1374 * configure, aclocal.m4, Makefile.in: Regenerate.
1375 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1377 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1379 * configure: Regenerate for libtool change.
1381 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1383 * configure: Regenerate for libtool reversion.
1385 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1387 * configure: Regenerate for libtool change.
1389 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1391 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1393 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1395 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1396 of jal. Use an absolute encoding for the frame information.
1398 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1400 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1401 * acinclude.m4: Delete logic for sincludes.
1402 * aclocal.m4, Makefile.in, configure: Regenerate.
1403 * include/Makefile: Likewise.
1404 * testsuite/Makefile: Likewise.
1406 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1408 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1409 on a 8-byte boundary.
1410 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1412 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1414 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1415 long long values. Round stack allocation to a multiple of 8 bytes
1416 for ATPCS compatibility.
1417 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1418 names. Handle returning long long types. Add Thumb and interworking
1419 support. Improve soft-float code.
1421 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1423 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1424 (libffi_exit): New function.
1425 (libffi_init): Build the testglue wrapper if needed.
1427 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1430 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1432 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1434 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1436 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1438 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1439 * testsuite/libffi.call/float3.c: New test case.
1441 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1443 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1444 the function returning a structure pointed with R2.
1445 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1446 the structure return value if T bit set. Emit position
1447 independent code and EH data if PIC.
1449 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1451 * Makefile.am: Add m32r support.
1452 * configure.ac: Likewise.
1453 * Makefile.in: Regenerate.
1454 * confiugre: Regenerate.
1455 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1456 (uint64, sint64, double, longdouble)
1457 * src/m32r: New directory.
1458 * src/m32r/ffi.c: New file.
1459 * src/m32r/sysv.S: Likewise.
1460 * src/m32r/ffitarget.h: Likewise.
1462 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1464 * testsuite/libffi.call/negint.c: New test case.
1466 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1469 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1470 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1471 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1474 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1476 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1477 * testsuite/libffi.call/promotion.c: Likewise.
1478 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1480 * testsuite/libffi.call/return_sc.c: Likewise.
1481 * testsuite/libffi.call/return_uc.c: Likewise.
1483 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1485 * src/powerpc/darwin.S: Fix comments and identation.
1486 * src/powerpc/darwin_closure.S: Likewise.
1488 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1490 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1491 (ffi_prep_args): Handle longdouble arguments.
1492 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1494 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1495 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1497 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1498 * src/types.c: Defined longdouble size and alignment for darwin.
1500 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1502 * src/powerpc/aix.S: Remove whitespaces.
1503 * src/powerpc/aix_closure.S: Likewise.
1504 * src/powerpc/asm.h: Likewise.
1505 * src/powerpc/ffi.c: Likewise.
1506 * src/powerpc/ffitarget.h: Likewise.
1507 * src/powerpc/linux64.S: Likewise.
1508 * src/powerpc/linux64_closure.S: Likewise.
1509 * src/powerpc/ppc_closure.S: Likewise.
1510 * src/powerpc/sysv.S: Likewise.
1512 2004-08-30 Anthony Green <green@redhat.com>
1514 * Makefile.am: Add frv support.
1515 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1516 * configure.ac: Read configure.host.
1517 * configure.in: Read configure.host.
1518 * configure.host: New file. frv-elf needs libgloss.
1519 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1520 alignment. This is needed to frv and shouldn't harm the others.
1521 * include/ffi_common.h (ALIGN_DOWN): New macro.
1522 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1524 2004-08-24 David Daney <daney@avtrex.com>
1526 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1527 * testsuite/libffi.call/closure_fn1.c: Likewise.
1528 * testsuite/libffi.call/closure_fn2.c Likewise.
1529 * testsuite/libffi.call/closure_fn3.c: Likewise.
1530 * testsuite/libffi.call/closure_fn4.c: Likewise.
1531 * testsuite/libffi.call/closure_fn5.c: Likewise.
1532 * testsuite/libffi.call/cls_18byte.c: Likewise.
1533 * testsuite/libffi.call/cls_19byte.c: Likewise.
1534 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1535 * testsuite/libffi.call/cls_20byte.c: Likewise.
1536 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1537 * testsuite/libffi.call/cls_24byte.c: Likewise.
1538 * testsuite/libffi.call/cls_2byte.c: Likewise.
1539 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1540 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1541 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1542 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1543 * testsuite/libffi.call/cls_4byte.c: Likewise.
1544 * testsuite/libffi.call/cls_64byte.c: Likewise.
1545 * testsuite/libffi.call/cls_6byte.c: Likewise.
1546 * testsuite/libffi.call/cls_7byte.c: Likewise.
1547 * testsuite/libffi.call/cls_8byte.c: Likewise.
1548 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1549 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1550 * testsuite/libffi.call/cls_align_double.c: Likewise.
1551 * testsuite/libffi.call/cls_align_float.c: Likewise.
1552 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1553 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1554 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1555 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1556 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1557 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1558 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1559 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1560 * testsuite/libffi.call/cls_double.c: Likewise.
1561 * testsuite/libffi.call/cls_float.c: Likewise.
1562 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1563 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1564 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1565 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1566 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1567 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1568 * testsuite/libffi.call/cls_schar.c: Likewise.
1569 * testsuite/libffi.call/cls_sint.c: Likewise.
1570 * testsuite/libffi.call/cls_sshort.c: Likewise.
1571 * testsuite/libffi.call/cls_uchar.c: Likewise.
1572 * testsuite/libffi.call/cls_uint.c: Likewise.
1573 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1574 * testsuite/libffi.call/cls_ushort.c: Likewise.
1575 * testsuite/libffi.call/nested_struct.c: Likewise.
1576 * testsuite/libffi.call/nested_struct1.c: Likewise.
1577 * testsuite/libffi.call/nested_struct2.c: Likewise.
1578 * testsuite/libffi.call/nested_struct3.c: Likewise.
1579 * testsuite/libffi.call/problem1.c: Likewise.
1580 * testsuite/libffi.special/unwindtest.cc: Likewise.
1581 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1583 * testsuite/libffi.call/cls_16byte.c: Likewise.
1584 * testsuite/libffi.call/cls_5byte.c: Likewise.
1586 2004-08-23 David Daney <daney@avtrex.com>
1589 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1590 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1591 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1592 parameters and return types.
1593 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1594 (ffi_prep_closure): Ditto.
1595 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1596 alignment calculations.
1597 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1598 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1600 2004-08-14 Casey Marshall <csm@gnu.org>
1602 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1603 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1604 integer (O32 ABI only).
1605 (ffi_prep_closure): new function.
1606 (ffi_closure_mips_inner_O32): new function.
1607 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1608 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1609 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1610 64 bit integers correctly.
1611 (ffi_closure_O32): new function.
1612 Added DWARF-2 unwind info for both functions.
1614 2004-08-10 Andrew Haley <aph@redhat.com>
1616 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1618 2004-08-01 Robert Millan <robertmh@gnu.org>
1620 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1621 * configure: Regenerate.
