1 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
3 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
5 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
8 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
9 signed/unsigned int8/16 for X86_DARWIN.
10 Updated copyright info.
11 Handle one and two byte structs with special cif->flags.
12 * src/x86/ffitarget.h: Add special types for one and two byte structs.
13 Updated copyright info.
14 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
16 Remove code to pop args from the stack after call.
17 Special-case signed/unsigned for int8/16, one and two byte structs.
18 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
19 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
21 Updated copyright info.
23 2007-12-08 David Daney <ddaney@avtrex.com>
25 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
26 SUBU, add with ADDU and use smaller code sequences.
28 2007-12-07 David Daney <ddaney@avtrex.com>
30 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
33 2007-12-06 David Daney <ddaney@avtrex.com>
35 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
37 (ffi_java_raw): New typedef.
38 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
39 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
41 (ffi_java_raw_closure) : Same.
42 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
44 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
46 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
47 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
48 sizeof(ffi_java_raw) for alignment calculations.
49 (ffi_java_ptrarray_to_raw): Same.
50 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
51 if FFI_SIZEOF_JAVA_RAW == 4.
52 (ffi_java_raw_to_rvalue): Same.
53 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
54 (ffi_java_translate_args): Same.
55 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
57 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
59 2007-12-06 David Daney <ddaney@avtrex.com>
61 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
64 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
67 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
68 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
69 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
70 set the NUM_FPR_ARG_REGISTERS according to.
71 Add support for potential soft-float support under hard-float
73 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
74 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
75 to the FFI_LINUX_SOFT_FLOAT ABI.
76 (ffi_prep_cif_machdep): Likewise.
77 (ffi_closure_helper_SYSV): Likewise.
78 * src/powerpc/ppc_closure.S: Make sure not to store float/double
79 on archs where __NO_FPRS__ is true.
80 Add FFI_TYPE_UINT128 support.
81 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
82 Adjust copyright notice.
84 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
86 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
87 * include/ffi_common.h: ... here.
90 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
92 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
93 * src/powerpc/linux64.S: Likewise.
94 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
95 * testsuite/libffi.call/return_ldl.c: New test.
97 2007-09-04 <aph@redhat.com>
99 * src/arm/sysv.S (UNWIND): New.
100 (Whole file): Conditionally compile unwinder directives.
101 * src/arm/sysv.S: Add unwinder directives.
103 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
104 Only treat r0 as a struct address if we're actually returning a
106 Only copy the bytes that are actually within a struct.
107 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
108 is returned in r0, not passed by address.
109 (ffi_call): Allocate a word-sized temporary for the case where
110 a composite is returned in r0.
111 (ffi_prep_incoming_args_SYSV): Align as necessary.
113 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
115 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
116 directly using the sys_cacheflush syscall.
118 2007-07-27 Andrew Haley <aph@redhat.com>
120 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
122 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
124 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
125 * configure.ac: Likewise.
126 * Makefile.in: Regenerate.
127 * include/Makefile.in: Likewise.
128 * testsuite/Makefile.in: Likewise.
129 * configure: Likewise.
131 2007-08-24 David Daney <ddaney@avtrex.com>
133 * testsuite/libffi.call/return_sl.c: New test.
135 2007-08-10 David Daney <ddaney@avtrex.com>
137 * testsuite/libffi.call/cls_multi_ushort.c,
138 testsuite/libffi.call/cls_align_uint16.c,
139 testsuite/libffi.call/nested_struct1.c,
140 testsuite/libffi.call/nested_struct3.c,
141 testsuite/libffi.call/cls_7_1_byte.c,
142 testsuite/libffi.call/nested_struct5.c,
143 testsuite/libffi.call/cls_double.c,
144 testsuite/libffi.call/nested_struct7.c,
145 testsuite/libffi.call/cls_sint.c,
146 testsuite/libffi.call/nested_struct9.c,
147 testsuite/libffi.call/cls_20byte1.c,
148 testsuite/libffi.call/cls_multi_sshortchar.c,
149 testsuite/libffi.call/cls_align_sint64.c,
150 testsuite/libffi.call/cls_3byte2.c,
151 testsuite/libffi.call/cls_multi_schar.c,
152 testsuite/libffi.call/cls_multi_uchar.c,
153 testsuite/libffi.call/cls_19byte.c,
154 testsuite/libffi.call/cls_9byte1.c,
155 testsuite/libffi.call/cls_align_float.c,
156 testsuite/libffi.call/closure_fn1.c,
157 testsuite/libffi.call/problem1.c,
158 testsuite/libffi.call/closure_fn3.c,
159 testsuite/libffi.call/cls_sshort.c,
160 testsuite/libffi.call/closure_fn5.c,
161 testsuite/libffi.call/cls_align_double.c,
162 testsuite/libffi.call/nested_struct.c,
163 testsuite/libffi.call/cls_2byte.c,
164 testsuite/libffi.call/nested_struct10.c,
165 testsuite/libffi.call/cls_4byte.c,
166 testsuite/libffi.call/cls_6byte.c,
167 testsuite/libffi.call/cls_8byte.c,
168 testsuite/libffi.call/cls_multi_sshort.c,
169 testsuite/libffi.call/cls_align_sint16.c,
170 testsuite/libffi.call/cls_align_uint32.c,
171 testsuite/libffi.call/cls_20byte.c,
172 testsuite/libffi.call/cls_float.c,
173 testsuite/libffi.call/nested_struct2.c,
174 testsuite/libffi.call/cls_5_1_byte.c,
175 testsuite/libffi.call/nested_struct4.c,
176 testsuite/libffi.call/cls_24byte.c,
177 testsuite/libffi.call/nested_struct6.c,
178 testsuite/libffi.call/cls_64byte.c,
179 testsuite/libffi.call/nested_struct8.c,
180 testsuite/libffi.call/cls_uint.c,
181 testsuite/libffi.call/cls_multi_ushortchar.c,
182 testsuite/libffi.call/cls_schar.c,
183 testsuite/libffi.call/cls_uchar.c,
184 testsuite/libffi.call/cls_align_uint64.c,
185 testsuite/libffi.call/cls_ulonglong.c,
186 testsuite/libffi.call/cls_align_longdouble.c,
187 testsuite/libffi.call/cls_1_1byte.c,
188 testsuite/libffi.call/cls_12byte.c,
189 testsuite/libffi.call/cls_3_1byte.c,
190 testsuite/libffi.call/cls_3byte1.c,
191 testsuite/libffi.call/cls_4_1byte.c,
192 testsuite/libffi.call/cls_6_1_byte.c,
193 testsuite/libffi.call/cls_16byte.c,
194 testsuite/libffi.call/cls_18byte.c,
195 testsuite/libffi.call/closure_fn0.c,
196 testsuite/libffi.call/cls_9byte2.c,
197 testsuite/libffi.call/closure_fn2.c,
198 testsuite/libffi.call/closure_fn4.c,
199 testsuite/libffi.call/cls_ushort.c,
200 testsuite/libffi.call/closure_fn6.c,
201 testsuite/libffi.call/cls_5byte.c,
202 testsuite/libffi.call/cls_align_pointer.c,
203 testsuite/libffi.call/cls_7byte.c,
204 testsuite/libffi.call/cls_align_sint32.c,
205 testsuite/libffi.special/unwindtest_ffi_call.cc,
206 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
208 2007-08-10 David Daney <ddaney@avtrex.com>
211 * configure.ac: Don't treat mips64 as a special case.
212 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
213 * configure: Regenerate
214 * Makefile.in: Ditto.
215 * fficonfig.h.in: Ditto.
216 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
217 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
218 (FFI_DEFAULT_ABI): Set for n64 case.
219 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
220 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
221 (ffi_closure_N32): New function.
