1 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
3 * configure: Regenerate.
4 * include/Makefile.in: Regenerate.
5 * testsuite/Makefile.in: Regenerate.
7 2008-06-07 Joseph Myers <joseph@codesourcery.com>
9 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
10 powerpc-*-beos*): Remove.
11 * configure: Regenerate.
13 2008-05-09 Julian Brown <julian@codesourcery.com>
15 * Makefile.am (LTLDFLAGS): New.
16 (libffi_la_LDFLAGS): Use above.
17 * Makefile.in: Regenerate.
19 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
22 * aclocal.m4: Regenerate.
23 * configure: Regenerate.
25 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
27 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
28 * src/sh64/sysv.S: Likewise.
30 2008-03-26 Daniel Jacobowitz <dan@debian.org>
32 * src/arm/sysv.S: Fix ARM comment marker.
34 2008-03-26 Jakub Jelinek <jakub@redhat.com>
36 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
37 * src/s390/sysv.S: Likewise.
38 * src/powerpc/ppc_closure.S: Likewise.
39 * src/powerpc/sysv.S: Likewise.
40 * src/x86/unix64.S: Likewise.
41 * src/x86/sysv.S: Likewise.
42 * src/sparc/v8.S: Likewise.
43 * src/sparc/v9.S: Likewise.
44 * src/m68k/sysv.S: Likewise.
45 * src/arm/sysv.S: Likewise.
47 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
49 * aclocal.m4: Regenerate.
50 * configure: Likewise.
51 * Makefile.in: Likewise.
52 * include/Makefile.in: Likewise.
53 * testsuite/Makefile.in: Likewise.
55 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
56 Andreas Tobler <a.tobler@schweiz.org>
58 * configure.ac: Add amd64-*-freebsd* target.
59 * configure: Regenerate.
61 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
64 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
67 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
70 2008-01-24 David Edelsohn <edelsohn@gnu.org>
72 * configure: Regenerate.
74 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
76 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
78 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
81 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
82 signed/unsigned int8/16 for X86_DARWIN.
83 Updated copyright info.
84 Handle one and two byte structs with special cif->flags.
85 * src/x86/ffitarget.h: Add special types for one and two byte structs.
86 Updated copyright info.
87 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
89 Remove code to pop args from the stack after call.
90 Special-case signed/unsigned for int8/16, one and two byte structs.
91 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
92 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
94 Updated copyright info.
96 2007-12-08 David Daney <ddaney@avtrex.com>
98 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
99 SUBU, add with ADDU and use smaller code sequences.
101 2007-12-07 David Daney <ddaney@avtrex.com>
103 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
106 2007-12-06 David Daney <ddaney@avtrex.com>
108 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
110 (ffi_java_raw): New typedef.
111 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
112 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
114 (ffi_java_raw_closure) : Same.
115 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
117 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
119 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
120 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
121 sizeof(ffi_java_raw) for alignment calculations.
122 (ffi_java_ptrarray_to_raw): Same.
123 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
124 if FFI_SIZEOF_JAVA_RAW == 4.
125 (ffi_java_raw_to_rvalue): Same.
126 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
127 (ffi_java_translate_args): Same.
128 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
130 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
132 2007-12-06 David Daney <ddaney@avtrex.com>
134 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
137 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
140 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
141 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
142 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
143 set the NUM_FPR_ARG_REGISTERS according to.
144 Add support for potential soft-float support under hard-float
146 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
147 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
148 to the FFI_LINUX_SOFT_FLOAT ABI.
149 (ffi_prep_cif_machdep): Likewise.
150 (ffi_closure_helper_SYSV): Likewise.
151 * src/powerpc/ppc_closure.S: Make sure not to store float/double
152 on archs where __NO_FPRS__ is true.
153 Add FFI_TYPE_UINT128 support.
154 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
155 Adjust copyright notice.
157 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
159 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
160 * include/ffi_common.h: ... here.
163 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
165 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
166 * src/powerpc/linux64.S: Likewise.
167 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
168 * testsuite/libffi.call/return_ldl.c: New test.
170 2007-09-04 <aph@redhat.com>
172 * src/arm/sysv.S (UNWIND): New.
173 (Whole file): Conditionally compile unwinder directives.
174 * src/arm/sysv.S: Add unwinder directives.
176 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
177 Only treat r0 as a struct address if we're actually returning a
179 Only copy the bytes that are actually within a struct.
180 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
181 is returned in r0, not passed by address.
182 (ffi_call): Allocate a word-sized temporary for the case where
183 a composite is returned in r0.
184 (ffi_prep_incoming_args_SYSV): Align as necessary.
186 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
188 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
189 directly using the sys_cacheflush syscall.
191 2007-07-27 Andrew Haley <aph@redhat.com>
193 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
195 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
197 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
198 * configure.ac: Likewise.
199 * Makefile.in: Regenerate.
200 * include/Makefile.in: Likewise.
201 * testsuite/Makefile.in: Likewise.
202 * configure: Likewise.
204 2007-08-24 David Daney <ddaney@avtrex.com>
206 * testsuite/libffi.call/return_sl.c: New test.
208 2007-08-10 David Daney <ddaney@avtrex.com>
210 * testsuite/libffi.call/cls_multi_ushort.c,
211 testsuite/libffi.call/cls_align_uint16.c,
212 testsuite/libffi.call/nested_struct1.c,
213 testsuite/libffi.call/nested_struct3.c,
214 testsuite/libffi.call/cls_7_1_byte.c,
215 testsuite/libffi.call/nested_struct5.c,
216 testsuite/libffi.call/cls_double.c,
217 testsuite/libffi.call/nested_struct7.c,
218 testsuite/libffi.call/cls_sint.c,
219 testsuite/libffi.call/nested_struct9.c,
220 testsuite/libffi.call/cls_20byte1.c,
221 testsuite/libffi.call/cls_multi_sshortchar.c,
222 testsuite/libffi.call/cls_align_sint64.c,
223 testsuite/libffi.call/cls_3byte2.c,
224 testsuite/libffi.call/cls_multi_schar.c,
225 testsuite/libffi.call/cls_multi_uchar.c,
226 testsuite/libffi.call/cls_19byte.c,
227 testsuite/libffi.call/cls_9byte1.c,
228 testsuite/libffi.call/cls_align_float.c,
229 testsuite/libffi.call/closure_fn1.c,
230 testsuite/libffi.call/problem1.c,
231 testsuite/libffi.call/closure_fn3.c,
232 testsuite/libffi.call/cls_sshort.c,
233 testsuite/libffi.call/closure_fn5.c,
234 testsuite/libffi.call/cls_align_double.c,
235 testsuite/libffi.call/nested_struct.c,
236 testsuite/libffi.call/cls_2byte.c,
237 testsuite/libffi.call/nested_struct10.c,
238 testsuite/libffi.call/cls_4byte.c,
239 testsuite/libffi.call/cls_6byte.c,
240 testsuite/libffi.call/cls_8byte.c,
241 testsuite/libffi.call/cls_multi_sshort.c,
242 testsuite/libffi.call/cls_align_sint16.c,
243 testsuite/libffi.call/cls_align_uint32.c,
244 testsuite/libffi.call/cls_20byte.c,
245 testsuite/libffi.call/cls_float.c,
246 testsuite/libffi.call/nested_struct2.c,
247 testsuite/libffi.call/cls_5_1_byte.c,
248 testsuite/libffi.call/nested_struct4.c,
249 testsuite/libffi.call/cls_24byte.c,
250 testsuite/libffi.call/nested_struct6.c,
251 testsuite/libffi.call/cls_64byte.c,
252 testsuite/libffi.call/nested_struct8.c,
253 testsuite/libffi.call/cls_uint.c,
254 testsuite/libffi.call/cls_multi_ushortchar.c,
255 testsuite/libffi.call/cls_schar.c,
256 testsuite/libffi.call/cls_uchar.c,
257 testsuite/libffi.call/cls_align_uint64.c,
258 testsuite/libffi.call/cls_ulonglong.c,
259 testsuite/libffi.call/cls_align_longdouble.c,
260 testsuite/libffi.call/cls_1_1byte.c,
261 testsuite/libffi.call/cls_12byte.c,
262 testsuite/libffi.call/cls_3_1byte.c,
263 testsuite/libffi.call/cls_3byte1.c,
264 testsuite/libffi.call/cls_4_1byte.c,
265 testsuite/libffi.call/cls_6_1_byte.c,
266 testsuite/libffi.call/cls_16byte.c,
267 testsuite/libffi.call/cls_18byte.c,
268 testsuite/libffi.call/closure_fn0.c,
269 testsuite/libffi.call/cls_9byte2.c,
270 testsuite/libffi.call/closure_fn2.c,
271 testsuite/libffi.call/closure_fn4.c,
272 testsuite/libffi.call/cls_ushort.c,
273 testsuite/libffi.call/closure_fn6.c,
274 testsuite/libffi.call/cls_5byte.c,
275 testsuite/libffi.call/cls_align_pointer.c,
276 testsuite/libffi.call/cls_7byte.c,
277 testsuite/libffi.call/cls_align_sint32.c,
278 testsuite/libffi.special/unwindtest_ffi_call.cc,
279 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
281 2007-08-10 David Daney <ddaney@avtrex.com>
284 * configure.ac: Don't treat mips64 as a special case.
285 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
286 * configure: Regenerate
287 * Makefile.in: Ditto.
288 * fficonfig.h.in: Ditto.
289 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
290 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
291 (FFI_DEFAULT_ABI): Set for n64 case.
292 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
293 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
294 (ffi_closure_N32): New function.
295 (.eh_frame): New section
296 * src/mips/o32.S: Clean up comments.
297 (ffi_closure_O32): Pass ffi_closure parameter in $12.
298 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
299 _MIPS_SIM == _ABIN32 throughout.
300 (FFI_MIPS_STOP_HERE): New, use in place of
302 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
303 to support n32/n64 ABIs.
304 (calc_n32_struct_flags): Rewrite.
305 (calc_n32_return_struct_flags): Remove unused variable. Reverse
306 position of flag bits.
307 (ffi_prep_cif_machdep): Rewrite n32 portion.
308 (ffi_call): Enable for n64. Add special handling for small structure
310 (ffi_prep_closure_loc): Add n32 and n64 support.
311 (ffi_closure_mips_inner_O32): Add cast to silence warning.
312 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
314 2007-08-08 David Daney <ddaney@avtrex.com>
316 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
317 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
319 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
320 * testsuite/libffi.call/cls_sint.c (main): Ditto.
321 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
322 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
323 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
324 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
325 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
326 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
327 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
328 * testsuite/libffi.call/return_fl3.c (main): Ditto.
329 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
330 * testsuite/libffi.call/nested_struct.c (main): Ditto.
331 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
332 * testsuite/libffi.call/return_ll1.c (main): Ditto.
333 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
334 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
335 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
336 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
337 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
338 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
339 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
340 * testsuite/libffi.call/cls_uint.c (main): Ditto.
341 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
342 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
343 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
344 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
345 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
346 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
347 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
348 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
349 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
351 2007-08-07 Andrew Haley <aph@redhat.com>
353 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
356 2007-08-06 Andrew Haley <aph@redhat.com>
359 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
360 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
363 2007-08-02 David Daney <ddaney@avtrex.com>
365 * testsuite/libffi.call/return_ul.c (main): Define return type as
366 ffi_arg. Use proper printf conversion specifier.
368 2007-07-30 Andrew Haley <aph@redhat.com>
371 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
372 signed/unsigned int8/16.
373 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
375 Remove code to pop args from the stack after call.
376 Special-case signed/unsigned int8/16.
377 * testsuite/libffi.call/return_sc.c (main): Revert.