1623 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1625 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1626 and mmap() explicitly instead of relying on preset autoconf cache
1628 * aclocal.m4: Regenerate.
1629 * configure: Regenerate.
1631 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1633 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1634 (ffi_check_float_struct): Remove unused prototype.
1636 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1638 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1639 character on Darwin, use '\n\t' instead.
1641 2004-06-26 Matthias Klose <doko@debian.org>
1643 * libtool-version: Fix typo in revision/age.
1645 2004-06-17 Matthias Klose <doko@debian.org>
1647 * libtool-version: New.
1648 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1649 * Makefile.in: Regenerate.
1651 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1653 * Makefile.am: Remove useless multilib rules.
1654 * Makefile.in: Regenerate.
1655 * aclocal.m4: Regenerate with automake 1.8.5.
1656 * configure.ac: Remove useless multilib configury.
1657 * configure: Regenerate.
1659 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1661 * .cvsignore: New file.
1663 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1665 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1667 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1670 2004-06-06 Sean McNeil <sean@mcneil.com>
1672 * configure.ac: Add x86_64-*-freebsd* support.
1673 * configure: Regenerate.
1675 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1678 * configure.ac: Test for existence of mmap and sys/mman.h before
1679 checking blacklist. Fix suggested by Jim Wilson.
1680 * configure: Regenerate.
1682 2004-04-26 Matt Austern <austern@apple.com>
1684 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1686 * src/powerpc/darwin_closure.S: Likewise.
1688 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1690 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1691 error. Reported by Thomas Heller <theller@python.net>.
1692 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1693 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1695 2004-03-20 Matthias Klose <doko@debian.org>
1697 * src/pa/linux.S: Fix typo.
1699 2004-03-19 Matthias Klose <doko@debian.org>
1701 * Makefile.am: Update.
1702 * Makefile.in: Regenerate.
1703 * src/pa/ffi.h.in: Remove.
1704 * src/pa/ffitarget.h: New file.
1706 2004-02-10 Randolph Chung <tausq@debian.org>
1708 * Makefile.am: Add PA support.
1709 * Makefile.in: Regenerate.
1710 * include/Makefile.in: Regenerate.
1711 * configure.ac: Add PA target.
1712 * configure: Regenerate.
1713 * src/pa/ffi.c: New file.
1714 * src/pa/ffi.h.in: Add PA support.
1715 * src/pa/linux.S: New file.
1716 * prep_cif.c: Add PA support.
1718 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1720 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1722 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1723 with ecif->cif->flags.
1724 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1726 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1727 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1728 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1729 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1731 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1733 * configure.in: Rename file to ...
1734 * configure.ac: ... this.
1735 * fficonfig.h.in: Regenerate.
1736 * Makefile.in: Regenerate.
1737 * include/Makefile.in: Regenerate.
1738 * testsuite/Makefile.in: Regenerate.
1740 2004-03-12 Matt Austern <austern@apple.com>
1742 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1743 changes in EH format resulting from addition of linkonce support.
1744 * src/powerpc/darwin_closure.S: Likewise.
1746 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1747 Paolo Bonzini <bonzini@gnu.org>
1749 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1750 Remove VPATH. Remove rules for object files. Remove multilib support.
1751 (AM_CCASFLAGS): Add.
1752 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1753 (AC_PREREQ): Bump version to 2.59.
1754 (AC_INIT): Fill with version info and bug address.
1755 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1756 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1757 De-precious CC so that the right flags are passed down to multilibs.
1758 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1759 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1760 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1761 * configure: Rebuilt.
1762 * aclocal.m4: Likewise.
1763 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1764 * fficonfig.h.in: Likewise.
1766 2004-03-11 Andreas Schwab <schwab@suse.de>
1768 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1769 arguments from fp registers only for the first 8 parameter slots.
1770 Don't convert a float parameter when passed in memory.
1772 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1774 * configure: Regenerate for config/accross.m4 correction.
1776 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1778 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1779 ecif->cif->bytes to bytes.
1780 (ffi_prep_cif_machdep): Add braces around nested if statement.
1782 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1784 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1786 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1787 (ffi_closure_helper_LINUX64): Fix typo.
1788 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1790 * testsuite/libffi.call/float.c: Likewise.
1791 * testsuite/libffi.call/float2.c: Likewise.
1793 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1795 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1796 long double function return and long double arg handling.
1797 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1798 Use "end_pfr" instead of "nf". Correct long double handling.
1800 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1802 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1803 space for long double return value. Adjust stack frame and offsets.
1804 Load f2 long double return.
1806 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1808 * src/types.c: Use 16 byte long double for POWERPC64.
1810 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1812 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1813 when the structure return address is passed in %o0.
1814 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1815 (ffi_v9_layout_struct): Align the field following a nested structure
1816 on a word boundary. Use memmove instead of memcpy.
1817 (ffi_call): Update call to ffi_V9_return_struct.
1818 (ffi_prep_closure): Define 'ctx' only for V8.
1819 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1820 and ffi_closure_sparc_inner_v9.
1821 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1822 Always skip the structure return address. For structures and long
1823 doubles, copy the argument directly.
1824 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1825 if required. Shift the maximum floating-point slot accordingly. For
1826 big structures, copy the argument directly; otherwise, left-justify the
1827 argument and call ffi_v9_layout_struct to lay out the structure on
1829 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1830 (ffi_closure_v8): Pass the structure return address. Update call to
1831 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1832 Skip the 'unimp' insn when returning long doubles and structures.
1833 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1834 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1835 FFI_TYPE_INT handling. Load structures both in integers and
1836 floating-point registers on return.
1837 * README: Update status of the SPARC port.
1839 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1841 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1843 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1845 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1847 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1848 value as of type ffi_arg, not unsigned int.
1850 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1852 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1855 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1857 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1860 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1862 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1865 2003-12-10 Richard Henderson <rth@redhat.com>
1867 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1868 size_t instead of int.
1870 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1872 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1873 * testsuite/libffi.call/many_win32.c (main): Replace variable
1874 int i with unsigned long ul.
1876 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1877 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1878 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1879 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1880 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1881 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1882 * testsuite/libffi.call/cls_align_float.c: Likewise.
1883 * testsuite/libffi.call/cls_align_double.c: Likewise.
1884 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1885 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1887 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1890 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1891 Align arguments to 32 bits.
1893 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1896 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1897 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1898 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1899 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1900 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1901 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1903 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1905 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1907 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1908 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1910 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1912 * Makefile.in: Rebuilt.
1913 * configure: Likewise.
1914 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1917 2003-11-21 Andreas Jaeger <aj@suse.de>
1918 Andreas Tobler <a.tobler@schweiz.ch>
1920 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1921 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1922 * Makefile.in: Rebuilt.
1923 * aclocal.m4: Likewise.
1924 * configure: Likewise.
1925 * fficonfig.h.in: Likewise.
1926 * testsuite/lib/libffi-dg.exp: Add include dir.
1927 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1928 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1929 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1930 for ffi_closure if available.
1931 * testsuite/libffi.call/closure_fn1.c: Likewise.
