1 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
4 * aclocal.m4: Regenerate.
5 * configure: Regenerate.
7 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
9 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
10 * src/sh64/sysv.S: Likewise.
12 2008-03-26 Daniel Jacobowitz <dan@debian.org>
14 * src/arm/sysv.S: Fix ARM comment marker.
16 2008-03-26 Jakub Jelinek <jakub@redhat.com>
18 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
19 * src/s390/sysv.S: Likewise.
20 * src/powerpc/ppc_closure.S: Likewise.
21 * src/powerpc/sysv.S: Likewise.
22 * src/x86/unix64.S: Likewise.
23 * src/x86/sysv.S: Likewise.
24 * src/sparc/v8.S: Likewise.
25 * src/sparc/v9.S: Likewise.
26 * src/m68k/sysv.S: Likewise.
27 * src/arm/sysv.S: Likewise.
29 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
31 * aclocal.m4: Regenerate.
32 * configure: Likewise.
33 * Makefile.in: Likewise.
34 * include/Makefile.in: Likewise.
35 * testsuite/Makefile.in: Likewise.
37 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
38 Andreas Tobler <a.tobler@schweiz.org>
40 * configure.ac: Add amd64-*-freebsd* target.
41 * configure: Regenerate.
43 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
46 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
49 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
52 2008-01-24 David Edelsohn <edelsohn@gnu.org>
54 * configure: Regenerate.
56 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
58 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
60 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
63 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
64 signed/unsigned int8/16 for X86_DARWIN.
65 Updated copyright info.
66 Handle one and two byte structs with special cif->flags.
67 * src/x86/ffitarget.h: Add special types for one and two byte structs.
68 Updated copyright info.
69 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
71 Remove code to pop args from the stack after call.
72 Special-case signed/unsigned for int8/16, one and two byte structs.
73 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
74 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
76 Updated copyright info.
78 2007-12-08 David Daney <ddaney@avtrex.com>
80 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
81 SUBU, add with ADDU and use smaller code sequences.
83 2007-12-07 David Daney <ddaney@avtrex.com>
85 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
88 2007-12-06 David Daney <ddaney@avtrex.com>
90 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
92 (ffi_java_raw): New typedef.
93 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
94 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
96 (ffi_java_raw_closure) : Same.
97 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
99 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
101 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
102 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
103 sizeof(ffi_java_raw) for alignment calculations.
104 (ffi_java_ptrarray_to_raw): Same.
105 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
106 if FFI_SIZEOF_JAVA_RAW == 4.
107 (ffi_java_raw_to_rvalue): Same.
108 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
109 (ffi_java_translate_args): Same.
110 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
112 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
114 2007-12-06 David Daney <ddaney@avtrex.com>
116 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
119 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
122 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
123 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
124 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
125 set the NUM_FPR_ARG_REGISTERS according to.
126 Add support for potential soft-float support under hard-float
128 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
129 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
130 to the FFI_LINUX_SOFT_FLOAT ABI.
131 (ffi_prep_cif_machdep): Likewise.
132 (ffi_closure_helper_SYSV): Likewise.
133 * src/powerpc/ppc_closure.S: Make sure not to store float/double
134 on archs where __NO_FPRS__ is true.
135 Add FFI_TYPE_UINT128 support.
136 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
137 Adjust copyright notice.
139 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
141 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
142 * include/ffi_common.h: ... here.
145 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
147 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
148 * src/powerpc/linux64.S: Likewise.
149 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
150 * testsuite/libffi.call/return_ldl.c: New test.
152 2007-09-04 <aph@redhat.com>
154 * src/arm/sysv.S (UNWIND): New.
155 (Whole file): Conditionally compile unwinder directives.
156 * src/arm/sysv.S: Add unwinder directives.
158 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
159 Only treat r0 as a struct address if we're actually returning a
161 Only copy the bytes that are actually within a struct.
162 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
163 is returned in r0, not passed by address.
164 (ffi_call): Allocate a word-sized temporary for the case where
165 a composite is returned in r0.
166 (ffi_prep_incoming_args_SYSV): Align as necessary.
168 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
170 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
171 directly using the sys_cacheflush syscall.
173 2007-07-27 Andrew Haley <aph@redhat.com>
175 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
177 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
179 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
180 * configure.ac: Likewise.
181 * Makefile.in: Regenerate.
182 * include/Makefile.in: Likewise.
183 * testsuite/Makefile.in: Likewise.
184 * configure: Likewise.
186 2007-08-24 David Daney <ddaney@avtrex.com>
188 * testsuite/libffi.call/return_sl.c: New test.
190 2007-08-10 David Daney <ddaney@avtrex.com>
192 * testsuite/libffi.call/cls_multi_ushort.c,
193 testsuite/libffi.call/cls_align_uint16.c,
194 testsuite/libffi.call/nested_struct1.c,
195 testsuite/libffi.call/nested_struct3.c,
196 testsuite/libffi.call/cls_7_1_byte.c,
197 testsuite/libffi.call/nested_struct5.c,
198 testsuite/libffi.call/cls_double.c,
199 testsuite/libffi.call/nested_struct7.c,
200 testsuite/libffi.call/cls_sint.c,
201 testsuite/libffi.call/nested_struct9.c,
202 testsuite/libffi.call/cls_20byte1.c,
203 testsuite/libffi.call/cls_multi_sshortchar.c,
204 testsuite/libffi.call/cls_align_sint64.c,
205 testsuite/libffi.call/cls_3byte2.c,
206 testsuite/libffi.call/cls_multi_schar.c,
207 testsuite/libffi.call/cls_multi_uchar.c,
208 testsuite/libffi.call/cls_19byte.c,
209 testsuite/libffi.call/cls_9byte1.c,
210 testsuite/libffi.call/cls_align_float.c,
211 testsuite/libffi.call/closure_fn1.c,
212 testsuite/libffi.call/problem1.c,
213 testsuite/libffi.call/closure_fn3.c,
214 testsuite/libffi.call/cls_sshort.c,
215 testsuite/libffi.call/closure_fn5.c,
216 testsuite/libffi.call/cls_align_double.c,
217 testsuite/libffi.call/nested_struct.c,
218 testsuite/libffi.call/cls_2byte.c,
219 testsuite/libffi.call/nested_struct10.c,
220 testsuite/libffi.call/cls_4byte.c,
221 testsuite/libffi.call/cls_6byte.c,
222 testsuite/libffi.call/cls_8byte.c,
223 testsuite/libffi.call/cls_multi_sshort.c,
224 testsuite/libffi.call/cls_align_sint16.c,
225 testsuite/libffi.call/cls_align_uint32.c,
226 testsuite/libffi.call/cls_20byte.c,
227 testsuite/libffi.call/cls_float.c,
228 testsuite/libffi.call/nested_struct2.c,
229 testsuite/libffi.call/cls_5_1_byte.c,
230 testsuite/libffi.call/nested_struct4.c,
231 testsuite/libffi.call/cls_24byte.c,
232 testsuite/libffi.call/nested_struct6.c,
233 testsuite/libffi.call/cls_64byte.c,
234 testsuite/libffi.call/nested_struct8.c,
235 testsuite/libffi.call/cls_uint.c,
236 testsuite/libffi.call/cls_multi_ushortchar.c,
237 testsuite/libffi.call/cls_schar.c,
238 testsuite/libffi.call/cls_uchar.c,
239 testsuite/libffi.call/cls_align_uint64.c,
240 testsuite/libffi.call/cls_ulonglong.c,
241 testsuite/libffi.call/cls_align_longdouble.c,
242 testsuite/libffi.call/cls_1_1byte.c,
243 testsuite/libffi.call/cls_12byte.c,
244 testsuite/libffi.call/cls_3_1byte.c,
245 testsuite/libffi.call/cls_3byte1.c,
246 testsuite/libffi.call/cls_4_1byte.c,
247 testsuite/libffi.call/cls_6_1_byte.c,
248 testsuite/libffi.call/cls_16byte.c,
249 testsuite/libffi.call/cls_18byte.c,
250 testsuite/libffi.call/closure_fn0.c,
251 testsuite/libffi.call/cls_9byte2.c,
252 testsuite/libffi.call/closure_fn2.c,
253 testsuite/libffi.call/closure_fn4.c,
254 testsuite/libffi.call/cls_ushort.c,
255 testsuite/libffi.call/closure_fn6.c,
256 testsuite/libffi.call/cls_5byte.c,
257 testsuite/libffi.call/cls_align_pointer.c,
258 testsuite/libffi.call/cls_7byte.c,
259 testsuite/libffi.call/cls_align_sint32.c,
260 testsuite/libffi.special/unwindtest_ffi_call.cc,
261 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
263 2007-08-10 David Daney <ddaney@avtrex.com>
266 * configure.ac: Don't treat mips64 as a special case.
267 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
268 * configure: Regenerate
269 * Makefile.in: Ditto.
270 * fficonfig.h.in: Ditto.
271 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
272 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
273 (FFI_DEFAULT_ABI): Set for n64 case.
274 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
275 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
276 (ffi_closure_N32): New function.
277 (.eh_frame): New section
278 * src/mips/o32.S: Clean up comments.
279 (ffi_closure_O32): Pass ffi_closure parameter in $12.
280 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
281 _MIPS_SIM == _ABIN32 throughout.
282 (FFI_MIPS_STOP_HERE): New, use in place of
284 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
285 to support n32/n64 ABIs.
286 (calc_n32_struct_flags): Rewrite.
287 (calc_n32_return_struct_flags): Remove unused variable. Reverse
288 position of flag bits.
289 (ffi_prep_cif_machdep): Rewrite n32 portion.
290 (ffi_call): Enable for n64. Add special handling for small structure
292 (ffi_prep_closure_loc): Add n32 and n64 support.
293 (ffi_closure_mips_inner_O32): Add cast to silence warning.
294 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
296 2007-08-08 David Daney <ddaney@avtrex.com>
298 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
299 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
301 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
302 * testsuite/libffi.call/cls_sint.c (main): Ditto.
303 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
304 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
305 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
306 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
307 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
308 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
309 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
310 * testsuite/libffi.call/return_fl3.c (main): Ditto.
311 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
312 * testsuite/libffi.call/nested_struct.c (main): Ditto.
313 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
314 * testsuite/libffi.call/return_ll1.c (main): Ditto.
315 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
316 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
317 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
318 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
319 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
320 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
321 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
322 * testsuite/libffi.call/cls_uint.c (main): Ditto.
323 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
324 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
325 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
326 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
327 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
328 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
329 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
330 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
331 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
333 2007-08-07 Andrew Haley <aph@redhat.com>
335 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
338 2007-08-06 Andrew Haley <aph@redhat.com>
341 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
342 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
345 2007-08-02 David Daney <ddaney@avtrex.com>
347 * testsuite/libffi.call/return_ul.c (main): Define return type as
348 ffi_arg. Use proper printf conversion specifier.
350 2007-07-30 Andrew Haley <aph@redhat.com>
353 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
354 signed/unsigned int8/16.
355 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
357 Remove code to pop args from the stack after call.
358 Special-case signed/unsigned int8/16.
359 * testsuite/libffi.call/return_sc.c (main): Revert.
361 2007-07-26 Richard Guenther <rguenther@suse.de>
364 * testsuite/libffi.call/return_sc.c (main): Verify call
365 result as signed char, not ffi_arg.
367 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
369 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
370 * configure: Regenerate.
372 2007-07-11 David Daney <ddaney@avtrex.com>
374 * src/mips/ffi.c: Don't include sys/cachectl.h.
375 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
378 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
380 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
381 from (ffi_prep_closure): ... this.