379 2007-07-26 Richard Guenther <rguenther@suse.de>
382 * testsuite/libffi.call/return_sc.c (main): Verify call
383 result as signed char, not ffi_arg.
385 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
387 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
388 * configure: Regenerate.
390 2007-07-11 David Daney <ddaney@avtrex.com>
392 * src/mips/ffi.c: Don't include sys/cachectl.h.
393 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
396 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
398 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
399 from (ffi_prep_closure): ... this.
400 (FFI_INIT_TRAMPOLINE): Adjust.
402 2005-12-31 Phil Blundell <pb@reciva.com>
404 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
405 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
406 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
407 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
408 (FFI_CLOSURES): Enable closure support.
410 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
412 * testsuite/libffi.call/cls_multi_ushort.c,
413 testsuite/libffi.call/cls_align_uint16.c,
414 testsuite/libffi.call/nested_struct1.c,
415 testsuite/libffi.call/nested_struct3.c,
416 testsuite/libffi.call/cls_7_1_byte.c,
417 testsuite/libffi.call/cls_double.c,
418 testsuite/libffi.call/nested_struct5.c,
419 testsuite/libffi.call/nested_struct7.c,
420 testsuite/libffi.call/cls_sint.c,
421 testsuite/libffi.call/nested_struct9.c,
422 testsuite/libffi.call/cls_20byte1.c,
423 testsuite/libffi.call/cls_multi_sshortchar.c,
424 testsuite/libffi.call/cls_align_sint64.c,
425 testsuite/libffi.call/cls_3byte2.c,
426 testsuite/libffi.call/cls_multi_schar.c,
427 testsuite/libffi.call/cls_multi_uchar.c,
428 testsuite/libffi.call/cls_19byte.c,
429 testsuite/libffi.call/cls_9byte1.c,
430 testsuite/libffi.call/cls_align_float.c,
431 testsuite/libffi.call/closure_fn1.c,
432 testsuite/libffi.call/problem1.c,
433 testsuite/libffi.call/closure_fn3.c,
434 testsuite/libffi.call/cls_sshort.c,
435 testsuite/libffi.call/closure_fn5.c,
436 testsuite/libffi.call/cls_align_double.c,
437 testsuite/libffi.call/cls_2byte.c,
438 testsuite/libffi.call/nested_struct.c,
439 testsuite/libffi.call/nested_struct10.c,
440 testsuite/libffi.call/cls_4byte.c,
441 testsuite/libffi.call/cls_6byte.c,
442 testsuite/libffi.call/cls_8byte.c,
443 testsuite/libffi.call/cls_multi_sshort.c,
444 testsuite/libffi.call/cls_align_uint32.c,
445 testsuite/libffi.call/cls_align_sint16.c,
446 testsuite/libffi.call/cls_float.c,
447 testsuite/libffi.call/cls_20byte.c,
448 testsuite/libffi.call/cls_5_1_byte.c,
449 testsuite/libffi.call/nested_struct2.c,
450 testsuite/libffi.call/cls_24byte.c,
451 testsuite/libffi.call/nested_struct4.c,
452 testsuite/libffi.call/nested_struct6.c,
453 testsuite/libffi.call/cls_64byte.c,
454 testsuite/libffi.call/nested_struct8.c,
455 testsuite/libffi.call/cls_uint.c,
456 testsuite/libffi.call/cls_multi_ushortchar.c,
457 testsuite/libffi.call/cls_schar.c,
458 testsuite/libffi.call/cls_uchar.c,
459 testsuite/libffi.call/cls_align_uint64.c,
460 testsuite/libffi.call/cls_ulonglong.c,
461 testsuite/libffi.call/cls_align_longdouble.c,
462 testsuite/libffi.call/cls_1_1byte.c,
463 testsuite/libffi.call/cls_12byte.c,
464 testsuite/libffi.call/cls_3_1byte.c,
465 testsuite/libffi.call/cls_3byte1.c,
466 testsuite/libffi.call/cls_4_1byte.c,
467 testsuite/libffi.call/cls_6_1_byte.c,
468 testsuite/libffi.call/cls_16byte.c,
469 testsuite/libffi.call/cls_18byte.c,
470 testsuite/libffi.call/closure_fn0.c,
471 testsuite/libffi.call/cls_9byte2.c,
472 testsuite/libffi.call/closure_fn2.c,
473 testsuite/libffi.call/closure_fn4.c,
474 testsuite/libffi.call/cls_ushort.c,
475 testsuite/libffi.call/closure_fn6.c,
476 testsuite/libffi.call/cls_5byte.c,
477 testsuite/libffi.call/cls_align_pointer.c,
478 testsuite/libffi.call/cls_7byte.c,
479 testsuite/libffi.call/cls_align_sint32.c,
480 testsuite/libffi.special/unwindtest_ffi_call.cc,
481 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
483 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
485 * aclocal.m4: Regenerated.
487 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
489 * configure: Regenerate.
491 2007-05-23 Steve Ellcey <sje@cup.hp.com>
493 * Makefile.in: Regenerate.
494 * configure: Regenerate.
495 * aclocal.m4: Regenerate.
496 * include/Makefile.in: Regenerate.
497 * testsuite/Makefile.in: Regenerate.
499 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
501 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
502 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
503 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
504 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
505 (FFI_CLOSURES): Enable closure support.
507 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
509 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
510 * configure: Regenerate.
511 * fficonfig.h.in: Regenerate.
512 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
513 CFI_OFFSET,CFI_DEF_CFA): New macros.
514 (ffi_call_SYSV): Add callframe annotation.
516 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
518 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
519 numerous test suite failures.
520 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
522 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
524 * Makefile.am (EXTRA_DIST): Bring up to date.
525 * Makefile.in: Regenerate.
526 * src/frv/eabi.S: Remove RCS keyword.
528 2007-04-06 Richard Henderson <rth@redhat.com>
530 * configure.ac: Tidy target case.
531 (HAVE_LONG_DOUBLE): Allow the target to override.
532 * configure: Regenerate.
533 * include/ffi.h.in: Don't define ffi_type_foo if
534 LIBFFI_HIDE_BASIC_TYPES is defined.
535 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
537 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
538 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
539 (ffi_type_longdouble): Special case for Alpha. Don't define
540 if long double == double.
542 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
543 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
544 (ffi_call, ffi_closure_osf_inner): Likewise.
545 (ffi_closure_osf_inner): Likewise. Mark hidden.
546 (ffi_call_osf, ffi_closure_osf): Mark hidden.
547 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
548 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
549 (load_table): Handle 128-bit long double.
551 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
553 2007-04-06 Tom Tromey <tromey@redhat.com>
556 * README: Fixed bug in example.
558 2007-04-03 Jakub Jelinek <jakub@redhat.com>
560 * src/closures.c: Include sys/statfs.h.
561 (_GNU_SOURCE): Define on Linux.
562 (FFI_MMAP_EXEC_SELINUX): Define.
563 (selinux_enabled): New variable.
564 (selinux_enabled_check): New function.
565 (is_selinux_enabled): Define.
568 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
570 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
571 Use 'volatile float sum' to create sum of floats to avoid false
572 negative due to excess precision on ix86 targets.
575 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
577 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
579 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
581 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
583 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
584 (ffi_prep_closure_loc): New.
585 (ffi_prep_raw_closure_loc): New.
586 (ffi_prep_java_raw_closure_loc): New.
587 * src/closures.c: New file.
588 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
589 Replace sflags with exec_offset.
590 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
591 sub_segment_exec_offset): New macros.
592 (get_segment_flags, set_segment_flags, check_segment_merge): New
594 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
595 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
596 destroy_mspace): Use new macros.
597 (sys_alloc): Silence warning.
598 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
599 * Makefile.in: Rebuilt.
600 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
601 terms of ffi_prep_closure_loc.
602 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
604 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
606 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
608 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
610 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
611 (ffi_prep_closure): ... this.
612 * src/pa/ffi.c: Likewise.
613 * src/cris/ffi.c: Likewise. Adjust.
614 * src/frv/ffi.c: Likewise.
615 * src/ia64/ffi.c: Likewise.
616 * src/mips/ffi.c: Likewise.
617 * src/powerpc/ffi_darwin.c: Likewise.
618 * src/s390/ffi.c: Likewise.
619 * src/sh/ffi.c: Likewise.
620 * src/sh64/ffi.c: Likewise.
621 * src/sparc/ffi.c: Likewise.
622 * src/x86/ffi64.c: Likewise.
623 * src/x86/ffi.c: Likewise.
624 (FFI_INIT_TRAMPOLINE): Adjust.
625 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
626 (ffi_prep_raw_closure): ... this.
627 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
628 (ffi_prep_closure): ... this.
629 (flush_icache): Adjust.
631 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
633 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
636 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
638 * Makefile.am: Add dummy install-pdf target.
639 * Makefile.in: Regenerate
641 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
643 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
644 ffi_closure_helper_SYSV): Add long double handling.
646 2007-02-02 Jakub Jelinek <jakub@redhat.com>
648 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
649 immediately after bctrl instruction.
651 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
653 * Makefile.am (all-recursive, install-recursive,
654 mostlyclean-recursive, clean-recursive, distclean-recursive,
655 maintainer-clean-recursive): Add missing targets.
656 * Makefile.in: Rebuilt.
658 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
660 * configure.ac: Add TARGET for x86_64-*-darwin*.
661 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
663 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
664 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
665 * configure: Regenerate.
666 * Makefile.in: Regenerate.
667 * include/Makefile.in: Regenerate.
668 * testsuite/Makefile.in: Regenerate.
669 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
672 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
674 * aclocal.m4: Regenerate with aclocal -I .. as written in the
677 2006-10-31 Geoffrey Keating <geoffk@apple.com>
679 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
680 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
682 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
683 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
685 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
686 Sandro Tolaini <tolaini@libero.it>
688 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
690 * configure: Regenerated.
691 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
692 (EXTRA_DIST): Add src/x86/darwin.S.
693 * Makefile.in: Regenerated.
694 * include/Makefile.in: Regenerated.
695 * testsuite/Makefile.in: Regenerated.
697 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
698 X86_WIN32, and additionally align stack to 16 bytes.
699 * src/x86/darwin.S: New, based on sysv.S.
700 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
702 2006-09-12 David Daney <ddaney@avtrex.com>
705 * include/Makefile.am: Install both ffi.h and ffitarget.h in
706 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
707 * aclocal.m4: Regenerated for automake 1.9.6.
708 * Makefile.in: Regenerated.
709 * include/Makefile.in: Regenerated.
710 * testsuite/Makefile.in: Regenerated.
712 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
714 * include/ffi_common.h (struct): Revert accidental commit.
716 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
718 * include/ffi_common.h: Remove lint directives.
719 * include/ffi.h.in: Likewise.
721 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
723 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
724 for 32-bit architectures.
725 * testsuite/libffi.call/return_ul.c: New test case.
727 2006-07-19 David Daney <ddaney@avtrex.com>
729 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
730 xfail remains for mips64.
732 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
734 * Makefile.am: Add install-html target. Add install-html to .PHONY
735 * Makefile.in: Regenerate.
736 * aclocal.m4: Regenerate.
737 * include/Makefile.in: Regenerate.
738 * testsuite/Makefile.in: Regenerate.
740 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
742 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
745 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
747 * README: Remove notice about 'Crazy Comments'.
748 * src/debug.c: Remove lint directives. Cleanup white spaces.
749 * src/java_raw_api.c: Likewise.
750 * src/prep_cif.c: Likewise.
751 * src/raw_api.c: Likewise.
752 * src/ffitest.c: Delete. No longer needed, all test cases migrated
754 * src/arm/ffi.c: Remove lint directives.