1932 * testsuite/libffi.call/closure_fn2.c: Likewise.
1933 * testsuite/libffi.call/closure_fn3.c: Likewise.
1934 * testsuite/libffi.call/closure_fn4.c: Likewise.
1935 * testsuite/libffi.call/closure_fn5.c: Likewise.
1936 * testsuite/libffi.call/cls_12byte.c: Likewise.
1937 * testsuite/libffi.call/cls_16byte.c: Likewise.
1938 * testsuite/libffi.call/cls_18byte.c: Likewise.
1939 * testsuite/libffi.call/cls_19byte.c: Likewise.
1940 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1941 * testsuite/libffi.call/cls_20byte.c: Likewise.
1942 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1943 * testsuite/libffi.call/cls_24byte.c: Likewise.
1944 * testsuite/libffi.call/cls_2byte.c: Likewise.
1945 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1946 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1947 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1948 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1949 * testsuite/libffi.call/cls_4byte.c: Likewise.
1950 * testsuite/libffi.call/cls_5byte.c: Likewise.
1951 * testsuite/libffi.call/cls_64byte.c: Likewise.
1952 * testsuite/libffi.call/cls_6byte.c: Likewise.
1953 * testsuite/libffi.call/cls_7byte.c: Likewise.
1954 * testsuite/libffi.call/cls_8byte.c: Likewise.
1955 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1956 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1957 * testsuite/libffi.call/cls_double.c: Likewise.
1958 * testsuite/libffi.call/cls_float.c: Likewise.
1959 * testsuite/libffi.call/cls_schar.c: Likewise.
1960 * testsuite/libffi.call/cls_sint.c: Likewise.
1961 * testsuite/libffi.call/cls_sshort.c: Likewise.
1962 * testsuite/libffi.call/cls_uchar.c: Likewise.
1963 * testsuite/libffi.call/cls_uint.c: Likewise.
1964 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1965 * testsuite/libffi.call/cls_ushort.c: Likewise.
1966 * testsuite/libffi.call/nested_struct.c: Likewise.
1967 * testsuite/libffi.call/nested_struct1.c: Likewise.
1968 * testsuite/libffi.call/nested_struct2.c: Likewise.
1969 * testsuite/libffi.call/nested_struct3.c: Likewise.
1970 * testsuite/libffi.call/problem1.c: Likewise.
1971 * testsuite/libffi.special/unwindtest.cc: Likewise.
1973 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
1975 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
1977 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
1979 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
1980 Add -lgcc_s to additional flags.
1982 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
1984 * configure.in, include/Makefile.am: PR libgcj/11147, install
1985 the ffitarget.h header file in a gcc versioned and target
1987 * configure: Regenerated.
1988 * Makefile.in, include/Makefile.in: Likewise.
1989 * testsuite/Makefile.in: Likewise.
1991 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
1993 * testsuite/libffi.call/closure_fn0.c: Print result and check
1994 with dg-output to make debugging easier.
1995 * testsuite/libffi.call/closure_fn1.c: Likewise.
1996 * testsuite/libffi.call/closure_fn2.c: Likewise.
1997 * testsuite/libffi.call/closure_fn3.c: Likewise.
1998 * testsuite/libffi.call/closure_fn4.c: Likewise.
1999 * testsuite/libffi.call/closure_fn5.c: Likewise.
2000 * testsuite/libffi.call/cls_12byte.c: Likewise.
2001 * testsuite/libffi.call/cls_16byte.c: Likewise.
2002 * testsuite/libffi.call/cls_18byte.c: Likewise.
2003 * testsuite/libffi.call/cls_19byte.c: Likewise.
2004 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2005 * testsuite/libffi.call/cls_20byte.c: Likewise.
2006 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2007 * testsuite/libffi.call/cls_24byte.c: Likewise.
2008 * testsuite/libffi.call/cls_2byte.c: Likewise.
2009 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2010 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2011 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2012 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2013 * testsuite/libffi.call/cls_4byte.c: Likewise.
2014 * testsuite/libffi.call/cls_5byte.c: Likewise.
2015 * testsuite/libffi.call/cls_64byte.c: Likewise.
2016 * testsuite/libffi.call/cls_6byte.c: Likewise.
2017 * testsuite/libffi.call/cls_7byte.c: Likewise.
2018 * testsuite/libffi.call/cls_8byte.c: Likewise.
2019 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2020 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2021 * testsuite/libffi.call/cls_double.c: Likewise.
2022 * testsuite/libffi.call/cls_float.c: Likewise.
2023 * testsuite/libffi.call/cls_schar.c: Likewise.
2024 * testsuite/libffi.call/cls_sint.c: Likewise.
2025 * testsuite/libffi.call/cls_sshort.c: Likewise.
2026 * testsuite/libffi.call/cls_uchar.c: Likewise.
2027 * testsuite/libffi.call/cls_uint.c: Likewise.
2028 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2029 * testsuite/libffi.call/cls_ushort.c: Likewise.
2030 * testsuite/libffi.call/problem1.c: Likewise.
2032 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2035 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2037 * testsuite/libffi.call/cls_9byte2.c: New test case.
2038 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2039 * testsuite/libffi.call/cls_64byte.c: Likewise.
2040 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2041 * testsuite/libffi.call/cls_19byte.c: Likewise.
2042 * testsuite/libffi.call/cls_18byte.c: Likewise.
2043 * testsuite/libffi.call/closure_fn4.c: Likewise.
2044 * testsuite/libffi.call/closure_fn5.c: Likewise.
2045 * testsuite/libffi.call/cls_schar.c: Likewise.
2046 * testsuite/libffi.call/cls_sint.c: Likewise.
2047 * testsuite/libffi.call/cls_sshort.c: Likewise.
2048 * testsuite/libffi.call/nested_struct2.c: Likewise.
2049 * testsuite/libffi.call/nested_struct3.c: Likewise.
2051 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2053 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2054 * testsuite/libffi.call/cls_uchar.c: Likewise.
2055 * testsuite/libffi.call/cls_uint.c: Likewise.
2056 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2057 * testsuite/libffi.call/cls_ushort.c: Likewise.
2058 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2060 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2062 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2065 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2067 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2068 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2070 2003-10-22 David Daney <ddaney@avtrex.com>
2072 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2073 __attribute__((__mode__(__SI__))) and friends.
2075 2003-10-22 Andreas Schwab <schwab@suse.de>
2077 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2079 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2081 * configure.in: AC_LINK_FILES(ffitarget.h).
2082 * configure: Regenerate.
2083 * Makefile.in: Likewise.
2084 * include/Makefile.in: Likewise.
2085 * testsuite/Makefile.in: Likewise.
2086 * fficonfig.h.in: Likewise.
2088 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2089 Richard Henderson <rth@redhat.com>
2091 Avoid that ffi.h includes fficonfig.h.
2093 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2094 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2095 (TARGET_SRC_MIPS_SGI): Removed.
2096 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2097 (MIPS_SGI): Removed.
2098 (CLEANFILES): Removed.