222 (.eh_frame): New section
223 * src/mips/o32.S: Clean up comments.
224 (ffi_closure_O32): Pass ffi_closure parameter in $12.
225 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
226 _MIPS_SIM == _ABIN32 throughout.
227 (FFI_MIPS_STOP_HERE): New, use in place of
229 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
230 to support n32/n64 ABIs.
231 (calc_n32_struct_flags): Rewrite.
232 (calc_n32_return_struct_flags): Remove unused variable. Reverse
233 position of flag bits.
234 (ffi_prep_cif_machdep): Rewrite n32 portion.
235 (ffi_call): Enable for n64. Add special handling for small structure
237 (ffi_prep_closure_loc): Add n32 and n64 support.
238 (ffi_closure_mips_inner_O32): Add cast to silence warning.
239 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
241 2007-08-08 David Daney <ddaney@avtrex.com>
243 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
244 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
246 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
247 * testsuite/libffi.call/cls_sint.c (main): Ditto.
248 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
249 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
250 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
251 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
252 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
253 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
254 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
255 * testsuite/libffi.call/return_fl3.c (main): Ditto.
256 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
257 * testsuite/libffi.call/nested_struct.c (main): Ditto.
258 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
259 * testsuite/libffi.call/return_ll1.c (main): Ditto.
260 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
261 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
262 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
263 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
264 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
265 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
266 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
267 * testsuite/libffi.call/cls_uint.c (main): Ditto.
268 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
269 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
270 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
271 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
272 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
273 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
274 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
275 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
276 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
278 2007-08-07 Andrew Haley <aph@redhat.com>
280 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
283 2007-08-06 Andrew Haley <aph@redhat.com>
286 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
287 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
290 2007-08-02 David Daney <ddaney@avtrex.com>
292 * testsuite/libffi.call/return_ul.c (main): Define return type as
293 ffi_arg. Use proper printf conversion specifier.
295 2007-07-30 Andrew Haley <aph@redhat.com>
298 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
299 signed/unsigned int8/16.
300 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
302 Remove code to pop args from the stack after call.
303 Special-case signed/unsigned int8/16.
304 * testsuite/libffi.call/return_sc.c (main): Revert.
306 2007-07-26 Richard Guenther <rguenther@suse.de>
309 * testsuite/libffi.call/return_sc.c (main): Verify call
310 result as signed char, not ffi_arg.
312 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
314 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
315 * configure: Regenerate.
317 2007-07-11 David Daney <ddaney@avtrex.com>
319 * src/mips/ffi.c: Don't include sys/cachectl.h.
320 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
323 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
325 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
326 from (ffi_prep_closure): ... this.
327 (FFI_INIT_TRAMPOLINE): Adjust.
329 2005-12-31 Phil Blundell <pb@reciva.com>
331 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
332 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
333 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
334 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
335 (FFI_CLOSURES): Enable closure support.
337 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
339 * testsuite/libffi.call/cls_multi_ushort.c,
340 testsuite/libffi.call/cls_align_uint16.c,
341 testsuite/libffi.call/nested_struct1.c,
342 testsuite/libffi.call/nested_struct3.c,
343 testsuite/libffi.call/cls_7_1_byte.c,
344 testsuite/libffi.call/cls_double.c,
345 testsuite/libffi.call/nested_struct5.c,
346 testsuite/libffi.call/nested_struct7.c,
347 testsuite/libffi.call/cls_sint.c,
348 testsuite/libffi.call/nested_struct9.c,
349 testsuite/libffi.call/cls_20byte1.c,
350 testsuite/libffi.call/cls_multi_sshortchar.c,
351 testsuite/libffi.call/cls_align_sint64.c,
352 testsuite/libffi.call/cls_3byte2.c,
353 testsuite/libffi.call/cls_multi_schar.c,
354 testsuite/libffi.call/cls_multi_uchar.c,
355 testsuite/libffi.call/cls_19byte.c,
356 testsuite/libffi.call/cls_9byte1.c,
357 testsuite/libffi.call/cls_align_float.c,
358 testsuite/libffi.call/closure_fn1.c,
359 testsuite/libffi.call/problem1.c,
360 testsuite/libffi.call/closure_fn3.c,
361 testsuite/libffi.call/cls_sshort.c,
362 testsuite/libffi.call/closure_fn5.c,
363 testsuite/libffi.call/cls_align_double.c,
364 testsuite/libffi.call/cls_2byte.c,
365 testsuite/libffi.call/nested_struct.c,
366 testsuite/libffi.call/nested_struct10.c,
367 testsuite/libffi.call/cls_4byte.c,
368 testsuite/libffi.call/cls_6byte.c,
369 testsuite/libffi.call/cls_8byte.c,
370 testsuite/libffi.call/cls_multi_sshort.c,
371 testsuite/libffi.call/cls_align_uint32.c,
372 testsuite/libffi.call/cls_align_sint16.c,
373 testsuite/libffi.call/cls_float.c,
374 testsuite/libffi.call/cls_20byte.c,
375 testsuite/libffi.call/cls_5_1_byte.c,
376 testsuite/libffi.call/nested_struct2.c,
377 testsuite/libffi.call/cls_24byte.c,
378 testsuite/libffi.call/nested_struct4.c,
379 testsuite/libffi.call/nested_struct6.c,
380 testsuite/libffi.call/cls_64byte.c,
381 testsuite/libffi.call/nested_struct8.c,
382 testsuite/libffi.call/cls_uint.c,
383 testsuite/libffi.call/cls_multi_ushortchar.c,
384 testsuite/libffi.call/cls_schar.c,
385 testsuite/libffi.call/cls_uchar.c,
386 testsuite/libffi.call/cls_align_uint64.c,
387 testsuite/libffi.call/cls_ulonglong.c,
388 testsuite/libffi.call/cls_align_longdouble.c,
389 testsuite/libffi.call/cls_1_1byte.c,
390 testsuite/libffi.call/cls_12byte.c,
391 testsuite/libffi.call/cls_3_1byte.c,
392 testsuite/libffi.call/cls_3byte1.c,
393 testsuite/libffi.call/cls_4_1byte.c,
394 testsuite/libffi.call/cls_6_1_byte.c,
395 testsuite/libffi.call/cls_16byte.c,
396 testsuite/libffi.call/cls_18byte.c,
397 testsuite/libffi.call/closure_fn0.c,
398 testsuite/libffi.call/cls_9byte2.c,
399 testsuite/libffi.call/closure_fn2.c,
400 testsuite/libffi.call/closure_fn4.c,
401 testsuite/libffi.call/cls_ushort.c,
402 testsuite/libffi.call/closure_fn6.c,
403 testsuite/libffi.call/cls_5byte.c,
404 testsuite/libffi.call/cls_align_pointer.c,
405 testsuite/libffi.call/cls_7byte.c,
406 testsuite/libffi.call/cls_align_sint32.c,
407 testsuite/libffi.special/unwindtest_ffi_call.cc,
408 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
410 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
412 * aclocal.m4: Regenerated.
414 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
416 * configure: Regenerate.
418 2007-05-23 Steve Ellcey <sje@cup.hp.com>
420 * Makefile.in: Regenerate.
421 * configure: Regenerate.
422 * aclocal.m4: Regenerate.
423 * include/Makefile.in: Regenerate.
424 * testsuite/Makefile.in: Regenerate.
426 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
428 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
429 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
430 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
431 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
432 (FFI_CLOSURES): Enable closure support.
434 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
436 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
437 * configure: Regenerate.
438 * fficonfig.h.in: Regenerate.
439 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
440 CFI_OFFSET,CFI_DEF_CFA): New macros.
441 (ffi_call_SYSV): Add callframe annotation.
443 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
445 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
446 numerous test suite failures.