382 (FFI_INIT_TRAMPOLINE): Adjust.
384 2005-12-31 Phil Blundell <pb@reciva.com>
386 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
387 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
388 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
389 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
390 (FFI_CLOSURES): Enable closure support.
392 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
394 * testsuite/libffi.call/cls_multi_ushort.c,
395 testsuite/libffi.call/cls_align_uint16.c,
396 testsuite/libffi.call/nested_struct1.c,
397 testsuite/libffi.call/nested_struct3.c,
398 testsuite/libffi.call/cls_7_1_byte.c,
399 testsuite/libffi.call/cls_double.c,
400 testsuite/libffi.call/nested_struct5.c,
401 testsuite/libffi.call/nested_struct7.c,
402 testsuite/libffi.call/cls_sint.c,
403 testsuite/libffi.call/nested_struct9.c,
404 testsuite/libffi.call/cls_20byte1.c,
405 testsuite/libffi.call/cls_multi_sshortchar.c,
406 testsuite/libffi.call/cls_align_sint64.c,
407 testsuite/libffi.call/cls_3byte2.c,
408 testsuite/libffi.call/cls_multi_schar.c,
409 testsuite/libffi.call/cls_multi_uchar.c,
410 testsuite/libffi.call/cls_19byte.c,
411 testsuite/libffi.call/cls_9byte1.c,
412 testsuite/libffi.call/cls_align_float.c,
413 testsuite/libffi.call/closure_fn1.c,
414 testsuite/libffi.call/problem1.c,
415 testsuite/libffi.call/closure_fn3.c,
416 testsuite/libffi.call/cls_sshort.c,
417 testsuite/libffi.call/closure_fn5.c,
418 testsuite/libffi.call/cls_align_double.c,
419 testsuite/libffi.call/cls_2byte.c,
420 testsuite/libffi.call/nested_struct.c,
421 testsuite/libffi.call/nested_struct10.c,
422 testsuite/libffi.call/cls_4byte.c,
423 testsuite/libffi.call/cls_6byte.c,
424 testsuite/libffi.call/cls_8byte.c,
425 testsuite/libffi.call/cls_multi_sshort.c,
426 testsuite/libffi.call/cls_align_uint32.c,
427 testsuite/libffi.call/cls_align_sint16.c,
428 testsuite/libffi.call/cls_float.c,
429 testsuite/libffi.call/cls_20byte.c,
430 testsuite/libffi.call/cls_5_1_byte.c,
431 testsuite/libffi.call/nested_struct2.c,
432 testsuite/libffi.call/cls_24byte.c,
433 testsuite/libffi.call/nested_struct4.c,
434 testsuite/libffi.call/nested_struct6.c,
435 testsuite/libffi.call/cls_64byte.c,
436 testsuite/libffi.call/nested_struct8.c,
437 testsuite/libffi.call/cls_uint.c,
438 testsuite/libffi.call/cls_multi_ushortchar.c,
439 testsuite/libffi.call/cls_schar.c,
440 testsuite/libffi.call/cls_uchar.c,
441 testsuite/libffi.call/cls_align_uint64.c,
442 testsuite/libffi.call/cls_ulonglong.c,
443 testsuite/libffi.call/cls_align_longdouble.c,
444 testsuite/libffi.call/cls_1_1byte.c,
445 testsuite/libffi.call/cls_12byte.c,
446 testsuite/libffi.call/cls_3_1byte.c,
447 testsuite/libffi.call/cls_3byte1.c,
448 testsuite/libffi.call/cls_4_1byte.c,
449 testsuite/libffi.call/cls_6_1_byte.c,
450 testsuite/libffi.call/cls_16byte.c,
451 testsuite/libffi.call/cls_18byte.c,
452 testsuite/libffi.call/closure_fn0.c,
453 testsuite/libffi.call/cls_9byte2.c,
454 testsuite/libffi.call/closure_fn2.c,
455 testsuite/libffi.call/closure_fn4.c,
456 testsuite/libffi.call/cls_ushort.c,
457 testsuite/libffi.call/closure_fn6.c,
458 testsuite/libffi.call/cls_5byte.c,
459 testsuite/libffi.call/cls_align_pointer.c,
460 testsuite/libffi.call/cls_7byte.c,
461 testsuite/libffi.call/cls_align_sint32.c,
462 testsuite/libffi.special/unwindtest_ffi_call.cc,
463 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
465 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
467 * aclocal.m4: Regenerated.
469 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
471 * configure: Regenerate.
473 2007-05-23 Steve Ellcey <sje@cup.hp.com>
475 * Makefile.in: Regenerate.
476 * configure: Regenerate.
477 * aclocal.m4: Regenerate.
478 * include/Makefile.in: Regenerate.
479 * testsuite/Makefile.in: Regenerate.
481 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
483 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
484 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
485 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
486 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
487 (FFI_CLOSURES): Enable closure support.
489 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
491 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
492 * configure: Regenerate.
493 * fficonfig.h.in: Regenerate.
494 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
495 CFI_OFFSET,CFI_DEF_CFA): New macros.
496 (ffi_call_SYSV): Add callframe annotation.
498 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
500 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
501 numerous test suite failures.
502 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
504 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
506 * Makefile.am (EXTRA_DIST): Bring up to date.
507 * Makefile.in: Regenerate.
508 * src/frv/eabi.S: Remove RCS keyword.
510 2007-04-06 Richard Henderson <rth@redhat.com>
512 * configure.ac: Tidy target case.
513 (HAVE_LONG_DOUBLE): Allow the target to override.
514 * configure: Regenerate.
515 * include/ffi.h.in: Don't define ffi_type_foo if
516 LIBFFI_HIDE_BASIC_TYPES is defined.
517 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
519 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
520 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
521 (ffi_type_longdouble): Special case for Alpha. Don't define
522 if long double == double.
524 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
525 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
526 (ffi_call, ffi_closure_osf_inner): Likewise.
527 (ffi_closure_osf_inner): Likewise. Mark hidden.
528 (ffi_call_osf, ffi_closure_osf): Mark hidden.
529 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
530 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
531 (load_table): Handle 128-bit long double.
533 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
535 2007-04-06 Tom Tromey <tromey@redhat.com>
538 * README: Fixed bug in example.
540 2007-04-03 Jakub Jelinek <jakub@redhat.com>
542 * src/closures.c: Include sys/statfs.h.
543 (_GNU_SOURCE): Define on Linux.
544 (FFI_MMAP_EXEC_SELINUX): Define.
545 (selinux_enabled): New variable.
546 (selinux_enabled_check): New function.
547 (is_selinux_enabled): Define.
550 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
552 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
553 Use 'volatile float sum' to create sum of floats to avoid false
554 negative due to excess precision on ix86 targets.
557 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
559 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
561 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
563 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
565 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
566 (ffi_prep_closure_loc): New.
567 (ffi_prep_raw_closure_loc): New.
568 (ffi_prep_java_raw_closure_loc): New.
569 * src/closures.c: New file.
570 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
571 Replace sflags with exec_offset.
572 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
573 sub_segment_exec_offset): New macros.
574 (get_segment_flags, set_segment_flags, check_segment_merge): New
576 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
577 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
578 destroy_mspace): Use new macros.
579 (sys_alloc): Silence warning.
580 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
581 * Makefile.in: Rebuilt.
582 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
583 terms of ffi_prep_closure_loc.
584 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
586 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
588 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
590 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
592 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
593 (ffi_prep_closure): ... this.
594 * src/pa/ffi.c: Likewise.
595 * src/cris/ffi.c: Likewise. Adjust.
596 * src/frv/ffi.c: Likewise.
597 * src/ia64/ffi.c: Likewise.
598 * src/mips/ffi.c: Likewise.
599 * src/powerpc/ffi_darwin.c: Likewise.
600 * src/s390/ffi.c: Likewise.
601 * src/sh/ffi.c: Likewise.
602 * src/sh64/ffi.c: Likewise.
603 * src/sparc/ffi.c: Likewise.
604 * src/x86/ffi64.c: Likewise.
605 * src/x86/ffi.c: Likewise.
606 (FFI_INIT_TRAMPOLINE): Adjust.
607 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
608 (ffi_prep_raw_closure): ... this.
609 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
610 (ffi_prep_closure): ... this.
611 (flush_icache): Adjust.
613 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
615 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
618 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
620 * Makefile.am: Add dummy install-pdf target.
621 * Makefile.in: Regenerate
623 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
625 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
626 ffi_closure_helper_SYSV): Add long double handling.
628 2007-02-02 Jakub Jelinek <jakub@redhat.com>
630 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
631 immediately after bctrl instruction.
633 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
635 * Makefile.am (all-recursive, install-recursive,
636 mostlyclean-recursive, clean-recursive, distclean-recursive,
637 maintainer-clean-recursive): Add missing targets.
638 * Makefile.in: Rebuilt.
640 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
642 * configure.ac: Add TARGET for x86_64-*-darwin*.
643 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
645 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
646 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
647 * configure: Regenerate.
648 * Makefile.in: Regenerate.
649 * include/Makefile.in: Regenerate.
650 * testsuite/Makefile.in: Regenerate.
651 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
654 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
656 * aclocal.m4: Regenerate with aclocal -I .. as written in the
659 2006-10-31 Geoffrey Keating <geoffk@apple.com>
661 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
662 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
664 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
665 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
667 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
668 Sandro Tolaini <tolaini@libero.it>
670 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
672 * configure: Regenerated.
673 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
674 (EXTRA_DIST): Add src/x86/darwin.S.
675 * Makefile.in: Regenerated.
676 * include/Makefile.in: Regenerated.
677 * testsuite/Makefile.in: Regenerated.
679 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
680 X86_WIN32, and additionally align stack to 16 bytes.
681 * src/x86/darwin.S: New, based on sysv.S.
682 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
684 2006-09-12 David Daney <ddaney@avtrex.com>
687 * include/Makefile.am: Install both ffi.h and ffitarget.h in
688 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
689 * aclocal.m4: Regenerated for automake 1.9.6.
690 * Makefile.in: Regenerated.
691 * include/Makefile.in: Regenerated.
692 * testsuite/Makefile.in: Regenerated.
694 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
696 * include/ffi_common.h (struct): Revert accidental commit.
698 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
700 * include/ffi_common.h: Remove lint directives.
701 * include/ffi.h.in: Likewise.
703 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
705 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
706 for 32-bit architectures.
707 * testsuite/libffi.call/return_ul.c: New test case.
709 2006-07-19 David Daney <ddaney@avtrex.com>
711 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
712 xfail remains for mips64.
714 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
716 * Makefile.am: Add install-html target. Add install-html to .PHONY
717 * Makefile.in: Regenerate.
718 * aclocal.m4: Regenerate.
719 * include/Makefile.in: Regenerate.
720 * testsuite/Makefile.in: Regenerate.
722 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
724 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
727 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
729 * README: Remove notice about 'Crazy Comments'.
730 * src/debug.c: Remove lint directives. Cleanup white spaces.
731 * src/java_raw_api.c: Likewise.
732 * src/prep_cif.c: Likewise.
733 * src/raw_api.c: Likewise.
734 * src/ffitest.c: Delete. No longer needed, all test cases migrated
736 * src/arm/ffi.c: Remove lint directives.
737 * src/m32r/ffi.c: Likewise.
738 * src/pa/ffi.c: Likewise.
739 * src/powerpc/ffi.c: Likewise.
740 * src/powerpc/ffi_darwin.c: Likewise.
741 * src/sh/ffi.c: Likewise.
742 * src/sh64/ffi.c: Likewise.
743 * src/x86/ffi.c: Likewise.