2099 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2101 * acconfig.h: Removed.
2102 * configure.in: Compute sizeofs only for double and long double.
2103 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2104 into AC_DEFINE instead of using acconfig.h. Create
2105 include/ffitarget.h instead of include/fficonfig.h. Rename
2106 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2107 AC_DEFINE EH_FRAME_FLAGS.
2108 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2109 (hack_DATA): Add ffitarget.h.
2110 * include/ffi.h.in: Remove all system specific definitions.
2111 Declare raw API even if it is not installed, why bother?
2112 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2113 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2114 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2115 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2116 * include/ffi_common.h (bool): Do not define.
2117 (ffi_assert): Accept failed assertion.
2118 (ffi_type_test): Return void and accept file/line.
2119 (FFI_ASSERT): Pass stringized failed assertion.
2120 (FFI_ASSERT_AT): New macro.
2121 (FFI_ASSERT_VALID_TYPE): New macro.
2122 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2123 UINT64, SINT64): Define here with gcc's __attribute__ macro
2125 (FLOAT32, ALIGN): Define here instead of in ffi.h
2126 * include/ffi-mips.h: Removed. Its content moved to
2127 src/mips/ffitarget.h after separating assembly and C sections.
2128 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2129 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2130 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2131 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2132 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2133 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2134 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2135 (ffi_type_test): Rewritten.
2136 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2137 FFI_ASSERT_VALID_TYPE.
2138 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2139 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2140 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2141 src/s390/ffitarget.h, src/sh/ffitarget.h,
2142 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2143 src/x86/ffitarget.h: New files.
2144 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2145 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2146 src/powerpc/aix.S, src/powerpc/darwin.S,
2147 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2148 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2149 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2150 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2151 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2154 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2156 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2157 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2159 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2161 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2162 Used when FFI_DEBUG = 1.
2164 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2166 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2169 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2171 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2172 FFI_MIPS_O32 for O32 ABI.
2174 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2176 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2177 SPARC64. Cleanup whitespaces.
2179 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2181 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2182 strongarm, xscale. Cleanup whitespaces.
2183 * testsuite/libffi.call/closure_fn1.c: Likewise.
2184 * testsuite/libffi.call/closure_fn2.c: Likewise.
2185 * testsuite/libffi.call/closure_fn3.c: Likewise.
2186 * testsuite/libffi.call/cls_12byte.c: Likewise.
2187 * testsuite/libffi.call/cls_16byte.c: Likewise.
2188 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2189 * testsuite/libffi.call/cls_20byte.c: Likewise.
2190 * testsuite/libffi.call/cls_24byte.c: Likewise.
2191 * testsuite/libffi.call/cls_2byte.c: Likewise.
2192 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2193 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2194 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2195 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2196 * testsuite/libffi.call/cls_4byte.c: Likewise.
2197 * testsuite/libffi.call/cls_5byte.c: Likewise.
2198 * testsuite/libffi.call/cls_6byte.c: Likewise.
2199 * testsuite/libffi.call/cls_7byte.c: Likewise.
2200 * testsuite/libffi.call/cls_8byte.c: Likewise.
2201 * testsuite/libffi.call/cls_double.c: Likewise.
2202 * testsuite/libffi.call/cls_float.c: Likewise.
2203 * testsuite/libffi.call/cls_uchar.c: Likewise.
2204 * testsuite/libffi.call/cls_uint.c: Likewise.
2205 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2206 * testsuite/libffi.call/cls_ushort.c: Likewise.
2207 * testsuite/libffi.call/nested_struct.c: Likewise.
2208 * testsuite/libffi.call/nested_struct1.c: Likewise.
2209 * testsuite/libffi.call/problem1.c: Likewise.
2210 * testsuite/libffi.special/unwindtest.cc: Likewise.
2211 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2213 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2215 * src/powerpc/aix.S: Cleanup whitespaces.
2216 * src/powerpc/aix_closure.S: Likewise.
2218 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2220 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2221 * src/powerpc/darwin_closure.S: Likewise.
2222 * src/powerpc/ffi_darwin.c: Likewise.
2224 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2225 David Edelsohn <edelsohn@gnu.org>
2227 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2228 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2230 * src/powerpc/darwin_closure.S: Likewise.
2231 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2232 according to the Darwin/AIX ABI.
2233 (ffi_prep_cif_machdep): Likewise.
2234 (ffi_closure_helper_DARWIN): Likewise.
2235 Remove the outgoing parameter stack logic. Simplify the evaluation
2236 of the different CASE types.
2237 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2238 statement in the trampoline code.
2240 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2242 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2243 for the register size.
2244 (ffi_closure_helper_SYSV): Handle the structure return value
2246 (ffi_closure_helper_SYSV): Return the appropriate type when
2247 the registers are used for the structure return value.
2248 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2249 the 64-bit return value. Update copyright years.
2251 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2253 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2254 srcdir for ffi_mips.h.
2256 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2258 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2260 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2261 placement of float result.
2262 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2263 cast of "resp" for big-endian 64 bit machines.
2265 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2267 * src/types.c (double, longdouble): Merge identical SH and ARM
2268 typedefs, and add POWERPC64.
2269 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2270 struct split over gpr and rest.
2271 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2272 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2274 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2276 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2279 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2281 * configure: Regenerate.
2283 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2285 * Makefile.am: Remove build rules for ffitest.
2286 * Makefile.in: Rebuilt.
2288 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2290 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2291 about implicit declaration of abort().
2293 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2295 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2296 * Makefile.in: Rebuilt.
2297 * configure.in: Add dejagnu test framework.
2298 * configure: Rebuilt.
2300 * testsuite/Makefile.am: New file.
2301 * testsuite/Makefile.in: Built
2302 * testsuite/lib/libffi-dg.exp: New file.
2303 * testsuite/config/default.exp: Likewise.
2304 * testsuite/libffi.call/call.exp: Likewise.
2305 * testsuite/libffi.call/ffitest.h: Likewise.
2306 * testsuite/libffi.call/closure_fn0.c: Likewise.
2307 * testsuite/libffi.call/closure_fn1.c: Likewise.
2308 * testsuite/libffi.call/closure_fn2.c: Likewise.
2309 * testsuite/libffi.call/closure_fn3.c: Likewise.
2310 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2311 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2312 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2313 * testsuite/libffi.call/cls_2byte.c: Likewise.
2314 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2315 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2316 * testsuite/libffi.call/cls_4byte.c: Likewise.
2317 * testsuite/libffi.call/cls_5byte.c: Likewise.
2318 * testsuite/libffi.call/cls_6byte.c: Likewise.
2319 * testsuite/libffi.call/cls_7byte.c: Likewise.
2320 * testsuite/libffi.call/cls_8byte.c: Likewise.
2321 * testsuite/libffi.call/cls_12byte.c: Likewise.
2322 * testsuite/libffi.call/cls_16byte.c: Likewise.
2323 * testsuite/libffi.call/cls_20byte.c: Likewise.
2324 * testsuite/libffi.call/cls_24byte.c: Likewise.