447 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
449 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
451 * Makefile.am (EXTRA_DIST): Bring up to date.
452 * Makefile.in: Regenerate.
453 * src/frv/eabi.S: Remove RCS keyword.
455 2007-04-06 Richard Henderson <rth@redhat.com>
457 * configure.ac: Tidy target case.
458 (HAVE_LONG_DOUBLE): Allow the target to override.
459 * configure: Regenerate.
460 * include/ffi.h.in: Don't define ffi_type_foo if
461 LIBFFI_HIDE_BASIC_TYPES is defined.
462 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
464 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
465 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
466 (ffi_type_longdouble): Special case for Alpha. Don't define
467 if long double == double.
469 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
470 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
471 (ffi_call, ffi_closure_osf_inner): Likewise.
472 (ffi_closure_osf_inner): Likewise. Mark hidden.
473 (ffi_call_osf, ffi_closure_osf): Mark hidden.
474 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
475 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
476 (load_table): Handle 128-bit long double.
478 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
480 2007-04-06 Tom Tromey <tromey@redhat.com>
483 * README: Fixed bug in example.
485 2007-04-03 Jakub Jelinek <jakub@redhat.com>
487 * src/closures.c: Include sys/statfs.h.
488 (_GNU_SOURCE): Define on Linux.
489 (FFI_MMAP_EXEC_SELINUX): Define.
490 (selinux_enabled): New variable.
491 (selinux_enabled_check): New function.
492 (is_selinux_enabled): Define.
495 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
497 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
498 Use 'volatile float sum' to create sum of floats to avoid false
499 negative due to excess precision on ix86 targets.
502 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
504 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
506 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
508 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
510 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
511 (ffi_prep_closure_loc): New.
512 (ffi_prep_raw_closure_loc): New.
513 (ffi_prep_java_raw_closure_loc): New.
514 * src/closures.c: New file.
515 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
516 Replace sflags with exec_offset.
517 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
518 sub_segment_exec_offset): New macros.
519 (get_segment_flags, set_segment_flags, check_segment_merge): New
521 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
522 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
523 destroy_mspace): Use new macros.
524 (sys_alloc): Silence warning.
525 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
526 * Makefile.in: Rebuilt.
527 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
528 terms of ffi_prep_closure_loc.
529 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
531 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
533 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
535 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
537 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
538 (ffi_prep_closure): ... this.
539 * src/pa/ffi.c: Likewise.
540 * src/cris/ffi.c: Likewise. Adjust.
541 * src/frv/ffi.c: Likewise.
542 * src/ia64/ffi.c: Likewise.
543 * src/mips/ffi.c: Likewise.
544 * src/powerpc/ffi_darwin.c: Likewise.
545 * src/s390/ffi.c: Likewise.
546 * src/sh/ffi.c: Likewise.
547 * src/sh64/ffi.c: Likewise.
548 * src/sparc/ffi.c: Likewise.
549 * src/x86/ffi64.c: Likewise.
550 * src/x86/ffi.c: Likewise.
551 (FFI_INIT_TRAMPOLINE): Adjust.
552 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
553 (ffi_prep_raw_closure): ... this.
554 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
555 (ffi_prep_closure): ... this.
556 (flush_icache): Adjust.
558 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
560 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
563 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
565 * Makefile.am: Add dummy install-pdf target.
566 * Makefile.in: Regenerate
568 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
570 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
571 ffi_closure_helper_SYSV): Add long double handling.
573 2007-02-02 Jakub Jelinek <jakub@redhat.com>
575 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
576 immediately after bctrl instruction.
578 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
580 * Makefile.am (all-recursive, install-recursive,
581 mostlyclean-recursive, clean-recursive, distclean-recursive,
582 maintainer-clean-recursive): Add missing targets.
583 * Makefile.in: Rebuilt.
585 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
587 * configure.ac: Add TARGET for x86_64-*-darwin*.
588 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
590 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
591 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
592 * configure: Regenerate.
593 * Makefile.in: Regenerate.
594 * include/Makefile.in: Regenerate.
595 * testsuite/Makefile.in: Regenerate.
596 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
599 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
601 * aclocal.m4: Regenerate with aclocal -I .. as written in the
604 2006-10-31 Geoffrey Keating <geoffk@apple.com>
606 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
607 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
609 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
610 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
612 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
613 Sandro Tolaini <tolaini@libero.it>
615 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
617 * configure: Regenerated.
618 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
619 (EXTRA_DIST): Add src/x86/darwin.S.
620 * Makefile.in: Regenerated.
621 * include/Makefile.in: Regenerated.
622 * testsuite/Makefile.in: Regenerated.
624 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
625 X86_WIN32, and additionally align stack to 16 bytes.
626 * src/x86/darwin.S: New, based on sysv.S.
627 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
629 2006-09-12 David Daney <ddaney@avtrex.com>
632 * include/Makefile.am: Install both ffi.h and ffitarget.h in
633 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
634 * aclocal.m4: Regenerated for automake 1.9.6.
635 * Makefile.in: Regenerated.
636 * include/Makefile.in: Regenerated.
637 * testsuite/Makefile.in: Regenerated.
639 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
641 * include/ffi_common.h (struct): Revert accidental commit.
643 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
645 * include/ffi_common.h: Remove lint directives.
646 * include/ffi.h.in: Likewise.
648 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
650 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
651 for 32-bit architectures.
652 * testsuite/libffi.call/return_ul.c: New test case.
654 2006-07-19 David Daney <ddaney@avtrex.com>
656 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
657 xfail remains for mips64.
659 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
661 * Makefile.am: Add install-html target. Add install-html to .PHONY
662 * Makefile.in: Regenerate.
663 * aclocal.m4: Regenerate.
664 * include/Makefile.in: Regenerate.
665 * testsuite/Makefile.in: Regenerate.
667 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
669 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
672 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
674 * README: Remove notice about 'Crazy Comments'.
675 * src/debug.c: Remove lint directives. Cleanup white spaces.
676 * src/java_raw_api.c: Likewise.
677 * src/prep_cif.c: Likewise.
678 * src/raw_api.c: Likewise.
679 * src/ffitest.c: Delete. No longer needed, all test cases migrated
681 * src/arm/ffi.c: Remove lint directives.
682 * src/m32r/ffi.c: Likewise.
683 * src/pa/ffi.c: Likewise.
684 * src/powerpc/ffi.c: Likewise.
685 * src/powerpc/ffi_darwin.c: Likewise.
686 * src/sh/ffi.c: Likewise.
687 * src/sh64/ffi.c: Likewise.
688 * src/x86/ffi.c: Likewise.
689 * testsuite/libffi.call/float2.c: Likewise.
690 * testsuite/libffi.call/promotion.c: Likewise.
691 * testsuite/libffi.call/struct1.c: Likewise.
693 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
695 * src/pa/hpux32.S: Correct unwind offset calculation for
697 * src/pa/linux.S: Likewise.
699 2006-04-12 James E Wilson <wilson@specifix.com>
702 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
703 (hfa_type_load): Call stf_spill.
704 (hfa_type_store): Call ldf_fill.
705 (ffi_call): Adjust calls to above routines. Add local temps for
708 2006-04-10 Matthias Klose <doko@debian.org>
710 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
711 directory names containing underscores.
713 2006-04-07 James E Wilson <wilson@specifix.com>
715 * testsuite/libffi.call/float4.c: New testcase.
717 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
718 Andreas Tobler <a.tobler@schweiz.ch>
720 * Makefile.am: Add PA_HPUX port.
721 * Makefile.in: Regenerate.
722 * include/Makefile.in: Likewise.
723 * testsuite/Makefile.in: Likewise.
724 * configure.ac: Add PA_HPUX rules.