744 * testsuite/libffi.call/float2.c: Likewise.
745 * testsuite/libffi.call/promotion.c: Likewise.
746 * testsuite/libffi.call/struct1.c: Likewise.
748 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
750 * src/pa/hpux32.S: Correct unwind offset calculation for
752 * src/pa/linux.S: Likewise.
754 2006-04-12 James E Wilson <wilson@specifix.com>
757 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
758 (hfa_type_load): Call stf_spill.
759 (hfa_type_store): Call ldf_fill.
760 (ffi_call): Adjust calls to above routines. Add local temps for
763 2006-04-10 Matthias Klose <doko@debian.org>
765 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
766 directory names containing underscores.
768 2006-04-07 James E Wilson <wilson@specifix.com>
770 * testsuite/libffi.call/float4.c: New testcase.
772 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
773 Andreas Tobler <a.tobler@schweiz.ch>
775 * Makefile.am: Add PA_HPUX port.
776 * Makefile.in: Regenerate.
777 * include/Makefile.in: Likewise.
778 * testsuite/Makefile.in: Likewise.
779 * configure.ac: Add PA_HPUX rules.
780 * configure: Regenerate.
781 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
782 Add PA_HPUX and PA64_HPUX.
783 Rename FFI_LINUX ABI to FFI_PA32 ABI.
784 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
785 (FFI_TYPE_SMALL_STRUCT2): Define.
786 (FFI_TYPE_SMALL_STRUCT4): Likewise.
787 (FFI_TYPE_SMALL_STRUCT8): Likewise.
788 (FFI_TYPE_SMALL_STRUCT3): Redefine.
789 (FFI_TYPE_SMALL_STRUCT5): Likewise.
790 (FFI_TYPE_SMALL_STRUCT6): Likewise.
791 (FFI_TYPE_SMALL_STRUCT7): Likewise.
792 * src/pa/ffi.c (ROUND_DOWN): Delete.
793 (fldw, fstw, fldd, fstd): Use '__asm__'.
794 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
795 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
796 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
797 Simplify incrementing of stack slot variable. Change type of local
799 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
801 (ffi_prep_cif_machdep): Likewise.
802 (ffi_call): Likewise.
803 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
804 return type to ffi_status. Simplify incrementing of stack slot
805 variable. Only copy floating point argument registers when PA_LINUX
806 is true. Reformat debug statement.
807 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
808 FFI_TYPE_SMALL_STRUCT8.
809 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
811 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
812 Add nops to cache flush. Add trampoline for PA_HPUX.
813 * src/pa/hpux32.S: New file.
814 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
815 ffi_prep_args_LINUX to ffi_prep_args_pa32.
816 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
817 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
818 argument type checks so that common argument types appear first.
819 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
820 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
822 2006-03-24 Alan Modra <amodra@bigpond.net.au>
824 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
825 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
826 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
828 (ffi_prep_args64): Assert using IBM extended double.
829 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
830 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
831 (ffi_call): Handle FFI_LINUX.
832 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
833 gpr3 return pointer as for struct return. Handle FFI_LINUX
834 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
836 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
837 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
838 Don't use r6 as pointer to results, instead use sp offset. Don't
839 make a special call to load lr with case table address, instead
840 use offset from previous call.
841 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
842 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
845 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
847 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
848 passed with FP registers correctly.
849 (ffi_closure_helper_SYSV): Likewise.
850 * src/sh64/sysv.S: Likewise.
852 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
854 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
855 args and userdata unused.
856 (closure_test_fn1): Mark cif and userdata unused.
857 (main): Remove unused res.
859 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
861 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
862 -O2, -O3, -Os and the warning flags -W -Wall.
863 * testsuite/libffi.special/special.exp: Likewise.
864 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
865 unused parameter unused for gcc or else do nothing.
866 * testsuite/libffi.special/ffitestcxx.h: Likewise.
867 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
869 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
870 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
871 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
872 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
873 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
874 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
875 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
876 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
877 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
878 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
879 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
880 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
881 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
882 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
883 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
884 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
885 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
886 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
887 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
888 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
889 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
890 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
891 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
892 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
894 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
896 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
898 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
899 void* to avoid compiler warning.
901 (cls_struct_align_gn): Mark cif and userdata unused.
902 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
904 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
906 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
908 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
910 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
912 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
913 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
914 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
916 (main): Cast res_call to silence gcc.
917 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
919 (main): Cast res_call to silence gcc.
920 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
922 (main): Cast res_call to silence gcc.
923 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
925 (main): Cast res_call to silence gcc.
926 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
928 (main): Cast res_call to silence gcc.
929 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
931 (main): Cast res_call to silence gcc.
932 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
934 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
935 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
937 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
938 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
940 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
941 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
943 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
944 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
946 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
947 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
949 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
951 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
952 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
953 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
954 Cleanup white spaces.
955 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
956 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
957 cif and userdata unused.
958 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
960 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
961 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
962 formatters to silence gcc.
963 (B_gn): Mark cif and userdata unused.
964 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
966 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
967 (B_gn): Mark cif and userdata unused.
968 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
970 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
971 (B_gn): Mark cif and userdata unused.
972 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
974 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
975 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
976 * testsuite/libffi.call/problem1.c (stub): Likewise.
977 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
979 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
980 in the last commit for this test case in the test case itself.
981 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
983 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
984 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
985 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
986 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
987 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
988 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
990 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
992 * src/sh/sysv.S: Fix register numbers in the FDE for
995 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
997 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
998 declaration to avoid a false negative on ix86. See PR323.
1000 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1002 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1003 and cast integer to void * if needed. Update the pointer to
1004 the FP register saved area correctly.
1006 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1008 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1010 * testsuite/libffi.call/nested_struct4.c: Likewise.
1012 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1014 * testsuite/libffi.call/return_dbl.c: New test case.
1015 * testsuite/libffi.call/return_dbl1.c: Likewise.
1016 * testsuite/libffi.call/return_dbl2.c: Likewise.
1017 * testsuite/libffi.call/return_fl.c: Likewise.
1018 * testsuite/libffi.call/return_fl1.c: Likewise.
1019 * testsuite/libffi.call/return_fl2.c: Likewise.
1020 * testsuite/libffi.call/return_fl3.c: Likewise.
1021 * testsuite/libffi.call/closure_fn6.c: Likewise.
1023 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1025 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1026 here to be used by other test cases too.
1028 * testsuite/libffi.call/nested_struct10.c: New test case.
1029 * testsuite/libffi.call/nested_struct9.c: Likewise.
1030 * testsuite/libffi.call/nested_struct8.c: Likewise.
1031 * testsuite/libffi.call/nested_struct7.c: Likewise.
1032 * testsuite/libffi.call/nested_struct6.c: Likewise.
1033 * testsuite/libffi.call/nested_struct5.c: Likewise.
1034 * testsuite/libffi.call/nested_struct4.c: Likewise.
1036 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1038 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1039 * configure: Rebuilt.
1041 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1043 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1044 instead do the shifting inline.
1045 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1046 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1047 and 8 byte structs, for the remaining struct sizes don't call
1048 __lshrdi3, instead do the shifting inline.
1050 2005-12-07 Thiemo Seufer <ths@networkno.de>
1052 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1053 missing parentheses.
1054 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1055 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1056 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1057 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1058 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1060 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1062 (ffi_prep_closure): Improve trampoline instruction scheduling.
1063 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1065 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1067 * src/powerpc/ffi.c: Formatting.
1068 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1069 (ffi_prep_args64): Likewise.
1071 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1073 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1074 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1076 2005-09-26 Tom Tromey <tromey@redhat.com>
1078 * testsuite/libffi.call/float1.c (value_type): New typedef.
1079 (CANARY): New define.
1080 (main): Check for result buffer overflow.
1081 * src/powerpc/linux64.S: Handle linux64 long double returns.
1082 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1083 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1085 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1088 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1090 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1092 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1094 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1095 (AH_BOTTOM): Add FFI_HIDDEN definition.
1096 * configure: Rebuilt.
1097 * fficonfig.h.in: Rebuilt.
1098 * src/powerpc/ffi.c (hidden): Remove.
1099 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1100 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1101 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1102 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1103 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1104 add FFI_HIDDEN to its prototype.
1105 (ffi_closure_SYSV_inner): New.
1106 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1107 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1109 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1112 * configure: Rebuilt.
1113 * configure.ac: Handle i*86-*-gnu*.
1115 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1117 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1118 DW_CFA_offset_extended_sf rather than
1119 DW_CFA_GNU_negative_offset_extended.
1120 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1122 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1124 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1126 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1127 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1128 partially on register.
1129 (ffi_closure_helper_SYSV): Likewise.
1130 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1132 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1134 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1136 * src/sh64/ffi.c (simple_type): Remove.
1137 (return_type): Handle small structures correctly.
1138 (ffi_prep_args): Likewise.
1139 (ffi_call): Likewise.
1140 (ffi_closure_helper_SYSV): Likewise.
1141 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1142 Emit position independent code if PIC and remove wrong datalabel
1143 prefixes from EH data.
1145 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1147 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1148 * Makefile.in: Regenerate.
1149 * include/Makefile.in: Likewise.
1150 * testsuite/Makefile.in: Likewise.
1151 * configure.ac: Add POWERPC_FREEBSD rules.
1152 * configure: Regenerate.
1153 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1154 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1155 * src/powerpc/ffi.c: Add flags to handle small structure returns
1157 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1159 (ffi_closure_helper_SYSV): Likewise.
1160 * src/powerpc/ppc_closure.S: Add return types for small structures.
1161 * src/powerpc/sysv.S: Add bits to handle small structures for
1164 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1166 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1167 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1168 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1170 2005-07-05 Randolph Chung <tausq@debian.org>
1172 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1173 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1174 structures. Kill compilation warnings.
1175 (ffi_closure_inner_LINUX): Print return values as hex in debug
1176 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1177 Properly handle 5-7 byte structure returns.
1178 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1179 (FFI_TYPE_SMALL_STRUCT2): Remove.
1180 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1181 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1182 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1183 (checksmst1, checksmst2): Remove.
1184 (checksmst3): Optimize handling of 3-byte struct returns.
1185 (checksmst567): Properly handle 5-7 byte struct returns.
1187 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1190 * src/mips/n32.S: Enforce PIC code.
1191 * src/mips/o32.S: Likewise.
1193 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1195 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1196 * configure: Regenerate.
1198 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1200 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1201 to call ffi_closure_helper_SYSV. Append @local instead.
1202 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1204 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1206 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1207 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1208 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1209 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1210 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1212 2005-05-09 Mike Stump <mrs@apple.com>
1214 * configure: Regenerate.
1216 2005-05-08 Richard Henderson <rth@redhat.com>
1219 * src/alpha/osf.S: Update unwind into to match code.
1221 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1222 Richard Henderson <rth@redhat.com>
1224 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1226 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1227 (ffi_prep_closure): Set carry bit if sse-used flag set.
1228 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1229 Only load sse registers if ssecount non-zero.
1230 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1232 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1234 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1235 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1236 * configure: Regenerate.
1238 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1240 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1241 have Tcl8.3-compatible intermediate variable.
1243 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1244 Hans-Peter Nilsson <hp@axis.com>
1246 * Makefile.am: Add CRIS support.
1247 * configure.ac: Likewise.
1248 * Makefile.in, configure, testsuite/Makefile.in,
1249 include/Makefile.in: Regenerate.
1250 * src/cris: New directory.
1251 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1252 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1254 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1255 \r?\n in output tests.
1257 2005-04-12 Mike Stump <mrs@apple.com>
1259 * configure: Regenerate.