2325 * testsuite/libffi.call/cls_double.c: Likewise.
2326 * testsuite/libffi.call/cls_float.c: Likewise.
2327 * testsuite/libffi.call/cls_uchar.c: Likewise.
2328 * testsuite/libffi.call/cls_uint.c: Likewise.
2329 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2330 * testsuite/libffi.call/cls_ushort.c: Likewise.
2331 * testsuite/libffi.call/float.c: Likewise.
2332 * testsuite/libffi.call/float1.c: Likewise.
2333 * testsuite/libffi.call/float2.c: Likewise.
2334 * testsuite/libffi.call/many.c: Likewise.
2335 * testsuite/libffi.call/many_win32.c: Likewise.
2336 * testsuite/libffi.call/nested_struct.c: Likewise.
2337 * testsuite/libffi.call/nested_struct1.c: Likewise.
2338 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2339 * testsuite/libffi.call/problem1.c: Likewise.
2340 * testsuite/libffi.call/promotion.c: Likewise.
2341 * testsuite/libffi.call/return_ll.c: Likewise.
2342 * testsuite/libffi.call/return_sc.c: Likewise.
2343 * testsuite/libffi.call/return_uc.c: Likewise.
2344 * testsuite/libffi.call/strlen.c: Likewise.
2345 * testsuite/libffi.call/strlen_win32.c: Likewise.
2346 * testsuite/libffi.call/struct1.c: Likewise.
2347 * testsuite/libffi.call/struct2.c: Likewise.
2348 * testsuite/libffi.call/struct3.c: Likewise.
2349 * testsuite/libffi.call/struct4.c: Likewise.
2350 * testsuite/libffi.call/struct5.c: Likewise.
2351 * testsuite/libffi.call/struct6.c: Likewise.
2352 * testsuite/libffi.call/struct7.c: Likewise.
2353 * testsuite/libffi.call/struct8.c: Likewise.
2354 * testsuite/libffi.call/struct9.c: Likewise.
2355 * testsuite/libffi.special/special.exp: New file.
2356 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2357 * testsuite/libffi.special/unwindtest.cc: Likewise.
2360 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2362 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2365 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2367 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2369 (ffi_closure_helper_LINUX64): Likewise.
2370 * src/powerpc/linux64.S: Remove code writing to parm save area.
2371 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2372 address in lr from ffi_closure_helper_LINUX64 call to calculate
2373 table address. Optimize function tail.
2375 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2377 * src/sparc/ffi.c: Handle all floating point registers.
2378 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2380 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2382 * README: Note that libffi is not part of GCC. Update the project
2385 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2387 * src/powerpc/ppc_closure.S: Include ffi.h.
2389 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2391 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2392 Use C style comments.
2394 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2396 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2397 * Makefile.in: Regenerate.
2398 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2399 * configure: Regenerate.
2400 * include/ffi.h.in: Add SHmedia support.
2401 * src/sh64/ffi.c: New file.
2402 * src/sh64/sysv.S: New file.
2404 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2406 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2407 should be read-only.
2408 * configure: Rebuilt.
2409 * fficonfig.h.in: Rebuilt.
2410 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2411 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2412 * src/powerpc/linux64.S: Likewise.
2413 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2414 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2415 if -fpic/-fPIC/-mrelocatable.
2416 * src/powerpc/powerpc_closure.S: Likewise.
2417 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2418 #write in .eh_frame flags.
2419 * src/sparc/v9.S: Likewise.
2420 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2421 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2422 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2424 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2426 Fixes PR bootstrap/10656
2427 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2428 support for .register pseudo-op.
2429 * src/sparc/v8.S: Use it.
2430 * fficonfig.h.in: Rebuilt.
2431 * configure: Rebuilt.
2433 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2435 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2436 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2437 Make it the default on POWERPC64.
2438 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2439 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2440 * configure: Rebuilt.
2441 * src/powerpc/ffi.c (hidden): Define.
2442 (ffi_prep_args_SYSV): Renamed from
2443 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2444 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2445 ASM_NEEDS_REGISTERS64): New.
2446 (ffi_prep_args64): New function.
2447 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2448 (ffi_call): Likewise.
2449 (ffi_prep_closure): Likewise.
2450 (flush_icache): Surround by #ifndef POWERPC64.
2451 (ffi_dblfl): New union type.
2452 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2453 (ffi_closure_helper_LINUX64): New function.
2454 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2456 * src/powerpc/sysv.S: Likewise.
2457 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2458 * src/powerpc/linux64.S: New file.
2459 * src/powerpc/linux64_closure.S: New file.
2460 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2461 src/powerpc/linux64_closure.S.
2462 (TARGET_SRC_POWERPC): Likewise.
2464 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2465 closure_test_fn3): Fix result printing on big-endian 64-bit
2467 (main): Print tst2_arg instead of uninitialized tst2_result.
2469 * src/ffitest.c (main): Hide what closure pointer really points to
2472 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2474 * configure.in (arm-*-netbsdelf*): Add configuration.
2475 (configure): Regenerated.
2477 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2479 * include/Makefile.in: Regenerate.
2481 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2483 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2485 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2486 Receive closure pointer through parameter, read args using
2487 __builtin_dwarf_cfa.
2488 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2490 2003-03-12 Andreas Schwab <schwab@suse.de>
2492 * configure.in: Avoid trailing /. in toolexeclibdir.
2493 * configure: Rebuilt.
2495 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2497 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2499 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2501 * libffi/src/powerpc/darwin_closure.S:
2502 Fix alignement bug, allocate 8 bytes for the result.
2503 * libffi/src/powerpc/aix_closure.S:
2505 * libffi/src/powerpc/ffi_darwin.c:
2506 Update stackframe description for aix/darwin_closure.S.
2508 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2510 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2513 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2514 Andreas Schwab <schwab@suse.de>
2516 * configure.in: Adjust command to source config-ml.in to account
2517 for changes to the libffi_basedir definition.
2518 (libffi_basedir): Remove ${srcdir} from value and include trailing
2521 * configure: Regenerate.
2523 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2525 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2527 2003-01-28 Andrew Haley <aph@redhat.com>
2529 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2530 * src/x86/ffi64.c (ffi_prep_closure): New.
2531 (ffi_closure_UNIX64_inner): New.
2532 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2534 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2536 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2537 Remove USE_LIBDIR conditional.
2538 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2539 * Makefile.in, configure: Rebuilt.
2541 2003-01027 David Edelsohn <edelsohn@gnu.org>
2543 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2544 * Makefile.in: Regenerate.
2546 2003-01-22 Andrew Haley <aph@redhat.com>
2548 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2551 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2553 * src/powerpc/darwin.S: Add unwind info.
2554 * src/powerpc/darwin_closure.S: Likewise.
2556 2003-01-14 Andrew Haley <aph@redhat.com>
2558 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2559 (ffi_prep_cif_machdep): Likewise.
2560 * src/x86/unix64.S: Add unwind info.