725 * configure: Regenerate.
726 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
727 Add PA_HPUX and PA64_HPUX.
728 Rename FFI_LINUX ABI to FFI_PA32 ABI.
729 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
730 (FFI_TYPE_SMALL_STRUCT2): Define.
731 (FFI_TYPE_SMALL_STRUCT4): Likewise.
732 (FFI_TYPE_SMALL_STRUCT8): Likewise.
733 (FFI_TYPE_SMALL_STRUCT3): Redefine.
734 (FFI_TYPE_SMALL_STRUCT5): Likewise.
735 (FFI_TYPE_SMALL_STRUCT6): Likewise.
736 (FFI_TYPE_SMALL_STRUCT7): Likewise.
737 * src/pa/ffi.c (ROUND_DOWN): Delete.
738 (fldw, fstw, fldd, fstd): Use '__asm__'.
739 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
740 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
741 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
742 Simplify incrementing of stack slot variable. Change type of local
744 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
746 (ffi_prep_cif_machdep): Likewise.
747 (ffi_call): Likewise.
748 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
749 return type to ffi_status. Simplify incrementing of stack slot
750 variable. Only copy floating point argument registers when PA_LINUX
751 is true. Reformat debug statement.
752 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
753 FFI_TYPE_SMALL_STRUCT8.
754 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
756 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
757 Add nops to cache flush. Add trampoline for PA_HPUX.
758 * src/pa/hpux32.S: New file.
759 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
760 ffi_prep_args_LINUX to ffi_prep_args_pa32.
761 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
762 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
763 argument type checks so that common argument types appear first.
764 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
765 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
767 2006-03-24 Alan Modra <amodra@bigpond.net.au>
769 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
770 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
771 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
773 (ffi_prep_args64): Assert using IBM extended double.
774 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
775 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
776 (ffi_call): Handle FFI_LINUX.
777 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
778 gpr3 return pointer as for struct return. Handle FFI_LINUX
779 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
781 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
782 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
783 Don't use r6 as pointer to results, instead use sp offset. Don't
784 make a special call to load lr with case table address, instead
785 use offset from previous call.
786 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
787 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
790 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
792 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
793 passed with FP registers correctly.
794 (ffi_closure_helper_SYSV): Likewise.
795 * src/sh64/sysv.S: Likewise.
797 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
799 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
800 args and userdata unused.
801 (closure_test_fn1): Mark cif and userdata unused.
802 (main): Remove unused res.
804 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
806 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
807 -O2, -O3, -Os and the warning flags -W -Wall.
808 * testsuite/libffi.special/special.exp: Likewise.
809 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
810 unused parameter unused for gcc or else do nothing.
811 * testsuite/libffi.special/ffitestcxx.h: Likewise.
812 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
814 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
815 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
816 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
817 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
818 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
819 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
820 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
821 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
822 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
823 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
824 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
825 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
826 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
827 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
828 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
829 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
830 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
831 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
832 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
833 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
834 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
835 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
836 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
837 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
839 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
841 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
843 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
844 void* to avoid compiler warning.
846 (cls_struct_align_gn): Mark cif and userdata unused.
847 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
849 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
851 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
853 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
855 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
857 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
858 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
859 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
861 (main): Cast res_call to silence gcc.
862 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
864 (main): Cast res_call to silence gcc.
865 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
867 (main): Cast res_call to silence gcc.
868 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
870 (main): Cast res_call to silence gcc.
871 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
873 (main): Cast res_call to silence gcc.
874 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
876 (main): Cast res_call to silence gcc.
877 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
879 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
880 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
882 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
883 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
885 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
886 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
888 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
889 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
891 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
892 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
894 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
896 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
897 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
898 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
899 Cleanup white spaces.
900 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
901 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
902 cif and userdata unused.
903 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
905 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
906 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
907 formatters to silence gcc.
908 (B_gn): Mark cif and userdata unused.
909 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
911 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
912 (B_gn): Mark cif and userdata unused.
913 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
915 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
916 (B_gn): Mark cif and userdata unused.
917 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
919 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
920 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
921 * testsuite/libffi.call/problem1.c (stub): Likewise.
922 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
924 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
925 in the last commit for this test case in the test case itself.
926 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
928 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
929 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
930 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
931 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
932 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
933 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
935 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
937 * src/sh/sysv.S: Fix register numbers in the FDE for
940 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
942 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
943 declaration to avoid a false negative on ix86. See PR323.
945 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
947 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
948 and cast integer to void * if needed. Update the pointer to
949 the FP register saved area correctly.
951 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
953 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
955 * testsuite/libffi.call/nested_struct4.c: Likewise.
957 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
959 * testsuite/libffi.call/return_dbl.c: New test case.
960 * testsuite/libffi.call/return_dbl1.c: Likewise.
961 * testsuite/libffi.call/return_dbl2.c: Likewise.
962 * testsuite/libffi.call/return_fl.c: Likewise.
963 * testsuite/libffi.call/return_fl1.c: Likewise.
964 * testsuite/libffi.call/return_fl2.c: Likewise.
965 * testsuite/libffi.call/return_fl3.c: Likewise.
966 * testsuite/libffi.call/closure_fn6.c: Likewise.
968 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
970 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
971 here to be used by other test cases too.
973 * testsuite/libffi.call/nested_struct10.c: New test case.
974 * testsuite/libffi.call/nested_struct9.c: Likewise.
975 * testsuite/libffi.call/nested_struct8.c: Likewise.
976 * testsuite/libffi.call/nested_struct7.c: Likewise.
977 * testsuite/libffi.call/nested_struct6.c: Likewise.
978 * testsuite/libffi.call/nested_struct5.c: Likewise.
979 * testsuite/libffi.call/nested_struct4.c: Likewise.
981 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
983 * configure.ac: Enable libffi for sparc64-*-freebsd*.
984 * configure: Rebuilt.
986 2006-01-18 Jakub Jelinek <jakub@redhat.com>
988 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
989 instead do the shifting inline.
990 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
991 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
992 and 8 byte structs, for the remaining struct sizes don't call
993 __lshrdi3, instead do the shifting inline.
995 2005-12-07 Thiemo Seufer <ths@networkno.de>
997 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
999 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1000 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1001 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1002 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1003 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1005 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1007 (ffi_prep_closure): Improve trampoline instruction scheduling.
1008 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1010 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1012 * src/powerpc/ffi.c: Formatting.
1013 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1014 (ffi_prep_args64): Likewise.
1016 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1018 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1019 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1021 2005-09-26 Tom Tromey <tromey@redhat.com>
1023 * testsuite/libffi.call/float1.c (value_type): New typedef.
1024 (CANARY): New define.
1025 (main): Check for result buffer overflow.
1026 * src/powerpc/linux64.S: Handle linux64 long double returns.
1027 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1028 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1030 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1033 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1035 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1037 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1039 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1040 (AH_BOTTOM): Add FFI_HIDDEN definition.
1041 * configure: Rebuilt.
1042 * fficonfig.h.in: Rebuilt.
1043 * src/powerpc/ffi.c (hidden): Remove.
1044 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1045 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1046 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1047 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1048 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1049 add FFI_HIDDEN to its prototype.
1050 (ffi_closure_SYSV_inner): New.
1051 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1052 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1054 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1057 * configure: Rebuilt.
1058 * configure.ac: Handle i*86-*-gnu*.
1060 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1062 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1063 DW_CFA_offset_extended_sf rather than
1064 DW_CFA_GNU_negative_offset_extended.
1065 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1067 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1069 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1071 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1072 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1073 partially on register.
1074 (ffi_closure_helper_SYSV): Likewise.
1075 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1077 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1079 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1081 * src/sh64/ffi.c (simple_type): Remove.