1261 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1263 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1265 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1267 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1269 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1270 to 64 bits in ILP32 mode.
1271 Load 64 bits even for short data.
1273 2005-03-23 Mike Stump <mrs@apple.com>
1275 * src/powerpc/darwin.S: Update for -m64 multilib.
1276 * src/powerpc/darwin_closure.S: Likewise.
1278 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1280 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1281 Do not set tool_include_dir.
1282 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1284 * include/Makefile.am: Set gcc_version and toollibffidir.
1285 * include/Makefile.in: Regenerate.
1287 2005-02-22 Andrew Haley <aph@redhat.com>
1289 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1290 odd-numbered register pairs for 64-bit integer types.
1292 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1295 * testsuite/libffi.call/return_ll1.c: New test case.
1297 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1299 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1300 * testsuite/libffi.call/float.c: Ditto.
1301 * testsuite/libffi.call/float2.c: Ditto.
1302 * testsuite/libffi.call/float3.c: Ditto.
1304 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1306 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1308 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1310 * testsuite/libffi.special/special.exp (cxx_options): Add
1313 2004-12-31 Richard Henderson <rth@redhat.com>
1315 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1316 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1317 offset parameters with a type parameter; deduce size and structure
1318 alignment. Update all users.
1320 2004-12-31 Richard Henderson <rth@redhat.com>
1322 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1323 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1324 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1325 into ffi_prep_closure.
1326 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1329 2004-12-27 Richard Henderson <rth@redhat.com>
1331 * src/x86/unix64.S: Fix typo in unwind info.
1333 2004-12-25 Richard Henderson <rth@redhat.com>
1335 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1336 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1337 (merge_classes): Check for it.
1339 (classify_argument): Pass byte_offset by value; perform all updates
1341 (examine_argument): Add classes argument; handle
1342 X86_64_COMPLEX_X87_CLASS.
1343 (ffi_prep_args): Merge into ...
1344 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1345 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1346 (ffi_fill_return_value): Remove.
1347 (ffi_prep_closure): Remove dead assert.
1348 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1349 Rewrite to use struct register_args instead of va_list. Create
1350 flags for handling structure returns.
1351 * src/x86/unix64.S: Remove dead strings.
1352 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1353 stack frame with ffi_call. Handle structure returns properly.
1354 (float2sse, floatfloat2sse, double2sse): Remove.
1355 (sse2float, sse2double, sse2floatfloat): Remove.
1356 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1357 to handle structure returns properly.
1359 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1361 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1363 * Makefile.in: Regenerated.
1365 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1367 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1368 * configure, aclocal.m4, Makefile.in: Regenerate.
1369 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1371 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1373 * configure: Regenerate for libtool change.
1375 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1377 * configure: Regenerate for libtool reversion.
1379 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1381 * configure: Regenerate for libtool change.
1383 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1385 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1387 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1389 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1390 of jal. Use an absolute encoding for the frame information.
1392 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1394 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1395 * acinclude.m4: Delete logic for sincludes.
1396 * aclocal.m4, Makefile.in, configure: Regenerate.
1397 * include/Makefile: Likewise.
1398 * testsuite/Makefile: Likewise.
1400 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1402 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1403 on a 8-byte boundary.
1404 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1406 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1408 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1409 long long values. Round stack allocation to a multiple of 8 bytes
1410 for ATPCS compatibility.
1411 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1412 names. Handle returning long long types. Add Thumb and interworking
1413 support. Improve soft-float code.
1415 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1417 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1418 (libffi_exit): New function.
1419 (libffi_init): Build the testglue wrapper if needed.
1421 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1424 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1426 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1428 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1430 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1432 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1433 * testsuite/libffi.call/float3.c: New test case.
1435 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1437 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1438 the function returning a structure pointed with R2.
1439 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1440 the structure return value if T bit set. Emit position
1441 independent code and EH data if PIC.
1443 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1445 * Makefile.am: Add m32r support.
1446 * configure.ac: Likewise.
1447 * Makefile.in: Regenerate.
1448 * confiugre: Regenerate.
1449 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1450 (uint64, sint64, double, longdouble)
1451 * src/m32r: New directory.
1452 * src/m32r/ffi.c: New file.
1453 * src/m32r/sysv.S: Likewise.
1454 * src/m32r/ffitarget.h: Likewise.
1456 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1458 * testsuite/libffi.call/negint.c: New test case.
1460 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1463 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1464 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1465 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1468 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1470 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1471 * testsuite/libffi.call/promotion.c: Likewise.
1472 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1474 * testsuite/libffi.call/return_sc.c: Likewise.
1475 * testsuite/libffi.call/return_uc.c: Likewise.
1477 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1479 * src/powerpc/darwin.S: Fix comments and identation.
1480 * src/powerpc/darwin_closure.S: Likewise.
1482 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1484 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1485 (ffi_prep_args): Handle longdouble arguments.
1486 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1488 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1489 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1491 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1492 * src/types.c: Defined longdouble size and alignment for darwin.
1494 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1496 * src/powerpc/aix.S: Remove whitespaces.
1497 * src/powerpc/aix_closure.S: Likewise.
1498 * src/powerpc/asm.h: Likewise.
1499 * src/powerpc/ffi.c: Likewise.
1500 * src/powerpc/ffitarget.h: Likewise.
1501 * src/powerpc/linux64.S: Likewise.
1502 * src/powerpc/linux64_closure.S: Likewise.
1503 * src/powerpc/ppc_closure.S: Likewise.
1504 * src/powerpc/sysv.S: Likewise.
1506 2004-08-30 Anthony Green <green@redhat.com>
1508 * Makefile.am: Add frv support.
1509 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1510 * configure.ac: Read configure.host.
1511 * configure.in: Read configure.host.
1512 * configure.host: New file. frv-elf needs libgloss.
1513 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1514 alignment. This is needed to frv and shouldn't harm the others.
1515 * include/ffi_common.h (ALIGN_DOWN): New macro.
1516 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1518 2004-08-24 David Daney <daney@avtrex.com>
1520 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1521 * testsuite/libffi.call/closure_fn1.c: Likewise.
1522 * testsuite/libffi.call/closure_fn2.c Likewise.
1523 * testsuite/libffi.call/closure_fn3.c: Likewise.
1524 * testsuite/libffi.call/closure_fn4.c: Likewise.
1525 * testsuite/libffi.call/closure_fn5.c: Likewise.
1526 * testsuite/libffi.call/cls_18byte.c: Likewise.
1527 * testsuite/libffi.call/cls_19byte.c: Likewise.
1528 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1529 * testsuite/libffi.call/cls_20byte.c: Likewise.
1530 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1531 * testsuite/libffi.call/cls_24byte.c: Likewise.
1532 * testsuite/libffi.call/cls_2byte.c: Likewise.
1533 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1534 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1535 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1536 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1537 * testsuite/libffi.call/cls_4byte.c: Likewise.
1538 * testsuite/libffi.call/cls_64byte.c: Likewise.
1539 * testsuite/libffi.call/cls_6byte.c: Likewise.
1540 * testsuite/libffi.call/cls_7byte.c: Likewise.
1541 * testsuite/libffi.call/cls_8byte.c: Likewise.
1542 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1543 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1544 * testsuite/libffi.call/cls_align_double.c: Likewise.
1545 * testsuite/libffi.call/cls_align_float.c: Likewise.
1546 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1547 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1548 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1549 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1550 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1551 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1552 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1553 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1554 * testsuite/libffi.call/cls_double.c: Likewise.
1555 * testsuite/libffi.call/cls_float.c: Likewise.
1556 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1557 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1558 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1559 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1560 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1561 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1562 * testsuite/libffi.call/cls_schar.c: Likewise.
1563 * testsuite/libffi.call/cls_sint.c: Likewise.
1564 * testsuite/libffi.call/cls_sshort.c: Likewise.
1565 * testsuite/libffi.call/cls_uchar.c: Likewise.
1566 * testsuite/libffi.call/cls_uint.c: Likewise.
1567 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1568 * testsuite/libffi.call/cls_ushort.c: Likewise.
1569 * testsuite/libffi.call/nested_struct.c: Likewise.
1570 * testsuite/libffi.call/nested_struct1.c: Likewise.
1571 * testsuite/libffi.call/nested_struct2.c: Likewise.
1572 * testsuite/libffi.call/nested_struct3.c: Likewise.
1573 * testsuite/libffi.call/problem1.c: Likewise.
1574 * testsuite/libffi.special/unwindtest.cc: Likewise.
1575 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1577 * testsuite/libffi.call/cls_16byte.c: Likewise.
1578 * testsuite/libffi.call/cls_5byte.c: Likewise.
1580 2004-08-23 David Daney <daney@avtrex.com>
1583 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1584 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1585 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1586 parameters and return types.
1587 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1588 (ffi_prep_closure): Ditto.
1589 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1590 alignment calculations.
1591 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1592 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1594 2004-08-14 Casey Marshall <csm@gnu.org>
1596 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1597 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1598 integer (O32 ABI only).
1599 (ffi_prep_closure): new function.
1600 (ffi_closure_mips_inner_O32): new function.
1601 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1602 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1603 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1604 64 bit integers correctly.
1605 (ffi_closure_O32): new function.
1606 Added DWARF-2 unwind info for both functions.
1608 2004-08-10 Andrew Haley <aph@redhat.com>
1610 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1612 2004-08-01 Robert Millan <robertmh@gnu.org>
1614 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1615 * configure: Regenerate.
1617 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1619 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1620 and mmap() explicitly instead of relying on preset autoconf cache
1622 * aclocal.m4: Regenerate.
1623 * configure: Regenerate.
1625 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1627 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1628 (ffi_check_float_struct): Remove unused prototype.
1630 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1632 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1633 character on Darwin, use '\n\t' instead.
1635 2004-06-26 Matthias Klose <doko@debian.org>
1637 * libtool-version: Fix typo in revision/age.
1639 2004-06-17 Matthias Klose <doko@debian.org>
1641 * libtool-version: New.
1642 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1643 * Makefile.in: Regenerate.
1645 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1647 * Makefile.am: Remove useless multilib rules.
1648 * Makefile.in: Regenerate.
1649 * aclocal.m4: Regenerate with automake 1.8.5.
1650 * configure.ac: Remove useless multilib configury.
1651 * configure: Regenerate.
1653 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1655 * .cvsignore: New file.
1657 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1659 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1661 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1664 2004-06-06 Sean McNeil <sean@mcneil.com>
1666 * configure.ac: Add x86_64-*-freebsd* support.
1667 * configure: Regenerate.
1669 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1672 * configure.ac: Test for existence of mmap and sys/mman.h before
1673 checking blacklist. Fix suggested by Jim Wilson.
1674 * configure: Regenerate.
1676 2004-04-26 Matt Austern <austern@apple.com>
1678 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1680 * src/powerpc/darwin_closure.S: Likewise.
1682 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1684 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1685 error. Reported by Thomas Heller <theller@python.net>.
1686 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1687 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1689 2004-03-20 Matthias Klose <doko@debian.org>
1691 * src/pa/linux.S: Fix typo.
1693 2004-03-19 Matthias Klose <doko@debian.org>
1695 * Makefile.am: Update.
1696 * Makefile.in: Regenerate.
1697 * src/pa/ffi.h.in: Remove.
1698 * src/pa/ffitarget.h: New file.
1700 2004-02-10 Randolph Chung <tausq@debian.org>
1702 * Makefile.am: Add PA support.
1703 * Makefile.in: Regenerate.
1704 * include/Makefile.in: Regenerate.
1705 * configure.ac: Add PA target.
1706 * configure: Regenerate.
1707 * src/pa/ffi.c: New file.