755 * src/m32r/ffi.c: Likewise.
756 * src/pa/ffi.c: Likewise.
757 * src/powerpc/ffi.c: Likewise.
758 * src/powerpc/ffi_darwin.c: Likewise.
759 * src/sh/ffi.c: Likewise.
760 * src/sh64/ffi.c: Likewise.
761 * src/x86/ffi.c: Likewise.
762 * testsuite/libffi.call/float2.c: Likewise.
763 * testsuite/libffi.call/promotion.c: Likewise.
764 * testsuite/libffi.call/struct1.c: Likewise.
766 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
768 * src/pa/hpux32.S: Correct unwind offset calculation for
770 * src/pa/linux.S: Likewise.
772 2006-04-12 James E Wilson <wilson@specifix.com>
775 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
776 (hfa_type_load): Call stf_spill.
777 (hfa_type_store): Call ldf_fill.
778 (ffi_call): Adjust calls to above routines. Add local temps for
781 2006-04-10 Matthias Klose <doko@debian.org>
783 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
784 directory names containing underscores.
786 2006-04-07 James E Wilson <wilson@specifix.com>
788 * testsuite/libffi.call/float4.c: New testcase.
790 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
791 Andreas Tobler <a.tobler@schweiz.ch>
793 * Makefile.am: Add PA_HPUX port.
794 * Makefile.in: Regenerate.
795 * include/Makefile.in: Likewise.
796 * testsuite/Makefile.in: Likewise.
797 * configure.ac: Add PA_HPUX rules.
798 * configure: Regenerate.
799 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
800 Add PA_HPUX and PA64_HPUX.
801 Rename FFI_LINUX ABI to FFI_PA32 ABI.
802 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
803 (FFI_TYPE_SMALL_STRUCT2): Define.
804 (FFI_TYPE_SMALL_STRUCT4): Likewise.
805 (FFI_TYPE_SMALL_STRUCT8): Likewise.
806 (FFI_TYPE_SMALL_STRUCT3): Redefine.
807 (FFI_TYPE_SMALL_STRUCT5): Likewise.
808 (FFI_TYPE_SMALL_STRUCT6): Likewise.
809 (FFI_TYPE_SMALL_STRUCT7): Likewise.
810 * src/pa/ffi.c (ROUND_DOWN): Delete.
811 (fldw, fstw, fldd, fstd): Use '__asm__'.
812 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
813 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
814 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
815 Simplify incrementing of stack slot variable. Change type of local
817 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
819 (ffi_prep_cif_machdep): Likewise.
820 (ffi_call): Likewise.
821 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
822 return type to ffi_status. Simplify incrementing of stack slot
823 variable. Only copy floating point argument registers when PA_LINUX
824 is true. Reformat debug statement.
825 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
826 FFI_TYPE_SMALL_STRUCT8.
827 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
829 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
830 Add nops to cache flush. Add trampoline for PA_HPUX.
831 * src/pa/hpux32.S: New file.
832 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
833 ffi_prep_args_LINUX to ffi_prep_args_pa32.
834 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
835 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
836 argument type checks so that common argument types appear first.
837 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
838 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
840 2006-03-24 Alan Modra <amodra@bigpond.net.au>
842 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
843 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
844 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
846 (ffi_prep_args64): Assert using IBM extended double.
847 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
848 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
849 (ffi_call): Handle FFI_LINUX.
850 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
851 gpr3 return pointer as for struct return. Handle FFI_LINUX
852 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
854 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
855 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
856 Don't use r6 as pointer to results, instead use sp offset. Don't
857 make a special call to load lr with case table address, instead
858 use offset from previous call.
859 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
860 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
863 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
865 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
866 passed with FP registers correctly.
867 (ffi_closure_helper_SYSV): Likewise.
868 * src/sh64/sysv.S: Likewise.
870 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
872 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
873 args and userdata unused.
874 (closure_test_fn1): Mark cif and userdata unused.
875 (main): Remove unused res.
877 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
879 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
880 -O2, -O3, -Os and the warning flags -W -Wall.
881 * testsuite/libffi.special/special.exp: Likewise.
882 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
883 unused parameter unused for gcc or else do nothing.
884 * testsuite/libffi.special/ffitestcxx.h: Likewise.
885 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
887 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
888 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
889 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
890 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
891 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
892 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
893 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
894 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
895 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
896 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
897 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
898 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
899 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
900 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
901 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
902 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
903 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
904 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
905 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
906 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
907 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
908 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
909 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
910 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
912 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
914 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
916 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
917 void* to avoid compiler warning.
919 (cls_struct_align_gn): Mark cif and userdata unused.
920 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
922 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
924 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
926 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
928 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
930 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
931 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
932 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
934 (main): Cast res_call to silence gcc.
935 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
937 (main): Cast res_call to silence gcc.
938 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
940 (main): Cast res_call to silence gcc.
941 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
943 (main): Cast res_call to silence gcc.
944 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
946 (main): Cast res_call to silence gcc.
947 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
949 (main): Cast res_call to silence gcc.
950 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
952 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
953 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
955 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
956 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
958 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
959 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
961 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
962 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
964 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
965 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
967 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
969 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
970 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
971 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
972 Cleanup white spaces.
973 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
974 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
975 cif and userdata unused.
976 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
978 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
979 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
980 formatters to silence gcc.
981 (B_gn): Mark cif and userdata unused.
982 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
984 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
985 (B_gn): Mark cif and userdata unused.
986 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
988 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
989 (B_gn): Mark cif and userdata unused.
990 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
992 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
993 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
994 * testsuite/libffi.call/problem1.c (stub): Likewise.
995 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
997 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
998 in the last commit for this test case in the test case itself.
999 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1001 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1002 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1003 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1004 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1005 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1006 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1008 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1010 * src/sh/sysv.S: Fix register numbers in the FDE for
1013 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1015 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1016 declaration to avoid a false negative on ix86. See PR323.
1018 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1020 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1021 and cast integer to void * if needed. Update the pointer to
1022 the FP register saved area correctly.
1024 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1026 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1028 * testsuite/libffi.call/nested_struct4.c: Likewise.
1030 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1032 * testsuite/libffi.call/return_dbl.c: New test case.
1033 * testsuite/libffi.call/return_dbl1.c: Likewise.
1034 * testsuite/libffi.call/return_dbl2.c: Likewise.
1035 * testsuite/libffi.call/return_fl.c: Likewise.
1036 * testsuite/libffi.call/return_fl1.c: Likewise.
1037 * testsuite/libffi.call/return_fl2.c: Likewise.
1038 * testsuite/libffi.call/return_fl3.c: Likewise.
1039 * testsuite/libffi.call/closure_fn6.c: Likewise.
1041 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1043 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1044 here to be used by other test cases too.
1046 * testsuite/libffi.call/nested_struct10.c: New test case.
1047 * testsuite/libffi.call/nested_struct9.c: Likewise.
1048 * testsuite/libffi.call/nested_struct8.c: Likewise.
1049 * testsuite/libffi.call/nested_struct7.c: Likewise.
1050 * testsuite/libffi.call/nested_struct6.c: Likewise.
1051 * testsuite/libffi.call/nested_struct5.c: Likewise.
1052 * testsuite/libffi.call/nested_struct4.c: Likewise.
1054 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1056 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1057 * configure: Rebuilt.
1059 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1061 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1062 instead do the shifting inline.
1063 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1064 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1065 and 8 byte structs, for the remaining struct sizes don't call
1066 __lshrdi3, instead do the shifting inline.
1068 2005-12-07 Thiemo Seufer <ths@networkno.de>
1070 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1071 missing parentheses.
1072 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1073 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1074 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1075 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1076 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1078 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1080 (ffi_prep_closure): Improve trampoline instruction scheduling.
1081 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1083 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1085 * src/powerpc/ffi.c: Formatting.
1086 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1087 (ffi_prep_args64): Likewise.
1089 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1091 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1092 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1094 2005-09-26 Tom Tromey <tromey@redhat.com>
1096 * testsuite/libffi.call/float1.c (value_type): New typedef.
1097 (CANARY): New define.
1098 (main): Check for result buffer overflow.
1099 * src/powerpc/linux64.S: Handle linux64 long double returns.
1100 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1101 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1103 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1106 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1108 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1110 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1112 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1113 (AH_BOTTOM): Add FFI_HIDDEN definition.
1114 * configure: Rebuilt.
1115 * fficonfig.h.in: Rebuilt.
1116 * src/powerpc/ffi.c (hidden): Remove.
1117 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1118 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1119 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1120 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1121 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1122 add FFI_HIDDEN to its prototype.
1123 (ffi_closure_SYSV_inner): New.
1124 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1125 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1127 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1130 * configure: Rebuilt.
1131 * configure.ac: Handle i*86-*-gnu*.
1133 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1135 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1136 DW_CFA_offset_extended_sf rather than
1137 DW_CFA_GNU_negative_offset_extended.
1138 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1140 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1142 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1144 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1145 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1146 partially on register.
1147 (ffi_closure_helper_SYSV): Likewise.
1148 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1150 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1152 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1154 * src/sh64/ffi.c (simple_type): Remove.
1155 (return_type): Handle small structures correctly.
1156 (ffi_prep_args): Likewise.
1157 (ffi_call): Likewise.
1158 (ffi_closure_helper_SYSV): Likewise.
1159 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1160 Emit position independent code if PIC and remove wrong datalabel
1161 prefixes from EH data.
1163 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1165 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1166 * Makefile.in: Regenerate.
1167 * include/Makefile.in: Likewise.
1168 * testsuite/Makefile.in: Likewise.
1169 * configure.ac: Add POWERPC_FREEBSD rules.
1170 * configure: Regenerate.
1171 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1172 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1173 * src/powerpc/ffi.c: Add flags to handle small structure returns
1175 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1177 (ffi_closure_helper_SYSV): Likewise.
1178 * src/powerpc/ppc_closure.S: Add return types for small structures.
1179 * src/powerpc/sysv.S: Add bits to handle small structures for
1182 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1184 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1185 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1186 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1188 2005-07-05 Randolph Chung <tausq@debian.org>
1190 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1191 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1192 structures. Kill compilation warnings.
1193 (ffi_closure_inner_LINUX): Print return values as hex in debug
1194 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1195 Properly handle 5-7 byte structure returns.
1196 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1197 (FFI_TYPE_SMALL_STRUCT2): Remove.
1198 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1199 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1200 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1201 (checksmst1, checksmst2): Remove.
1202 (checksmst3): Optimize handling of 3-byte struct returns.
1203 (checksmst567): Properly handle 5-7 byte struct returns.
1205 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1208 * src/mips/n32.S: Enforce PIC code.
1209 * src/mips/o32.S: Likewise.
1211 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1213 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1214 * configure: Regenerate.
1216 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1218 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1219 to call ffi_closure_helper_SYSV. Append @local instead.
1220 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1222 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1224 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1225 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1226 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1227 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1228 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1230 2005-05-09 Mike Stump <mrs@apple.com>
1232 * configure: Regenerate.
1234 2005-05-08 Richard Henderson <rth@redhat.com>
1237 * src/alpha/osf.S: Update unwind into to match code.
1239 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1240 Richard Henderson <rth@redhat.com>
1242 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1244 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1245 (ffi_prep_closure): Set carry bit if sse-used flag set.
1246 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1247 Only load sse registers if ssecount non-zero.
1248 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1250 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1252 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1253 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1254 * configure: Regenerate.
1256 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1258 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1259 have Tcl8.3-compatible intermediate variable.
1261 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1262 Hans-Peter Nilsson <hp@axis.com>
1264 * Makefile.am: Add CRIS support.
1265 * configure.ac: Likewise.
