1 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
3 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
4 * src/sh64/sysv.S: Likewise.
6 2008-03-26 Daniel Jacobowitz <dan@debian.org>
8 * src/arm/sysv.S: Fix ARM comment marker.
10 2008-03-26 Jakub Jelinek <jakub@redhat.com>
12 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
13 * src/s390/sysv.S: Likewise.
14 * src/powerpc/ppc_closure.S: Likewise.
15 * src/powerpc/sysv.S: Likewise.
16 * src/x86/unix64.S: Likewise.
17 * src/x86/sysv.S: Likewise.
18 * src/sparc/v8.S: Likewise.
19 * src/sparc/v9.S: Likewise.
20 * src/m68k/sysv.S: Likewise.
21 * src/arm/sysv.S: Likewise.
23 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
25 * aclocal.m4: Regenerate.
26 * configure: Likewise.
27 * Makefile.in: Likewise.
28 * include/Makefile.in: Likewise.
29 * testsuite/Makefile.in: Likewise.
31 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
32 Andreas Tobler <a.tobler@schweiz.org>
34 * configure.ac: Add amd64-*-freebsd* target.
35 * configure: Regenerate.
37 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
40 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
43 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
46 2008-01-24 David Edelsohn <edelsohn@gnu.org>
48 * configure: Regenerate.
50 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
52 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
54 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
57 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
58 signed/unsigned int8/16 for X86_DARWIN.
59 Updated copyright info.
60 Handle one and two byte structs with special cif->flags.
61 * src/x86/ffitarget.h: Add special types for one and two byte structs.
62 Updated copyright info.
63 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
65 Remove code to pop args from the stack after call.
66 Special-case signed/unsigned for int8/16, one and two byte structs.
67 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
68 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
70 Updated copyright info.
72 2007-12-08 David Daney <ddaney@avtrex.com>
74 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
75 SUBU, add with ADDU and use smaller code sequences.
77 2007-12-07 David Daney <ddaney@avtrex.com>
79 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
82 2007-12-06 David Daney <ddaney@avtrex.com>
84 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
86 (ffi_java_raw): New typedef.
87 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
88 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
90 (ffi_java_raw_closure) : Same.
91 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
93 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
95 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
96 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
97 sizeof(ffi_java_raw) for alignment calculations.
98 (ffi_java_ptrarray_to_raw): Same.
99 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
100 if FFI_SIZEOF_JAVA_RAW == 4.
101 (ffi_java_raw_to_rvalue): Same.
102 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
103 (ffi_java_translate_args): Same.
104 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
106 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
108 2007-12-06 David Daney <ddaney@avtrex.com>
110 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
113 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
116 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
117 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
118 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
119 set the NUM_FPR_ARG_REGISTERS according to.
120 Add support for potential soft-float support under hard-float
122 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
123 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
124 to the FFI_LINUX_SOFT_FLOAT ABI.
125 (ffi_prep_cif_machdep): Likewise.
126 (ffi_closure_helper_SYSV): Likewise.
127 * src/powerpc/ppc_closure.S: Make sure not to store float/double
128 on archs where __NO_FPRS__ is true.
129 Add FFI_TYPE_UINT128 support.
130 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
131 Adjust copyright notice.
133 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
135 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
136 * include/ffi_common.h: ... here.
139 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
141 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
142 * src/powerpc/linux64.S: Likewise.
143 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
144 * testsuite/libffi.call/return_ldl.c: New test.
146 2007-09-04 <aph@redhat.com>
148 * src/arm/sysv.S (UNWIND): New.
149 (Whole file): Conditionally compile unwinder directives.
150 * src/arm/sysv.S: Add unwinder directives.
152 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
153 Only treat r0 as a struct address if we're actually returning a
155 Only copy the bytes that are actually within a struct.
156 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
157 is returned in r0, not passed by address.
158 (ffi_call): Allocate a word-sized temporary for the case where
159 a composite is returned in r0.
160 (ffi_prep_incoming_args_SYSV): Align as necessary.
162 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
164 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
165 directly using the sys_cacheflush syscall.
167 2007-07-27 Andrew Haley <aph@redhat.com>
169 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
171 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
173 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
174 * configure.ac: Likewise.
175 * Makefile.in: Regenerate.
176 * include/Makefile.in: Likewise.
177 * testsuite/Makefile.in: Likewise.
178 * configure: Likewise.
180 2007-08-24 David Daney <ddaney@avtrex.com>
182 * testsuite/libffi.call/return_sl.c: New test.
184 2007-08-10 David Daney <ddaney@avtrex.com>
186 * testsuite/libffi.call/cls_multi_ushort.c,
187 testsuite/libffi.call/cls_align_uint16.c,
188 testsuite/libffi.call/nested_struct1.c,
189 testsuite/libffi.call/nested_struct3.c,
190 testsuite/libffi.call/cls_7_1_byte.c,
191 testsuite/libffi.call/nested_struct5.c,
192 testsuite/libffi.call/cls_double.c,
193 testsuite/libffi.call/nested_struct7.c,
194 testsuite/libffi.call/cls_sint.c,
195 testsuite/libffi.call/nested_struct9.c,
196 testsuite/libffi.call/cls_20byte1.c,
197 testsuite/libffi.call/cls_multi_sshortchar.c,
198 testsuite/libffi.call/cls_align_sint64.c,
199 testsuite/libffi.call/cls_3byte2.c,
200 testsuite/libffi.call/cls_multi_schar.c,
201 testsuite/libffi.call/cls_multi_uchar.c,
202 testsuite/libffi.call/cls_19byte.c,
203 testsuite/libffi.call/cls_9byte1.c,
204 testsuite/libffi.call/cls_align_float.c,
205 testsuite/libffi.call/closure_fn1.c,
206 testsuite/libffi.call/problem1.c,
207 testsuite/libffi.call/closure_fn3.c,
208 testsuite/libffi.call/cls_sshort.c,
209 testsuite/libffi.call/closure_fn5.c,
210 testsuite/libffi.call/cls_align_double.c,
211 testsuite/libffi.call/nested_struct.c,
212 testsuite/libffi.call/cls_2byte.c,
213 testsuite/libffi.call/nested_struct10.c,
214 testsuite/libffi.call/cls_4byte.c,
215 testsuite/libffi.call/cls_6byte.c,
216 testsuite/libffi.call/cls_8byte.c,
217 testsuite/libffi.call/cls_multi_sshort.c,
218 testsuite/libffi.call/cls_align_sint16.c,
219 testsuite/libffi.call/cls_align_uint32.c,
220 testsuite/libffi.call/cls_20byte.c,
221 testsuite/libffi.call/cls_float.c,
222 testsuite/libffi.call/nested_struct2.c,
223 testsuite/libffi.call/cls_5_1_byte.c,
224 testsuite/libffi.call/nested_struct4.c,
225 testsuite/libffi.call/cls_24byte.c,
226 testsuite/libffi.call/nested_struct6.c,
227 testsuite/libffi.call/cls_64byte.c,
228 testsuite/libffi.call/nested_struct8.c,
229 testsuite/libffi.call/cls_uint.c,
230 testsuite/libffi.call/cls_multi_ushortchar.c,
231 testsuite/libffi.call/cls_schar.c,
232 testsuite/libffi.call/cls_uchar.c,
233 testsuite/libffi.call/cls_align_uint64.c,
234 testsuite/libffi.call/cls_ulonglong.c,
235 testsuite/libffi.call/cls_align_longdouble.c,
236 testsuite/libffi.call/cls_1_1byte.c,
237 testsuite/libffi.call/cls_12byte.c,
238 testsuite/libffi.call/cls_3_1byte.c,
239 testsuite/libffi.call/cls_3byte1.c,
240 testsuite/libffi.call/cls_4_1byte.c,
241 testsuite/libffi.call/cls_6_1_byte.c,
242 testsuite/libffi.call/cls_16byte.c,
243 testsuite/libffi.call/cls_18byte.c,
244 testsuite/libffi.call/closure_fn0.c,
245 testsuite/libffi.call/cls_9byte2.c,
246 testsuite/libffi.call/closure_fn2.c,
247 testsuite/libffi.call/closure_fn4.c,
248 testsuite/libffi.call/cls_ushort.c,
249 testsuite/libffi.call/closure_fn6.c,
250 testsuite/libffi.call/cls_5byte.c,
251 testsuite/libffi.call/cls_align_pointer.c,
252 testsuite/libffi.call/cls_7byte.c,
253 testsuite/libffi.call/cls_align_sint32.c,
254 testsuite/libffi.special/unwindtest_ffi_call.cc,
255 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
257 2007-08-10 David Daney <ddaney@avtrex.com>
260 * configure.ac: Don't treat mips64 as a special case.
261 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
262 * configure: Regenerate
263 * Makefile.in: Ditto.
264 * fficonfig.h.in: Ditto.
265 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
266 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
267 (FFI_DEFAULT_ABI): Set for n64 case.
268 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
269 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
270 (ffi_closure_N32): New function.
271 (.eh_frame): New section
272 * src/mips/o32.S: Clean up comments.
273 (ffi_closure_O32): Pass ffi_closure parameter in $12.
274 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
275 _MIPS_SIM == _ABIN32 throughout.
276 (FFI_MIPS_STOP_HERE): New, use in place of
278 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
279 to support n32/n64 ABIs.
280 (calc_n32_struct_flags): Rewrite.
281 (calc_n32_return_struct_flags): Remove unused variable. Reverse
282 position of flag bits.
283 (ffi_prep_cif_machdep): Rewrite n32 portion.
284 (ffi_call): Enable for n64. Add special handling for small structure
286 (ffi_prep_closure_loc): Add n32 and n64 support.
287 (ffi_closure_mips_inner_O32): Add cast to silence warning.
288 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
290 2007-08-08 David Daney <ddaney@avtrex.com>
292 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
293 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
295 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
296 * testsuite/libffi.call/cls_sint.c (main): Ditto.
297 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
298 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
299 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
300 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
301 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
302 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
303 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
304 * testsuite/libffi.call/return_fl3.c (main): Ditto.
305 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
306 * testsuite/libffi.call/nested_struct.c (main): Ditto.
307 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
308 * testsuite/libffi.call/return_ll1.c (main): Ditto.
309 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
310 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
311 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
312 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
313 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
314 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
315 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
316 * testsuite/libffi.call/cls_uint.c (main): Ditto.
317 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
318 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
319 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
320 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
321 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
322 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
323 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
324 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
325 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
327 2007-08-07 Andrew Haley <aph@redhat.com>
329 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
332 2007-08-06 Andrew Haley <aph@redhat.com>
335 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
336 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
339 2007-08-02 David Daney <ddaney@avtrex.com>
341 * testsuite/libffi.call/return_ul.c (main): Define return type as
342 ffi_arg. Use proper printf conversion specifier.
344 2007-07-30 Andrew Haley <aph@redhat.com>
347 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
348 signed/unsigned int8/16.
349 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
351 Remove code to pop args from the stack after call.
352 Special-case signed/unsigned int8/16.
353 * testsuite/libffi.call/return_sc.c (main): Revert.
355 2007-07-26 Richard Guenther <rguenther@suse.de>
358 * testsuite/libffi.call/return_sc.c (main): Verify call
359 result as signed char, not ffi_arg.
361 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
363 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
364 * configure: Regenerate.
366 2007-07-11 David Daney <ddaney@avtrex.com>
368 * src/mips/ffi.c: Don't include sys/cachectl.h.
369 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
372 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
374 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
375 from (ffi_prep_closure): ... this.
376 (FFI_INIT_TRAMPOLINE): Adjust.