1082 (return_type): Handle small structures correctly.
1083 (ffi_prep_args): Likewise.
1084 (ffi_call): Likewise.
1085 (ffi_closure_helper_SYSV): Likewise.
1086 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1087 Emit position independent code if PIC and remove wrong datalabel
1088 prefixes from EH data.
1090 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1092 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1093 * Makefile.in: Regenerate.
1094 * include/Makefile.in: Likewise.
1095 * testsuite/Makefile.in: Likewise.
1096 * configure.ac: Add POWERPC_FREEBSD rules.
1097 * configure: Regenerate.
1098 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1099 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1100 * src/powerpc/ffi.c: Add flags to handle small structure returns
1102 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1104 (ffi_closure_helper_SYSV): Likewise.
1105 * src/powerpc/ppc_closure.S: Add return types for small structures.
1106 * src/powerpc/sysv.S: Add bits to handle small structures for
1109 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1111 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1112 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1113 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1115 2005-07-05 Randolph Chung <tausq@debian.org>
1117 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1118 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1119 structures. Kill compilation warnings.
1120 (ffi_closure_inner_LINUX): Print return values as hex in debug
1121 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1122 Properly handle 5-7 byte structure returns.
1123 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1124 (FFI_TYPE_SMALL_STRUCT2): Remove.
1125 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1126 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1127 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1128 (checksmst1, checksmst2): Remove.
1129 (checksmst3): Optimize handling of 3-byte struct returns.
1130 (checksmst567): Properly handle 5-7 byte struct returns.
1132 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1135 * src/mips/n32.S: Enforce PIC code.
1136 * src/mips/o32.S: Likewise.
1138 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1140 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1141 * configure: Regenerate.
1143 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1145 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1146 to call ffi_closure_helper_SYSV. Append @local instead.
1147 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1149 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1151 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1152 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1153 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1154 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1155 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1157 2005-05-09 Mike Stump <mrs@apple.com>
1159 * configure: Regenerate.
1161 2005-05-08 Richard Henderson <rth@redhat.com>
1164 * src/alpha/osf.S: Update unwind into to match code.
1166 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1167 Richard Henderson <rth@redhat.com>
1169 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1171 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1172 (ffi_prep_closure): Set carry bit if sse-used flag set.
1173 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1174 Only load sse registers if ssecount non-zero.
1175 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1177 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1179 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1180 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1181 * configure: Regenerate.
1183 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1185 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1186 have Tcl8.3-compatible intermediate variable.
1188 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1189 Hans-Peter Nilsson <hp@axis.com>
1191 * Makefile.am: Add CRIS support.
1192 * configure.ac: Likewise.
1193 * Makefile.in, configure, testsuite/Makefile.in,
1194 include/Makefile.in: Regenerate.
1195 * src/cris: New directory.
1196 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1197 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1199 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1200 \r?\n in output tests.
1202 2005-04-12 Mike Stump <mrs@apple.com>
1204 * configure: Regenerate.
1206 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1208 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1210 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1212 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1214 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1215 to 64 bits in ILP32 mode.
1216 Load 64 bits even for short data.
1218 2005-03-23 Mike Stump <mrs@apple.com>
1220 * src/powerpc/darwin.S: Update for -m64 multilib.
1221 * src/powerpc/darwin_closure.S: Likewise.
1223 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1225 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1226 Do not set tool_include_dir.
1227 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1229 * include/Makefile.am: Set gcc_version and toollibffidir.
1230 * include/Makefile.in: Regenerate.
1232 2005-02-22 Andrew Haley <aph@redhat.com>
1234 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1235 odd-numbered register pairs for 64-bit integer types.
1237 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1240 * testsuite/libffi.call/return_ll1.c: New test case.
1242 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1244 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1245 * testsuite/libffi.call/float.c: Ditto.
1246 * testsuite/libffi.call/float2.c: Ditto.
1247 * testsuite/libffi.call/float3.c: Ditto.
1249 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1251 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1253 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1255 * testsuite/libffi.special/special.exp (cxx_options): Add
1258 2004-12-31 Richard Henderson <rth@redhat.com>
1260 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1261 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1262 offset parameters with a type parameter; deduce size and structure
1263 alignment. Update all users.
1265 2004-12-31 Richard Henderson <rth@redhat.com>
1267 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1268 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1269 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1270 into ffi_prep_closure.
1271 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1274 2004-12-27 Richard Henderson <rth@redhat.com>
1276 * src/x86/unix64.S: Fix typo in unwind info.
1278 2004-12-25 Richard Henderson <rth@redhat.com>
1280 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1281 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1282 (merge_classes): Check for it.
1284 (classify_argument): Pass byte_offset by value; perform all updates
1286 (examine_argument): Add classes argument; handle
1287 X86_64_COMPLEX_X87_CLASS.
1288 (ffi_prep_args): Merge into ...
1289 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1290 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1291 (ffi_fill_return_value): Remove.
1292 (ffi_prep_closure): Remove dead assert.
1293 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1294 Rewrite to use struct register_args instead of va_list. Create
1295 flags for handling structure returns.
1296 * src/x86/unix64.S: Remove dead strings.
1297 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1298 stack frame with ffi_call. Handle structure returns properly.
1299 (float2sse, floatfloat2sse, double2sse): Remove.
1300 (sse2float, sse2double, sse2floatfloat): Remove.
1301 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1302 to handle structure returns properly.
1304 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1306 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1308 * Makefile.in: Regenerated.
1310 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1312 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1313 * configure, aclocal.m4, Makefile.in: Regenerate.
1314 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1316 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1318 * configure: Regenerate for libtool change.
1320 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1322 * configure: Regenerate for libtool reversion.
1324 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1326 * configure: Regenerate for libtool change.
1328 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1330 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1332 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1334 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1335 of jal. Use an absolute encoding for the frame information.
1337 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1339 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1340 * acinclude.m4: Delete logic for sincludes.
1341 * aclocal.m4, Makefile.in, configure: Regenerate.
1342 * include/Makefile: Likewise.
1343 * testsuite/Makefile: Likewise.
1345 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1347 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1348 on a 8-byte boundary.
1349 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1351 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1353 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1354 long long values. Round stack allocation to a multiple of 8 bytes
1355 for ATPCS compatibility.
1356 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1357 names. Handle returning long long types. Add Thumb and interworking
1358 support. Improve soft-float code.
1360 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1362 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1363 (libffi_exit): New function.
1364 (libffi_init): Build the testglue wrapper if needed.
1366 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1369 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1371 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1373 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1375 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1377 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1378 * testsuite/libffi.call/float3.c: New test case.
1380 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1382 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1383 the function returning a structure pointed with R2.
1384 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1385 the structure return value if T bit set. Emit position
1386 independent code and EH data if PIC.
1388 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1390 * Makefile.am: Add m32r support.
1391 * configure.ac: Likewise.
1392 * Makefile.in: Regenerate.
1393 * confiugre: Regenerate.
1394 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1395 (uint64, sint64, double, longdouble)
1396 * src/m32r: New directory.
1397 * src/m32r/ffi.c: New file.
1398 * src/m32r/sysv.S: Likewise.
1399 * src/m32r/ffitarget.h: Likewise.
1401 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1403 * testsuite/libffi.call/negint.c: New test case.
1405 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1408 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1409 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1410 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1413 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1415 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1416 * testsuite/libffi.call/promotion.c: Likewise.
1417 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1419 * testsuite/libffi.call/return_sc.c: Likewise.
1420 * testsuite/libffi.call/return_uc.c: Likewise.
1422 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1424 * src/powerpc/darwin.S: Fix comments and identation.
1425 * src/powerpc/darwin_closure.S: Likewise.
1427 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1429 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1430 (ffi_prep_args): Handle longdouble arguments.