1266 * Makefile.in, configure, testsuite/Makefile.in,
1267 include/Makefile.in: Regenerate.
1268 * src/cris: New directory.
1269 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1270 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1272 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1273 \r?\n in output tests.
1275 2005-04-12 Mike Stump <mrs@apple.com>
1277 * configure: Regenerate.
1279 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1281 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1283 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1285 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1287 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1288 to 64 bits in ILP32 mode.
1289 Load 64 bits even for short data.
1291 2005-03-23 Mike Stump <mrs@apple.com>
1293 * src/powerpc/darwin.S: Update for -m64 multilib.
1294 * src/powerpc/darwin_closure.S: Likewise.
1296 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1298 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1299 Do not set tool_include_dir.
1300 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1302 * include/Makefile.am: Set gcc_version and toollibffidir.
1303 * include/Makefile.in: Regenerate.
1305 2005-02-22 Andrew Haley <aph@redhat.com>
1307 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1308 odd-numbered register pairs for 64-bit integer types.
1310 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1313 * testsuite/libffi.call/return_ll1.c: New test case.
1315 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1317 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1318 * testsuite/libffi.call/float.c: Ditto.
1319 * testsuite/libffi.call/float2.c: Ditto.
1320 * testsuite/libffi.call/float3.c: Ditto.
1322 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1324 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1326 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1328 * testsuite/libffi.special/special.exp (cxx_options): Add
1331 2004-12-31 Richard Henderson <rth@redhat.com>
1333 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1334 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1335 offset parameters with a type parameter; deduce size and structure
1336 alignment. Update all users.
1338 2004-12-31 Richard Henderson <rth@redhat.com>
1340 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1341 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1342 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1343 into ffi_prep_closure.
1344 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1347 2004-12-27 Richard Henderson <rth@redhat.com>
1349 * src/x86/unix64.S: Fix typo in unwind info.
1351 2004-12-25 Richard Henderson <rth@redhat.com>
1353 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1354 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1355 (merge_classes): Check for it.
1357 (classify_argument): Pass byte_offset by value; perform all updates
1359 (examine_argument): Add classes argument; handle
1360 X86_64_COMPLEX_X87_CLASS.
1361 (ffi_prep_args): Merge into ...
1362 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1363 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1364 (ffi_fill_return_value): Remove.
1365 (ffi_prep_closure): Remove dead assert.
1366 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1367 Rewrite to use struct register_args instead of va_list. Create
1368 flags for handling structure returns.
1369 * src/x86/unix64.S: Remove dead strings.
1370 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1371 stack frame with ffi_call. Handle structure returns properly.
1372 (float2sse, floatfloat2sse, double2sse): Remove.
1373 (sse2float, sse2double, sse2floatfloat): Remove.
1374 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1375 to handle structure returns properly.
1377 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1379 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1381 * Makefile.in: Regenerated.
1383 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1385 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1386 * configure, aclocal.m4, Makefile.in: Regenerate.
1387 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1389 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1391 * configure: Regenerate for libtool change.
1393 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1395 * configure: Regenerate for libtool reversion.
1397 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1399 * configure: Regenerate for libtool change.
1401 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1403 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1405 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1407 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1408 of jal. Use an absolute encoding for the frame information.
1410 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1412 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1413 * acinclude.m4: Delete logic for sincludes.
1414 * aclocal.m4, Makefile.in, configure: Regenerate.
1415 * include/Makefile: Likewise.
1416 * testsuite/Makefile: Likewise.
1418 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1420 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1421 on a 8-byte boundary.
1422 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1424 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1426 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1427 long long values. Round stack allocation to a multiple of 8 bytes
1428 for ATPCS compatibility.
1429 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1430 names. Handle returning long long types. Add Thumb and interworking
1431 support. Improve soft-float code.
1433 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1435 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1436 (libffi_exit): New function.
1437 (libffi_init): Build the testglue wrapper if needed.
1439 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1442 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1444 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1446 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1448 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1450 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1451 * testsuite/libffi.call/float3.c: New test case.
1453 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1455 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1456 the function returning a structure pointed with R2.
1457 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1458 the structure return value if T bit set. Emit position
1459 independent code and EH data if PIC.
1461 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1463 * Makefile.am: Add m32r support.
1464 * configure.ac: Likewise.
1465 * Makefile.in: Regenerate.
1466 * confiugre: Regenerate.
1467 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1468 (uint64, sint64, double, longdouble)
1469 * src/m32r: New directory.
1470 * src/m32r/ffi.c: New file.
1471 * src/m32r/sysv.S: Likewise.
1472 * src/m32r/ffitarget.h: Likewise.
1474 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1476 * testsuite/libffi.call/negint.c: New test case.
1478 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1481 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1482 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1483 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1486 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1488 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1489 * testsuite/libffi.call/promotion.c: Likewise.
1490 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1492 * testsuite/libffi.call/return_sc.c: Likewise.
1493 * testsuite/libffi.call/return_uc.c: Likewise.
1495 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1497 * src/powerpc/darwin.S: Fix comments and identation.
1498 * src/powerpc/darwin_closure.S: Likewise.
1500 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1502 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1503 (ffi_prep_args): Handle longdouble arguments.
1504 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1506 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1507 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1509 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1510 * src/types.c: Defined longdouble size and alignment for darwin.
1512 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1514 * src/powerpc/aix.S: Remove whitespaces.
1515 * src/powerpc/aix_closure.S: Likewise.
1516 * src/powerpc/asm.h: Likewise.
1517 * src/powerpc/ffi.c: Likewise.
1518 * src/powerpc/ffitarget.h: Likewise.
1519 * src/powerpc/linux64.S: Likewise.
1520 * src/powerpc/linux64_closure.S: Likewise.
1521 * src/powerpc/ppc_closure.S: Likewise.
1522 * src/powerpc/sysv.S: Likewise.
1524 2004-08-30 Anthony Green <green@redhat.com>
1526 * Makefile.am: Add frv support.
1527 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1528 * configure.ac: Read configure.host.
1529 * configure.in: Read configure.host.
1530 * configure.host: New file. frv-elf needs libgloss.
1531 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1532 alignment. This is needed to frv and shouldn't harm the others.
1533 * include/ffi_common.h (ALIGN_DOWN): New macro.
1534 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1536 2004-08-24 David Daney <daney@avtrex.com>
1538 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1539 * testsuite/libffi.call/closure_fn1.c: Likewise.
1540 * testsuite/libffi.call/closure_fn2.c Likewise.
1541 * testsuite/libffi.call/closure_fn3.c: Likewise.
1542 * testsuite/libffi.call/closure_fn4.c: Likewise.
1543 * testsuite/libffi.call/closure_fn5.c: Likewise.
1544 * testsuite/libffi.call/cls_18byte.c: Likewise.
1545 * testsuite/libffi.call/cls_19byte.c: Likewise.
1546 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1547 * testsuite/libffi.call/cls_20byte.c: Likewise.
1548 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1549 * testsuite/libffi.call/cls_24byte.c: Likewise.
1550 * testsuite/libffi.call/cls_2byte.c: Likewise.
1551 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1552 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1553 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1554 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1555 * testsuite/libffi.call/cls_4byte.c: Likewise.
1556 * testsuite/libffi.call/cls_64byte.c: Likewise.
1557 * testsuite/libffi.call/cls_6byte.c: Likewise.
1558 * testsuite/libffi.call/cls_7byte.c: Likewise.
1559 * testsuite/libffi.call/cls_8byte.c: Likewise.
1560 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1561 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1562 * testsuite/libffi.call/cls_align_double.c: Likewise.
1563 * testsuite/libffi.call/cls_align_float.c: Likewise.
1564 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1565 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1566 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1567 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1568 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1569 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1570 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1571 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1572 * testsuite/libffi.call/cls_double.c: Likewise.
1573 * testsuite/libffi.call/cls_float.c: Likewise.
1574 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1575 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1576 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1577 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1578 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1579 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1580 * testsuite/libffi.call/cls_schar.c: Likewise.
1581 * testsuite/libffi.call/cls_sint.c: Likewise.
1582 * testsuite/libffi.call/cls_sshort.c: Likewise.
1583 * testsuite/libffi.call/cls_uchar.c: Likewise.
1584 * testsuite/libffi.call/cls_uint.c: Likewise.
1585 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1586 * testsuite/libffi.call/cls_ushort.c: Likewise.
1587 * testsuite/libffi.call/nested_struct.c: Likewise.
1588 * testsuite/libffi.call/nested_struct1.c: Likewise.
1589 * testsuite/libffi.call/nested_struct2.c: Likewise.
1590 * testsuite/libffi.call/nested_struct3.c: Likewise.
1591 * testsuite/libffi.call/problem1.c: Likewise.
1592 * testsuite/libffi.special/unwindtest.cc: Likewise.
1593 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1595 * testsuite/libffi.call/cls_16byte.c: Likewise.
1596 * testsuite/libffi.call/cls_5byte.c: Likewise.
1598 2004-08-23 David Daney <daney@avtrex.com>
1601 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1602 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1603 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1604 parameters and return types.
1605 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1606 (ffi_prep_closure): Ditto.
1607 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1608 alignment calculations.
1609 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1610 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1612 2004-08-14 Casey Marshall <csm@gnu.org>
1614 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1615 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1616 integer (O32 ABI only).
1617 (ffi_prep_closure): new function.
1618 (ffi_closure_mips_inner_O32): new function.
1619 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1620 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1621 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1622 64 bit integers correctly.
1623 (ffi_closure_O32): new function.
1624 Added DWARF-2 unwind info for both functions.
1626 2004-08-10 Andrew Haley <aph@redhat.com>
1628 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1630 2004-08-01 Robert Millan <robertmh@gnu.org>
1632 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1633 * configure: Regenerate.
1635 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1637 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1638 and mmap() explicitly instead of relying on preset autoconf cache
1640 * aclocal.m4: Regenerate.
1641 * configure: Regenerate.
1643 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1645 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1646 (ffi_check_float_struct): Remove unused prototype.
1648 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1650 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1651 character on Darwin, use '\n\t' instead.
1653 2004-06-26 Matthias Klose <doko@debian.org>
1655 * libtool-version: Fix typo in revision/age.
1657 2004-06-17 Matthias Klose <doko@debian.org>
1659 * libtool-version: New.
1660 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1661 * Makefile.in: Regenerate.
1663 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1665 * Makefile.am: Remove useless multilib rules.
1666 * Makefile.in: Regenerate.
1667 * aclocal.m4: Regenerate with automake 1.8.5.
1668 * configure.ac: Remove useless multilib configury.
1669 * configure: Regenerate.
1671 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1673 * .cvsignore: New file.
1675 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1677 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1679 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1682 2004-06-06 Sean McNeil <sean@mcneil.com>
1684 * configure.ac: Add x86_64-*-freebsd* support.
1685 * configure: Regenerate.
1687 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1690 * configure.ac: Test for existence of mmap and sys/mman.h before
1691 checking blacklist. Fix suggested by Jim Wilson.
1692 * configure: Regenerate.
1694 2004-04-26 Matt Austern <austern@apple.com>
1696 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1698 * src/powerpc/darwin_closure.S: Likewise.
1700 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1702 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1703 error. Reported by Thomas Heller <theller@python.net>.
1704 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1705 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1707 2004-03-20 Matthias Klose <doko@debian.org>
1709 * src/pa/linux.S: Fix typo.
1711 2004-03-19 Matthias Klose <doko@debian.org>
1713 * Makefile.am: Update.
1714 * Makefile.in: Regenerate.
1715 * src/pa/ffi.h.in: Remove.
1716 * src/pa/ffitarget.h: New file.