1708 * src/pa/ffi.h.in: Add PA support.
1709 * src/pa/linux.S: New file.
1710 * prep_cif.c: Add PA support.
1712 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1714 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1716 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1717 with ecif->cif->flags.
1718 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1720 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1721 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1722 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1723 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1725 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1727 * configure.in: Rename file to ...
1728 * configure.ac: ... this.
1729 * fficonfig.h.in: Regenerate.
1730 * Makefile.in: Regenerate.
1731 * include/Makefile.in: Regenerate.
1732 * testsuite/Makefile.in: Regenerate.
1734 2004-03-12 Matt Austern <austern@apple.com>
1736 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1737 changes in EH format resulting from addition of linkonce support.
1738 * src/powerpc/darwin_closure.S: Likewise.
1740 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1741 Paolo Bonzini <bonzini@gnu.org>
1743 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1744 Remove VPATH. Remove rules for object files. Remove multilib support.
1745 (AM_CCASFLAGS): Add.
1746 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1747 (AC_PREREQ): Bump version to 2.59.
1748 (AC_INIT): Fill with version info and bug address.
1749 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1750 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1751 De-precious CC so that the right flags are passed down to multilibs.
1752 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1753 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1754 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1755 * configure: Rebuilt.
1756 * aclocal.m4: Likewise.
1757 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1758 * fficonfig.h.in: Likewise.
1760 2004-03-11 Andreas Schwab <schwab@suse.de>
1762 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1763 arguments from fp registers only for the first 8 parameter slots.
1764 Don't convert a float parameter when passed in memory.
1766 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1768 * configure: Regenerate for config/accross.m4 correction.
1770 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1772 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1773 ecif->cif->bytes to bytes.
1774 (ffi_prep_cif_machdep): Add braces around nested if statement.
1776 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1778 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1780 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1781 (ffi_closure_helper_LINUX64): Fix typo.
1782 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1784 * testsuite/libffi.call/float.c: Likewise.
1785 * testsuite/libffi.call/float2.c: Likewise.
1787 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1789 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1790 long double function return and long double arg handling.
1791 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1792 Use "end_pfr" instead of "nf". Correct long double handling.
1794 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1796 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1797 space for long double return value. Adjust stack frame and offsets.
1798 Load f2 long double return.
1800 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1802 * src/types.c: Use 16 byte long double for POWERPC64.
1804 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1806 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1807 when the structure return address is passed in %o0.
1808 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1809 (ffi_v9_layout_struct): Align the field following a nested structure
1810 on a word boundary. Use memmove instead of memcpy.
1811 (ffi_call): Update call to ffi_V9_return_struct.
1812 (ffi_prep_closure): Define 'ctx' only for V8.
1813 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1814 and ffi_closure_sparc_inner_v9.
1815 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1816 Always skip the structure return address. For structures and long
1817 doubles, copy the argument directly.
1818 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1819 if required. Shift the maximum floating-point slot accordingly. For
1820 big structures, copy the argument directly; otherwise, left-justify the
1821 argument and call ffi_v9_layout_struct to lay out the structure on
1823 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1824 (ffi_closure_v8): Pass the structure return address. Update call to
1825 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1826 Skip the 'unimp' insn when returning long doubles and structures.
1827 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1828 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1829 FFI_TYPE_INT handling. Load structures both in integers and
1830 floating-point registers on return.
1831 * README: Update status of the SPARC port.
1833 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1835 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1837 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1839 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1841 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1842 value as of type ffi_arg, not unsigned int.
1844 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1846 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1849 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1851 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1854 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1856 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1859 2003-12-10 Richard Henderson <rth@redhat.com>
1861 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1862 size_t instead of int.
1864 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1866 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1867 * testsuite/libffi.call/many_win32.c (main): Replace variable
1868 int i with unsigned long ul.
1870 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1871 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1872 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1873 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1874 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1875 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1876 * testsuite/libffi.call/cls_align_float.c: Likewise.
1877 * testsuite/libffi.call/cls_align_double.c: Likewise.
1878 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1879 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1881 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1884 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1885 Align arguments to 32 bits.
1887 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1890 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1891 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1892 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1893 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1894 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1895 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1897 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1899 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1901 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1902 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1904 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1906 * Makefile.in: Rebuilt.
1907 * configure: Likewise.
1908 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1911 2003-11-21 Andreas Jaeger <aj@suse.de>
1912 Andreas Tobler <a.tobler@schweiz.ch>
1914 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1915 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1916 * Makefile.in: Rebuilt.
1917 * aclocal.m4: Likewise.
1918 * configure: Likewise.
1919 * fficonfig.h.in: Likewise.
1920 * testsuite/lib/libffi-dg.exp: Add include dir.
1921 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1922 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1923 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1924 for ffi_closure if available.
1925 * testsuite/libffi.call/closure_fn1.c: Likewise.
1926 * testsuite/libffi.call/closure_fn2.c: Likewise.
1927 * testsuite/libffi.call/closure_fn3.c: Likewise.
1928 * testsuite/libffi.call/closure_fn4.c: Likewise.
1929 * testsuite/libffi.call/closure_fn5.c: Likewise.
1930 * testsuite/libffi.call/cls_12byte.c: Likewise.
1931 * testsuite/libffi.call/cls_16byte.c: Likewise.
1932 * testsuite/libffi.call/cls_18byte.c: Likewise.
1933 * testsuite/libffi.call/cls_19byte.c: Likewise.
1934 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1935 * testsuite/libffi.call/cls_20byte.c: Likewise.
1936 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1937 * testsuite/libffi.call/cls_24byte.c: Likewise.
1938 * testsuite/libffi.call/cls_2byte.c: Likewise.
1939 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1940 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1941 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1942 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1943 * testsuite/libffi.call/cls_4byte.c: Likewise.
1944 * testsuite/libffi.call/cls_5byte.c: Likewise.
1945 * testsuite/libffi.call/cls_64byte.c: Likewise.
1946 * testsuite/libffi.call/cls_6byte.c: Likewise.
1947 * testsuite/libffi.call/cls_7byte.c: Likewise.
1948 * testsuite/libffi.call/cls_8byte.c: Likewise.
1949 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1950 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1951 * testsuite/libffi.call/cls_double.c: Likewise.
1952 * testsuite/libffi.call/cls_float.c: Likewise.
1953 * testsuite/libffi.call/cls_schar.c: Likewise.
1954 * testsuite/libffi.call/cls_sint.c: Likewise.
1955 * testsuite/libffi.call/cls_sshort.c: Likewise.
1956 * testsuite/libffi.call/cls_uchar.c: Likewise.
1957 * testsuite/libffi.call/cls_uint.c: Likewise.
1958 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1959 * testsuite/libffi.call/cls_ushort.c: Likewise.
1960 * testsuite/libffi.call/nested_struct.c: Likewise.
1961 * testsuite/libffi.call/nested_struct1.c: Likewise.
1962 * testsuite/libffi.call/nested_struct2.c: Likewise.
1963 * testsuite/libffi.call/nested_struct3.c: Likewise.
1964 * testsuite/libffi.call/problem1.c: Likewise.
1965 * testsuite/libffi.special/unwindtest.cc: Likewise.
1967 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
1969 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
1971 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
1973 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
1974 Add -lgcc_s to additional flags.
1976 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
1978 * configure.in, include/Makefile.am: PR libgcj/11147, install
1979 the ffitarget.h header file in a gcc versioned and target
1981 * configure: Regenerated.
1982 * Makefile.in, include/Makefile.in: Likewise.
1983 * testsuite/Makefile.in: Likewise.
1985 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
1987 * testsuite/libffi.call/closure_fn0.c: Print result and check
1988 with dg-output to make debugging easier.
1989 * testsuite/libffi.call/closure_fn1.c: Likewise.
1990 * testsuite/libffi.call/closure_fn2.c: Likewise.
1991 * testsuite/libffi.call/closure_fn3.c: Likewise.
1992 * testsuite/libffi.call/closure_fn4.c: Likewise.
1993 * testsuite/libffi.call/closure_fn5.c: Likewise.
1994 * testsuite/libffi.call/cls_12byte.c: Likewise.
1995 * testsuite/libffi.call/cls_16byte.c: Likewise.
1996 * testsuite/libffi.call/cls_18byte.c: Likewise.
1997 * testsuite/libffi.call/cls_19byte.c: Likewise.
1998 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1999 * testsuite/libffi.call/cls_20byte.c: Likewise.
2000 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2001 * testsuite/libffi.call/cls_24byte.c: Likewise.
2002 * testsuite/libffi.call/cls_2byte.c: Likewise.
2003 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2004 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2005 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2006 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2007 * testsuite/libffi.call/cls_4byte.c: Likewise.
2008 * testsuite/libffi.call/cls_5byte.c: Likewise.
2009 * testsuite/libffi.call/cls_64byte.c: Likewise.
2010 * testsuite/libffi.call/cls_6byte.c: Likewise.
2011 * testsuite/libffi.call/cls_7byte.c: Likewise.
2012 * testsuite/libffi.call/cls_8byte.c: Likewise.
2013 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2014 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2015 * testsuite/libffi.call/cls_double.c: Likewise.
2016 * testsuite/libffi.call/cls_float.c: Likewise.
2017 * testsuite/libffi.call/cls_schar.c: Likewise.
2018 * testsuite/libffi.call/cls_sint.c: Likewise.
2019 * testsuite/libffi.call/cls_sshort.c: Likewise.
2020 * testsuite/libffi.call/cls_uchar.c: Likewise.
2021 * testsuite/libffi.call/cls_uint.c: Likewise.
2022 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2023 * testsuite/libffi.call/cls_ushort.c: Likewise.
2024 * testsuite/libffi.call/problem1.c: Likewise.
2026 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2029 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2031 * testsuite/libffi.call/cls_9byte2.c: New test case.
2032 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2033 * testsuite/libffi.call/cls_64byte.c: Likewise.
2034 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2035 * testsuite/libffi.call/cls_19byte.c: Likewise.
2036 * testsuite/libffi.call/cls_18byte.c: Likewise.
2037 * testsuite/libffi.call/closure_fn4.c: Likewise.
2038 * testsuite/libffi.call/closure_fn5.c: Likewise.
2039 * testsuite/libffi.call/cls_schar.c: Likewise.
2040 * testsuite/libffi.call/cls_sint.c: Likewise.
2041 * testsuite/libffi.call/cls_sshort.c: Likewise.
2042 * testsuite/libffi.call/nested_struct2.c: Likewise.
2043 * testsuite/libffi.call/nested_struct3.c: Likewise.
2045 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2047 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2048 * testsuite/libffi.call/cls_uchar.c: Likewise.
2049 * testsuite/libffi.call/cls_uint.c: Likewise.
2050 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2051 * testsuite/libffi.call/cls_ushort.c: Likewise.
2052 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2054 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2056 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2059 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2061 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2062 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2064 2003-10-22 David Daney <ddaney@avtrex.com>
2066 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2067 __attribute__((__mode__(__SI__))) and friends.
2069 2003-10-22 Andreas Schwab <schwab@suse.de>
2071 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2073 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2075 * configure.in: AC_LINK_FILES(ffitarget.h).
2076 * configure: Regenerate.
2077 * Makefile.in: Likewise.
2078 * include/Makefile.in: Likewise.
2079 * testsuite/Makefile.in: Likewise.
2080 * fficonfig.h.in: Likewise.
2082 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2083 Richard Henderson <rth@redhat.com>
2085 Avoid that ffi.h includes fficonfig.h.
2087 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2088 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2089 (TARGET_SRC_MIPS_SGI): Removed.