378 2005-12-31 Phil Blundell <pb@reciva.com>
380 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
381 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
382 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
383 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
384 (FFI_CLOSURES): Enable closure support.
386 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
388 * testsuite/libffi.call/cls_multi_ushort.c,
389 testsuite/libffi.call/cls_align_uint16.c,
390 testsuite/libffi.call/nested_struct1.c,
391 testsuite/libffi.call/nested_struct3.c,
392 testsuite/libffi.call/cls_7_1_byte.c,
393 testsuite/libffi.call/cls_double.c,
394 testsuite/libffi.call/nested_struct5.c,
395 testsuite/libffi.call/nested_struct7.c,
396 testsuite/libffi.call/cls_sint.c,
397 testsuite/libffi.call/nested_struct9.c,
398 testsuite/libffi.call/cls_20byte1.c,
399 testsuite/libffi.call/cls_multi_sshortchar.c,
400 testsuite/libffi.call/cls_align_sint64.c,
401 testsuite/libffi.call/cls_3byte2.c,
402 testsuite/libffi.call/cls_multi_schar.c,
403 testsuite/libffi.call/cls_multi_uchar.c,
404 testsuite/libffi.call/cls_19byte.c,
405 testsuite/libffi.call/cls_9byte1.c,
406 testsuite/libffi.call/cls_align_float.c,
407 testsuite/libffi.call/closure_fn1.c,
408 testsuite/libffi.call/problem1.c,
409 testsuite/libffi.call/closure_fn3.c,
410 testsuite/libffi.call/cls_sshort.c,
411 testsuite/libffi.call/closure_fn5.c,
412 testsuite/libffi.call/cls_align_double.c,
413 testsuite/libffi.call/cls_2byte.c,
414 testsuite/libffi.call/nested_struct.c,
415 testsuite/libffi.call/nested_struct10.c,
416 testsuite/libffi.call/cls_4byte.c,
417 testsuite/libffi.call/cls_6byte.c,
418 testsuite/libffi.call/cls_8byte.c,
419 testsuite/libffi.call/cls_multi_sshort.c,
420 testsuite/libffi.call/cls_align_uint32.c,
421 testsuite/libffi.call/cls_align_sint16.c,
422 testsuite/libffi.call/cls_float.c,
423 testsuite/libffi.call/cls_20byte.c,
424 testsuite/libffi.call/cls_5_1_byte.c,
425 testsuite/libffi.call/nested_struct2.c,
426 testsuite/libffi.call/cls_24byte.c,
427 testsuite/libffi.call/nested_struct4.c,
428 testsuite/libffi.call/nested_struct6.c,
429 testsuite/libffi.call/cls_64byte.c,
430 testsuite/libffi.call/nested_struct8.c,
431 testsuite/libffi.call/cls_uint.c,
432 testsuite/libffi.call/cls_multi_ushortchar.c,
433 testsuite/libffi.call/cls_schar.c,
434 testsuite/libffi.call/cls_uchar.c,
435 testsuite/libffi.call/cls_align_uint64.c,
436 testsuite/libffi.call/cls_ulonglong.c,
437 testsuite/libffi.call/cls_align_longdouble.c,
438 testsuite/libffi.call/cls_1_1byte.c,
439 testsuite/libffi.call/cls_12byte.c,
440 testsuite/libffi.call/cls_3_1byte.c,
441 testsuite/libffi.call/cls_3byte1.c,
442 testsuite/libffi.call/cls_4_1byte.c,
443 testsuite/libffi.call/cls_6_1_byte.c,
444 testsuite/libffi.call/cls_16byte.c,
445 testsuite/libffi.call/cls_18byte.c,
446 testsuite/libffi.call/closure_fn0.c,
447 testsuite/libffi.call/cls_9byte2.c,
448 testsuite/libffi.call/closure_fn2.c,
449 testsuite/libffi.call/closure_fn4.c,
450 testsuite/libffi.call/cls_ushort.c,
451 testsuite/libffi.call/closure_fn6.c,
452 testsuite/libffi.call/cls_5byte.c,
453 testsuite/libffi.call/cls_align_pointer.c,
454 testsuite/libffi.call/cls_7byte.c,
455 testsuite/libffi.call/cls_align_sint32.c,
456 testsuite/libffi.special/unwindtest_ffi_call.cc,
457 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
459 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
461 * aclocal.m4: Regenerated.
463 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
465 * configure: Regenerate.
467 2007-05-23 Steve Ellcey <sje@cup.hp.com>
469 * Makefile.in: Regenerate.
470 * configure: Regenerate.
471 * aclocal.m4: Regenerate.
472 * include/Makefile.in: Regenerate.
473 * testsuite/Makefile.in: Regenerate.
475 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
477 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
478 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
479 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
480 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
481 (FFI_CLOSURES): Enable closure support.
483 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
485 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
486 * configure: Regenerate.
487 * fficonfig.h.in: Regenerate.
488 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
489 CFI_OFFSET,CFI_DEF_CFA): New macros.
490 (ffi_call_SYSV): Add callframe annotation.
492 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
494 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
495 numerous test suite failures.
496 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
498 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
500 * Makefile.am (EXTRA_DIST): Bring up to date.
501 * Makefile.in: Regenerate.
502 * src/frv/eabi.S: Remove RCS keyword.
504 2007-04-06 Richard Henderson <rth@redhat.com>
506 * configure.ac: Tidy target case.
507 (HAVE_LONG_DOUBLE): Allow the target to override.
508 * configure: Regenerate.
509 * include/ffi.h.in: Don't define ffi_type_foo if
510 LIBFFI_HIDE_BASIC_TYPES is defined.
511 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
513 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
514 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
515 (ffi_type_longdouble): Special case for Alpha. Don't define
516 if long double == double.
518 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
519 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
520 (ffi_call, ffi_closure_osf_inner): Likewise.
521 (ffi_closure_osf_inner): Likewise. Mark hidden.
522 (ffi_call_osf, ffi_closure_osf): Mark hidden.
523 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
524 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
525 (load_table): Handle 128-bit long double.
527 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
529 2007-04-06 Tom Tromey <tromey@redhat.com>
532 * README: Fixed bug in example.
534 2007-04-03 Jakub Jelinek <jakub@redhat.com>
536 * src/closures.c: Include sys/statfs.h.
537 (_GNU_SOURCE): Define on Linux.
538 (FFI_MMAP_EXEC_SELINUX): Define.
539 (selinux_enabled): New variable.
540 (selinux_enabled_check): New function.
541 (is_selinux_enabled): Define.
544 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
546 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
547 Use 'volatile float sum' to create sum of floats to avoid false
548 negative due to excess precision on ix86 targets.
551 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
553 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
555 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
557 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
559 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
560 (ffi_prep_closure_loc): New.
561 (ffi_prep_raw_closure_loc): New.
562 (ffi_prep_java_raw_closure_loc): New.
563 * src/closures.c: New file.
564 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
565 Replace sflags with exec_offset.
566 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
567 sub_segment_exec_offset): New macros.
568 (get_segment_flags, set_segment_flags, check_segment_merge): New
570 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
571 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
572 destroy_mspace): Use new macros.
573 (sys_alloc): Silence warning.
574 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
575 * Makefile.in: Rebuilt.
576 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
577 terms of ffi_prep_closure_loc.
578 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
580 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
582 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
584 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
586 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
587 (ffi_prep_closure): ... this.
588 * src/pa/ffi.c: Likewise.
589 * src/cris/ffi.c: Likewise. Adjust.
590 * src/frv/ffi.c: Likewise.
591 * src/ia64/ffi.c: Likewise.
592 * src/mips/ffi.c: Likewise.
593 * src/powerpc/ffi_darwin.c: Likewise.
594 * src/s390/ffi.c: Likewise.
595 * src/sh/ffi.c: Likewise.
596 * src/sh64/ffi.c: Likewise.
597 * src/sparc/ffi.c: Likewise.
598 * src/x86/ffi64.c: Likewise.
599 * src/x86/ffi.c: Likewise.
600 (FFI_INIT_TRAMPOLINE): Adjust.
601 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
602 (ffi_prep_raw_closure): ... this.
603 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
604 (ffi_prep_closure): ... this.
605 (flush_icache): Adjust.
607 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
609 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
612 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
614 * Makefile.am: Add dummy install-pdf target.
615 * Makefile.in: Regenerate
617 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
619 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
620 ffi_closure_helper_SYSV): Add long double handling.
622 2007-02-02 Jakub Jelinek <jakub@redhat.com>
624 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
625 immediately after bctrl instruction.
627 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
629 * Makefile.am (all-recursive, install-recursive,
630 mostlyclean-recursive, clean-recursive, distclean-recursive,
631 maintainer-clean-recursive): Add missing targets.
632 * Makefile.in: Rebuilt.
634 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
636 * configure.ac: Add TARGET for x86_64-*-darwin*.
637 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
639 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
640 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
641 * configure: Regenerate.
642 * Makefile.in: Regenerate.
643 * include/Makefile.in: Regenerate.
644 * testsuite/Makefile.in: Regenerate.
645 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
648 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
650 * aclocal.m4: Regenerate with aclocal -I .. as written in the
653 2006-10-31 Geoffrey Keating <geoffk@apple.com>
655 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
656 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
658 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
659 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
661 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
662 Sandro Tolaini <tolaini@libero.it>
664 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
666 * configure: Regenerated.
667 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
668 (EXTRA_DIST): Add src/x86/darwin.S.
669 * Makefile.in: Regenerated.
670 * include/Makefile.in: Regenerated.
671 * testsuite/Makefile.in: Regenerated.
673 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
674 X86_WIN32, and additionally align stack to 16 bytes.
675 * src/x86/darwin.S: New, based on sysv.S.
676 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
678 2006-09-12 David Daney <ddaney@avtrex.com>
681 * include/Makefile.am: Install both ffi.h and ffitarget.h in
682 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
683 * aclocal.m4: Regenerated for automake 1.9.6.
684 * Makefile.in: Regenerated.
685 * include/Makefile.in: Regenerated.
686 * testsuite/Makefile.in: Regenerated.
688 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
690 * include/ffi_common.h (struct): Revert accidental commit.
692 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
694 * include/ffi_common.h: Remove lint directives.
695 * include/ffi.h.in: Likewise.
697 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
699 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
700 for 32-bit architectures.
701 * testsuite/libffi.call/return_ul.c: New test case.
703 2006-07-19 David Daney <ddaney@avtrex.com>
705 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
706 xfail remains for mips64.
708 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
710 * Makefile.am: Add install-html target. Add install-html to .PHONY
711 * Makefile.in: Regenerate.
712 * aclocal.m4: Regenerate.
713 * include/Makefile.in: Regenerate.
714 * testsuite/Makefile.in: Regenerate.
716 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
718 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
721 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
723 * README: Remove notice about 'Crazy Comments'.
724 * src/debug.c: Remove lint directives. Cleanup white spaces.
725 * src/java_raw_api.c: Likewise.
726 * src/prep_cif.c: Likewise.
727 * src/raw_api.c: Likewise.
728 * src/ffitest.c: Delete. No longer needed, all test cases migrated
730 * src/arm/ffi.c: Remove lint directives.
731 * src/m32r/ffi.c: Likewise.
732 * src/pa/ffi.c: Likewise.
733 * src/powerpc/ffi.c: Likewise.
734 * src/powerpc/ffi_darwin.c: Likewise.
735 * src/sh/ffi.c: Likewise.
736 * src/sh64/ffi.c: Likewise.
737 * src/x86/ffi.c: Likewise.
738 * testsuite/libffi.call/float2.c: Likewise.