1718 2004-02-10 Randolph Chung <tausq@debian.org>
1720 * Makefile.am: Add PA support.
1721 * Makefile.in: Regenerate.
1722 * include/Makefile.in: Regenerate.
1723 * configure.ac: Add PA target.
1724 * configure: Regenerate.
1725 * src/pa/ffi.c: New file.
1726 * src/pa/ffi.h.in: Add PA support.
1727 * src/pa/linux.S: New file.
1728 * prep_cif.c: Add PA support.
1730 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1732 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1734 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1735 with ecif->cif->flags.
1736 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1738 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1739 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1740 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1741 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1743 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1745 * configure.in: Rename file to ...
1746 * configure.ac: ... this.
1747 * fficonfig.h.in: Regenerate.
1748 * Makefile.in: Regenerate.
1749 * include/Makefile.in: Regenerate.
1750 * testsuite/Makefile.in: Regenerate.
1752 2004-03-12 Matt Austern <austern@apple.com>
1754 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1755 changes in EH format resulting from addition of linkonce support.
1756 * src/powerpc/darwin_closure.S: Likewise.
1758 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1759 Paolo Bonzini <bonzini@gnu.org>
1761 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1762 Remove VPATH. Remove rules for object files. Remove multilib support.
1763 (AM_CCASFLAGS): Add.
1764 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1765 (AC_PREREQ): Bump version to 2.59.
1766 (AC_INIT): Fill with version info and bug address.
1767 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1768 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1769 De-precious CC so that the right flags are passed down to multilibs.
1770 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1771 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1772 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1773 * configure: Rebuilt.
1774 * aclocal.m4: Likewise.
1775 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1776 * fficonfig.h.in: Likewise.
1778 2004-03-11 Andreas Schwab <schwab@suse.de>
1780 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1781 arguments from fp registers only for the first 8 parameter slots.
1782 Don't convert a float parameter when passed in memory.
1784 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1786 * configure: Regenerate for config/accross.m4 correction.
1788 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1790 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1791 ecif->cif->bytes to bytes.
1792 (ffi_prep_cif_machdep): Add braces around nested if statement.
1794 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1796 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1798 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1799 (ffi_closure_helper_LINUX64): Fix typo.
1800 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1802 * testsuite/libffi.call/float.c: Likewise.
1803 * testsuite/libffi.call/float2.c: Likewise.
1805 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1807 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1808 long double function return and long double arg handling.
1809 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1810 Use "end_pfr" instead of "nf". Correct long double handling.
1812 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1814 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1815 space for long double return value. Adjust stack frame and offsets.
1816 Load f2 long double return.
1818 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1820 * src/types.c: Use 16 byte long double for POWERPC64.
1822 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1824 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1825 when the structure return address is passed in %o0.
1826 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1827 (ffi_v9_layout_struct): Align the field following a nested structure
1828 on a word boundary. Use memmove instead of memcpy.
1829 (ffi_call): Update call to ffi_V9_return_struct.
1830 (ffi_prep_closure): Define 'ctx' only for V8.
1831 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1832 and ffi_closure_sparc_inner_v9.
1833 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1834 Always skip the structure return address. For structures and long
1835 doubles, copy the argument directly.
1836 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1837 if required. Shift the maximum floating-point slot accordingly. For
1838 big structures, copy the argument directly; otherwise, left-justify the
1839 argument and call ffi_v9_layout_struct to lay out the structure on
1841 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1842 (ffi_closure_v8): Pass the structure return address. Update call to
1843 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1844 Skip the 'unimp' insn when returning long doubles and structures.
1845 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1846 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1847 FFI_TYPE_INT handling. Load structures both in integers and
1848 floating-point registers on return.
1849 * README: Update status of the SPARC port.
1851 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1853 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1855 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1857 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1859 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1860 value as of type ffi_arg, not unsigned int.
1862 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1864 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1867 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1869 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1872 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1874 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1877 2003-12-10 Richard Henderson <rth@redhat.com>
1879 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1880 size_t instead of int.
1882 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1884 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1885 * testsuite/libffi.call/many_win32.c (main): Replace variable
1886 int i with unsigned long ul.
1888 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1889 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1890 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1891 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1892 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1893 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1894 * testsuite/libffi.call/cls_align_float.c: Likewise.
1895 * testsuite/libffi.call/cls_align_double.c: Likewise.
1896 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1897 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1899 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1902 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1903 Align arguments to 32 bits.
1905 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1908 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1909 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1910 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1911 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1912 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1913 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1915 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1917 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1919 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1920 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1922 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1924 * Makefile.in: Rebuilt.
1925 * configure: Likewise.
1926 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1929 2003-11-21 Andreas Jaeger <aj@suse.de>
1930 Andreas Tobler <a.tobler@schweiz.ch>
1932 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1933 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1934 * Makefile.in: Rebuilt.
1935 * aclocal.m4: Likewise.
1936 * configure: Likewise.
1937 * fficonfig.h.in: Likewise.
1938 * testsuite/lib/libffi-dg.exp: Add include dir.
1939 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1940 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1941 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1942 for ffi_closure if available.
1943 * testsuite/libffi.call/closure_fn1.c: Likewise.
1944 * testsuite/libffi.call/closure_fn2.c: Likewise.
1945 * testsuite/libffi.call/closure_fn3.c: Likewise.
1946 * testsuite/libffi.call/closure_fn4.c: Likewise.
1947 * testsuite/libffi.call/closure_fn5.c: Likewise.
1948 * testsuite/libffi.call/cls_12byte.c: Likewise.
1949 * testsuite/libffi.call/cls_16byte.c: Likewise.
1950 * testsuite/libffi.call/cls_18byte.c: Likewise.
1951 * testsuite/libffi.call/cls_19byte.c: Likewise.
1952 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1953 * testsuite/libffi.call/cls_20byte.c: Likewise.
1954 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1955 * testsuite/libffi.call/cls_24byte.c: Likewise.
1956 * testsuite/libffi.call/cls_2byte.c: Likewise.
1957 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1958 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1959 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1960 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1961 * testsuite/libffi.call/cls_4byte.c: Likewise.
1962 * testsuite/libffi.call/cls_5byte.c: Likewise.
1963 * testsuite/libffi.call/cls_64byte.c: Likewise.
1964 * testsuite/libffi.call/cls_6byte.c: Likewise.
1965 * testsuite/libffi.call/cls_7byte.c: Likewise.
1966 * testsuite/libffi.call/cls_8byte.c: Likewise.
1967 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1968 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1969 * testsuite/libffi.call/cls_double.c: Likewise.
1970 * testsuite/libffi.call/cls_float.c: Likewise.
1971 * testsuite/libffi.call/cls_schar.c: Likewise.
1972 * testsuite/libffi.call/cls_sint.c: Likewise.
1973 * testsuite/libffi.call/cls_sshort.c: Likewise.
1974 * testsuite/libffi.call/cls_uchar.c: Likewise.
1975 * testsuite/libffi.call/cls_uint.c: Likewise.
1976 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1977 * testsuite/libffi.call/cls_ushort.c: Likewise.
1978 * testsuite/libffi.call/nested_struct.c: Likewise.
1979 * testsuite/libffi.call/nested_struct1.c: Likewise.
1980 * testsuite/libffi.call/nested_struct2.c: Likewise.
1981 * testsuite/libffi.call/nested_struct3.c: Likewise.
1982 * testsuite/libffi.call/problem1.c: Likewise.
1983 * testsuite/libffi.special/unwindtest.cc: Likewise.
1985 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
1987 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
1989 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
1991 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
1992 Add -lgcc_s to additional flags.
1994 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
1996 * configure.in, include/Makefile.am: PR libgcj/11147, install
1997 the ffitarget.h header file in a gcc versioned and target
1999 * configure: Regenerated.
2000 * Makefile.in, include/Makefile.in: Likewise.
2001 * testsuite/Makefile.in: Likewise.
2003 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2005 * testsuite/libffi.call/closure_fn0.c: Print result and check
2006 with dg-output to make debugging easier.
2007 * testsuite/libffi.call/closure_fn1.c: Likewise.
2008 * testsuite/libffi.call/closure_fn2.c: Likewise.
2009 * testsuite/libffi.call/closure_fn3.c: Likewise.
2010 * testsuite/libffi.call/closure_fn4.c: Likewise.
2011 * testsuite/libffi.call/closure_fn5.c: Likewise.
2012 * testsuite/libffi.call/cls_12byte.c: Likewise.
2013 * testsuite/libffi.call/cls_16byte.c: Likewise.
2014 * testsuite/libffi.call/cls_18byte.c: Likewise.
2015 * testsuite/libffi.call/cls_19byte.c: Likewise.
2016 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2017 * testsuite/libffi.call/cls_20byte.c: Likewise.
2018 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2019 * testsuite/libffi.call/cls_24byte.c: Likewise.
2020 * testsuite/libffi.call/cls_2byte.c: Likewise.
2021 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2022 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2023 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2024 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2025 * testsuite/libffi.call/cls_4byte.c: Likewise.
2026 * testsuite/libffi.call/cls_5byte.c: Likewise.
2027 * testsuite/libffi.call/cls_64byte.c: Likewise.
2028 * testsuite/libffi.call/cls_6byte.c: Likewise.
2029 * testsuite/libffi.call/cls_7byte.c: Likewise.
2030 * testsuite/libffi.call/cls_8byte.c: Likewise.
2031 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2032 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2033 * testsuite/libffi.call/cls_double.c: Likewise.
2034 * testsuite/libffi.call/cls_float.c: Likewise.
2035 * testsuite/libffi.call/cls_schar.c: Likewise.
2036 * testsuite/libffi.call/cls_sint.c: Likewise.
2037 * testsuite/libffi.call/cls_sshort.c: Likewise.
2038 * testsuite/libffi.call/cls_uchar.c: Likewise.
2039 * testsuite/libffi.call/cls_uint.c: Likewise.
2040 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2041 * testsuite/libffi.call/cls_ushort.c: Likewise.
2042 * testsuite/libffi.call/problem1.c: Likewise.
2044 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2047 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2049 * testsuite/libffi.call/cls_9byte2.c: New test case.
2050 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2051 * testsuite/libffi.call/cls_64byte.c: Likewise.
2052 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2053 * testsuite/libffi.call/cls_19byte.c: Likewise.
2054 * testsuite/libffi.call/cls_18byte.c: Likewise.
2055 * testsuite/libffi.call/closure_fn4.c: Likewise.
2056 * testsuite/libffi.call/closure_fn5.c: Likewise.
2057 * testsuite/libffi.call/cls_schar.c: Likewise.
2058 * testsuite/libffi.call/cls_sint.c: Likewise.
2059 * testsuite/libffi.call/cls_sshort.c: Likewise.
2060 * testsuite/libffi.call/nested_struct2.c: Likewise.
2061 * testsuite/libffi.call/nested_struct3.c: Likewise.
2063 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2065 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2066 * testsuite/libffi.call/cls_uchar.c: Likewise.
2067 * testsuite/libffi.call/cls_uint.c: Likewise.
2068 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2069 * testsuite/libffi.call/cls_ushort.c: Likewise.
2070 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2072 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2074 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2077 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2079 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2080 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2082 2003-10-22 David Daney <ddaney@avtrex.com>
2084 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2085 __attribute__((__mode__(__SI__))) and friends.
2087 2003-10-22 Andreas Schwab <schwab@suse.de>
2089 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2091 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2093 * configure.in: AC_LINK_FILES(ffitarget.h).
2094 * configure: Regenerate.
2095 * Makefile.in: Likewise.
2096 * include/Makefile.in: Likewise.
2097 * testsuite/Makefile.in: Likewise.