2090 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2091 (MIPS_SGI): Removed.
2092 (CLEANFILES): Removed.
2093 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2095 * acconfig.h: Removed.
2096 * configure.in: Compute sizeofs only for double and long double.
2097 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2098 into AC_DEFINE instead of using acconfig.h. Create
2099 include/ffitarget.h instead of include/fficonfig.h. Rename
2100 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2101 AC_DEFINE EH_FRAME_FLAGS.
2102 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2103 (hack_DATA): Add ffitarget.h.
2104 * include/ffi.h.in: Remove all system specific definitions.
2105 Declare raw API even if it is not installed, why bother?
2106 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2107 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2108 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2109 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2110 * include/ffi_common.h (bool): Do not define.
2111 (ffi_assert): Accept failed assertion.
2112 (ffi_type_test): Return void and accept file/line.
2113 (FFI_ASSERT): Pass stringized failed assertion.
2114 (FFI_ASSERT_AT): New macro.
2115 (FFI_ASSERT_VALID_TYPE): New macro.
2116 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2117 UINT64, SINT64): Define here with gcc's __attribute__ macro
2119 (FLOAT32, ALIGN): Define here instead of in ffi.h
2120 * include/ffi-mips.h: Removed. Its content moved to
2121 src/mips/ffitarget.h after separating assembly and C sections.
2122 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2123 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2124 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2125 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2126 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2127 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2128 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2129 (ffi_type_test): Rewritten.
2130 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2131 FFI_ASSERT_VALID_TYPE.
2132 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2133 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2134 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2135 src/s390/ffitarget.h, src/sh/ffitarget.h,
2136 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2137 src/x86/ffitarget.h: New files.
2138 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2139 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2140 src/powerpc/aix.S, src/powerpc/darwin.S,
2141 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2142 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2143 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2144 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2145 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2148 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2150 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2151 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2153 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2155 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2156 Used when FFI_DEBUG = 1.
2158 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2160 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2163 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2165 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2166 FFI_MIPS_O32 for O32 ABI.
2168 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2170 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2171 SPARC64. Cleanup whitespaces.
2173 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2175 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2176 strongarm, xscale. Cleanup whitespaces.
2177 * testsuite/libffi.call/closure_fn1.c: Likewise.
2178 * testsuite/libffi.call/closure_fn2.c: Likewise.
2179 * testsuite/libffi.call/closure_fn3.c: Likewise.
2180 * testsuite/libffi.call/cls_12byte.c: Likewise.
2181 * testsuite/libffi.call/cls_16byte.c: Likewise.
2182 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2183 * testsuite/libffi.call/cls_20byte.c: Likewise.
2184 * testsuite/libffi.call/cls_24byte.c: Likewise.
2185 * testsuite/libffi.call/cls_2byte.c: Likewise.
2186 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2187 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2188 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2189 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2190 * testsuite/libffi.call/cls_4byte.c: Likewise.
2191 * testsuite/libffi.call/cls_5byte.c: Likewise.
2192 * testsuite/libffi.call/cls_6byte.c: Likewise.
2193 * testsuite/libffi.call/cls_7byte.c: Likewise.
2194 * testsuite/libffi.call/cls_8byte.c: Likewise.
2195 * testsuite/libffi.call/cls_double.c: Likewise.
2196 * testsuite/libffi.call/cls_float.c: Likewise.
2197 * testsuite/libffi.call/cls_uchar.c: Likewise.
2198 * testsuite/libffi.call/cls_uint.c: Likewise.
2199 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2200 * testsuite/libffi.call/cls_ushort.c: Likewise.
2201 * testsuite/libffi.call/nested_struct.c: Likewise.
2202 * testsuite/libffi.call/nested_struct1.c: Likewise.
2203 * testsuite/libffi.call/problem1.c: Likewise.
2204 * testsuite/libffi.special/unwindtest.cc: Likewise.
2205 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2207 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2209 * src/powerpc/aix.S: Cleanup whitespaces.
2210 * src/powerpc/aix_closure.S: Likewise.
2212 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2214 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2215 * src/powerpc/darwin_closure.S: Likewise.
2216 * src/powerpc/ffi_darwin.c: Likewise.
2218 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2219 David Edelsohn <edelsohn@gnu.org>
2221 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2222 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2224 * src/powerpc/darwin_closure.S: Likewise.
2225 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2226 according to the Darwin/AIX ABI.
2227 (ffi_prep_cif_machdep): Likewise.
2228 (ffi_closure_helper_DARWIN): Likewise.
2229 Remove the outgoing parameter stack logic. Simplify the evaluation
2230 of the different CASE types.
2231 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2232 statement in the trampoline code.
2234 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2236 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2237 for the register size.
2238 (ffi_closure_helper_SYSV): Handle the structure return value
2240 (ffi_closure_helper_SYSV): Return the appropriate type when
2241 the registers are used for the structure return value.
2242 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2243 the 64-bit return value. Update copyright years.
2245 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2247 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2248 srcdir for ffi_mips.h.
2250 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2252 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2254 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2255 placement of float result.
2256 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2257 cast of "resp" for big-endian 64 bit machines.
2259 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2261 * src/types.c (double, longdouble): Merge identical SH and ARM
2262 typedefs, and add POWERPC64.
2263 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2264 struct split over gpr and rest.
2265 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2266 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2268 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2270 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2273 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2275 * configure: Regenerate.
2277 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2279 * Makefile.am: Remove build rules for ffitest.
2280 * Makefile.in: Rebuilt.
2282 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2284 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2285 about implicit declaration of abort().
2287 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2289 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2290 * Makefile.in: Rebuilt.
2291 * configure.in: Add dejagnu test framework.
2292 * configure: Rebuilt.
2294 * testsuite/Makefile.am: New file.
2295 * testsuite/Makefile.in: Built
2296 * testsuite/lib/libffi-dg.exp: New file.
2297 * testsuite/config/default.exp: Likewise.
2298 * testsuite/libffi.call/call.exp: Likewise.
2299 * testsuite/libffi.call/ffitest.h: Likewise.
2300 * testsuite/libffi.call/closure_fn0.c: Likewise.
2301 * testsuite/libffi.call/closure_fn1.c: Likewise.
2302 * testsuite/libffi.call/closure_fn2.c: Likewise.
2303 * testsuite/libffi.call/closure_fn3.c: Likewise.
2304 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2305 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2306 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2307 * testsuite/libffi.call/cls_2byte.c: Likewise.
2308 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2309 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2310 * testsuite/libffi.call/cls_4byte.c: Likewise.
2311 * testsuite/libffi.call/cls_5byte.c: Likewise.
2312 * testsuite/libffi.call/cls_6byte.c: Likewise.
2313 * testsuite/libffi.call/cls_7byte.c: Likewise.
2314 * testsuite/libffi.call/cls_8byte.c: Likewise.
2315 * testsuite/libffi.call/cls_12byte.c: Likewise.
2316 * testsuite/libffi.call/cls_16byte.c: Likewise.
2317 * testsuite/libffi.call/cls_20byte.c: Likewise.
2318 * testsuite/libffi.call/cls_24byte.c: Likewise.
2319 * testsuite/libffi.call/cls_double.c: Likewise.
2320 * testsuite/libffi.call/cls_float.c: Likewise.
2321 * testsuite/libffi.call/cls_uchar.c: Likewise.
2322 * testsuite/libffi.call/cls_uint.c: Likewise.
2323 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2324 * testsuite/libffi.call/cls_ushort.c: Likewise.
2325 * testsuite/libffi.call/float.c: Likewise.
2326 * testsuite/libffi.call/float1.c: Likewise.
2327 * testsuite/libffi.call/float2.c: Likewise.
2328 * testsuite/libffi.call/many.c: Likewise.
2329 * testsuite/libffi.call/many_win32.c: Likewise.
2330 * testsuite/libffi.call/nested_struct.c: Likewise.
2331 * testsuite/libffi.call/nested_struct1.c: Likewise.
2332 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2333 * testsuite/libffi.call/problem1.c: Likewise.
2334 * testsuite/libffi.call/promotion.c: Likewise.
2335 * testsuite/libffi.call/return_ll.c: Likewise.
2336 * testsuite/libffi.call/return_sc.c: Likewise.
2337 * testsuite/libffi.call/return_uc.c: Likewise.
2338 * testsuite/libffi.call/strlen.c: Likewise.
2339 * testsuite/libffi.call/strlen_win32.c: Likewise.
2340 * testsuite/libffi.call/struct1.c: Likewise.
2341 * testsuite/libffi.call/struct2.c: Likewise.
2342 * testsuite/libffi.call/struct3.c: Likewise.
2343 * testsuite/libffi.call/struct4.c: Likewise.
2344 * testsuite/libffi.call/struct5.c: Likewise.
2345 * testsuite/libffi.call/struct6.c: Likewise.
2346 * testsuite/libffi.call/struct7.c: Likewise.
2347 * testsuite/libffi.call/struct8.c: Likewise.
2348 * testsuite/libffi.call/struct9.c: Likewise.
2349 * testsuite/libffi.special/special.exp: New file.
2350 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2351 * testsuite/libffi.special/unwindtest.cc: Likewise.
2354 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2356 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2359 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2361 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2363 (ffi_closure_helper_LINUX64): Likewise.
2364 * src/powerpc/linux64.S: Remove code writing to parm save area.
2365 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2366 address in lr from ffi_closure_helper_LINUX64 call to calculate
2367 table address. Optimize function tail.
2369 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2371 * src/sparc/ffi.c: Handle all floating point registers.
2372 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2374 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2376 * README: Note that libffi is not part of GCC. Update the project
2379 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2381 * src/powerpc/ppc_closure.S: Include ffi.h.
2383 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2385 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2386 Use C style comments.
2388 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2390 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2391 * Makefile.in: Regenerate.
2392 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2393 * configure: Regenerate.
2394 * include/ffi.h.in: Add SHmedia support.
2395 * src/sh64/ffi.c: New file.
2396 * src/sh64/sysv.S: New file.
2398 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2400 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2401 should be read-only.
2402 * configure: Rebuilt.
2403 * fficonfig.h.in: Rebuilt.
2404 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2405 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2406 * src/powerpc/linux64.S: Likewise.
2407 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2408 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2409 if -fpic/-fPIC/-mrelocatable.
2410 * src/powerpc/powerpc_closure.S: Likewise.
2411 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2412 #write in .eh_frame flags.
2413 * src/sparc/v9.S: Likewise.
2414 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2415 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2416 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2418 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2420 Fixes PR bootstrap/10656
2421 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2422 support for .register pseudo-op.
2423 * src/sparc/v8.S: Use it.
2424 * fficonfig.h.in: Rebuilt.
2425 * configure: Rebuilt.
2427 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2429 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2430 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2431 Make it the default on POWERPC64.
2432 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2433 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2434 * configure: Rebuilt.
2435 * src/powerpc/ffi.c (hidden): Define.
2436 (ffi_prep_args_SYSV): Renamed from
2437 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2438 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2439 ASM_NEEDS_REGISTERS64): New.
2440 (ffi_prep_args64): New function.
2441 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2442 (ffi_call): Likewise.
2443 (ffi_prep_closure): Likewise.
2444 (flush_icache): Surround by #ifndef POWERPC64.
2445 (ffi_dblfl): New union type.
2446 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2447 (ffi_closure_helper_LINUX64): New function.
2448 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2450 * src/powerpc/sysv.S: Likewise.
2451 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2452 * src/powerpc/linux64.S: New file.
2453 * src/powerpc/linux64_closure.S: New file.
2454 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2455 src/powerpc/linux64_closure.S.