2562 2003-01-14 Andreas Jaeger <aj@suse.de>
2564 * src/ffitest.c (main): Only use ffi_closures if those are
2567 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2569 * libffi/src/ffitest.c
2570 add closure testcases
2572 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2574 * libffi/src/powerpc/ffi.c
2575 fix alignment bug for float (4 byte aligned iso 8 byte)
2577 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2579 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2580 * src/powerpc/darwin.S: Remove RCS version string.
2582 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2584 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2585 * src/ffitest.c (main): Use static storage for closure.
2586 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2587 * src/sparc/v8.S (ffi_closure_v8): New.
2588 * src/sparc/v9.S (ffi_closure_v9): New.
2590 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2592 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2593 enumeration for X86_WIN32.
2594 * src/x86/win32.S: Added ffi_call_STDCALL function
2596 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2597 switch cases for recognising FFI_STDCALL and
2598 calling ffi_call_STDCALL if target is X86_WIN32.
2599 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2600 stdcall versions of the "my_strlen" and "many"
2601 test functions (for X86_WIN32).
2602 Added test cases to test stdcall invocation using
2605 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2607 * src/sh/sysv.S: Add DWARF2 unwind info.
2609 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2611 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2613 2002-11-26 Jim Wilson <wilson@redhat.com>
2615 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2617 2002-11-23 H.J. Lu <hjl@gnu.org>
2619 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2620 Include ../config/accross.m4.
2621 * aclocal.m4; Rebuild.
2622 * configure: Likewise.
2624 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2626 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2628 2002-11-11 DJ Delorie <dj@redhat.com>
2630 * configure.in: Look for common files in the right place.
2632 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2634 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2635 raw data as _Jv_word values, not ffi_raw.
2636 (ffi_java_ptrarray_to_raw): Likewise.
2637 (ffi_java_rvalue_to_raw): New function.
2638 (ffi_java_raw_call): Call it.
2639 (ffi_java_raw_to_rvalue): New function.
2640 (ffi_java_translate_args): Call it.
2641 * src/ffitest.c (closure_test_fn): Interpret return value
2642 as ffi_arg, not int.
2643 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2644 FFI_TYPE_POINTER case.
2645 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2646 values extended to word size.
2648 2002-10-02 Andreas Jaeger <aj@suse.de>
2650 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2652 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2654 * include/ffi.h.in: Fix i386 win32 compilation.
2656 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2658 * configure.in: Add s390x-*-linux-* target.
2659 * configure: Regenerate.
2660 * include/ffi.h.in: Define S390X for s390x targets.
2661 (FFI_CLOSURES): Define for s390/s390x.
2662 (FFI_TRAMPOLINE_SIZE): Likewise.
2663 (FFI_NATIVE_RAW_API): Likewise.
2664 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2665 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2666 * src/s390/ffi.c: Major rework of existing code. Add support for
2667 s390x targets. Add closure support.
2668 * src/s390/sysv.S: Likewise.
2670 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2672 * src/arm/sysv.S: Fix typo.
2674 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2676 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2677 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2678 (ffi_call_SYSV): Handle soft-float.
2680 2002-09-27 Bo Thorsen <bo@suse.de>
2682 * include/ffi.h.in: Fix multilib x86-64 support.
2684 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2686 * Makefile.am (all-multi): Fix multilib parallel build.
2688 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2690 * configure.in (sh[34]*-*-linux*): Add brackets.
2691 * configure: Regenerate.
2693 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2695 * Makefile.am: Add SH support.
2696 * Makefile.in: Regenerate.
2697 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2698 * configure: Regenerate.
2699 * include/ffi.h.in: Add SH support.
2700 * src/sh/ffi.c: New file.
2701 * src/sh/sysv.S: New file.
2702 * src/types.c: Add SH support.
2704 2002-07-16 Bo Thorsen <bo@suse.de>
2706 * src/x86/ffi64.c: New file that adds x86-64 support.
2707 * src/x86/unix64.S: New file that handles argument setup for
2709 * src/x86/sysv.S: Don't use this on x86-64.
2710 * src/x86/ffi.c: Don't use this on x86-64.
2712 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2714 * src/ffitest.c (struct6): New test that tests a special case in
2716 (struct7): Likewise.
2717 (struct8): Likewise.
2718 (struct9): Likewise.
2719 (closure_test_fn): Silence warning about this when it's not used.
2720 (main): Add the new tests.
2721 (main): Fix a couple of wrong casts and silence some compiler warnings.
2722 * include/ffi.h.in: Add x86-64 ABI definition.
2723 * fficonfig.h.in: Regenerate.
2724 * Makefile.am: Add x86-64 support.
2725 * configure.in: Likewise.
2726 * Makefile.in: Regenerate.
2727 * configure: Likewise.
2729 2002-06-24 Bo Thorsen <bo@suse.de>
2731 * src/types.c: Merge settings for similar architectures.
2732 Add x86-64 sizes and alignments.
2734 2002-06-23 Bo Thorsen <bo@suse.de>
2736 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2737 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2738 * src/mips/ffi.c (ffi_prep_args): Likewise.
2739 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2741 2002-07-18 H.J. Lu (hjl@gnu.org)
2743 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2744 (libffi_la_SOURCES): Support MIPS_LINUX.
2745 (libffi_convenience_la_SOURCES): Likewise.
2746 * Makefile.in: Regenerated.
2748 * configure.in (mips64*-*): Skip.
2749 (mips*-*-linux*): New.
2750 * configure: Regenerated.
2752 * src/mips/ffi.c: Include <sgidefs.h>.
2754 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2756 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2758 2002-05-27 Roger Sayle <roger@eyesopen.com>
2760 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2762 2002-05-27 Bo Thorsen <bo@suse.de>
2764 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2767 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2769 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2770 beginning of function (for older apple cc).
2772 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2774 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2775 script entry, and set LD to it when configuring multilibs.
2776 * configure: Rebuilt.
2778 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2780 * configure.in (sparc64-*-netbsd*): Add target.
2781 (sparc-*-netbsdelf*): Likewise.
2782 * configure: Regenerate.
2784 2002-04-28 David S. Miller <davem@redhat.com>
2786 * configure.in, configure: Fix SPARC test in previous change.
2788 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2790 * Makefile.am: Add Linux for S/390 support.
2791 * Makefile.in: Regenerate.
2792 * configure.in: Add Linux for S/390 support.
2793 * configure: Regenerate.
2794 * include/ffi.h.in: Add Linux for S/390 support.
2795 * src/s390/ffi.c: New file from libffi CVS tree.
2796 * src/s390/sysv.S: New file from libffi CVS tree.
2798 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2800 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2802 * src/sparc/v8.S: Use it.
2803 * src/sparc/v9.S: Likewise.
2804 * fficonfig.h.in: Rebuilt.
2805 * configure: Rebuilt.
2807 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2809 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2811 * src/ia64/unix.S: Add unwind information. Fix comments.
2812 Save sp in a way that's compatible with unwind info.
2813 (ffi_call_unix): Correctly restore sp in all cases.
2814 * src/ia64/ffi.c: Add, fix comments.