739 * testsuite/libffi.call/promotion.c: Likewise.
740 * testsuite/libffi.call/struct1.c: Likewise.
742 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
744 * src/pa/hpux32.S: Correct unwind offset calculation for
746 * src/pa/linux.S: Likewise.
748 2006-04-12 James E Wilson <wilson@specifix.com>
751 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
752 (hfa_type_load): Call stf_spill.
753 (hfa_type_store): Call ldf_fill.
754 (ffi_call): Adjust calls to above routines. Add local temps for
757 2006-04-10 Matthias Klose <doko@debian.org>
759 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
760 directory names containing underscores.
762 2006-04-07 James E Wilson <wilson@specifix.com>
764 * testsuite/libffi.call/float4.c: New testcase.
766 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
767 Andreas Tobler <a.tobler@schweiz.ch>
769 * Makefile.am: Add PA_HPUX port.
770 * Makefile.in: Regenerate.
771 * include/Makefile.in: Likewise.
772 * testsuite/Makefile.in: Likewise.
773 * configure.ac: Add PA_HPUX rules.
774 * configure: Regenerate.
775 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
776 Add PA_HPUX and PA64_HPUX.
777 Rename FFI_LINUX ABI to FFI_PA32 ABI.
778 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
779 (FFI_TYPE_SMALL_STRUCT2): Define.
780 (FFI_TYPE_SMALL_STRUCT4): Likewise.
781 (FFI_TYPE_SMALL_STRUCT8): Likewise.
782 (FFI_TYPE_SMALL_STRUCT3): Redefine.
783 (FFI_TYPE_SMALL_STRUCT5): Likewise.
784 (FFI_TYPE_SMALL_STRUCT6): Likewise.
785 (FFI_TYPE_SMALL_STRUCT7): Likewise.
786 * src/pa/ffi.c (ROUND_DOWN): Delete.
787 (fldw, fstw, fldd, fstd): Use '__asm__'.
788 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
789 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
790 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
791 Simplify incrementing of stack slot variable. Change type of local
793 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
795 (ffi_prep_cif_machdep): Likewise.
796 (ffi_call): Likewise.
797 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
798 return type to ffi_status. Simplify incrementing of stack slot
799 variable. Only copy floating point argument registers when PA_LINUX
800 is true. Reformat debug statement.
801 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
802 FFI_TYPE_SMALL_STRUCT8.
803 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
805 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
806 Add nops to cache flush. Add trampoline for PA_HPUX.
807 * src/pa/hpux32.S: New file.
808 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
809 ffi_prep_args_LINUX to ffi_prep_args_pa32.
810 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
811 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
812 argument type checks so that common argument types appear first.
813 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
814 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
816 2006-03-24 Alan Modra <amodra@bigpond.net.au>
818 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
819 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
820 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
822 (ffi_prep_args64): Assert using IBM extended double.
823 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
824 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
825 (ffi_call): Handle FFI_LINUX.
826 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
827 gpr3 return pointer as for struct return. Handle FFI_LINUX
828 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
830 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
831 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
832 Don't use r6 as pointer to results, instead use sp offset. Don't
833 make a special call to load lr with case table address, instead
834 use offset from previous call.
835 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
836 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
839 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
841 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
842 passed with FP registers correctly.
843 (ffi_closure_helper_SYSV): Likewise.
844 * src/sh64/sysv.S: Likewise.
846 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
848 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
849 args and userdata unused.
850 (closure_test_fn1): Mark cif and userdata unused.
851 (main): Remove unused res.
853 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
855 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
856 -O2, -O3, -Os and the warning flags -W -Wall.
857 * testsuite/libffi.special/special.exp: Likewise.
858 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
859 unused parameter unused for gcc or else do nothing.
860 * testsuite/libffi.special/ffitestcxx.h: Likewise.
861 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
863 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
864 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
865 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
866 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
867 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
868 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
869 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
870 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
871 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
872 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
873 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
874 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
875 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
876 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
877 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
878 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
879 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
880 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
881 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
882 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
883 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
884 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
885 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
886 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
888 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
890 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
892 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
893 void* to avoid compiler warning.
895 (cls_struct_align_gn): Mark cif and userdata unused.
896 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
898 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
900 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
902 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
904 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
906 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
907 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
908 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
910 (main): Cast res_call to silence gcc.
911 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
913 (main): Cast res_call to silence gcc.
914 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
916 (main): Cast res_call to silence gcc.
917 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
919 (main): Cast res_call to silence gcc.
920 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
922 (main): Cast res_call to silence gcc.
923 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
925 (main): Cast res_call to silence gcc.
926 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
928 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
929 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
931 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
932 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
934 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
935 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
937 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
938 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
940 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
941 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
943 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
945 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
946 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
947 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
948 Cleanup white spaces.
949 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
950 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
951 cif and userdata unused.
952 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
954 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
955 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
956 formatters to silence gcc.
957 (B_gn): Mark cif and userdata unused.
958 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
960 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
961 (B_gn): Mark cif and userdata unused.
962 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
964 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
965 (B_gn): Mark cif and userdata unused.
966 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
968 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
969 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
970 * testsuite/libffi.call/problem1.c (stub): Likewise.
971 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
973 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
974 in the last commit for this test case in the test case itself.
975 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
977 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
978 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
979 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
980 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
981 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
982 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
984 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
986 * src/sh/sysv.S: Fix register numbers in the FDE for
989 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
991 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
992 declaration to avoid a false negative on ix86. See PR323.
994 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
996 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
997 and cast integer to void * if needed. Update the pointer to
998 the FP register saved area correctly.
1000 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1002 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1004 * testsuite/libffi.call/nested_struct4.c: Likewise.
1006 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1008 * testsuite/libffi.call/return_dbl.c: New test case.
1009 * testsuite/libffi.call/return_dbl1.c: Likewise.
1010 * testsuite/libffi.call/return_dbl2.c: Likewise.
1011 * testsuite/libffi.call/return_fl.c: Likewise.
1012 * testsuite/libffi.call/return_fl1.c: Likewise.
1013 * testsuite/libffi.call/return_fl2.c: Likewise.
1014 * testsuite/libffi.call/return_fl3.c: Likewise.
1015 * testsuite/libffi.call/closure_fn6.c: Likewise.
1017 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1019 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1020 here to be used by other test cases too.
1022 * testsuite/libffi.call/nested_struct10.c: New test case.
1023 * testsuite/libffi.call/nested_struct9.c: Likewise.
1024 * testsuite/libffi.call/nested_struct8.c: Likewise.
1025 * testsuite/libffi.call/nested_struct7.c: Likewise.
1026 * testsuite/libffi.call/nested_struct6.c: Likewise.
1027 * testsuite/libffi.call/nested_struct5.c: Likewise.
1028 * testsuite/libffi.call/nested_struct4.c: Likewise.
1030 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1032 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1033 * configure: Rebuilt.
1035 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1037 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1038 instead do the shifting inline.
1039 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1040 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1041 and 8 byte structs, for the remaining struct sizes don't call
1042 __lshrdi3, instead do the shifting inline.
1044 2005-12-07 Thiemo Seufer <ths@networkno.de>
1046 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1047 missing parentheses.
1048 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1049 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1050 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1051 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1052 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1054 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1056 (ffi_prep_closure): Improve trampoline instruction scheduling.
1057 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1059 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1061 * src/powerpc/ffi.c: Formatting.
1062 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1063 (ffi_prep_args64): Likewise.
1065 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1067 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1068 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1070 2005-09-26 Tom Tromey <tromey@redhat.com>
1072 * testsuite/libffi.call/float1.c (value_type): New typedef.
1073 (CANARY): New define.
1074 (main): Check for result buffer overflow.
1075 * src/powerpc/linux64.S: Handle linux64 long double returns.
1076 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1077 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1079 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1082 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1084 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1086 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1088 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1089 (AH_BOTTOM): Add FFI_HIDDEN definition.
1090 * configure: Rebuilt.
1091 * fficonfig.h.in: Rebuilt.
1092 * src/powerpc/ffi.c (hidden): Remove.
1093 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1094 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1095 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1096 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1097 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1098 add FFI_HIDDEN to its prototype.
1099 (ffi_closure_SYSV_inner): New.
1100 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1101 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1103 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1106 * configure: Rebuilt.
1107 * configure.ac: Handle i*86-*-gnu*.
1109 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1111 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1112 DW_CFA_offset_extended_sf rather than
1113 DW_CFA_GNU_negative_offset_extended.
1114 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1116 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1118 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1120 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1121 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1122 partially on register.
1123 (ffi_closure_helper_SYSV): Likewise.
1124 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1126 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1128 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1130 * src/sh64/ffi.c (simple_type): Remove.
1131 (return_type): Handle small structures correctly.
1132 (ffi_prep_args): Likewise.
1133 (ffi_call): Likewise.
1134 (ffi_closure_helper_SYSV): Likewise.
1135 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1136 Emit position independent code if PIC and remove wrong datalabel
1137 prefixes from EH data.
1139 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1141 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1142 * Makefile.in: Regenerate.
1143 * include/Makefile.in: Likewise.
1144 * testsuite/Makefile.in: Likewise.
1145 * configure.ac: Add POWERPC_FREEBSD rules.
1146 * configure: Regenerate.
1147 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1148 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1149 * src/powerpc/ffi.c: Add flags to handle small structure returns
1151 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1153 (ffi_closure_helper_SYSV): Likewise.
1154 * src/powerpc/ppc_closure.S: Add return types for small structures.
1155 * src/powerpc/sysv.S: Add bits to handle small structures for
1158 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1160 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1161 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1162 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1164 2005-07-05 Randolph Chung <tausq@debian.org>
1166 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1167 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1168 structures. Kill compilation warnings.
1169 (ffi_closure_inner_LINUX): Print return values as hex in debug
1170 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1171 Properly handle 5-7 byte structure returns.
1172 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1173 (FFI_TYPE_SMALL_STRUCT2): Remove.
1174 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1175 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1176 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1177 (checksmst1, checksmst2): Remove.
1178 (checksmst3): Optimize handling of 3-byte struct returns.
1179 (checksmst567): Properly handle 5-7 byte struct returns.
1181 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1184 * src/mips/n32.S: Enforce PIC code.
1185 * src/mips/o32.S: Likewise.
1187 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1189 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1190 * configure: Regenerate.
1192 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1194 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1195 to call ffi_closure_helper_SYSV. Append @local instead.
1196 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1198 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1200 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1201 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1202 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1203 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1204 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1206 2005-05-09 Mike Stump <mrs@apple.com>
1208 * configure: Regenerate.
1210 2005-05-08 Richard Henderson <rth@redhat.com>
1213 * src/alpha/osf.S: Update unwind into to match code.
1215 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1216 Richard Henderson <rth@redhat.com>
1218 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1220 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1221 (ffi_prep_closure): Set carry bit if sse-used flag set.
1222 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1223 Only load sse registers if ssecount non-zero.
1224 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1226 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1228 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1229 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1230 * configure: Regenerate.
1232 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1234 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1235 have Tcl8.3-compatible intermediate variable.
1237 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1238 Hans-Peter Nilsson <hp@axis.com>
1240 * Makefile.am: Add CRIS support.
1241 * configure.ac: Likewise.
1242 * Makefile.in, configure, testsuite/Makefile.in,
1243 include/Makefile.in: Regenerate.
1244 * src/cris: New directory.
1245 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1246 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1248 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1249 \r?\n in output tests.
1251 2005-04-12 Mike Stump <mrs@apple.com>
1253 * configure: Regenerate.
1255 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1257 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1259 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1261 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1263 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1264 to 64 bits in ILP32 mode.