2098 * fficonfig.h.in: Likewise.
2100 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2101 Richard Henderson <rth@redhat.com>
2103 Avoid that ffi.h includes fficonfig.h.
2105 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2106 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2107 (TARGET_SRC_MIPS_SGI): Removed.
2108 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2109 (MIPS_SGI): Removed.
2110 (CLEANFILES): Removed.
2111 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2113 * acconfig.h: Removed.
2114 * configure.in: Compute sizeofs only for double and long double.
2115 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2116 into AC_DEFINE instead of using acconfig.h. Create
2117 include/ffitarget.h instead of include/fficonfig.h. Rename
2118 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2119 AC_DEFINE EH_FRAME_FLAGS.
2120 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2121 (hack_DATA): Add ffitarget.h.
2122 * include/ffi.h.in: Remove all system specific definitions.
2123 Declare raw API even if it is not installed, why bother?
2124 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2125 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2126 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2127 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2128 * include/ffi_common.h (bool): Do not define.
2129 (ffi_assert): Accept failed assertion.
2130 (ffi_type_test): Return void and accept file/line.
2131 (FFI_ASSERT): Pass stringized failed assertion.
2132 (FFI_ASSERT_AT): New macro.
2133 (FFI_ASSERT_VALID_TYPE): New macro.
2134 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2135 UINT64, SINT64): Define here with gcc's __attribute__ macro
2137 (FLOAT32, ALIGN): Define here instead of in ffi.h
2138 * include/ffi-mips.h: Removed. Its content moved to
2139 src/mips/ffitarget.h after separating assembly and C sections.
2140 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2141 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2142 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2143 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2144 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2145 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2146 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2147 (ffi_type_test): Rewritten.
2148 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2149 FFI_ASSERT_VALID_TYPE.
2150 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2151 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2152 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2153 src/s390/ffitarget.h, src/sh/ffitarget.h,
2154 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2155 src/x86/ffitarget.h: New files.
2156 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2157 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2158 src/powerpc/aix.S, src/powerpc/darwin.S,
2159 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2160 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2161 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2162 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2163 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2166 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2168 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2169 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2171 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2173 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2174 Used when FFI_DEBUG = 1.
2176 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2178 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2181 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2183 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2184 FFI_MIPS_O32 for O32 ABI.
2186 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2188 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2189 SPARC64. Cleanup whitespaces.
2191 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2193 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2194 strongarm, xscale. Cleanup whitespaces.
2195 * testsuite/libffi.call/closure_fn1.c: Likewise.
2196 * testsuite/libffi.call/closure_fn2.c: Likewise.
2197 * testsuite/libffi.call/closure_fn3.c: Likewise.
2198 * testsuite/libffi.call/cls_12byte.c: Likewise.
2199 * testsuite/libffi.call/cls_16byte.c: Likewise.
2200 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2201 * testsuite/libffi.call/cls_20byte.c: Likewise.
2202 * testsuite/libffi.call/cls_24byte.c: Likewise.
2203 * testsuite/libffi.call/cls_2byte.c: Likewise.
2204 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2205 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2206 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2207 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2208 * testsuite/libffi.call/cls_4byte.c: Likewise.
2209 * testsuite/libffi.call/cls_5byte.c: Likewise.
2210 * testsuite/libffi.call/cls_6byte.c: Likewise.
2211 * testsuite/libffi.call/cls_7byte.c: Likewise.
2212 * testsuite/libffi.call/cls_8byte.c: Likewise.
2213 * testsuite/libffi.call/cls_double.c: Likewise.
2214 * testsuite/libffi.call/cls_float.c: Likewise.
2215 * testsuite/libffi.call/cls_uchar.c: Likewise.
2216 * testsuite/libffi.call/cls_uint.c: Likewise.
2217 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2218 * testsuite/libffi.call/cls_ushort.c: Likewise.
2219 * testsuite/libffi.call/nested_struct.c: Likewise.
2220 * testsuite/libffi.call/nested_struct1.c: Likewise.
2221 * testsuite/libffi.call/problem1.c: Likewise.
2222 * testsuite/libffi.special/unwindtest.cc: Likewise.
2223 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2225 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2227 * src/powerpc/aix.S: Cleanup whitespaces.
2228 * src/powerpc/aix_closure.S: Likewise.
2230 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2232 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2233 * src/powerpc/darwin_closure.S: Likewise.
2234 * src/powerpc/ffi_darwin.c: Likewise.
2236 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2237 David Edelsohn <edelsohn@gnu.org>
2239 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2240 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2242 * src/powerpc/darwin_closure.S: Likewise.
2243 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2244 according to the Darwin/AIX ABI.
2245 (ffi_prep_cif_machdep): Likewise.
2246 (ffi_closure_helper_DARWIN): Likewise.
2247 Remove the outgoing parameter stack logic. Simplify the evaluation
2248 of the different CASE types.
2249 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2250 statement in the trampoline code.
2252 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2254 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2255 for the register size.
2256 (ffi_closure_helper_SYSV): Handle the structure return value
2258 (ffi_closure_helper_SYSV): Return the appropriate type when
2259 the registers are used for the structure return value.
2260 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2261 the 64-bit return value. Update copyright years.
2263 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2265 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2266 srcdir for ffi_mips.h.
2268 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2270 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2272 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2273 placement of float result.
2274 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2275 cast of "resp" for big-endian 64 bit machines.
2277 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2279 * src/types.c (double, longdouble): Merge identical SH and ARM
2280 typedefs, and add POWERPC64.
2281 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2282 struct split over gpr and rest.
2283 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2284 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2286 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2288 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2291 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2293 * configure: Regenerate.
2295 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2297 * Makefile.am: Remove build rules for ffitest.
2298 * Makefile.in: Rebuilt.
2300 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2302 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2303 about implicit declaration of abort().
2305 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2307 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2308 * Makefile.in: Rebuilt.
2309 * configure.in: Add dejagnu test framework.
2310 * configure: Rebuilt.
2312 * testsuite/Makefile.am: New file.
2313 * testsuite/Makefile.in: Built
2314 * testsuite/lib/libffi-dg.exp: New file.
2315 * testsuite/config/default.exp: Likewise.
2316 * testsuite/libffi.call/call.exp: Likewise.
2317 * testsuite/libffi.call/ffitest.h: Likewise.
2318 * testsuite/libffi.call/closure_fn0.c: Likewise.
2319 * testsuite/libffi.call/closure_fn1.c: Likewise.
2320 * testsuite/libffi.call/closure_fn2.c: Likewise.
2321 * testsuite/libffi.call/closure_fn3.c: Likewise.
2322 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2323 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2324 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2325 * testsuite/libffi.call/cls_2byte.c: Likewise.
2326 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2327 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2328 * testsuite/libffi.call/cls_4byte.c: Likewise.
2329 * testsuite/libffi.call/cls_5byte.c: Likewise.
2330 * testsuite/libffi.call/cls_6byte.c: Likewise.
2331 * testsuite/libffi.call/cls_7byte.c: Likewise.
2332 * testsuite/libffi.call/cls_8byte.c: Likewise.
2333 * testsuite/libffi.call/cls_12byte.c: Likewise.
2334 * testsuite/libffi.call/cls_16byte.c: Likewise.
2335 * testsuite/libffi.call/cls_20byte.c: Likewise.
2336 * testsuite/libffi.call/cls_24byte.c: Likewise.
2337 * testsuite/libffi.call/cls_double.c: Likewise.
2338 * testsuite/libffi.call/cls_float.c: Likewise.
2339 * testsuite/libffi.call/cls_uchar.c: Likewise.
2340 * testsuite/libffi.call/cls_uint.c: Likewise.
2341 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2342 * testsuite/libffi.call/cls_ushort.c: Likewise.
2343 * testsuite/libffi.call/float.c: Likewise.
2344 * testsuite/libffi.call/float1.c: Likewise.
2345 * testsuite/libffi.call/float2.c: Likewise.
2346 * testsuite/libffi.call/many.c: Likewise.
2347 * testsuite/libffi.call/many_win32.c: Likewise.
2348 * testsuite/libffi.call/nested_struct.c: Likewise.
2349 * testsuite/libffi.call/nested_struct1.c: Likewise.
2350 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2351 * testsuite/libffi.call/problem1.c: Likewise.
2352 * testsuite/libffi.call/promotion.c: Likewise.
2353 * testsuite/libffi.call/return_ll.c: Likewise.
2354 * testsuite/libffi.call/return_sc.c: Likewise.
2355 * testsuite/libffi.call/return_uc.c: Likewise.
2356 * testsuite/libffi.call/strlen.c: Likewise.
2357 * testsuite/libffi.call/strlen_win32.c: Likewise.
2358 * testsuite/libffi.call/struct1.c: Likewise.
2359 * testsuite/libffi.call/struct2.c: Likewise.
2360 * testsuite/libffi.call/struct3.c: Likewise.
2361 * testsuite/libffi.call/struct4.c: Likewise.
2362 * testsuite/libffi.call/struct5.c: Likewise.
2363 * testsuite/libffi.call/struct6.c: Likewise.
2364 * testsuite/libffi.call/struct7.c: Likewise.
2365 * testsuite/libffi.call/struct8.c: Likewise.
2366 * testsuite/libffi.call/struct9.c: Likewise.
2367 * testsuite/libffi.special/special.exp: New file.
2368 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2369 * testsuite/libffi.special/unwindtest.cc: Likewise.
2372 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2374 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2377 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2379 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2381 (ffi_closure_helper_LINUX64): Likewise.
2382 * src/powerpc/linux64.S: Remove code writing to parm save area.
2383 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2384 address in lr from ffi_closure_helper_LINUX64 call to calculate
2385 table address. Optimize function tail.
2387 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2389 * src/sparc/ffi.c: Handle all floating point registers.
2390 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2392 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2394 * README: Note that libffi is not part of GCC. Update the project
2397 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2399 * src/powerpc/ppc_closure.S: Include ffi.h.
2401 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2403 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2404 Use C style comments.
2406 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2408 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2409 * Makefile.in: Regenerate.
2410 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2411 * configure: Regenerate.
2412 * include/ffi.h.in: Add SHmedia support.
2413 * src/sh64/ffi.c: New file.
2414 * src/sh64/sysv.S: New file.
2416 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2418 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2419 should be read-only.
2420 * configure: Rebuilt.
2421 * fficonfig.h.in: Rebuilt.
2422 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2423 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2424 * src/powerpc/linux64.S: Likewise.
2425 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2426 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2427 if -fpic/-fPIC/-mrelocatable.
2428 * src/powerpc/powerpc_closure.S: Likewise.
2429 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2430 #write in .eh_frame flags.
2431 * src/sparc/v9.S: Likewise.
2432 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2433 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2434 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2436 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2438 Fixes PR bootstrap/10656
2439 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2440 support for .register pseudo-op.
2441 * src/sparc/v8.S: Use it.
2442 * fficonfig.h.in: Rebuilt.
2443 * configure: Rebuilt.
2445 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2447 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2448 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2449 Make it the default on POWERPC64.
2450 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2451 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2452 * configure: Rebuilt.
2453 * src/powerpc/ffi.c (hidden): Define.
2454 (ffi_prep_args_SYSV): Renamed from
2455 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2456 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2457 ASM_NEEDS_REGISTERS64): New.
2458 (ffi_prep_args64): New function.
2459 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2460 (ffi_call): Likewise.
2461 (ffi_prep_closure): Likewise.
2462 (flush_icache): Surround by #ifndef POWERPC64.
2463 (ffi_dblfl): New union type.
2464 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2465 (ffi_closure_helper_LINUX64): New function.
2466 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2468 * src/powerpc/sysv.S: Likewise.