2816 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2818 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2820 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2822 * configure.in (alpha*-*-netbsd*): Add target.
2823 * configure: Regenerate.
2825 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2827 * src/sparc/v8.S: Add unwind info.
2828 * src/sparc/v9.S: Likewise.
2830 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2832 * configure.in: Enable i*86-*-netbsdelf*.
2833 * configure: Rebuilt.
2835 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2838 * src/mips/n32.s: Delete
2839 * src/mips/o32.s: Delete
2841 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2843 * configure.in: Enable alpha*-*-freebsd*.
2844 * configure: Rebuilt.
2846 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2848 * Makefile.am: libfficonvenience -> libffi_convenience.
2849 * Makefile.in: Rebuilt.
2851 * Makefile.am: Define ffitest_OBJECTS.
2852 * Makefile.in: Rebuilt.
2854 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2855 David Edelsohn <edelsohn@gnu.org>
2857 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2858 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2859 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2860 * Makefile.in: Regenerate.
2861 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2862 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2863 (flush_icache, flush_range): New functions.
2864 (ffi_closure_helper_DARWIN): New function.
2865 * src/powerpc/aix_closure.S: New file.
2866 * src/powerpc/darwin_closure.S: New file.
2868 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2870 * include/ffi.h.in: Add typedef for ffi_arg.
2871 * src/ffitest.c (main): Declare rint with ffi_arg.
2873 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2875 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2876 number of GPRs for floating-point arguments.
2878 2002-01-31 Anthony Green <green@redhat.com>
2880 * configure: Rebuilt.
2881 * configure.in: Replace CHECK_SIZEOF and endian tests with
2882 cross-compiler friendly macros.
2883 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2886 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2888 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2889 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2891 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2893 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2894 (TARGET_SRC_POWERPC_AIX): New.
2895 (POWERPC_AIX): New stanza.
2896 * Makefile.in: Regenerate.
2897 * configure.in: Add AIX case.
2898 * configure: Regenerate.
2899 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2900 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2901 size. Fix "long double" support.
2902 (ffi_call): Add FFI_AIX case.
2903 * src/powerpc/aix.S: New.
2905 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2907 Implement Darwin PowerPC ABI.
2908 * configure.in: Handle powerpc-*-darwin*.
2909 * Makefile.am: Set source files for POWERPC_DARWIN.
2910 * configure: Rebuilt.
2911 * Makefile.in: Rebuilt.
2912 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2914 * src/powerpc/darwin.S: New file.
2915 * src/powerpc/ffi_darwin.c: New file.
2917 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2919 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2921 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2923 * src/x86/sysv.S: Avoid gas-only .balign directive.
2924 Use C style comments.
2926 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2928 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2929 Fixes PR bootstrap/3563.
2931 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2933 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2935 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2937 * configure.in: Recognize sparc*-sun-* host.
2938 * configure: Regenerate.
2940 2001-06-06 Andrew Haley <aph@redhat.com>
2942 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
2944 2001-06-03 Andrew Haley <aph@redhat.com>
2946 * src/alpha/osf.S: Add unwind info.
2947 * src/powerpc/sysv.S: Add unwind info.
2948 * src/powerpc/ppc_closure.S: Likewise.
2950 2000-05-31 Jeff Sturm <jsturm@one-point.com>
2952 * configure.in: Fix AC_ARG_ENABLE usage.
2953 * configure: Rebuilt.
2955 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2957 * configure.in: Remove warning about beta code.
2958 * configure: Rebuilt.
2960 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
2962 * src/ia64/unix.S: Restore stack pointer when returning from
2964 * src/ia64/ffi.c: Fix typo in comment.
2966 2001-04-18 Jim Wilson <wilson@redhat.com>
2968 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
2969 to eliminate RAW DV.
2971 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
2973 * Makefile.am: Make a libtool convenience library.
2974 * Makefile.in: Rebuilt.
2976 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
2978 * configure.in: Use different syntax for subdirectory creation.
2979 * configure: Rebuilt.
2981 2001-03-27 Jon Beniston <jon@beniston.com>
2983 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
2984 * configure: Rebuilt.
2985 * Makefile.am: Added X86_WIN32 target support.
2986 * Makefile.in: Rebuilt.
2988 * include/ffi.h.in: Added X86_WIN32 target support.
2990 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
2991 * src/types.c: Added X86_WIN32 target support.
2993 * src/x86/win32.S: New file. Based on sysv.S, but with EH
2994 stuff removed and made to work with CygWin's gas.
2996 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
2998 * configure.in: Make target subdirectory in build dir.
2999 * Makefile.am: Override suffix based rules to specify correct output
3001 * Makefile.in: Rebuilt.
3002 * configure: Rebuilt.
3004 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3006 * src/powerpc/ppc_closure.S: New file.
3007 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3008 involving long long and register pairs.
3009 (ffi_prep_closure): New function.
3010 (flush_icache): Likewise.
3011 (ffi_closure_helper_SYSV): Likewise.
3012 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3013 (FFI_TRAMPOLINE_SIZE): Likewise.
3014 (FFI_NATIVE_RAW_API): Likewise.
3015 * Makefile.in: Rebuilt.
3016 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3017 (TARGET_SRC_POWERPC): Likewise.
3019 2001-03-19 Tom Tromey <tromey@redhat.com>
3021 * Makefile.in: Rebuilt.
3022 * Makefile.am (ffitest_LDFLAGS): New macro.
3024 2001-03-02 Nick Clifton <nickc@redhat.com>
3026 * include/ffi.h.in: Remove RCS ident string.
3027 * include/ffi_mips.h: Remove RCS ident string.
3028 * src/debug.c: Remove RCS ident string.
3029 * src/ffitest.c: Remove RCS ident string.
3030 * src/prep_cif.c: Remove RCS ident string.
3031 * src/types.c: Remove RCS ident string.
3032 * src/alpha/ffi.c: Remove RCS ident string.
3033 * src/alpha/osf.S: Remove RCS ident string.
3034 * src/arm/ffi.c: Remove RCS ident string.
3035 * src/arm/sysv.S: Remove RCS ident string.
3036 * src/mips/ffi.c: Remove RCS ident string.
3037 * src/mips/n32.S: Remove RCS ident string.
3038 * src/mips/o32.S: Remove RCS ident string.
3039 * src/sparc/ffi.c: Remove RCS ident string.
3040 * src/sparc/v8.S: Remove RCS ident string.
3041 * src/sparc/v9.S: Remove RCS ident string.
3042 * src/x86/ffi.c: Remove RCS ident string.
3043 * src/x86/sysv.S: Remove RCS ident string.
3045 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3047 * include/ffi.h.in: Change sourceware.cygnus.com references to
3050 2000-12-09 Richard Henderson <rth@redhat.com>
3052 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3053 (ffi_closure_osf_inner): Index rather than increment avalue
3054 and arg_types. Give ffi_closure_osf the raw return value type.
3055 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3058 2000-12-07 Richard Henderson <rth@redhat.com>
3060 * src/raw_api.c (ffi_translate_args): Fix typo.