1265 Load 64 bits even for short data.
1267 2005-03-23 Mike Stump <mrs@apple.com>
1269 * src/powerpc/darwin.S: Update for -m64 multilib.
1270 * src/powerpc/darwin_closure.S: Likewise.
1272 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1274 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1275 Do not set tool_include_dir.
1276 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1278 * include/Makefile.am: Set gcc_version and toollibffidir.
1279 * include/Makefile.in: Regenerate.
1281 2005-02-22 Andrew Haley <aph@redhat.com>
1283 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1284 odd-numbered register pairs for 64-bit integer types.
1286 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1289 * testsuite/libffi.call/return_ll1.c: New test case.
1291 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1293 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1294 * testsuite/libffi.call/float.c: Ditto.
1295 * testsuite/libffi.call/float2.c: Ditto.
1296 * testsuite/libffi.call/float3.c: Ditto.
1298 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1300 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1302 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1304 * testsuite/libffi.special/special.exp (cxx_options): Add
1307 2004-12-31 Richard Henderson <rth@redhat.com>
1309 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1310 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1311 offset parameters with a type parameter; deduce size and structure
1312 alignment. Update all users.
1314 2004-12-31 Richard Henderson <rth@redhat.com>
1316 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1317 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1318 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1319 into ffi_prep_closure.
1320 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1323 2004-12-27 Richard Henderson <rth@redhat.com>
1325 * src/x86/unix64.S: Fix typo in unwind info.
1327 2004-12-25 Richard Henderson <rth@redhat.com>
1329 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1330 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1331 (merge_classes): Check for it.
1333 (classify_argument): Pass byte_offset by value; perform all updates
1335 (examine_argument): Add classes argument; handle
1336 X86_64_COMPLEX_X87_CLASS.
1337 (ffi_prep_args): Merge into ...
1338 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1339 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1340 (ffi_fill_return_value): Remove.
1341 (ffi_prep_closure): Remove dead assert.
1342 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1343 Rewrite to use struct register_args instead of va_list. Create
1344 flags for handling structure returns.
1345 * src/x86/unix64.S: Remove dead strings.
1346 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1347 stack frame with ffi_call. Handle structure returns properly.
1348 (float2sse, floatfloat2sse, double2sse): Remove.
1349 (sse2float, sse2double, sse2floatfloat): Remove.
1350 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1351 to handle structure returns properly.
1353 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1355 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1357 * Makefile.in: Regenerated.
1359 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1361 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1362 * configure, aclocal.m4, Makefile.in: Regenerate.
1363 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1365 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1367 * configure: Regenerate for libtool change.
1369 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1371 * configure: Regenerate for libtool reversion.
1373 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1375 * configure: Regenerate for libtool change.
1377 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1379 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1381 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1383 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1384 of jal. Use an absolute encoding for the frame information.
1386 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1388 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1389 * acinclude.m4: Delete logic for sincludes.
1390 * aclocal.m4, Makefile.in, configure: Regenerate.
1391 * include/Makefile: Likewise.
1392 * testsuite/Makefile: Likewise.
1394 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1396 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1397 on a 8-byte boundary.
1398 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1400 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1402 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1403 long long values. Round stack allocation to a multiple of 8 bytes
1404 for ATPCS compatibility.
1405 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1406 names. Handle returning long long types. Add Thumb and interworking
1407 support. Improve soft-float code.
1409 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1411 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1412 (libffi_exit): New function.
1413 (libffi_init): Build the testglue wrapper if needed.
1415 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1418 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1420 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1422 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1424 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1426 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1427 * testsuite/libffi.call/float3.c: New test case.
1429 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1431 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1432 the function returning a structure pointed with R2.
1433 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1434 the structure return value if T bit set. Emit position
1435 independent code and EH data if PIC.
1437 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1439 * Makefile.am: Add m32r support.
1440 * configure.ac: Likewise.
1441 * Makefile.in: Regenerate.
1442 * confiugre: Regenerate.
1443 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1444 (uint64, sint64, double, longdouble)
1445 * src/m32r: New directory.
1446 * src/m32r/ffi.c: New file.
1447 * src/m32r/sysv.S: Likewise.
1448 * src/m32r/ffitarget.h: Likewise.
1450 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1452 * testsuite/libffi.call/negint.c: New test case.
1454 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1457 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1458 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1459 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1462 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1464 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1465 * testsuite/libffi.call/promotion.c: Likewise.
1466 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1468 * testsuite/libffi.call/return_sc.c: Likewise.
1469 * testsuite/libffi.call/return_uc.c: Likewise.
1471 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1473 * src/powerpc/darwin.S: Fix comments and identation.
1474 * src/powerpc/darwin_closure.S: Likewise.
1476 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1478 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1479 (ffi_prep_args): Handle longdouble arguments.
1480 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1482 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1483 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1485 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1486 * src/types.c: Defined longdouble size and alignment for darwin.
1488 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1490 * src/powerpc/aix.S: Remove whitespaces.
1491 * src/powerpc/aix_closure.S: Likewise.
1492 * src/powerpc/asm.h: Likewise.
1493 * src/powerpc/ffi.c: Likewise.
1494 * src/powerpc/ffitarget.h: Likewise.
1495 * src/powerpc/linux64.S: Likewise.
1496 * src/powerpc/linux64_closure.S: Likewise.
1497 * src/powerpc/ppc_closure.S: Likewise.
1498 * src/powerpc/sysv.S: Likewise.
1500 2004-08-30 Anthony Green <green@redhat.com>
1502 * Makefile.am: Add frv support.
1503 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1504 * configure.ac: Read configure.host.
1505 * configure.in: Read configure.host.
1506 * configure.host: New file. frv-elf needs libgloss.
1507 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1508 alignment. This is needed to frv and shouldn't harm the others.
1509 * include/ffi_common.h (ALIGN_DOWN): New macro.
1510 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1512 2004-08-24 David Daney <daney@avtrex.com>
1514 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1515 * testsuite/libffi.call/closure_fn1.c: Likewise.
1516 * testsuite/libffi.call/closure_fn2.c Likewise.
1517 * testsuite/libffi.call/closure_fn3.c: Likewise.
1518 * testsuite/libffi.call/closure_fn4.c: Likewise.
1519 * testsuite/libffi.call/closure_fn5.c: Likewise.
1520 * testsuite/libffi.call/cls_18byte.c: Likewise.
1521 * testsuite/libffi.call/cls_19byte.c: Likewise.
1522 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1523 * testsuite/libffi.call/cls_20byte.c: Likewise.
1524 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1525 * testsuite/libffi.call/cls_24byte.c: Likewise.
1526 * testsuite/libffi.call/cls_2byte.c: Likewise.
1527 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1528 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1529 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1530 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1531 * testsuite/libffi.call/cls_4byte.c: Likewise.
1532 * testsuite/libffi.call/cls_64byte.c: Likewise.
1533 * testsuite/libffi.call/cls_6byte.c: Likewise.
1534 * testsuite/libffi.call/cls_7byte.c: Likewise.
1535 * testsuite/libffi.call/cls_8byte.c: Likewise.
1536 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1537 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1538 * testsuite/libffi.call/cls_align_double.c: Likewise.
1539 * testsuite/libffi.call/cls_align_float.c: Likewise.
1540 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1541 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1542 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1543 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1544 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1545 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1546 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1547 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1548 * testsuite/libffi.call/cls_double.c: Likewise.
1549 * testsuite/libffi.call/cls_float.c: Likewise.
1550 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1551 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1552 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1553 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1554 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1555 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1556 * testsuite/libffi.call/cls_schar.c: Likewise.
1557 * testsuite/libffi.call/cls_sint.c: Likewise.
1558 * testsuite/libffi.call/cls_sshort.c: Likewise.
1559 * testsuite/libffi.call/cls_uchar.c: Likewise.
1560 * testsuite/libffi.call/cls_uint.c: Likewise.
1561 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1562 * testsuite/libffi.call/cls_ushort.c: Likewise.
1563 * testsuite/libffi.call/nested_struct.c: Likewise.
1564 * testsuite/libffi.call/nested_struct1.c: Likewise.
1565 * testsuite/libffi.call/nested_struct2.c: Likewise.
1566 * testsuite/libffi.call/nested_struct3.c: Likewise.
1567 * testsuite/libffi.call/problem1.c: Likewise.
1568 * testsuite/libffi.special/unwindtest.cc: Likewise.
1569 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1571 * testsuite/libffi.call/cls_16byte.c: Likewise.
1572 * testsuite/libffi.call/cls_5byte.c: Likewise.
1574 2004-08-23 David Daney <daney@avtrex.com>
1577 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1578 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1579 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1580 parameters and return types.
1581 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1582 (ffi_prep_closure): Ditto.
1583 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1584 alignment calculations.
1585 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1586 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1588 2004-08-14 Casey Marshall <csm@gnu.org>
1590 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1591 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1592 integer (O32 ABI only).
1593 (ffi_prep_closure): new function.
1594 (ffi_closure_mips_inner_O32): new function.
1595 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1596 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1597 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1598 64 bit integers correctly.
1599 (ffi_closure_O32): new function.
1600 Added DWARF-2 unwind info for both functions.
1602 2004-08-10 Andrew Haley <aph@redhat.com>
1604 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1606 2004-08-01 Robert Millan <robertmh@gnu.org>
1608 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1609 * configure: Regenerate.
1611 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1613 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1614 and mmap() explicitly instead of relying on preset autoconf cache
1616 * aclocal.m4: Regenerate.
1617 * configure: Regenerate.
1619 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1621 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1622 (ffi_check_float_struct): Remove unused prototype.
1624 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1626 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1627 character on Darwin, use '\n\t' instead.
1629 2004-06-26 Matthias Klose <doko@debian.org>
1631 * libtool-version: Fix typo in revision/age.
1633 2004-06-17 Matthias Klose <doko@debian.org>
1635 * libtool-version: New.
1636 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1637 * Makefile.in: Regenerate.
1639 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1641 * Makefile.am: Remove useless multilib rules.
1642 * Makefile.in: Regenerate.
1643 * aclocal.m4: Regenerate with automake 1.8.5.
1644 * configure.ac: Remove useless multilib configury.
1645 * configure: Regenerate.
1647 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1649 * .cvsignore: New file.
1651 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1653 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1655 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1658 2004-06-06 Sean McNeil <sean@mcneil.com>
1660 * configure.ac: Add x86_64-*-freebsd* support.
1661 * configure: Regenerate.
1663 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1666 * configure.ac: Test for existence of mmap and sys/mman.h before
1667 checking blacklist. Fix suggested by Jim Wilson.
1668 * configure: Regenerate.
1670 2004-04-26 Matt Austern <austern@apple.com>
1672 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1674 * src/powerpc/darwin_closure.S: Likewise.
1676 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1678 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1679 error. Reported by Thomas Heller <theller@python.net>.
1680 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1681 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1683 2004-03-20 Matthias Klose <doko@debian.org>
1685 * src/pa/linux.S: Fix typo.
1687 2004-03-19 Matthias Klose <doko@debian.org>
1689 * Makefile.am: Update.
1690 * Makefile.in: Regenerate.
1691 * src/pa/ffi.h.in: Remove.
1692 * src/pa/ffitarget.h: New file.
1694 2004-02-10 Randolph Chung <tausq@debian.org>
1696 * Makefile.am: Add PA support.
1697 * Makefile.in: Regenerate.
1698 * include/Makefile.in: Regenerate.
1699 * configure.ac: Add PA target.
1700 * configure: Regenerate.
1701 * src/pa/ffi.c: New file.
1702 * src/pa/ffi.h.in: Add PA support.