1431 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1433 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1434 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1436 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1437 * src/types.c: Defined longdouble size and alignment for darwin.
1439 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1441 * src/powerpc/aix.S: Remove whitespaces.
1442 * src/powerpc/aix_closure.S: Likewise.
1443 * src/powerpc/asm.h: Likewise.
1444 * src/powerpc/ffi.c: Likewise.
1445 * src/powerpc/ffitarget.h: Likewise.
1446 * src/powerpc/linux64.S: Likewise.
1447 * src/powerpc/linux64_closure.S: Likewise.
1448 * src/powerpc/ppc_closure.S: Likewise.
1449 * src/powerpc/sysv.S: Likewise.
1451 2004-08-30 Anthony Green <green@redhat.com>
1453 * Makefile.am: Add frv support.
1454 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1455 * configure.ac: Read configure.host.
1456 * configure.in: Read configure.host.
1457 * configure.host: New file. frv-elf needs libgloss.
1458 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1459 alignment. This is needed to frv and shouldn't harm the others.
1460 * include/ffi_common.h (ALIGN_DOWN): New macro.
1461 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1463 2004-08-24 David Daney <daney@avtrex.com>
1465 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1466 * testsuite/libffi.call/closure_fn1.c: Likewise.
1467 * testsuite/libffi.call/closure_fn2.c Likewise.
1468 * testsuite/libffi.call/closure_fn3.c: Likewise.
1469 * testsuite/libffi.call/closure_fn4.c: Likewise.
1470 * testsuite/libffi.call/closure_fn5.c: Likewise.
1471 * testsuite/libffi.call/cls_18byte.c: Likewise.
1472 * testsuite/libffi.call/cls_19byte.c: Likewise.
1473 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1474 * testsuite/libffi.call/cls_20byte.c: Likewise.
1475 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1476 * testsuite/libffi.call/cls_24byte.c: Likewise.
1477 * testsuite/libffi.call/cls_2byte.c: Likewise.
1478 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1479 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1480 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1481 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1482 * testsuite/libffi.call/cls_4byte.c: Likewise.
1483 * testsuite/libffi.call/cls_64byte.c: Likewise.
1484 * testsuite/libffi.call/cls_6byte.c: Likewise.
1485 * testsuite/libffi.call/cls_7byte.c: Likewise.
1486 * testsuite/libffi.call/cls_8byte.c: Likewise.
1487 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1488 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1489 * testsuite/libffi.call/cls_align_double.c: Likewise.
1490 * testsuite/libffi.call/cls_align_float.c: Likewise.
1491 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1492 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1493 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1494 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1495 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1496 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1497 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1498 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1499 * testsuite/libffi.call/cls_double.c: Likewise.
1500 * testsuite/libffi.call/cls_float.c: Likewise.
1501 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1502 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1503 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1504 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1505 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1506 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1507 * testsuite/libffi.call/cls_schar.c: Likewise.
1508 * testsuite/libffi.call/cls_sint.c: Likewise.
1509 * testsuite/libffi.call/cls_sshort.c: Likewise.
1510 * testsuite/libffi.call/cls_uchar.c: Likewise.
1511 * testsuite/libffi.call/cls_uint.c: Likewise.
1512 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1513 * testsuite/libffi.call/cls_ushort.c: Likewise.
1514 * testsuite/libffi.call/nested_struct.c: Likewise.
1515 * testsuite/libffi.call/nested_struct1.c: Likewise.
1516 * testsuite/libffi.call/nested_struct2.c: Likewise.
1517 * testsuite/libffi.call/nested_struct3.c: Likewise.
1518 * testsuite/libffi.call/problem1.c: Likewise.
1519 * testsuite/libffi.special/unwindtest.cc: Likewise.
1520 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1522 * testsuite/libffi.call/cls_16byte.c: Likewise.
1523 * testsuite/libffi.call/cls_5byte.c: Likewise.
1525 2004-08-23 David Daney <daney@avtrex.com>
1528 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1529 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1530 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1531 parameters and return types.
1532 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1533 (ffi_prep_closure): Ditto.
1534 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1535 alignment calculations.
1536 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1537 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1539 2004-08-14 Casey Marshall <csm@gnu.org>
1541 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1542 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1543 integer (O32 ABI only).
1544 (ffi_prep_closure): new function.
1545 (ffi_closure_mips_inner_O32): new function.
1546 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1547 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1548 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1549 64 bit integers correctly.
1550 (ffi_closure_O32): new function.
1551 Added DWARF-2 unwind info for both functions.
1553 2004-08-10 Andrew Haley <aph@redhat.com>
1555 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1557 2004-08-01 Robert Millan <robertmh@gnu.org>
1559 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1560 * configure: Regenerate.
1562 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1564 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1565 and mmap() explicitly instead of relying on preset autoconf cache
1567 * aclocal.m4: Regenerate.
1568 * configure: Regenerate.
1570 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1572 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1573 (ffi_check_float_struct): Remove unused prototype.
1575 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1577 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1578 character on Darwin, use '\n\t' instead.
1580 2004-06-26 Matthias Klose <doko@debian.org>
1582 * libtool-version: Fix typo in revision/age.
1584 2004-06-17 Matthias Klose <doko@debian.org>
1586 * libtool-version: New.
1587 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1588 * Makefile.in: Regenerate.
1590 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1592 * Makefile.am: Remove useless multilib rules.
1593 * Makefile.in: Regenerate.
1594 * aclocal.m4: Regenerate with automake 1.8.5.
1595 * configure.ac: Remove useless multilib configury.
1596 * configure: Regenerate.
1598 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1600 * .cvsignore: New file.
1602 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1604 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1606 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1609 2004-06-06 Sean McNeil <sean@mcneil.com>
1611 * configure.ac: Add x86_64-*-freebsd* support.
1612 * configure: Regenerate.
1614 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1617 * configure.ac: Test for existence of mmap and sys/mman.h before
1618 checking blacklist. Fix suggested by Jim Wilson.
1619 * configure: Regenerate.
1621 2004-04-26 Matt Austern <austern@apple.com>
1623 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1625 * src/powerpc/darwin_closure.S: Likewise.
1627 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1629 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1630 error. Reported by Thomas Heller <theller@python.net>.
1631 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1632 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1634 2004-03-20 Matthias Klose <doko@debian.org>
1636 * src/pa/linux.S: Fix typo.
1638 2004-03-19 Matthias Klose <doko@debian.org>
1640 * Makefile.am: Update.
1641 * Makefile.in: Regenerate.
1642 * src/pa/ffi.h.in: Remove.
1643 * src/pa/ffitarget.h: New file.
1645 2004-02-10 Randolph Chung <tausq@debian.org>
1647 * Makefile.am: Add PA support.
1648 * Makefile.in: Regenerate.
1649 * include/Makefile.in: Regenerate.
1650 * configure.ac: Add PA target.
1651 * configure: Regenerate.
1652 * src/pa/ffi.c: New file.
1653 * src/pa/ffi.h.in: Add PA support.
1654 * src/pa/linux.S: New file.
1655 * prep_cif.c: Add PA support.
1657 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1659 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1661 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1662 with ecif->cif->flags.
1663 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1665 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1666 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1667 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1668 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1670 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1672 * configure.in: Rename file to ...
1673 * configure.ac: ... this.
1674 * fficonfig.h.in: Regenerate.
1675 * Makefile.in: Regenerate.
1676 * include/Makefile.in: Regenerate.
1677 * testsuite/Makefile.in: Regenerate.
1679 2004-03-12 Matt Austern <austern@apple.com>
1681 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1682 changes in EH format resulting from addition of linkonce support.
1683 * src/powerpc/darwin_closure.S: Likewise.