2469 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2470 * src/powerpc/linux64.S: New file.
2471 * src/powerpc/linux64_closure.S: New file.
2472 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2473 src/powerpc/linux64_closure.S.
2474 (TARGET_SRC_POWERPC): Likewise.
2476 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2477 closure_test_fn3): Fix result printing on big-endian 64-bit
2479 (main): Print tst2_arg instead of uninitialized tst2_result.
2481 * src/ffitest.c (main): Hide what closure pointer really points to
2484 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2486 * configure.in (arm-*-netbsdelf*): Add configuration.
2487 (configure): Regenerated.
2489 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2491 * include/Makefile.in: Regenerate.
2493 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2495 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2497 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2498 Receive closure pointer through parameter, read args using
2499 __builtin_dwarf_cfa.
2500 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2502 2003-03-12 Andreas Schwab <schwab@suse.de>
2504 * configure.in: Avoid trailing /. in toolexeclibdir.
2505 * configure: Rebuilt.
2507 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2509 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2511 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2513 * libffi/src/powerpc/darwin_closure.S:
2514 Fix alignement bug, allocate 8 bytes for the result.
2515 * libffi/src/powerpc/aix_closure.S:
2517 * libffi/src/powerpc/ffi_darwin.c:
2518 Update stackframe description for aix/darwin_closure.S.
2520 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2522 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2525 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2526 Andreas Schwab <schwab@suse.de>
2528 * configure.in: Adjust command to source config-ml.in to account
2529 for changes to the libffi_basedir definition.
2530 (libffi_basedir): Remove ${srcdir} from value and include trailing
2533 * configure: Regenerate.
2535 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2537 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2539 2003-01-28 Andrew Haley <aph@redhat.com>
2541 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2542 * src/x86/ffi64.c (ffi_prep_closure): New.
2543 (ffi_closure_UNIX64_inner): New.
2544 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2546 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2548 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2549 Remove USE_LIBDIR conditional.
2550 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2551 * Makefile.in, configure: Rebuilt.
2553 2003-01027 David Edelsohn <edelsohn@gnu.org>
2555 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2556 * Makefile.in: Regenerate.
2558 2003-01-22 Andrew Haley <aph@redhat.com>
2560 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2563 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2565 * src/powerpc/darwin.S: Add unwind info.
2566 * src/powerpc/darwin_closure.S: Likewise.
2568 2003-01-14 Andrew Haley <aph@redhat.com>
2570 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2571 (ffi_prep_cif_machdep): Likewise.
2572 * src/x86/unix64.S: Add unwind info.
2574 2003-01-14 Andreas Jaeger <aj@suse.de>
2576 * src/ffitest.c (main): Only use ffi_closures if those are
2579 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2581 * libffi/src/ffitest.c
2582 add closure testcases
2584 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2586 * libffi/src/powerpc/ffi.c
2587 fix alignment bug for float (4 byte aligned iso 8 byte)
2589 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2591 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2592 * src/powerpc/darwin.S: Remove RCS version string.
2594 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2596 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2597 * src/ffitest.c (main): Use static storage for closure.
2598 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2599 * src/sparc/v8.S (ffi_closure_v8): New.
2600 * src/sparc/v9.S (ffi_closure_v9): New.
2602 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2604 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2605 enumeration for X86_WIN32.
2606 * src/x86/win32.S: Added ffi_call_STDCALL function
2608 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2609 switch cases for recognising FFI_STDCALL and
2610 calling ffi_call_STDCALL if target is X86_WIN32.
2611 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2612 stdcall versions of the "my_strlen" and "many"
2613 test functions (for X86_WIN32).
2614 Added test cases to test stdcall invocation using
2617 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2619 * src/sh/sysv.S: Add DWARF2 unwind info.
2621 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2623 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2625 2002-11-26 Jim Wilson <wilson@redhat.com>
2627 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2629 2002-11-23 H.J. Lu <hjl@gnu.org>
2631 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2632 Include ../config/accross.m4.
2633 * aclocal.m4; Rebuild.
2634 * configure: Likewise.
2636 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2638 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2640 2002-11-11 DJ Delorie <dj@redhat.com>
2642 * configure.in: Look for common files in the right place.
2644 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2646 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2647 raw data as _Jv_word values, not ffi_raw.
2648 (ffi_java_ptrarray_to_raw): Likewise.
2649 (ffi_java_rvalue_to_raw): New function.
2650 (ffi_java_raw_call): Call it.
2651 (ffi_java_raw_to_rvalue): New function.
2652 (ffi_java_translate_args): Call it.
2653 * src/ffitest.c (closure_test_fn): Interpret return value
2654 as ffi_arg, not int.
2655 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2656 FFI_TYPE_POINTER case.
2657 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2658 values extended to word size.
2660 2002-10-02 Andreas Jaeger <aj@suse.de>
2662 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2664 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2666 * include/ffi.h.in: Fix i386 win32 compilation.
2668 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2670 * configure.in: Add s390x-*-linux-* target.
2671 * configure: Regenerate.
2672 * include/ffi.h.in: Define S390X for s390x targets.
2673 (FFI_CLOSURES): Define for s390/s390x.
2674 (FFI_TRAMPOLINE_SIZE): Likewise.
2675 (FFI_NATIVE_RAW_API): Likewise.
2676 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2677 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2678 * src/s390/ffi.c: Major rework of existing code. Add support for
2679 s390x targets. Add closure support.
2680 * src/s390/sysv.S: Likewise.
2682 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2684 * src/arm/sysv.S: Fix typo.
2686 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2688 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2689 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2690 (ffi_call_SYSV): Handle soft-float.
2692 2002-09-27 Bo Thorsen <bo@suse.de>
2694 * include/ffi.h.in: Fix multilib x86-64 support.
2696 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2698 * Makefile.am (all-multi): Fix multilib parallel build.
2700 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2702 * configure.in (sh[34]*-*-linux*): Add brackets.
2703 * configure: Regenerate.
2705 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2707 * Makefile.am: Add SH support.
2708 * Makefile.in: Regenerate.
2709 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2710 * configure: Regenerate.
2711 * include/ffi.h.in: Add SH support.
2712 * src/sh/ffi.c: New file.
2713 * src/sh/sysv.S: New file.
2714 * src/types.c: Add SH support.
2716 2002-07-16 Bo Thorsen <bo@suse.de>
2718 * src/x86/ffi64.c: New file that adds x86-64 support.
2719 * src/x86/unix64.S: New file that handles argument setup for
2721 * src/x86/sysv.S: Don't use this on x86-64.
2722 * src/x86/ffi.c: Don't use this on x86-64.
2724 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2726 * src/ffitest.c (struct6): New test that tests a special case in
2728 (struct7): Likewise.
2729 (struct8): Likewise.
2730 (struct9): Likewise.
2731 (closure_test_fn): Silence warning about this when it's not used.
2732 (main): Add the new tests.
2733 (main): Fix a couple of wrong casts and silence some compiler warnings.
2734 * include/ffi.h.in: Add x86-64 ABI definition.
2735 * fficonfig.h.in: Regenerate.
2736 * Makefile.am: Add x86-64 support.
2737 * configure.in: Likewise.
2738 * Makefile.in: Regenerate.
2739 * configure: Likewise.
2741 2002-06-24 Bo Thorsen <bo@suse.de>
2743 * src/types.c: Merge settings for similar architectures.
2744 Add x86-64 sizes and alignments.
2746 2002-06-23 Bo Thorsen <bo@suse.de>
2748 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2749 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2750 * src/mips/ffi.c (ffi_prep_args): Likewise.
2751 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2753 2002-07-18 H.J. Lu (hjl@gnu.org)
2755 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2756 (libffi_la_SOURCES): Support MIPS_LINUX.
2757 (libffi_convenience_la_SOURCES): Likewise.
2758 * Makefile.in: Regenerated.
2760 * configure.in (mips64*-*): Skip.
2761 (mips*-*-linux*): New.
2762 * configure: Regenerated.
2764 * src/mips/ffi.c: Include <sgidefs.h>.
2766 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2768 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2770 2002-05-27 Roger Sayle <roger@eyesopen.com>
2772 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2774 2002-05-27 Bo Thorsen <bo@suse.de>
2776 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2779 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2781 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2782 beginning of function (for older apple cc).
2784 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2786 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2787 script entry, and set LD to it when configuring multilibs.
2788 * configure: Rebuilt.
2790 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2792 * configure.in (sparc64-*-netbsd*): Add target.
2793 (sparc-*-netbsdelf*): Likewise.
2794 * configure: Regenerate.
2796 2002-04-28 David S. Miller <davem@redhat.com>
2798 * configure.in, configure: Fix SPARC test in previous change.
2800 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2802 * Makefile.am: Add Linux for S/390 support.
2803 * Makefile.in: Regenerate.
2804 * configure.in: Add Linux for S/390 support.
2805 * configure: Regenerate.
2806 * include/ffi.h.in: Add Linux for S/390 support.
2807 * src/s390/ffi.c: New file from libffi CVS tree.
2808 * src/s390/sysv.S: New file from libffi CVS tree.
2810 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2812 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2814 * src/sparc/v8.S: Use it.
2815 * src/sparc/v9.S: Likewise.
2816 * fficonfig.h.in: Rebuilt.
2817 * configure: Rebuilt.
2819 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2821 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2823 * src/ia64/unix.S: Add unwind information. Fix comments.
2824 Save sp in a way that's compatible with unwind info.
2825 (ffi_call_unix): Correctly restore sp in all cases.
2826 * src/ia64/ffi.c: Add, fix comments.
2828 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2830 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2832 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2834 * configure.in (alpha*-*-netbsd*): Add target.
2835 * configure: Regenerate.
2837 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2839 * src/sparc/v8.S: Add unwind info.
2840 * src/sparc/v9.S: Likewise.
2842 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2844 * configure.in: Enable i*86-*-netbsdelf*.
2845 * configure: Rebuilt.
2847 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2850 * src/mips/n32.s: Delete
2851 * src/mips/o32.s: Delete
2853 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2855 * configure.in: Enable alpha*-*-freebsd*.
2856 * configure: Rebuilt.
2858 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2860 * Makefile.am: libfficonvenience -> libffi_convenience.
2861 * Makefile.in: Rebuilt.
2863 * Makefile.am: Define ffitest_OBJECTS.
2864 * Makefile.in: Rebuilt.
2866 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2867 David Edelsohn <edelsohn@gnu.org>
2869 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2870 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2871 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2872 * Makefile.in: Regenerate.
2873 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2874 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2875 (flush_icache, flush_range): New functions.
2876 (ffi_closure_helper_DARWIN): New function.
2877 * src/powerpc/aix_closure.S: New file.
2878 * src/powerpc/darwin_closure.S: New file.
2880 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2882 * include/ffi.h.in: Add typedef for ffi_arg.
2883 * src/ffitest.c (main): Declare rint with ffi_arg.
2885 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2887 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2888 number of GPRs for floating-point arguments.
2890 2002-01-31 Anthony Green <green@redhat.com>
2892 * configure: Rebuilt.
2893 * configure.in: Replace CHECK_SIZEOF and endian tests with
2894 cross-compiler friendly macros.
2895 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2898 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2900 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2901 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2903 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2905 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2906 (TARGET_SRC_POWERPC_AIX): New.
2907 (POWERPC_AIX): New stanza.
2908 * Makefile.in: Regenerate.
2909 * configure.in: Add AIX case.
2910 * configure: Regenerate.
2911 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2912 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2913 size. Fix "long double" support.
2914 (ffi_call): Add FFI_AIX case.
2915 * src/powerpc/aix.S: New.
2917 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2919 Implement Darwin PowerPC ABI.