1703 * src/pa/linux.S: New file.
1704 * prep_cif.c: Add PA support.
1706 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1708 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1710 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1711 with ecif->cif->flags.
1712 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1714 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1715 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1716 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1717 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1719 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1721 * configure.in: Rename file to ...
1722 * configure.ac: ... this.
1723 * fficonfig.h.in: Regenerate.
1724 * Makefile.in: Regenerate.
1725 * include/Makefile.in: Regenerate.
1726 * testsuite/Makefile.in: Regenerate.
1728 2004-03-12 Matt Austern <austern@apple.com>
1730 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1731 changes in EH format resulting from addition of linkonce support.
1732 * src/powerpc/darwin_closure.S: Likewise.
1734 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1735 Paolo Bonzini <bonzini@gnu.org>
1737 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1738 Remove VPATH. Remove rules for object files. Remove multilib support.
1739 (AM_CCASFLAGS): Add.
1740 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1741 (AC_PREREQ): Bump version to 2.59.
1742 (AC_INIT): Fill with version info and bug address.
1743 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1744 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1745 De-precious CC so that the right flags are passed down to multilibs.
1746 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1747 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1748 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1749 * configure: Rebuilt.
1750 * aclocal.m4: Likewise.
1751 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1752 * fficonfig.h.in: Likewise.
1754 2004-03-11 Andreas Schwab <schwab@suse.de>
1756 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1757 arguments from fp registers only for the first 8 parameter slots.
1758 Don't convert a float parameter when passed in memory.
1760 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1762 * configure: Regenerate for config/accross.m4 correction.
1764 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1766 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1767 ecif->cif->bytes to bytes.
1768 (ffi_prep_cif_machdep): Add braces around nested if statement.
1770 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1772 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1774 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1775 (ffi_closure_helper_LINUX64): Fix typo.
1776 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1778 * testsuite/libffi.call/float.c: Likewise.
1779 * testsuite/libffi.call/float2.c: Likewise.
1781 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1783 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1784 long double function return and long double arg handling.
1785 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1786 Use "end_pfr" instead of "nf". Correct long double handling.
1788 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1790 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1791 space for long double return value. Adjust stack frame and offsets.
1792 Load f2 long double return.
1794 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1796 * src/types.c: Use 16 byte long double for POWERPC64.
1798 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1800 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1801 when the structure return address is passed in %o0.
1802 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1803 (ffi_v9_layout_struct): Align the field following a nested structure
1804 on a word boundary. Use memmove instead of memcpy.
1805 (ffi_call): Update call to ffi_V9_return_struct.
1806 (ffi_prep_closure): Define 'ctx' only for V8.
1807 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1808 and ffi_closure_sparc_inner_v9.
1809 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1810 Always skip the structure return address. For structures and long
1811 doubles, copy the argument directly.
1812 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1813 if required. Shift the maximum floating-point slot accordingly. For
1814 big structures, copy the argument directly; otherwise, left-justify the
1815 argument and call ffi_v9_layout_struct to lay out the structure on
1817 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1818 (ffi_closure_v8): Pass the structure return address. Update call to
1819 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1820 Skip the 'unimp' insn when returning long doubles and structures.
1821 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1822 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1823 FFI_TYPE_INT handling. Load structures both in integers and
1824 floating-point registers on return.
1825 * README: Update status of the SPARC port.
1827 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1829 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1831 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1833 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1835 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1836 value as of type ffi_arg, not unsigned int.
1838 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1840 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1843 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1845 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1848 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1850 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1853 2003-12-10 Richard Henderson <rth@redhat.com>
1855 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1856 size_t instead of int.
1858 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1860 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1861 * testsuite/libffi.call/many_win32.c (main): Replace variable
1862 int i with unsigned long ul.
1864 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1865 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1866 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1867 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1868 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1869 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1870 * testsuite/libffi.call/cls_align_float.c: Likewise.
1871 * testsuite/libffi.call/cls_align_double.c: Likewise.
1872 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1873 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1875 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1878 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1879 Align arguments to 32 bits.
1881 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1884 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1885 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1886 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1887 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1888 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1889 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1891 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1893 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1895 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1896 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1898 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1900 * Makefile.in: Rebuilt.
1901 * configure: Likewise.
1902 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1905 2003-11-21 Andreas Jaeger <aj@suse.de>
1906 Andreas Tobler <a.tobler@schweiz.ch>
1908 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1909 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1910 * Makefile.in: Rebuilt.
1911 * aclocal.m4: Likewise.
1912 * configure: Likewise.
1913 * fficonfig.h.in: Likewise.
1914 * testsuite/lib/libffi-dg.exp: Add include dir.
1915 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1916 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1917 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1918 for ffi_closure if available.
1919 * testsuite/libffi.call/closure_fn1.c: Likewise.
1920 * testsuite/libffi.call/closure_fn2.c: Likewise.
1921 * testsuite/libffi.call/closure_fn3.c: Likewise.
1922 * testsuite/libffi.call/closure_fn4.c: Likewise.
1923 * testsuite/libffi.call/closure_fn5.c: Likewise.
1924 * testsuite/libffi.call/cls_12byte.c: Likewise.
1925 * testsuite/libffi.call/cls_16byte.c: Likewise.
1926 * testsuite/libffi.call/cls_18byte.c: Likewise.
1927 * testsuite/libffi.call/cls_19byte.c: Likewise.
1928 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1929 * testsuite/libffi.call/cls_20byte.c: Likewise.
1930 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1931 * testsuite/libffi.call/cls_24byte.c: Likewise.
1932 * testsuite/libffi.call/cls_2byte.c: Likewise.
1933 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1934 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1935 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1936 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1937 * testsuite/libffi.call/cls_4byte.c: Likewise.
1938 * testsuite/libffi.call/cls_5byte.c: Likewise.
1939 * testsuite/libffi.call/cls_64byte.c: Likewise.
1940 * testsuite/libffi.call/cls_6byte.c: Likewise.
1941 * testsuite/libffi.call/cls_7byte.c: Likewise.
1942 * testsuite/libffi.call/cls_8byte.c: Likewise.
1943 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1944 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1945 * testsuite/libffi.call/cls_double.c: Likewise.
1946 * testsuite/libffi.call/cls_float.c: Likewise.
1947 * testsuite/libffi.call/cls_schar.c: Likewise.
1948 * testsuite/libffi.call/cls_sint.c: Likewise.
1949 * testsuite/libffi.call/cls_sshort.c: Likewise.
1950 * testsuite/libffi.call/cls_uchar.c: Likewise.
1951 * testsuite/libffi.call/cls_uint.c: Likewise.
1952 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1953 * testsuite/libffi.call/cls_ushort.c: Likewise.
1954 * testsuite/libffi.call/nested_struct.c: Likewise.
1955 * testsuite/libffi.call/nested_struct1.c: Likewise.
1956 * testsuite/libffi.call/nested_struct2.c: Likewise.
1957 * testsuite/libffi.call/nested_struct3.c: Likewise.
1958 * testsuite/libffi.call/problem1.c: Likewise.
1959 * testsuite/libffi.special/unwindtest.cc: Likewise.
1961 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
1963 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
1965 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
1967 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
1968 Add -lgcc_s to additional flags.
1970 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
1972 * configure.in, include/Makefile.am: PR libgcj/11147, install
1973 the ffitarget.h header file in a gcc versioned and target
1975 * configure: Regenerated.
1976 * Makefile.in, include/Makefile.in: Likewise.
1977 * testsuite/Makefile.in: Likewise.
1979 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
1981 * testsuite/libffi.call/closure_fn0.c: Print result and check
1982 with dg-output to make debugging easier.
1983 * testsuite/libffi.call/closure_fn1.c: Likewise.
1984 * testsuite/libffi.call/closure_fn2.c: Likewise.
1985 * testsuite/libffi.call/closure_fn3.c: Likewise.
1986 * testsuite/libffi.call/closure_fn4.c: Likewise.
1987 * testsuite/libffi.call/closure_fn5.c: Likewise.
1988 * testsuite/libffi.call/cls_12byte.c: Likewise.
1989 * testsuite/libffi.call/cls_16byte.c: Likewise.
1990 * testsuite/libffi.call/cls_18byte.c: Likewise.
1991 * testsuite/libffi.call/cls_19byte.c: Likewise.
1992 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1993 * testsuite/libffi.call/cls_20byte.c: Likewise.
1994 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1995 * testsuite/libffi.call/cls_24byte.c: Likewise.
1996 * testsuite/libffi.call/cls_2byte.c: Likewise.
1997 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1998 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1999 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2000 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2001 * testsuite/libffi.call/cls_4byte.c: Likewise.
2002 * testsuite/libffi.call/cls_5byte.c: Likewise.
2003 * testsuite/libffi.call/cls_64byte.c: Likewise.
2004 * testsuite/libffi.call/cls_6byte.c: Likewise.
2005 * testsuite/libffi.call/cls_7byte.c: Likewise.
2006 * testsuite/libffi.call/cls_8byte.c: Likewise.
2007 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2008 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2009 * testsuite/libffi.call/cls_double.c: Likewise.
2010 * testsuite/libffi.call/cls_float.c: Likewise.
2011 * testsuite/libffi.call/cls_schar.c: Likewise.
2012 * testsuite/libffi.call/cls_sint.c: Likewise.
2013 * testsuite/libffi.call/cls_sshort.c: Likewise.
2014 * testsuite/libffi.call/cls_uchar.c: Likewise.
2015 * testsuite/libffi.call/cls_uint.c: Likewise.
2016 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2017 * testsuite/libffi.call/cls_ushort.c: Likewise.
2018 * testsuite/libffi.call/problem1.c: Likewise.
2020 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2023 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2025 * testsuite/libffi.call/cls_9byte2.c: New test case.
2026 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2027 * testsuite/libffi.call/cls_64byte.c: Likewise.
2028 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2029 * testsuite/libffi.call/cls_19byte.c: Likewise.
2030 * testsuite/libffi.call/cls_18byte.c: Likewise.
2031 * testsuite/libffi.call/closure_fn4.c: Likewise.
2032 * testsuite/libffi.call/closure_fn5.c: Likewise.
2033 * testsuite/libffi.call/cls_schar.c: Likewise.
2034 * testsuite/libffi.call/cls_sint.c: Likewise.
2035 * testsuite/libffi.call/cls_sshort.c: Likewise.
2036 * testsuite/libffi.call/nested_struct2.c: Likewise.
2037 * testsuite/libffi.call/nested_struct3.c: Likewise.
2039 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2041 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2042 * testsuite/libffi.call/cls_uchar.c: Likewise.
2043 * testsuite/libffi.call/cls_uint.c: Likewise.
2044 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2045 * testsuite/libffi.call/cls_ushort.c: Likewise.
2046 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2048 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2050 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2053 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2055 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2056 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2058 2003-10-22 David Daney <ddaney@avtrex.com>
2060 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2061 __attribute__((__mode__(__SI__))) and friends.
2063 2003-10-22 Andreas Schwab <schwab@suse.de>
2065 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2067 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2069 * configure.in: AC_LINK_FILES(ffitarget.h).
2070 * configure: Regenerate.
2071 * Makefile.in: Likewise.
2072 * include/Makefile.in: Likewise.
2073 * testsuite/Makefile.in: Likewise.
2074 * fficonfig.h.in: Likewise.
2076 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2077 Richard Henderson <rth@redhat.com>
2079 Avoid that ffi.h includes fficonfig.h.
2081 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2082 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2083 (TARGET_SRC_MIPS_SGI): Removed.
2084 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2085 (MIPS_SGI): Removed.