1685 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1686 Paolo Bonzini <bonzini@gnu.org>
1688 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1689 Remove VPATH. Remove rules for object files. Remove multilib support.
1690 (AM_CCASFLAGS): Add.
1691 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1692 (AC_PREREQ): Bump version to 2.59.
1693 (AC_INIT): Fill with version info and bug address.
1694 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1695 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1696 De-precious CC so that the right flags are passed down to multilibs.
1697 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1698 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1699 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1700 * configure: Rebuilt.
1701 * aclocal.m4: Likewise.
1702 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1703 * fficonfig.h.in: Likewise.
1705 2004-03-11 Andreas Schwab <schwab@suse.de>
1707 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1708 arguments from fp registers only for the first 8 parameter slots.
1709 Don't convert a float parameter when passed in memory.
1711 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1713 * configure: Regenerate for config/accross.m4 correction.
1715 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1717 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1718 ecif->cif->bytes to bytes.
1719 (ffi_prep_cif_machdep): Add braces around nested if statement.
1721 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1723 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1725 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1726 (ffi_closure_helper_LINUX64): Fix typo.
1727 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1729 * testsuite/libffi.call/float.c: Likewise.
1730 * testsuite/libffi.call/float2.c: Likewise.
1732 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1734 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1735 long double function return and long double arg handling.
1736 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1737 Use "end_pfr" instead of "nf". Correct long double handling.
1739 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1741 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1742 space for long double return value. Adjust stack frame and offsets.
1743 Load f2 long double return.
1745 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1747 * src/types.c: Use 16 byte long double for POWERPC64.
1749 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1751 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1752 when the structure return address is passed in %o0.
1753 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1754 (ffi_v9_layout_struct): Align the field following a nested structure
1755 on a word boundary. Use memmove instead of memcpy.
1756 (ffi_call): Update call to ffi_V9_return_struct.
1757 (ffi_prep_closure): Define 'ctx' only for V8.
1758 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1759 and ffi_closure_sparc_inner_v9.
1760 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1761 Always skip the structure return address. For structures and long
1762 doubles, copy the argument directly.
1763 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1764 if required. Shift the maximum floating-point slot accordingly. For
1765 big structures, copy the argument directly; otherwise, left-justify the
1766 argument and call ffi_v9_layout_struct to lay out the structure on
1768 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1769 (ffi_closure_v8): Pass the structure return address. Update call to
1770 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1771 Skip the 'unimp' insn when returning long doubles and structures.
1772 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1773 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1774 FFI_TYPE_INT handling. Load structures both in integers and
1775 floating-point registers on return.
1776 * README: Update status of the SPARC port.
1778 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1780 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1782 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1784 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1786 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1787 value as of type ffi_arg, not unsigned int.
1789 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1791 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1794 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1796 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1799 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1801 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1804 2003-12-10 Richard Henderson <rth@redhat.com>
1806 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1807 size_t instead of int.
1809 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1811 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1812 * testsuite/libffi.call/many_win32.c (main): Replace variable
1813 int i with unsigned long ul.
1815 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1816 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1817 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1818 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1819 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1820 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1821 * testsuite/libffi.call/cls_align_float.c: Likewise.
1822 * testsuite/libffi.call/cls_align_double.c: Likewise.
1823 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1824 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1826 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1829 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1830 Align arguments to 32 bits.
1832 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1835 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1836 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1837 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1838 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1839 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1840 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1842 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1844 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1846 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1847 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1849 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1851 * Makefile.in: Rebuilt.
1852 * configure: Likewise.
1853 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1856 2003-11-21 Andreas Jaeger <aj@suse.de>
1857 Andreas Tobler <a.tobler@schweiz.ch>
1859 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1860 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1861 * Makefile.in: Rebuilt.
1862 * aclocal.m4: Likewise.
1863 * configure: Likewise.
1864 * fficonfig.h.in: Likewise.
1865 * testsuite/lib/libffi-dg.exp: Add include dir.
1866 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1867 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1868 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1869 for ffi_closure if available.
1870 * testsuite/libffi.call/closure_fn1.c: Likewise.
1871 * testsuite/libffi.call/closure_fn2.c: Likewise.
1872 * testsuite/libffi.call/closure_fn3.c: Likewise.
1873 * testsuite/libffi.call/closure_fn4.c: Likewise.
1874 * testsuite/libffi.call/closure_fn5.c: Likewise.
1875 * testsuite/libffi.call/cls_12byte.c: Likewise.
1876 * testsuite/libffi.call/cls_16byte.c: Likewise.
1877 * testsuite/libffi.call/cls_18byte.c: Likewise.
1878 * testsuite/libffi.call/cls_19byte.c: Likewise.
1879 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1880 * testsuite/libffi.call/cls_20byte.c: Likewise.
1881 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1882 * testsuite/libffi.call/cls_24byte.c: Likewise.
1883 * testsuite/libffi.call/cls_2byte.c: Likewise.
1884 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1885 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1886 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1887 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1888 * testsuite/libffi.call/cls_4byte.c: Likewise.
1889 * testsuite/libffi.call/cls_5byte.c: Likewise.
1890 * testsuite/libffi.call/cls_64byte.c: Likewise.
1891 * testsuite/libffi.call/cls_6byte.c: Likewise.
1892 * testsuite/libffi.call/cls_7byte.c: Likewise.
1893 * testsuite/libffi.call/cls_8byte.c: Likewise.
1894 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1895 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1896 * testsuite/libffi.call/cls_double.c: Likewise.
1897 * testsuite/libffi.call/cls_float.c: Likewise.
1898 * testsuite/libffi.call/cls_schar.c: Likewise.
1899 * testsuite/libffi.call/cls_sint.c: Likewise.
1900 * testsuite/libffi.call/cls_sshort.c: Likewise.
1901 * testsuite/libffi.call/cls_uchar.c: Likewise.
1902 * testsuite/libffi.call/cls_uint.c: Likewise.
1903 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1904 * testsuite/libffi.call/cls_ushort.c: Likewise.
1905 * testsuite/libffi.call/nested_struct.c: Likewise.
1906 * testsuite/libffi.call/nested_struct1.c: Likewise.
1907 * testsuite/libffi.call/nested_struct2.c: Likewise.
1908 * testsuite/libffi.call/nested_struct3.c: Likewise.
1909 * testsuite/libffi.call/problem1.c: Likewise.
1910 * testsuite/libffi.special/unwindtest.cc: Likewise.
1912 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
1914 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
1916 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
1918 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
1919 Add -lgcc_s to additional flags.
1921 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
1923 * configure.in, include/Makefile.am: PR libgcj/11147, install
1924 the ffitarget.h header file in a gcc versioned and target
1926 * configure: Regenerated.
1927 * Makefile.in, include/Makefile.in: Likewise.
1928 * testsuite/Makefile.in: Likewise.
1930 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
1932 * testsuite/libffi.call/closure_fn0.c: Print result and check
1933 with dg-output to make debugging easier.
1934 * testsuite/libffi.call/closure_fn1.c: Likewise.
1935 * testsuite/libffi.call/closure_fn2.c: Likewise.
1936 * testsuite/libffi.call/closure_fn3.c: Likewise.
1937 * testsuite/libffi.call/closure_fn4.c: Likewise.
1938 * testsuite/libffi.call/closure_fn5.c: Likewise.
1939 * testsuite/libffi.call/cls_12byte.c: Likewise.
1940 * testsuite/libffi.call/cls_16byte.c: Likewise.
1941 * testsuite/libffi.call/cls_18byte.c: Likewise.
1942 * testsuite/libffi.call/cls_19byte.c: Likewise.
1943 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1944 * testsuite/libffi.call/cls_20byte.c: Likewise.