2456 (TARGET_SRC_POWERPC): Likewise.
2458 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2459 closure_test_fn3): Fix result printing on big-endian 64-bit
2461 (main): Print tst2_arg instead of uninitialized tst2_result.
2463 * src/ffitest.c (main): Hide what closure pointer really points to
2466 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2468 * configure.in (arm-*-netbsdelf*): Add configuration.
2469 (configure): Regenerated.
2471 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2473 * include/Makefile.in: Regenerate.
2475 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2477 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2479 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2480 Receive closure pointer through parameter, read args using
2481 __builtin_dwarf_cfa.
2482 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2484 2003-03-12 Andreas Schwab <schwab@suse.de>
2486 * configure.in: Avoid trailing /. in toolexeclibdir.
2487 * configure: Rebuilt.
2489 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2491 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2493 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2495 * libffi/src/powerpc/darwin_closure.S:
2496 Fix alignement bug, allocate 8 bytes for the result.
2497 * libffi/src/powerpc/aix_closure.S:
2499 * libffi/src/powerpc/ffi_darwin.c:
2500 Update stackframe description for aix/darwin_closure.S.
2502 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2504 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2507 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2508 Andreas Schwab <schwab@suse.de>
2510 * configure.in: Adjust command to source config-ml.in to account
2511 for changes to the libffi_basedir definition.
2512 (libffi_basedir): Remove ${srcdir} from value and include trailing
2515 * configure: Regenerate.
2517 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2519 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2521 2003-01-28 Andrew Haley <aph@redhat.com>
2523 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2524 * src/x86/ffi64.c (ffi_prep_closure): New.
2525 (ffi_closure_UNIX64_inner): New.
2526 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2528 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2530 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2531 Remove USE_LIBDIR conditional.
2532 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2533 * Makefile.in, configure: Rebuilt.
2535 2003-01027 David Edelsohn <edelsohn@gnu.org>
2537 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2538 * Makefile.in: Regenerate.
2540 2003-01-22 Andrew Haley <aph@redhat.com>
2542 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2545 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2547 * src/powerpc/darwin.S: Add unwind info.
2548 * src/powerpc/darwin_closure.S: Likewise.
2550 2003-01-14 Andrew Haley <aph@redhat.com>
2552 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2553 (ffi_prep_cif_machdep): Likewise.
2554 * src/x86/unix64.S: Add unwind info.
2556 2003-01-14 Andreas Jaeger <aj@suse.de>
2558 * src/ffitest.c (main): Only use ffi_closures if those are
2561 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2563 * libffi/src/ffitest.c
2564 add closure testcases
2566 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2568 * libffi/src/powerpc/ffi.c
2569 fix alignment bug for float (4 byte aligned iso 8 byte)
2571 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2573 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2574 * src/powerpc/darwin.S: Remove RCS version string.
2576 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2578 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2579 * src/ffitest.c (main): Use static storage for closure.
2580 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2581 * src/sparc/v8.S (ffi_closure_v8): New.
2582 * src/sparc/v9.S (ffi_closure_v9): New.
2584 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2586 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2587 enumeration for X86_WIN32.
2588 * src/x86/win32.S: Added ffi_call_STDCALL function
2590 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2591 switch cases for recognising FFI_STDCALL and
2592 calling ffi_call_STDCALL if target is X86_WIN32.
2593 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2594 stdcall versions of the "my_strlen" and "many"
2595 test functions (for X86_WIN32).
2596 Added test cases to test stdcall invocation using
2599 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2601 * src/sh/sysv.S: Add DWARF2 unwind info.
2603 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2605 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2607 2002-11-26 Jim Wilson <wilson@redhat.com>
2609 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2611 2002-11-23 H.J. Lu <hjl@gnu.org>
2613 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2614 Include ../config/accross.m4.
2615 * aclocal.m4; Rebuild.
2616 * configure: Likewise.
2618 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2620 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2622 2002-11-11 DJ Delorie <dj@redhat.com>
2624 * configure.in: Look for common files in the right place.
2626 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2628 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2629 raw data as _Jv_word values, not ffi_raw.
2630 (ffi_java_ptrarray_to_raw): Likewise.
2631 (ffi_java_rvalue_to_raw): New function.
2632 (ffi_java_raw_call): Call it.
2633 (ffi_java_raw_to_rvalue): New function.
2634 (ffi_java_translate_args): Call it.
2635 * src/ffitest.c (closure_test_fn): Interpret return value
2636 as ffi_arg, not int.
2637 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2638 FFI_TYPE_POINTER case.
2639 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2640 values extended to word size.
2642 2002-10-02 Andreas Jaeger <aj@suse.de>
2644 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2646 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2648 * include/ffi.h.in: Fix i386 win32 compilation.
2650 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2652 * configure.in: Add s390x-*-linux-* target.
2653 * configure: Regenerate.
2654 * include/ffi.h.in: Define S390X for s390x targets.
2655 (FFI_CLOSURES): Define for s390/s390x.
2656 (FFI_TRAMPOLINE_SIZE): Likewise.
2657 (FFI_NATIVE_RAW_API): Likewise.
2658 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2659 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2660 * src/s390/ffi.c: Major rework of existing code. Add support for
2661 s390x targets. Add closure support.
2662 * src/s390/sysv.S: Likewise.
2664 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2666 * src/arm/sysv.S: Fix typo.
2668 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2670 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2671 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2672 (ffi_call_SYSV): Handle soft-float.
2674 2002-09-27 Bo Thorsen <bo@suse.de>
2676 * include/ffi.h.in: Fix multilib x86-64 support.
2678 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2680 * Makefile.am (all-multi): Fix multilib parallel build.
2682 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2684 * configure.in (sh[34]*-*-linux*): Add brackets.
2685 * configure: Regenerate.
2687 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2689 * Makefile.am: Add SH support.
2690 * Makefile.in: Regenerate.
2691 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2692 * configure: Regenerate.
2693 * include/ffi.h.in: Add SH support.
2694 * src/sh/ffi.c: New file.
2695 * src/sh/sysv.S: New file.
2696 * src/types.c: Add SH support.
2698 2002-07-16 Bo Thorsen <bo@suse.de>
2700 * src/x86/ffi64.c: New file that adds x86-64 support.
2701 * src/x86/unix64.S: New file that handles argument setup for
2703 * src/x86/sysv.S: Don't use this on x86-64.
2704 * src/x86/ffi.c: Don't use this on x86-64.
2706 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2708 * src/ffitest.c (struct6): New test that tests a special case in
2710 (struct7): Likewise.
2711 (struct8): Likewise.
2712 (struct9): Likewise.
2713 (closure_test_fn): Silence warning about this when it's not used.
2714 (main): Add the new tests.
2715 (main): Fix a couple of wrong casts and silence some compiler warnings.
2716 * include/ffi.h.in: Add x86-64 ABI definition.
2717 * fficonfig.h.in: Regenerate.
2718 * Makefile.am: Add x86-64 support.
2719 * configure.in: Likewise.
2720 * Makefile.in: Regenerate.
2721 * configure: Likewise.
2723 2002-06-24 Bo Thorsen <bo@suse.de>
2725 * src/types.c: Merge settings for similar architectures.
2726 Add x86-64 sizes and alignments.
2728 2002-06-23 Bo Thorsen <bo@suse.de>
2730 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2731 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2732 * src/mips/ffi.c (ffi_prep_args): Likewise.
2733 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2735 2002-07-18 H.J. Lu (hjl@gnu.org)
2737 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2738 (libffi_la_SOURCES): Support MIPS_LINUX.
2739 (libffi_convenience_la_SOURCES): Likewise.
2740 * Makefile.in: Regenerated.
2742 * configure.in (mips64*-*): Skip.
2743 (mips*-*-linux*): New.
2744 * configure: Regenerated.
2746 * src/mips/ffi.c: Include <sgidefs.h>.
2748 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2750 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2752 2002-05-27 Roger Sayle <roger@eyesopen.com>
2754 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2756 2002-05-27 Bo Thorsen <bo@suse.de>
2758 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2761 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2763 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2764 beginning of function (for older apple cc).
2766 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2768 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2769 script entry, and set LD to it when configuring multilibs.
2770 * configure: Rebuilt.
2772 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2774 * configure.in (sparc64-*-netbsd*): Add target.
2775 (sparc-*-netbsdelf*): Likewise.
2776 * configure: Regenerate.
2778 2002-04-28 David S. Miller <davem@redhat.com>
2780 * configure.in, configure: Fix SPARC test in previous change.
2782 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2784 * Makefile.am: Add Linux for S/390 support.
2785 * Makefile.in: Regenerate.
2786 * configure.in: Add Linux for S/390 support.
2787 * configure: Regenerate.
2788 * include/ffi.h.in: Add Linux for S/390 support.
2789 * src/s390/ffi.c: New file from libffi CVS tree.
2790 * src/s390/sysv.S: New file from libffi CVS tree.
2792 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2794 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2796 * src/sparc/v8.S: Use it.
2797 * src/sparc/v9.S: Likewise.
2798 * fficonfig.h.in: Rebuilt.
2799 * configure: Rebuilt.
2801 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2803 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2805 * src/ia64/unix.S: Add unwind information. Fix comments.
2806 Save sp in a way that's compatible with unwind info.
2807 (ffi_call_unix): Correctly restore sp in all cases.
2808 * src/ia64/ffi.c: Add, fix comments.
2810 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2812 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2814 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2816 * configure.in (alpha*-*-netbsd*): Add target.
2817 * configure: Regenerate.
2819 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2821 * src/sparc/v8.S: Add unwind info.
2822 * src/sparc/v9.S: Likewise.
2824 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2826 * configure.in: Enable i*86-*-netbsdelf*.
2827 * configure: Rebuilt.
2829 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2832 * src/mips/n32.s: Delete
2833 * src/mips/o32.s: Delete
2835 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2837 * configure.in: Enable alpha*-*-freebsd*.
2838 * configure: Rebuilt.
2840 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2842 * Makefile.am: libfficonvenience -> libffi_convenience.
2843 * Makefile.in: Rebuilt.
2845 * Makefile.am: Define ffitest_OBJECTS.
2846 * Makefile.in: Rebuilt.
2848 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2849 David Edelsohn <edelsohn@gnu.org>
2851 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2852 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2853 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2854 * Makefile.in: Regenerate.
2855 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2856 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2857 (flush_icache, flush_range): New functions.
2858 (ffi_closure_helper_DARWIN): New function.
2859 * src/powerpc/aix_closure.S: New file.
2860 * src/powerpc/darwin_closure.S: New file.
2862 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2864 * include/ffi.h.in: Add typedef for ffi_arg.
2865 * src/ffitest.c (main): Declare rint with ffi_arg.
2867 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2869 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2870 number of GPRs for floating-point arguments.
2872 2002-01-31 Anthony Green <green@redhat.com>
2874 * configure: Rebuilt.
2875 * configure.in: Replace CHECK_SIZEOF and endian tests with
2876 cross-compiler friendly macros.
2877 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2880 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2882 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2883 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2885 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2887 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2888 (TARGET_SRC_POWERPC_AIX): New.
2889 (POWERPC_AIX): New stanza.
2890 * Makefile.in: Regenerate.
2891 * configure.in: Add AIX case.
2892 * configure: Regenerate.
2893 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2894 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2895 size. Fix "long double" support.
2896 (ffi_call): Add FFI_AIX case.
2897 * src/powerpc/aix.S: New.
2899 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2901 Implement Darwin PowerPC ABI.
2902 * configure.in: Handle powerpc-*-darwin*.
2903 * Makefile.am: Set source files for POWERPC_DARWIN.