2086 (CLEANFILES): Removed.
2087 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2089 * acconfig.h: Removed.
2090 * configure.in: Compute sizeofs only for double and long double.
2091 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2092 into AC_DEFINE instead of using acconfig.h. Create
2093 include/ffitarget.h instead of include/fficonfig.h. Rename
2094 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2095 AC_DEFINE EH_FRAME_FLAGS.
2096 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2097 (hack_DATA): Add ffitarget.h.
2098 * include/ffi.h.in: Remove all system specific definitions.
2099 Declare raw API even if it is not installed, why bother?
2100 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2101 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2102 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2103 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2104 * include/ffi_common.h (bool): Do not define.
2105 (ffi_assert): Accept failed assertion.
2106 (ffi_type_test): Return void and accept file/line.
2107 (FFI_ASSERT): Pass stringized failed assertion.
2108 (FFI_ASSERT_AT): New macro.
2109 (FFI_ASSERT_VALID_TYPE): New macro.
2110 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2111 UINT64, SINT64): Define here with gcc's __attribute__ macro
2113 (FLOAT32, ALIGN): Define here instead of in ffi.h
2114 * include/ffi-mips.h: Removed. Its content moved to
2115 src/mips/ffitarget.h after separating assembly and C sections.
2116 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2117 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2118 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2119 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2120 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2121 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2122 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2123 (ffi_type_test): Rewritten.
2124 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2125 FFI_ASSERT_VALID_TYPE.
2126 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2127 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2128 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2129 src/s390/ffitarget.h, src/sh/ffitarget.h,
2130 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2131 src/x86/ffitarget.h: New files.
2132 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2133 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2134 src/powerpc/aix.S, src/powerpc/darwin.S,
2135 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2136 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2137 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2138 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2139 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2142 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2144 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2145 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2147 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2149 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2150 Used when FFI_DEBUG = 1.
2152 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2154 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2157 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2159 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2160 FFI_MIPS_O32 for O32 ABI.
2162 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2164 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2165 SPARC64. Cleanup whitespaces.
2167 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2169 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2170 strongarm, xscale. Cleanup whitespaces.
2171 * testsuite/libffi.call/closure_fn1.c: Likewise.
2172 * testsuite/libffi.call/closure_fn2.c: Likewise.
2173 * testsuite/libffi.call/closure_fn3.c: Likewise.
2174 * testsuite/libffi.call/cls_12byte.c: Likewise.
2175 * testsuite/libffi.call/cls_16byte.c: Likewise.
2176 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2177 * testsuite/libffi.call/cls_20byte.c: Likewise.
2178 * testsuite/libffi.call/cls_24byte.c: Likewise.
2179 * testsuite/libffi.call/cls_2byte.c: Likewise.
2180 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2181 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2182 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2183 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2184 * testsuite/libffi.call/cls_4byte.c: Likewise.
2185 * testsuite/libffi.call/cls_5byte.c: Likewise.
2186 * testsuite/libffi.call/cls_6byte.c: Likewise.
2187 * testsuite/libffi.call/cls_7byte.c: Likewise.
2188 * testsuite/libffi.call/cls_8byte.c: Likewise.
2189 * testsuite/libffi.call/cls_double.c: Likewise.
2190 * testsuite/libffi.call/cls_float.c: Likewise.
2191 * testsuite/libffi.call/cls_uchar.c: Likewise.
2192 * testsuite/libffi.call/cls_uint.c: Likewise.
2193 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2194 * testsuite/libffi.call/cls_ushort.c: Likewise.
2195 * testsuite/libffi.call/nested_struct.c: Likewise.
2196 * testsuite/libffi.call/nested_struct1.c: Likewise.
2197 * testsuite/libffi.call/problem1.c: Likewise.
2198 * testsuite/libffi.special/unwindtest.cc: Likewise.
2199 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2201 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2203 * src/powerpc/aix.S: Cleanup whitespaces.
2204 * src/powerpc/aix_closure.S: Likewise.
2206 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2208 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2209 * src/powerpc/darwin_closure.S: Likewise.
2210 * src/powerpc/ffi_darwin.c: Likewise.
2212 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2213 David Edelsohn <edelsohn@gnu.org>
2215 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2216 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2218 * src/powerpc/darwin_closure.S: Likewise.
2219 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2220 according to the Darwin/AIX ABI.
2221 (ffi_prep_cif_machdep): Likewise.
2222 (ffi_closure_helper_DARWIN): Likewise.
2223 Remove the outgoing parameter stack logic. Simplify the evaluation
2224 of the different CASE types.
2225 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2226 statement in the trampoline code.
2228 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2230 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2231 for the register size.
2232 (ffi_closure_helper_SYSV): Handle the structure return value
2234 (ffi_closure_helper_SYSV): Return the appropriate type when
2235 the registers are used for the structure return value.
2236 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2237 the 64-bit return value. Update copyright years.
2239 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2241 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2242 srcdir for ffi_mips.h.
2244 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2246 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2248 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2249 placement of float result.
2250 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2251 cast of "resp" for big-endian 64 bit machines.
2253 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2255 * src/types.c (double, longdouble): Merge identical SH and ARM
2256 typedefs, and add POWERPC64.
2257 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2258 struct split over gpr and rest.
2259 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2260 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2262 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2264 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2267 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2269 * configure: Regenerate.
2271 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2273 * Makefile.am: Remove build rules for ffitest.
2274 * Makefile.in: Rebuilt.
2276 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2278 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2279 about implicit declaration of abort().
2281 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2283 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2284 * Makefile.in: Rebuilt.
2285 * configure.in: Add dejagnu test framework.
2286 * configure: Rebuilt.
2288 * testsuite/Makefile.am: New file.
2289 * testsuite/Makefile.in: Built
2290 * testsuite/lib/libffi-dg.exp: New file.
2291 * testsuite/config/default.exp: Likewise.
2292 * testsuite/libffi.call/call.exp: Likewise.
2293 * testsuite/libffi.call/ffitest.h: Likewise.
2294 * testsuite/libffi.call/closure_fn0.c: Likewise.
2295 * testsuite/libffi.call/closure_fn1.c: Likewise.
2296 * testsuite/libffi.call/closure_fn2.c: Likewise.
2297 * testsuite/libffi.call/closure_fn3.c: Likewise.
2298 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2299 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2300 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2301 * testsuite/libffi.call/cls_2byte.c: Likewise.
2302 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2303 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2304 * testsuite/libffi.call/cls_4byte.c: Likewise.
2305 * testsuite/libffi.call/cls_5byte.c: Likewise.
2306 * testsuite/libffi.call/cls_6byte.c: Likewise.
2307 * testsuite/libffi.call/cls_7byte.c: Likewise.
2308 * testsuite/libffi.call/cls_8byte.c: Likewise.
2309 * testsuite/libffi.call/cls_12byte.c: Likewise.
2310 * testsuite/libffi.call/cls_16byte.c: Likewise.
2311 * testsuite/libffi.call/cls_20byte.c: Likewise.
2312 * testsuite/libffi.call/cls_24byte.c: Likewise.
2313 * testsuite/libffi.call/cls_double.c: Likewise.
2314 * testsuite/libffi.call/cls_float.c: Likewise.
2315 * testsuite/libffi.call/cls_uchar.c: Likewise.
2316 * testsuite/libffi.call/cls_uint.c: Likewise.
2317 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2318 * testsuite/libffi.call/cls_ushort.c: Likewise.
2319 * testsuite/libffi.call/float.c: Likewise.
2320 * testsuite/libffi.call/float1.c: Likewise.
2321 * testsuite/libffi.call/float2.c: Likewise.
2322 * testsuite/libffi.call/many.c: Likewise.
2323 * testsuite/libffi.call/many_win32.c: Likewise.
2324 * testsuite/libffi.call/nested_struct.c: Likewise.
2325 * testsuite/libffi.call/nested_struct1.c: Likewise.
2326 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2327 * testsuite/libffi.call/problem1.c: Likewise.
2328 * testsuite/libffi.call/promotion.c: Likewise.
2329 * testsuite/libffi.call/return_ll.c: Likewise.
2330 * testsuite/libffi.call/return_sc.c: Likewise.
2331 * testsuite/libffi.call/return_uc.c: Likewise.
2332 * testsuite/libffi.call/strlen.c: Likewise.
2333 * testsuite/libffi.call/strlen_win32.c: Likewise.
2334 * testsuite/libffi.call/struct1.c: Likewise.
2335 * testsuite/libffi.call/struct2.c: Likewise.
2336 * testsuite/libffi.call/struct3.c: Likewise.
2337 * testsuite/libffi.call/struct4.c: Likewise.
2338 * testsuite/libffi.call/struct5.c: Likewise.
2339 * testsuite/libffi.call/struct6.c: Likewise.
2340 * testsuite/libffi.call/struct7.c: Likewise.
2341 * testsuite/libffi.call/struct8.c: Likewise.
2342 * testsuite/libffi.call/struct9.c: Likewise.
2343 * testsuite/libffi.special/special.exp: New file.
2344 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2345 * testsuite/libffi.special/unwindtest.cc: Likewise.
2348 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2350 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2353 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2355 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2357 (ffi_closure_helper_LINUX64): Likewise.
2358 * src/powerpc/linux64.S: Remove code writing to parm save area.
2359 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2360 address in lr from ffi_closure_helper_LINUX64 call to calculate
2361 table address. Optimize function tail.
2363 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2365 * src/sparc/ffi.c: Handle all floating point registers.
2366 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2368 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2370 * README: Note that libffi is not part of GCC. Update the project
2373 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2375 * src/powerpc/ppc_closure.S: Include ffi.h.
2377 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2379 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2380 Use C style comments.
2382 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2384 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2385 * Makefile.in: Regenerate.
2386 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2387 * configure: Regenerate.
2388 * include/ffi.h.in: Add SHmedia support.
2389 * src/sh64/ffi.c: New file.
2390 * src/sh64/sysv.S: New file.
2392 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2394 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2395 should be read-only.
2396 * configure: Rebuilt.
2397 * fficonfig.h.in: Rebuilt.
2398 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2399 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2400 * src/powerpc/linux64.S: Likewise.
2401 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2402 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2403 if -fpic/-fPIC/-mrelocatable.
2404 * src/powerpc/powerpc_closure.S: Likewise.
2405 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2406 #write in .eh_frame flags.
2407 * src/sparc/v9.S: Likewise.
2408 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2409 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2410 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2412 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2414 Fixes PR bootstrap/10656
2415 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2416 support for .register pseudo-op.
2417 * src/sparc/v8.S: Use it.
2418 * fficonfig.h.in: Rebuilt.
2419 * configure: Rebuilt.
2421 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2423 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2424 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2425 Make it the default on POWERPC64.
2426 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2427 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2428 * configure: Rebuilt.
2429 * src/powerpc/ffi.c (hidden): Define.
2430 (ffi_prep_args_SYSV): Renamed from
2431 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2432 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2433 ASM_NEEDS_REGISTERS64): New.
2434 (ffi_prep_args64): New function.
2435 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2436 (ffi_call): Likewise.
2437 (ffi_prep_closure): Likewise.
2438 (flush_icache): Surround by #ifndef POWERPC64.
2439 (ffi_dblfl): New union type.
2440 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2441 (ffi_closure_helper_LINUX64): New function.
2442 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2444 * src/powerpc/sysv.S: Likewise.
2445 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2446 * src/powerpc/linux64.S: New file.
2447 * src/powerpc/linux64_closure.S: New file.
2448 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2449 src/powerpc/linux64_closure.S.
2450 (TARGET_SRC_POWERPC): Likewise.