2920 * configure.in: Handle powerpc-*-darwin*.
2921 * Makefile.am: Set source files for POWERPC_DARWIN.
2922 * configure: Rebuilt.
2923 * Makefile.in: Rebuilt.
2924 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2926 * src/powerpc/darwin.S: New file.
2927 * src/powerpc/ffi_darwin.c: New file.
2929 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2931 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2933 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2935 * src/x86/sysv.S: Avoid gas-only .balign directive.
2936 Use C style comments.
2938 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2940 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2941 Fixes PR bootstrap/3563.
2943 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2945 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2947 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2949 * configure.in: Recognize sparc*-sun-* host.
2950 * configure: Regenerate.
2952 2001-06-06 Andrew Haley <aph@redhat.com>
2954 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
2956 2001-06-03 Andrew Haley <aph@redhat.com>
2958 * src/alpha/osf.S: Add unwind info.
2959 * src/powerpc/sysv.S: Add unwind info.
2960 * src/powerpc/ppc_closure.S: Likewise.
2962 2000-05-31 Jeff Sturm <jsturm@one-point.com>
2964 * configure.in: Fix AC_ARG_ENABLE usage.
2965 * configure: Rebuilt.
2967 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2969 * configure.in: Remove warning about beta code.
2970 * configure: Rebuilt.
2972 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
2974 * src/ia64/unix.S: Restore stack pointer when returning from
2976 * src/ia64/ffi.c: Fix typo in comment.
2978 2001-04-18 Jim Wilson <wilson@redhat.com>
2980 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
2981 to eliminate RAW DV.
2983 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
2985 * Makefile.am: Make a libtool convenience library.
2986 * Makefile.in: Rebuilt.
2988 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
2990 * configure.in: Use different syntax for subdirectory creation.
2991 * configure: Rebuilt.
2993 2001-03-27 Jon Beniston <jon@beniston.com>
2995 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
2996 * configure: Rebuilt.
2997 * Makefile.am: Added X86_WIN32 target support.
2998 * Makefile.in: Rebuilt.
3000 * include/ffi.h.in: Added X86_WIN32 target support.
3002 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3003 * src/types.c: Added X86_WIN32 target support.
3005 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3006 stuff removed and made to work with CygWin's gas.
3008 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3010 * configure.in: Make target subdirectory in build dir.
3011 * Makefile.am: Override suffix based rules to specify correct output
3013 * Makefile.in: Rebuilt.
3014 * configure: Rebuilt.
3016 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3018 * src/powerpc/ppc_closure.S: New file.
3019 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3020 involving long long and register pairs.
3021 (ffi_prep_closure): New function.
3022 (flush_icache): Likewise.
3023 (ffi_closure_helper_SYSV): Likewise.
3024 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3025 (FFI_TRAMPOLINE_SIZE): Likewise.
3026 (FFI_NATIVE_RAW_API): Likewise.
3027 * Makefile.in: Rebuilt.
3028 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3029 (TARGET_SRC_POWERPC): Likewise.
3031 2001-03-19 Tom Tromey <tromey@redhat.com>
3033 * Makefile.in: Rebuilt.
3034 * Makefile.am (ffitest_LDFLAGS): New macro.
3036 2001-03-02 Nick Clifton <nickc@redhat.com>
3038 * include/ffi.h.in: Remove RCS ident string.
3039 * include/ffi_mips.h: Remove RCS ident string.
3040 * src/debug.c: Remove RCS ident string.
3041 * src/ffitest.c: Remove RCS ident string.
3042 * src/prep_cif.c: Remove RCS ident string.
3043 * src/types.c: Remove RCS ident string.
3044 * src/alpha/ffi.c: Remove RCS ident string.
3045 * src/alpha/osf.S: Remove RCS ident string.
3046 * src/arm/ffi.c: Remove RCS ident string.
3047 * src/arm/sysv.S: Remove RCS ident string.
3048 * src/mips/ffi.c: Remove RCS ident string.
3049 * src/mips/n32.S: Remove RCS ident string.
3050 * src/mips/o32.S: Remove RCS ident string.
3051 * src/sparc/ffi.c: Remove RCS ident string.
3052 * src/sparc/v8.S: Remove RCS ident string.
3053 * src/sparc/v9.S: Remove RCS ident string.
3054 * src/x86/ffi.c: Remove RCS ident string.
3055 * src/x86/sysv.S: Remove RCS ident string.
3057 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3059 * include/ffi.h.in: Change sourceware.cygnus.com references to
3062 2000-12-09 Richard Henderson <rth@redhat.com>
3064 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3065 (ffi_closure_osf_inner): Index rather than increment avalue
3066 and arg_types. Give ffi_closure_osf the raw return value type.
3067 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3070 2000-12-07 Richard Henderson <rth@redhat.com>
3072 * src/raw_api.c (ffi_translate_args): Fix typo.
3073 (ffi_prep_closure): Likewise.
3075 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3076 FFI_TRAMPOLINE_SIZE.
3077 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3078 cif->bytes for new ffi_call_osf implementation.
3079 (ffi_prep_args): Absorb into ...
3080 (ffi_call): ... here. Do all stack allocation here and
3081 avoid a callback function.
3082 (ffi_prep_closure, ffi_closure_osf_inner): New.
3083 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3084 (ffi_closure_osf): New.
3086 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3088 * config.guess, config.sub, install-sh: Removed.
3089 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3090 * Makefile.in: Rebuilt.
3092 * acinclude.m4: Include libtool macros from the top level.
3093 * aclocal.m4, configure: Rebuilt.
3095 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3097 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3098 * configure: Rebuilt.
3100 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3102 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3103 memory correctly. Use conditional instructions, not branches where
3106 2000-05-04 Tom Tromey <tromey@cygnus.com>
3108 * configure: Rebuilt.
3109 * configure.in: Match `arm*-*-linux-*'.
3110 From Chris Dornan <cdornan@arm.com>.
3112 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3114 * Makefile.am (SUBDIRS): Define.
3115 (AM_MAKEFLAGS): Likewise.
3116 (Multilib support.): Add section.
3117 * Makefile.in: Rebuilt.
3118 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3119 New variables. Set for gcc using -print-multi-lib. Export them
3121 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3122 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3124 (extra_compiler_flags_value, extra_compiler_flags): Check these
3125 for extra compiler options which need to be passed down in
3128 2000-04-16 Anthony Green <green@redhat.com>
3130 * configure: Rebuilt.
3131 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3133 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3135 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3136 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3137 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3138 Replace all void * sizeofs with sizeof(int).
3139 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3140 different than DOUBLE.
3141 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3142 (ffi_prep_args_v9): New function.
3143 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3144 (ffi_V9_return_struct): New function.
3145 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3146 32bit code (not yet cross-arch calls).
3147 * src/sparc/v8.S: Add struct return delay nop.
3149 * src/sparc/v9.S: New file.
3150 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3151 is used on sparc64 only for structures larger than 32 bytes.
3152 Pass by reference for structures is done for structure arguments
3153 larger than 16 bytes.
3154 * src/ffitest.c (main): Use 64bit rint on sparc64.
3155 Run long long tests on sparc.
3156 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3158 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3160 * configure.in (sparc-*-linux*): New supported target.
3161 (sparc64-*-linux*): Likewise.
3162 * configure: Rebuilt.
3163 * Makefile.am: Add v9.S to SPARC files.
3164 * Makefile.in: Likewise.
3165 (LINK): Surround $(CCLD) into double quotes, so that multilib
3166 compiles work correctly.
3168 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3170 * configure: Rebuilt.
3171 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3172 proposed by Bryce McKinlay.
3174 2000-03-20 Tom Tromey <tromey@cygnus.com>
3176 * Makefile.in: Hand edit for java_raw_api.lo.
3178 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3180 * config.guess, config.sub: Update from the gcc tree.
3181 Fix for PR libgcj/168.
3183 2000-03-03 Tom Tromey <tromey@cygnus.com>
3185 * Makefile.in: Fixed ia64 by hand.
3187 * configure: Rebuilt.
3188 * configure.in (--enable-multilib): New option.
3189 (libffi_basedir): New subst.
3190 (AC_OUTPUT): Added multilib code.
3192 2000-03-02 Tom Tromey <tromey@cygnus.com>
3194 * Makefile.in: Rebuilt.
3195 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3198 2000-02-25 Hans Boehm <boehm@acm.org>
3200 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3202 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3204 (ffi_prep_raw_closure): Use ffi_translate_args, not
3205 ffi_closure_translate.
3206 * src/java_raw_api.c: New file.
3207 * src/ffitest.c (closure_test_fn): New function.
3208 (main): Define `rint' as long long on IA64. Added new test when
3209 FFI_CLOSURES is defined.
3210 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3211 (ffi_abi): Recognize IA64.
3212 (ffi_raw): Added `flt' field.
3213 Added "Java raw API" code.
3214 * configure.in: Recognize ia64.
3215 * Makefile.am (TARGET_SRC_IA64): New macro.
3216 (libffi_la_common_SOURCES): Added java_raw_api.c.
3217 (libffi_la_SOURCES): Define in IA64 case.
3219 2000-01-04 Tom Tromey <tromey@cygnus.com>
3221 * Makefile.in: Rebuilt with newer automake.
3223 1999-12-31 Tom Tromey <tromey@cygnus.com>
3225 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3227 1999-09-01 Tom Tromey <tromey@cygnus.com>
3229 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3231 * fficonfig.h.in: Rebuilt.
3232 * configure: Rebuilt.
3233 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3234 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3235 * acconfig.h: Don't #undef PACKAGE or VERSION.
3237 1999-08-09 Anthony Green <green@cygnus.com>
3239 * include/ffi.h.in: Try to work around messy header problem
3240 with PACKAGE and VERSION.
3242 * configure: Rebuilt.
3243 * configure.in: Change version to 2.00-beta.
3245 * fficonfig.h.in: Rebuilt.
3246 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3248 * src/x86/ffi.c (ffi_raw_call): Rename.
3250 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3252 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3253 (ffi_prep_incoming_args_SYSV): Ditto.
3254 (ffi_prep_closure): Ditto.
3255 (ffi_closure_raw_SYSV): Ditto.
3256 (ffi_prep_raw_closure): More ditto.
3257 (ffi_call_raw): Final ditto.
3259 * include/ffi.h.in: Add definitions for closure and raw API.
3261 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3264 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3266 * src/raw_api.c: New file.
3268 * include/ffi.h.in (ffi_raw): New type.
3269 (UINT_ARG, SINT_ARG): New defines.
3270 (ffi_closure, ffi_raw_closure): New types.
3271 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3273 * configure.in: Add check for endianness and sizeof void*.
3275 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3276 instead of directly.
3278 * configure: Rebuilt.
3280 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3282 * configure.in: Add x86 and powerpc BeOS configurations.
3283 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3285 1999-05-09 Anthony Green <green@cygnus.com>
3287 * configure.in: Add warning about this being beta code.
3288 Remove src/Makefile.am from the picture.
3289 * configure: Rebuilt.
3291 * Makefile.am: Move logic from src/Makefile.am. Add changes
3292 to support libffi as a target library.
3293 * Makefile.in: Rebuilt.
3295 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3296 Upgraded to new autoconf, automake, libtool.
3300 * LICENSE: Update copyright date.
3302 * src/Makefile.am, src/Makefile.in: Removed.
3304 1998-11-29 Anthony Green <green@cygnus.com>
3306 * include/ChangeLog: Removed.
3307 * src/ChangeLog: Removed.
3308 * src/mips/ChangeLog: Removed.
3309 * src/sparc/ChangeLog: Remboved.
3310 * src/x86/ChangeLog: Removed.
3312 * ChangeLog.v1: Created.