3061 (ffi_prep_closure): Likewise.
3063 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3064 FFI_TRAMPOLINE_SIZE.
3065 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3066 cif->bytes for new ffi_call_osf implementation.
3067 (ffi_prep_args): Absorb into ...
3068 (ffi_call): ... here. Do all stack allocation here and
3069 avoid a callback function.
3070 (ffi_prep_closure, ffi_closure_osf_inner): New.
3071 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3072 (ffi_closure_osf): New.
3074 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3076 * config.guess, config.sub, install-sh: Removed.
3077 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3078 * Makefile.in: Rebuilt.
3080 * acinclude.m4: Include libtool macros from the top level.
3081 * aclocal.m4, configure: Rebuilt.
3083 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3085 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3086 * configure: Rebuilt.
3088 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3090 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3091 memory correctly. Use conditional instructions, not branches where
3094 2000-05-04 Tom Tromey <tromey@cygnus.com>
3096 * configure: Rebuilt.
3097 * configure.in: Match `arm*-*-linux-*'.
3098 From Chris Dornan <cdornan@arm.com>.
3100 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3102 * Makefile.am (SUBDIRS): Define.
3103 (AM_MAKEFLAGS): Likewise.
3104 (Multilib support.): Add section.
3105 * Makefile.in: Rebuilt.
3106 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3107 New variables. Set for gcc using -print-multi-lib. Export them
3109 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3110 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3112 (extra_compiler_flags_value, extra_compiler_flags): Check these
3113 for extra compiler options which need to be passed down in
3116 2000-04-16 Anthony Green <green@redhat.com>
3118 * configure: Rebuilt.
3119 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3121 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3123 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3124 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3125 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3126 Replace all void * sizeofs with sizeof(int).
3127 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3128 different than DOUBLE.
3129 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3130 (ffi_prep_args_v9): New function.
3131 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3132 (ffi_V9_return_struct): New function.
3133 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3134 32bit code (not yet cross-arch calls).
3135 * src/sparc/v8.S: Add struct return delay nop.
3137 * src/sparc/v9.S: New file.
3138 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3139 is used on sparc64 only for structures larger than 32 bytes.
3140 Pass by reference for structures is done for structure arguments
3141 larger than 16 bytes.
3142 * src/ffitest.c (main): Use 64bit rint on sparc64.
3143 Run long long tests on sparc.
3144 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3146 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3148 * configure.in (sparc-*-linux*): New supported target.
3149 (sparc64-*-linux*): Likewise.
3150 * configure: Rebuilt.
3151 * Makefile.am: Add v9.S to SPARC files.
3152 * Makefile.in: Likewise.
3153 (LINK): Surround $(CCLD) into double quotes, so that multilib
3154 compiles work correctly.
3156 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3158 * configure: Rebuilt.
3159 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3160 proposed by Bryce McKinlay.
3162 2000-03-20 Tom Tromey <tromey@cygnus.com>
3164 * Makefile.in: Hand edit for java_raw_api.lo.
3166 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3168 * config.guess, config.sub: Update from the gcc tree.
3169 Fix for PR libgcj/168.
3171 2000-03-03 Tom Tromey <tromey@cygnus.com>
3173 * Makefile.in: Fixed ia64 by hand.
3175 * configure: Rebuilt.
3176 * configure.in (--enable-multilib): New option.
3177 (libffi_basedir): New subst.
3178 (AC_OUTPUT): Added multilib code.
3180 2000-03-02 Tom Tromey <tromey@cygnus.com>
3182 * Makefile.in: Rebuilt.
3183 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3186 2000-02-25 Hans Boehm <boehm@acm.org>
3188 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3190 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3192 (ffi_prep_raw_closure): Use ffi_translate_args, not
3193 ffi_closure_translate.
3194 * src/java_raw_api.c: New file.
3195 * src/ffitest.c (closure_test_fn): New function.
3196 (main): Define `rint' as long long on IA64. Added new test when
3197 FFI_CLOSURES is defined.
3198 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3199 (ffi_abi): Recognize IA64.
3200 (ffi_raw): Added `flt' field.
3201 Added "Java raw API" code.
3202 * configure.in: Recognize ia64.
3203 * Makefile.am (TARGET_SRC_IA64): New macro.
3204 (libffi_la_common_SOURCES): Added java_raw_api.c.
3205 (libffi_la_SOURCES): Define in IA64 case.
3207 2000-01-04 Tom Tromey <tromey@cygnus.com>
3209 * Makefile.in: Rebuilt with newer automake.
3211 1999-12-31 Tom Tromey <tromey@cygnus.com>
3213 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3215 1999-09-01 Tom Tromey <tromey@cygnus.com>
3217 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3219 * fficonfig.h.in: Rebuilt.
3220 * configure: Rebuilt.
3221 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3222 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3223 * acconfig.h: Don't #undef PACKAGE or VERSION.
3225 1999-08-09 Anthony Green <green@cygnus.com>
3227 * include/ffi.h.in: Try to work around messy header problem
3228 with PACKAGE and VERSION.
3230 * configure: Rebuilt.
3231 * configure.in: Change version to 2.00-beta.
3233 * fficonfig.h.in: Rebuilt.
3234 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3236 * src/x86/ffi.c (ffi_raw_call): Rename.
3238 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3240 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3241 (ffi_prep_incoming_args_SYSV): Ditto.
3242 (ffi_prep_closure): Ditto.
3243 (ffi_closure_raw_SYSV): Ditto.
3244 (ffi_prep_raw_closure): More ditto.
3245 (ffi_call_raw): Final ditto.
3247 * include/ffi.h.in: Add definitions for closure and raw API.
3249 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3252 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3254 * src/raw_api.c: New file.
3256 * include/ffi.h.in (ffi_raw): New type.
3257 (UINT_ARG, SINT_ARG): New defines.
3258 (ffi_closure, ffi_raw_closure): New types.
3259 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3261 * configure.in: Add check for endianness and sizeof void*.
3263 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3264 instead of directly.
3266 * configure: Rebuilt.
3268 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3270 * configure.in: Add x86 and powerpc BeOS configurations.
3271 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3273 1999-05-09 Anthony Green <green@cygnus.com>
3275 * configure.in: Add warning about this being beta code.
3276 Remove src/Makefile.am from the picture.
3277 * configure: Rebuilt.
3279 * Makefile.am: Move logic from src/Makefile.am. Add changes
3280 to support libffi as a target library.
3281 * Makefile.in: Rebuilt.
3283 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3284 Upgraded to new autoconf, automake, libtool.
3288 * LICENSE: Update copyright date.
3290 * src/Makefile.am, src/Makefile.in: Removed.
3292 1998-11-29 Anthony Green <green@cygnus.com>
3294 * include/ChangeLog: Removed.
3295 * src/ChangeLog: Removed.
3296 * src/mips/ChangeLog: Removed.
3297 * src/sparc/ChangeLog: Remboved.
3298 * src/x86/ChangeLog: Removed.
3300 * ChangeLog.v1: Created.