2452 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2453 closure_test_fn3): Fix result printing on big-endian 64-bit
2455 (main): Print tst2_arg instead of uninitialized tst2_result.
2457 * src/ffitest.c (main): Hide what closure pointer really points to
2460 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2462 * configure.in (arm-*-netbsdelf*): Add configuration.
2463 (configure): Regenerated.
2465 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2467 * include/Makefile.in: Regenerate.
2469 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2471 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2473 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2474 Receive closure pointer through parameter, read args using
2475 __builtin_dwarf_cfa.
2476 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2478 2003-03-12 Andreas Schwab <schwab@suse.de>
2480 * configure.in: Avoid trailing /. in toolexeclibdir.
2481 * configure: Rebuilt.
2483 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2485 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2487 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2489 * libffi/src/powerpc/darwin_closure.S:
2490 Fix alignement bug, allocate 8 bytes for the result.
2491 * libffi/src/powerpc/aix_closure.S:
2493 * libffi/src/powerpc/ffi_darwin.c:
2494 Update stackframe description for aix/darwin_closure.S.
2496 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2498 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2501 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2502 Andreas Schwab <schwab@suse.de>
2504 * configure.in: Adjust command to source config-ml.in to account
2505 for changes to the libffi_basedir definition.
2506 (libffi_basedir): Remove ${srcdir} from value and include trailing
2509 * configure: Regenerate.
2511 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2513 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2515 2003-01-28 Andrew Haley <aph@redhat.com>
2517 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2518 * src/x86/ffi64.c (ffi_prep_closure): New.
2519 (ffi_closure_UNIX64_inner): New.
2520 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2522 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2524 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2525 Remove USE_LIBDIR conditional.
2526 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2527 * Makefile.in, configure: Rebuilt.
2529 2003-01027 David Edelsohn <edelsohn@gnu.org>
2531 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2532 * Makefile.in: Regenerate.
2534 2003-01-22 Andrew Haley <aph@redhat.com>
2536 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2539 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2541 * src/powerpc/darwin.S: Add unwind info.
2542 * src/powerpc/darwin_closure.S: Likewise.
2544 2003-01-14 Andrew Haley <aph@redhat.com>
2546 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2547 (ffi_prep_cif_machdep): Likewise.
2548 * src/x86/unix64.S: Add unwind info.
2550 2003-01-14 Andreas Jaeger <aj@suse.de>
2552 * src/ffitest.c (main): Only use ffi_closures if those are
2555 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2557 * libffi/src/ffitest.c
2558 add closure testcases
2560 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2562 * libffi/src/powerpc/ffi.c
2563 fix alignment bug for float (4 byte aligned iso 8 byte)
2565 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2567 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2568 * src/powerpc/darwin.S: Remove RCS version string.
2570 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2572 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2573 * src/ffitest.c (main): Use static storage for closure.
2574 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2575 * src/sparc/v8.S (ffi_closure_v8): New.
2576 * src/sparc/v9.S (ffi_closure_v9): New.
2578 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2580 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2581 enumeration for X86_WIN32.
2582 * src/x86/win32.S: Added ffi_call_STDCALL function
2584 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2585 switch cases for recognising FFI_STDCALL and
2586 calling ffi_call_STDCALL if target is X86_WIN32.
2587 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2588 stdcall versions of the "my_strlen" and "many"
2589 test functions (for X86_WIN32).
2590 Added test cases to test stdcall invocation using
2593 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2595 * src/sh/sysv.S: Add DWARF2 unwind info.
2597 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2599 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2601 2002-11-26 Jim Wilson <wilson@redhat.com>
2603 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2605 2002-11-23 H.J. Lu <hjl@gnu.org>
2607 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2608 Include ../config/accross.m4.
2609 * aclocal.m4; Rebuild.
2610 * configure: Likewise.
2612 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2614 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2616 2002-11-11 DJ Delorie <dj@redhat.com>
2618 * configure.in: Look for common files in the right place.
2620 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2622 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2623 raw data as _Jv_word values, not ffi_raw.
2624 (ffi_java_ptrarray_to_raw): Likewise.
2625 (ffi_java_rvalue_to_raw): New function.
2626 (ffi_java_raw_call): Call it.
2627 (ffi_java_raw_to_rvalue): New function.
2628 (ffi_java_translate_args): Call it.
2629 * src/ffitest.c (closure_test_fn): Interpret return value
2630 as ffi_arg, not int.
2631 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2632 FFI_TYPE_POINTER case.
2633 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2634 values extended to word size.
2636 2002-10-02 Andreas Jaeger <aj@suse.de>
2638 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2640 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2642 * include/ffi.h.in: Fix i386 win32 compilation.
2644 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2646 * configure.in: Add s390x-*-linux-* target.
2647 * configure: Regenerate.
2648 * include/ffi.h.in: Define S390X for s390x targets.
2649 (FFI_CLOSURES): Define for s390/s390x.
2650 (FFI_TRAMPOLINE_SIZE): Likewise.
2651 (FFI_NATIVE_RAW_API): Likewise.
2652 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2653 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2654 * src/s390/ffi.c: Major rework of existing code. Add support for
2655 s390x targets. Add closure support.
2656 * src/s390/sysv.S: Likewise.
2658 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2660 * src/arm/sysv.S: Fix typo.
2662 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2664 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2665 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2666 (ffi_call_SYSV): Handle soft-float.
2668 2002-09-27 Bo Thorsen <bo@suse.de>
2670 * include/ffi.h.in: Fix multilib x86-64 support.
2672 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2674 * Makefile.am (all-multi): Fix multilib parallel build.
2676 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2678 * configure.in (sh[34]*-*-linux*): Add brackets.
2679 * configure: Regenerate.
2681 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2683 * Makefile.am: Add SH support.
2684 * Makefile.in: Regenerate.
2685 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2686 * configure: Regenerate.
2687 * include/ffi.h.in: Add SH support.
2688 * src/sh/ffi.c: New file.
2689 * src/sh/sysv.S: New file.
2690 * src/types.c: Add SH support.
2692 2002-07-16 Bo Thorsen <bo@suse.de>
2694 * src/x86/ffi64.c: New file that adds x86-64 support.
2695 * src/x86/unix64.S: New file that handles argument setup for
2697 * src/x86/sysv.S: Don't use this on x86-64.
2698 * src/x86/ffi.c: Don't use this on x86-64.
2700 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2702 * src/ffitest.c (struct6): New test that tests a special case in
2704 (struct7): Likewise.
2705 (struct8): Likewise.
2706 (struct9): Likewise.
2707 (closure_test_fn): Silence warning about this when it's not used.
2708 (main): Add the new tests.
2709 (main): Fix a couple of wrong casts and silence some compiler warnings.
2710 * include/ffi.h.in: Add x86-64 ABI definition.
2711 * fficonfig.h.in: Regenerate.
2712 * Makefile.am: Add x86-64 support.
2713 * configure.in: Likewise.
2714 * Makefile.in: Regenerate.
2715 * configure: Likewise.
2717 2002-06-24 Bo Thorsen <bo@suse.de>
2719 * src/types.c: Merge settings for similar architectures.
2720 Add x86-64 sizes and alignments.
2722 2002-06-23 Bo Thorsen <bo@suse.de>
2724 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2725 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2726 * src/mips/ffi.c (ffi_prep_args): Likewise.
2727 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2729 2002-07-18 H.J. Lu (hjl@gnu.org)
2731 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2732 (libffi_la_SOURCES): Support MIPS_LINUX.
2733 (libffi_convenience_la_SOURCES): Likewise.
2734 * Makefile.in: Regenerated.
2736 * configure.in (mips64*-*): Skip.
2737 (mips*-*-linux*): New.
2738 * configure: Regenerated.
2740 * src/mips/ffi.c: Include <sgidefs.h>.
2742 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2744 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2746 2002-05-27 Roger Sayle <roger@eyesopen.com>
2748 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2750 2002-05-27 Bo Thorsen <bo@suse.de>
2752 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2755 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2757 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2758 beginning of function (for older apple cc).
2760 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2762 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2763 script entry, and set LD to it when configuring multilibs.
2764 * configure: Rebuilt.
2766 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2768 * configure.in (sparc64-*-netbsd*): Add target.
2769 (sparc-*-netbsdelf*): Likewise.
2770 * configure: Regenerate.
2772 2002-04-28 David S. Miller <davem@redhat.com>
2774 * configure.in, configure: Fix SPARC test in previous change.
2776 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2778 * Makefile.am: Add Linux for S/390 support.
2779 * Makefile.in: Regenerate.
2780 * configure.in: Add Linux for S/390 support.
2781 * configure: Regenerate.
2782 * include/ffi.h.in: Add Linux for S/390 support.
2783 * src/s390/ffi.c: New file from libffi CVS tree.
2784 * src/s390/sysv.S: New file from libffi CVS tree.
2786 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2788 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2790 * src/sparc/v8.S: Use it.
2791 * src/sparc/v9.S: Likewise.
2792 * fficonfig.h.in: Rebuilt.
2793 * configure: Rebuilt.
2795 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2797 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2799 * src/ia64/unix.S: Add unwind information. Fix comments.
2800 Save sp in a way that's compatible with unwind info.
2801 (ffi_call_unix): Correctly restore sp in all cases.
2802 * src/ia64/ffi.c: Add, fix comments.
2804 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2806 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2808 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2810 * configure.in (alpha*-*-netbsd*): Add target.
2811 * configure: Regenerate.
2813 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2815 * src/sparc/v8.S: Add unwind info.
2816 * src/sparc/v9.S: Likewise.
2818 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2820 * configure.in: Enable i*86-*-netbsdelf*.
2821 * configure: Rebuilt.
2823 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2826 * src/mips/n32.s: Delete
2827 * src/mips/o32.s: Delete
2829 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2831 * configure.in: Enable alpha*-*-freebsd*.
2832 * configure: Rebuilt.
2834 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2836 * Makefile.am: libfficonvenience -> libffi_convenience.
2837 * Makefile.in: Rebuilt.
2839 * Makefile.am: Define ffitest_OBJECTS.
2840 * Makefile.in: Rebuilt.
2842 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2843 David Edelsohn <edelsohn@gnu.org>
2845 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2846 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2847 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2848 * Makefile.in: Regenerate.
2849 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2850 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2851 (flush_icache, flush_range): New functions.
2852 (ffi_closure_helper_DARWIN): New function.
2853 * src/powerpc/aix_closure.S: New file.
2854 * src/powerpc/darwin_closure.S: New file.
2856 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2858 * include/ffi.h.in: Add typedef for ffi_arg.
2859 * src/ffitest.c (main): Declare rint with ffi_arg.
2861 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2863 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2864 number of GPRs for floating-point arguments.
2866 2002-01-31 Anthony Green <green@redhat.com>
2868 * configure: Rebuilt.
2869 * configure.in: Replace CHECK_SIZEOF and endian tests with
2870 cross-compiler friendly macros.
2871 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2874 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2876 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2877 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2879 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2881 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2882 (TARGET_SRC_POWERPC_AIX): New.
2883 (POWERPC_AIX): New stanza.
2884 * Makefile.in: Regenerate.
2885 * configure.in: Add AIX case.
2886 * configure: Regenerate.
2887 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2888 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2889 size. Fix "long double" support.
2890 (ffi_call): Add FFI_AIX case.
2891 * src/powerpc/aix.S: New.
2893 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2895 Implement Darwin PowerPC ABI.
2896 * configure.in: Handle powerpc-*-darwin*.
2897 * Makefile.am: Set source files for POWERPC_DARWIN.
2898 * configure: Rebuilt.