2904 * configure: Rebuilt.
2905 * Makefile.in: Rebuilt.
2906 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2908 * src/powerpc/darwin.S: New file.
2909 * src/powerpc/ffi_darwin.c: New file.
2911 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2913 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2915 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2917 * src/x86/sysv.S: Avoid gas-only .balign directive.
2918 Use C style comments.
2920 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2922 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2923 Fixes PR bootstrap/3563.
2925 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2927 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2929 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2931 * configure.in: Recognize sparc*-sun-* host.
2932 * configure: Regenerate.
2934 2001-06-06 Andrew Haley <aph@redhat.com>
2936 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
2938 2001-06-03 Andrew Haley <aph@redhat.com>
2940 * src/alpha/osf.S: Add unwind info.
2941 * src/powerpc/sysv.S: Add unwind info.
2942 * src/powerpc/ppc_closure.S: Likewise.
2944 2000-05-31 Jeff Sturm <jsturm@one-point.com>
2946 * configure.in: Fix AC_ARG_ENABLE usage.
2947 * configure: Rebuilt.
2949 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2951 * configure.in: Remove warning about beta code.
2952 * configure: Rebuilt.
2954 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
2956 * src/ia64/unix.S: Restore stack pointer when returning from
2958 * src/ia64/ffi.c: Fix typo in comment.
2960 2001-04-18 Jim Wilson <wilson@redhat.com>
2962 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
2963 to eliminate RAW DV.
2965 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
2967 * Makefile.am: Make a libtool convenience library.
2968 * Makefile.in: Rebuilt.
2970 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
2972 * configure.in: Use different syntax for subdirectory creation.
2973 * configure: Rebuilt.
2975 2001-03-27 Jon Beniston <jon@beniston.com>
2977 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
2978 * configure: Rebuilt.
2979 * Makefile.am: Added X86_WIN32 target support.
2980 * Makefile.in: Rebuilt.
2982 * include/ffi.h.in: Added X86_WIN32 target support.
2984 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
2985 * src/types.c: Added X86_WIN32 target support.
2987 * src/x86/win32.S: New file. Based on sysv.S, but with EH
2988 stuff removed and made to work with CygWin's gas.
2990 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
2992 * configure.in: Make target subdirectory in build dir.
2993 * Makefile.am: Override suffix based rules to specify correct output
2995 * Makefile.in: Rebuilt.
2996 * configure: Rebuilt.
2998 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3000 * src/powerpc/ppc_closure.S: New file.
3001 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3002 involving long long and register pairs.
3003 (ffi_prep_closure): New function.
3004 (flush_icache): Likewise.
3005 (ffi_closure_helper_SYSV): Likewise.
3006 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3007 (FFI_TRAMPOLINE_SIZE): Likewise.
3008 (FFI_NATIVE_RAW_API): Likewise.
3009 * Makefile.in: Rebuilt.
3010 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3011 (TARGET_SRC_POWERPC): Likewise.
3013 2001-03-19 Tom Tromey <tromey@redhat.com>
3015 * Makefile.in: Rebuilt.
3016 * Makefile.am (ffitest_LDFLAGS): New macro.
3018 2001-03-02 Nick Clifton <nickc@redhat.com>
3020 * include/ffi.h.in: Remove RCS ident string.
3021 * include/ffi_mips.h: Remove RCS ident string.
3022 * src/debug.c: Remove RCS ident string.
3023 * src/ffitest.c: Remove RCS ident string.
3024 * src/prep_cif.c: Remove RCS ident string.
3025 * src/types.c: Remove RCS ident string.
3026 * src/alpha/ffi.c: Remove RCS ident string.
3027 * src/alpha/osf.S: Remove RCS ident string.
3028 * src/arm/ffi.c: Remove RCS ident string.
3029 * src/arm/sysv.S: Remove RCS ident string.
3030 * src/mips/ffi.c: Remove RCS ident string.
3031 * src/mips/n32.S: Remove RCS ident string.
3032 * src/mips/o32.S: Remove RCS ident string.
3033 * src/sparc/ffi.c: Remove RCS ident string.
3034 * src/sparc/v8.S: Remove RCS ident string.
3035 * src/sparc/v9.S: Remove RCS ident string.
3036 * src/x86/ffi.c: Remove RCS ident string.
3037 * src/x86/sysv.S: Remove RCS ident string.
3039 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3041 * include/ffi.h.in: Change sourceware.cygnus.com references to
3044 2000-12-09 Richard Henderson <rth@redhat.com>
3046 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3047 (ffi_closure_osf_inner): Index rather than increment avalue
3048 and arg_types. Give ffi_closure_osf the raw return value type.
3049 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3052 2000-12-07 Richard Henderson <rth@redhat.com>
3054 * src/raw_api.c (ffi_translate_args): Fix typo.
3055 (ffi_prep_closure): Likewise.
3057 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3058 FFI_TRAMPOLINE_SIZE.
3059 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3060 cif->bytes for new ffi_call_osf implementation.
3061 (ffi_prep_args): Absorb into ...
3062 (ffi_call): ... here. Do all stack allocation here and
3063 avoid a callback function.
3064 (ffi_prep_closure, ffi_closure_osf_inner): New.
3065 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3066 (ffi_closure_osf): New.
3068 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3070 * config.guess, config.sub, install-sh: Removed.
3071 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3072 * Makefile.in: Rebuilt.
3074 * acinclude.m4: Include libtool macros from the top level.
3075 * aclocal.m4, configure: Rebuilt.
3077 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3079 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3080 * configure: Rebuilt.
3082 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3084 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3085 memory correctly. Use conditional instructions, not branches where
3088 2000-05-04 Tom Tromey <tromey@cygnus.com>
3090 * configure: Rebuilt.
3091 * configure.in: Match `arm*-*-linux-*'.
3092 From Chris Dornan <cdornan@arm.com>.
3094 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3096 * Makefile.am (SUBDIRS): Define.
3097 (AM_MAKEFLAGS): Likewise.
3098 (Multilib support.): Add section.
3099 * Makefile.in: Rebuilt.
3100 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3101 New variables. Set for gcc using -print-multi-lib. Export them
3103 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3104 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3106 (extra_compiler_flags_value, extra_compiler_flags): Check these
3107 for extra compiler options which need to be passed down in
3110 2000-04-16 Anthony Green <green@redhat.com>
3112 * configure: Rebuilt.
3113 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3115 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3117 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3118 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3119 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3120 Replace all void * sizeofs with sizeof(int).
3121 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3122 different than DOUBLE.
3123 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3124 (ffi_prep_args_v9): New function.
3125 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3126 (ffi_V9_return_struct): New function.
3127 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3128 32bit code (not yet cross-arch calls).
3129 * src/sparc/v8.S: Add struct return delay nop.
3131 * src/sparc/v9.S: New file.
3132 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3133 is used on sparc64 only for structures larger than 32 bytes.
3134 Pass by reference for structures is done for structure arguments
3135 larger than 16 bytes.
3136 * src/ffitest.c (main): Use 64bit rint on sparc64.
3137 Run long long tests on sparc.
3138 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3140 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3142 * configure.in (sparc-*-linux*): New supported target.
3143 (sparc64-*-linux*): Likewise.
3144 * configure: Rebuilt.
3145 * Makefile.am: Add v9.S to SPARC files.
3146 * Makefile.in: Likewise.
3147 (LINK): Surround $(CCLD) into double quotes, so that multilib
3148 compiles work correctly.
3150 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3152 * configure: Rebuilt.
3153 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3154 proposed by Bryce McKinlay.
3156 2000-03-20 Tom Tromey <tromey@cygnus.com>
3158 * Makefile.in: Hand edit for java_raw_api.lo.
3160 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3162 * config.guess, config.sub: Update from the gcc tree.
3163 Fix for PR libgcj/168.
3165 2000-03-03 Tom Tromey <tromey@cygnus.com>
3167 * Makefile.in: Fixed ia64 by hand.
3169 * configure: Rebuilt.
3170 * configure.in (--enable-multilib): New option.
3171 (libffi_basedir): New subst.
3172 (AC_OUTPUT): Added multilib code.
3174 2000-03-02 Tom Tromey <tromey@cygnus.com>
3176 * Makefile.in: Rebuilt.
3177 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3180 2000-02-25 Hans Boehm <boehm@acm.org>
3182 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3184 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3186 (ffi_prep_raw_closure): Use ffi_translate_args, not
3187 ffi_closure_translate.
3188 * src/java_raw_api.c: New file.
3189 * src/ffitest.c (closure_test_fn): New function.
3190 (main): Define `rint' as long long on IA64. Added new test when
3191 FFI_CLOSURES is defined.
3192 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3193 (ffi_abi): Recognize IA64.
3194 (ffi_raw): Added `flt' field.
3195 Added "Java raw API" code.
3196 * configure.in: Recognize ia64.
3197 * Makefile.am (TARGET_SRC_IA64): New macro.
3198 (libffi_la_common_SOURCES): Added java_raw_api.c.
3199 (libffi_la_SOURCES): Define in IA64 case.
3201 2000-01-04 Tom Tromey <tromey@cygnus.com>
3203 * Makefile.in: Rebuilt with newer automake.
3205 1999-12-31 Tom Tromey <tromey@cygnus.com>
3207 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3209 1999-09-01 Tom Tromey <tromey@cygnus.com>
3211 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3213 * fficonfig.h.in: Rebuilt.
3214 * configure: Rebuilt.
3215 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3216 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3217 * acconfig.h: Don't #undef PACKAGE or VERSION.
3219 1999-08-09 Anthony Green <green@cygnus.com>
3221 * include/ffi.h.in: Try to work around messy header problem
3222 with PACKAGE and VERSION.
3224 * configure: Rebuilt.
3225 * configure.in: Change version to 2.00-beta.
3227 * fficonfig.h.in: Rebuilt.
3228 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3230 * src/x86/ffi.c (ffi_raw_call): Rename.
3232 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3234 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3235 (ffi_prep_incoming_args_SYSV): Ditto.
3236 (ffi_prep_closure): Ditto.
3237 (ffi_closure_raw_SYSV): Ditto.
3238 (ffi_prep_raw_closure): More ditto.
3239 (ffi_call_raw): Final ditto.
3241 * include/ffi.h.in: Add definitions for closure and raw API.
3243 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3246 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3248 * src/raw_api.c: New file.
3250 * include/ffi.h.in (ffi_raw): New type.
3251 (UINT_ARG, SINT_ARG): New defines.
3252 (ffi_closure, ffi_raw_closure): New types.
3253 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3255 * configure.in: Add check for endianness and sizeof void*.
3257 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3258 instead of directly.
3260 * configure: Rebuilt.
3262 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3264 * configure.in: Add x86 and powerpc BeOS configurations.
3265 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3267 1999-05-09 Anthony Green <green@cygnus.com>
3269 * configure.in: Add warning about this being beta code.
3270 Remove src/Makefile.am from the picture.
3271 * configure: Rebuilt.
3273 * Makefile.am: Move logic from src/Makefile.am. Add changes
3274 to support libffi as a target library.
3275 * Makefile.in: Rebuilt.
3277 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3278 Upgraded to new autoconf, automake, libtool.
3282 * LICENSE: Update copyright date.
3284 * src/Makefile.am, src/Makefile.in: Removed.
3286 1998-11-29 Anthony Green <green@cygnus.com>
3288 * include/ChangeLog: Removed.
3289 * src/ChangeLog: Removed.
3290 * src/mips/ChangeLog: Removed.
3291 * src/sparc/ChangeLog: Remboved.
3292 * src/x86/ChangeLog: Removed.
3294 * ChangeLog.v1: Created.