2899 * Makefile.in: Rebuilt.
2900 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2902 * src/powerpc/darwin.S: New file.
2903 * src/powerpc/ffi_darwin.c: New file.
2905 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2907 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2909 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2911 * src/x86/sysv.S: Avoid gas-only .balign directive.
2912 Use C style comments.
2914 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2916 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2917 Fixes PR bootstrap/3563.
2919 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2921 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2923 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2925 * configure.in: Recognize sparc*-sun-* host.
2926 * configure: Regenerate.
2928 2001-06-06 Andrew Haley <aph@redhat.com>
2930 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
2932 2001-06-03 Andrew Haley <aph@redhat.com>
2934 * src/alpha/osf.S: Add unwind info.
2935 * src/powerpc/sysv.S: Add unwind info.
2936 * src/powerpc/ppc_closure.S: Likewise.
2938 2000-05-31 Jeff Sturm <jsturm@one-point.com>
2940 * configure.in: Fix AC_ARG_ENABLE usage.
2941 * configure: Rebuilt.
2943 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2945 * configure.in: Remove warning about beta code.
2946 * configure: Rebuilt.
2948 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
2950 * src/ia64/unix.S: Restore stack pointer when returning from
2952 * src/ia64/ffi.c: Fix typo in comment.
2954 2001-04-18 Jim Wilson <wilson@redhat.com>
2956 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
2957 to eliminate RAW DV.
2959 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
2961 * Makefile.am: Make a libtool convenience library.
2962 * Makefile.in: Rebuilt.
2964 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
2966 * configure.in: Use different syntax for subdirectory creation.
2967 * configure: Rebuilt.
2969 2001-03-27 Jon Beniston <jon@beniston.com>
2971 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
2972 * configure: Rebuilt.
2973 * Makefile.am: Added X86_WIN32 target support.
2974 * Makefile.in: Rebuilt.
2976 * include/ffi.h.in: Added X86_WIN32 target support.
2978 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
2979 * src/types.c: Added X86_WIN32 target support.
2981 * src/x86/win32.S: New file. Based on sysv.S, but with EH
2982 stuff removed and made to work with CygWin's gas.
2984 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
2986 * configure.in: Make target subdirectory in build dir.
2987 * Makefile.am: Override suffix based rules to specify correct output
2989 * Makefile.in: Rebuilt.
2990 * configure: Rebuilt.
2992 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
2994 * src/powerpc/ppc_closure.S: New file.
2995 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
2996 involving long long and register pairs.
2997 (ffi_prep_closure): New function.
2998 (flush_icache): Likewise.
2999 (ffi_closure_helper_SYSV): Likewise.
3000 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3001 (FFI_TRAMPOLINE_SIZE): Likewise.
3002 (FFI_NATIVE_RAW_API): Likewise.
3003 * Makefile.in: Rebuilt.
3004 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3005 (TARGET_SRC_POWERPC): Likewise.
3007 2001-03-19 Tom Tromey <tromey@redhat.com>
3009 * Makefile.in: Rebuilt.
3010 * Makefile.am (ffitest_LDFLAGS): New macro.
3012 2001-03-02 Nick Clifton <nickc@redhat.com>
3014 * include/ffi.h.in: Remove RCS ident string.
3015 * include/ffi_mips.h: Remove RCS ident string.
3016 * src/debug.c: Remove RCS ident string.
3017 * src/ffitest.c: Remove RCS ident string.
3018 * src/prep_cif.c: Remove RCS ident string.
3019 * src/types.c: Remove RCS ident string.
3020 * src/alpha/ffi.c: Remove RCS ident string.
3021 * src/alpha/osf.S: Remove RCS ident string.
3022 * src/arm/ffi.c: Remove RCS ident string.
3023 * src/arm/sysv.S: Remove RCS ident string.
3024 * src/mips/ffi.c: Remove RCS ident string.
3025 * src/mips/n32.S: Remove RCS ident string.
3026 * src/mips/o32.S: Remove RCS ident string.
3027 * src/sparc/ffi.c: Remove RCS ident string.
3028 * src/sparc/v8.S: Remove RCS ident string.
3029 * src/sparc/v9.S: Remove RCS ident string.
3030 * src/x86/ffi.c: Remove RCS ident string.
3031 * src/x86/sysv.S: Remove RCS ident string.
3033 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3035 * include/ffi.h.in: Change sourceware.cygnus.com references to
3038 2000-12-09 Richard Henderson <rth@redhat.com>
3040 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3041 (ffi_closure_osf_inner): Index rather than increment avalue
3042 and arg_types. Give ffi_closure_osf the raw return value type.
3043 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3046 2000-12-07 Richard Henderson <rth@redhat.com>
3048 * src/raw_api.c (ffi_translate_args): Fix typo.
3049 (ffi_prep_closure): Likewise.
3051 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3052 FFI_TRAMPOLINE_SIZE.
3053 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3054 cif->bytes for new ffi_call_osf implementation.
3055 (ffi_prep_args): Absorb into ...
3056 (ffi_call): ... here. Do all stack allocation here and
3057 avoid a callback function.
3058 (ffi_prep_closure, ffi_closure_osf_inner): New.
3059 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3060 (ffi_closure_osf): New.
3062 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3064 * config.guess, config.sub, install-sh: Removed.
3065 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3066 * Makefile.in: Rebuilt.
3068 * acinclude.m4: Include libtool macros from the top level.
3069 * aclocal.m4, configure: Rebuilt.
3071 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3073 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3074 * configure: Rebuilt.
3076 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3078 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3079 memory correctly. Use conditional instructions, not branches where
3082 2000-05-04 Tom Tromey <tromey@cygnus.com>
3084 * configure: Rebuilt.
3085 * configure.in: Match `arm*-*-linux-*'.
3086 From Chris Dornan <cdornan@arm.com>.
3088 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3090 * Makefile.am (SUBDIRS): Define.
3091 (AM_MAKEFLAGS): Likewise.
3092 (Multilib support.): Add section.
3093 * Makefile.in: Rebuilt.
3094 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3095 New variables. Set for gcc using -print-multi-lib. Export them
3097 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3098 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3100 (extra_compiler_flags_value, extra_compiler_flags): Check these
3101 for extra compiler options which need to be passed down in
3104 2000-04-16 Anthony Green <green@redhat.com>
3106 * configure: Rebuilt.
3107 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3109 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3111 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3112 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3113 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3114 Replace all void * sizeofs with sizeof(int).
3115 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3116 different than DOUBLE.
3117 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3118 (ffi_prep_args_v9): New function.
3119 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3120 (ffi_V9_return_struct): New function.
3121 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3122 32bit code (not yet cross-arch calls).
3123 * src/sparc/v8.S: Add struct return delay nop.
3125 * src/sparc/v9.S: New file.
3126 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3127 is used on sparc64 only for structures larger than 32 bytes.
3128 Pass by reference for structures is done for structure arguments
3129 larger than 16 bytes.
3130 * src/ffitest.c (main): Use 64bit rint on sparc64.
3131 Run long long tests on sparc.
3132 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3134 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3136 * configure.in (sparc-*-linux*): New supported target.
3137 (sparc64-*-linux*): Likewise.
3138 * configure: Rebuilt.
3139 * Makefile.am: Add v9.S to SPARC files.
3140 * Makefile.in: Likewise.
3141 (LINK): Surround $(CCLD) into double quotes, so that multilib
3142 compiles work correctly.
3144 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3146 * configure: Rebuilt.
3147 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3148 proposed by Bryce McKinlay.
3150 2000-03-20 Tom Tromey <tromey@cygnus.com>
3152 * Makefile.in: Hand edit for java_raw_api.lo.
3154 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3156 * config.guess, config.sub: Update from the gcc tree.
3157 Fix for PR libgcj/168.
3159 2000-03-03 Tom Tromey <tromey@cygnus.com>
3161 * Makefile.in: Fixed ia64 by hand.
3163 * configure: Rebuilt.
3164 * configure.in (--enable-multilib): New option.
3165 (libffi_basedir): New subst.
3166 (AC_OUTPUT): Added multilib code.
3168 2000-03-02 Tom Tromey <tromey@cygnus.com>
3170 * Makefile.in: Rebuilt.
3171 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3174 2000-02-25 Hans Boehm <boehm@acm.org>
3176 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3178 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3180 (ffi_prep_raw_closure): Use ffi_translate_args, not
3181 ffi_closure_translate.
3182 * src/java_raw_api.c: New file.
3183 * src/ffitest.c (closure_test_fn): New function.
3184 (main): Define `rint' as long long on IA64. Added new test when
3185 FFI_CLOSURES is defined.
3186 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3187 (ffi_abi): Recognize IA64.
3188 (ffi_raw): Added `flt' field.
3189 Added "Java raw API" code.
3190 * configure.in: Recognize ia64.
3191 * Makefile.am (TARGET_SRC_IA64): New macro.
3192 (libffi_la_common_SOURCES): Added java_raw_api.c.
3193 (libffi_la_SOURCES): Define in IA64 case.
3195 2000-01-04 Tom Tromey <tromey@cygnus.com>
3197 * Makefile.in: Rebuilt with newer automake.
3199 1999-12-31 Tom Tromey <tromey@cygnus.com>
3201 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3203 1999-09-01 Tom Tromey <tromey@cygnus.com>
3205 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3207 * fficonfig.h.in: Rebuilt.
3208 * configure: Rebuilt.
3209 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3210 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3211 * acconfig.h: Don't #undef PACKAGE or VERSION.
3213 1999-08-09 Anthony Green <green@cygnus.com>
3215 * include/ffi.h.in: Try to work around messy header problem
3216 with PACKAGE and VERSION.
3218 * configure: Rebuilt.
3219 * configure.in: Change version to 2.00-beta.
3221 * fficonfig.h.in: Rebuilt.
3222 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3224 * src/x86/ffi.c (ffi_raw_call): Rename.
3226 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3228 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3229 (ffi_prep_incoming_args_SYSV): Ditto.
3230 (ffi_prep_closure): Ditto.
3231 (ffi_closure_raw_SYSV): Ditto.
3232 (ffi_prep_raw_closure): More ditto.
3233 (ffi_call_raw): Final ditto.
3235 * include/ffi.h.in: Add definitions for closure and raw API.
3237 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3240 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3242 * src/raw_api.c: New file.
3244 * include/ffi.h.in (ffi_raw): New type.
3245 (UINT_ARG, SINT_ARG): New defines.
3246 (ffi_closure, ffi_raw_closure): New types.
3247 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3249 * configure.in: Add check for endianness and sizeof void*.
3251 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3252 instead of directly.
3254 * configure: Rebuilt.
3256 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3258 * configure.in: Add x86 and powerpc BeOS configurations.
3259 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3261 1999-05-09 Anthony Green <green@cygnus.com>
3263 * configure.in: Add warning about this being beta code.
3264 Remove src/Makefile.am from the picture.
3265 * configure: Rebuilt.
3267 * Makefile.am: Move logic from src/Makefile.am. Add changes
3268 to support libffi as a target library.
3269 * Makefile.in: Rebuilt.
3271 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3272 Upgraded to new autoconf, automake, libtool.
3276 * LICENSE: Update copyright date.
3278 * src/Makefile.am, src/Makefile.in: Removed.
3280 1998-11-29 Anthony Green <green@cygnus.com>
3282 * include/ChangeLog: Removed.
3283 * src/ChangeLog: Removed.
3284 * src/mips/ChangeLog: Removed.
3285 * src/sparc/ChangeLog: Remboved.
3286 * src/x86/ChangeLog: Removed.
3288 * ChangeLog.v1: Created.