1945 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1946 * testsuite/libffi.call/cls_24byte.c: Likewise.
1947 * testsuite/libffi.call/cls_2byte.c: Likewise.
1948 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1949 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1950 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1951 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1952 * testsuite/libffi.call/cls_4byte.c: Likewise.
1953 * testsuite/libffi.call/cls_5byte.c: Likewise.
1954 * testsuite/libffi.call/cls_64byte.c: Likewise.
1955 * testsuite/libffi.call/cls_6byte.c: Likewise.
1956 * testsuite/libffi.call/cls_7byte.c: Likewise.
1957 * testsuite/libffi.call/cls_8byte.c: Likewise.
1958 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1959 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1960 * testsuite/libffi.call/cls_double.c: Likewise.
1961 * testsuite/libffi.call/cls_float.c: Likewise.
1962 * testsuite/libffi.call/cls_schar.c: Likewise.
1963 * testsuite/libffi.call/cls_sint.c: Likewise.
1964 * testsuite/libffi.call/cls_sshort.c: Likewise.
1965 * testsuite/libffi.call/cls_uchar.c: Likewise.
1966 * testsuite/libffi.call/cls_uint.c: Likewise.
1967 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1968 * testsuite/libffi.call/cls_ushort.c: Likewise.
1969 * testsuite/libffi.call/problem1.c: Likewise.
1971 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
1974 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
1976 * testsuite/libffi.call/cls_9byte2.c: New test case.
1977 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1978 * testsuite/libffi.call/cls_64byte.c: Likewise.
1979 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1980 * testsuite/libffi.call/cls_19byte.c: Likewise.
1981 * testsuite/libffi.call/cls_18byte.c: Likewise.
1982 * testsuite/libffi.call/closure_fn4.c: Likewise.
1983 * testsuite/libffi.call/closure_fn5.c: Likewise.
1984 * testsuite/libffi.call/cls_schar.c: Likewise.
1985 * testsuite/libffi.call/cls_sint.c: Likewise.
1986 * testsuite/libffi.call/cls_sshort.c: Likewise.
1987 * testsuite/libffi.call/nested_struct2.c: Likewise.
1988 * testsuite/libffi.call/nested_struct3.c: Likewise.
1990 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
1992 * testsuite/libffi.call/cls_double.c: Do a check on the result.
1993 * testsuite/libffi.call/cls_uchar.c: Likewise.
1994 * testsuite/libffi.call/cls_uint.c: Likewise.
1995 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1996 * testsuite/libffi.call/cls_ushort.c: Likewise.
1997 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
1999 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2001 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2004 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2006 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2007 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2009 2003-10-22 David Daney <ddaney@avtrex.com>
2011 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2012 __attribute__((__mode__(__SI__))) and friends.
2014 2003-10-22 Andreas Schwab <schwab@suse.de>
2016 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2018 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2020 * configure.in: AC_LINK_FILES(ffitarget.h).
2021 * configure: Regenerate.
2022 * Makefile.in: Likewise.
2023 * include/Makefile.in: Likewise.
2024 * testsuite/Makefile.in: Likewise.
2025 * fficonfig.h.in: Likewise.
2027 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2028 Richard Henderson <rth@redhat.com>
2030 Avoid that ffi.h includes fficonfig.h.
2032 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2033 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2034 (TARGET_SRC_MIPS_SGI): Removed.
2035 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2036 (MIPS_SGI): Removed.
2037 (CLEANFILES): Removed.
2038 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2040 * acconfig.h: Removed.
2041 * configure.in: Compute sizeofs only for double and long double.
2042 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2043 into AC_DEFINE instead of using acconfig.h. Create
2044 include/ffitarget.h instead of include/fficonfig.h. Rename
2045 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2046 AC_DEFINE EH_FRAME_FLAGS.
2047 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2048 (hack_DATA): Add ffitarget.h.
2049 * include/ffi.h.in: Remove all system specific definitions.
2050 Declare raw API even if it is not installed, why bother?
2051 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2052 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2053 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2054 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2055 * include/ffi_common.h (bool): Do not define.
2056 (ffi_assert): Accept failed assertion.
2057 (ffi_type_test): Return void and accept file/line.
2058 (FFI_ASSERT): Pass stringized failed assertion.
2059 (FFI_ASSERT_AT): New macro.
2060 (FFI_ASSERT_VALID_TYPE): New macro.
2061 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2062 UINT64, SINT64): Define here with gcc's __attribute__ macro
2064 (FLOAT32, ALIGN): Define here instead of in ffi.h
2065 * include/ffi-mips.h: Removed. Its content moved to
2066 src/mips/ffitarget.h after separating assembly and C sections.
2067 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2068 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2069 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2070 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2071 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2072 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2073 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2074 (ffi_type_test): Rewritten.
2075 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2076 FFI_ASSERT_VALID_TYPE.
2077 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2078 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2079 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2080 src/s390/ffitarget.h, src/sh/ffitarget.h,
2081 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2082 src/x86/ffitarget.h: New files.
2083 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2084 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2085 src/powerpc/aix.S, src/powerpc/darwin.S,
2086 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2087 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2088 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2089 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2090 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2093 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2095 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2096 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2098 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2100 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2101 Used when FFI_DEBUG = 1.
2103 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2105 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2108 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2110 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2111 FFI_MIPS_O32 for O32 ABI.
2113 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2115 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2116 SPARC64. Cleanup whitespaces.
2118 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2120 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2121 strongarm, xscale. Cleanup whitespaces.
2122 * testsuite/libffi.call/closure_fn1.c: Likewise.
2123 * testsuite/libffi.call/closure_fn2.c: Likewise.
2124 * testsuite/libffi.call/closure_fn3.c: Likewise.
2125 * testsuite/libffi.call/cls_12byte.c: Likewise.
2126 * testsuite/libffi.call/cls_16byte.c: Likewise.
2127 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2128 * testsuite/libffi.call/cls_20byte.c: Likewise.
2129 * testsuite/libffi.call/cls_24byte.c: Likewise.
2130 * testsuite/libffi.call/cls_2byte.c: Likewise.
2131 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2132 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2133 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2134 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2135 * testsuite/libffi.call/cls_4byte.c: Likewise.
2136 * testsuite/libffi.call/cls_5byte.c: Likewise.
2137 * testsuite/libffi.call/cls_6byte.c: Likewise.
2138 * testsuite/libffi.call/cls_7byte.c: Likewise.
2139 * testsuite/libffi.call/cls_8byte.c: Likewise.
2140 * testsuite/libffi.call/cls_double.c: Likewise.
2141 * testsuite/libffi.call/cls_float.c: Likewise.
2142 * testsuite/libffi.call/cls_uchar.c: Likewise.
2143 * testsuite/libffi.call/cls_uint.c: Likewise.
2144 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2145 * testsuite/libffi.call/cls_ushort.c: Likewise.
2146 * testsuite/libffi.call/nested_struct.c: Likewise.
2147 * testsuite/libffi.call/nested_struct1.c: Likewise.
2148 * testsuite/libffi.call/problem1.c: Likewise.
2149 * testsuite/libffi.special/unwindtest.cc: Likewise.
2150 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2152 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2154 * src/powerpc/aix.S: Cleanup whitespaces.
2155 * src/powerpc/aix_closure.S: Likewise.
2157 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2159 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2160 * src/powerpc/darwin_closure.S: Likewise.
2161 * src/powerpc/ffi_darwin.c: Likewise.
2163 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2164 David Edelsohn <edelsohn@gnu.org>
2166 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2167 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2169 * src/powerpc/darwin_closure.S: Likewise.
2170 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2171 according to the Darwin/AIX ABI.
2172 (ffi_prep_cif_machdep): Likewise.
2173 (ffi_closure_helper_DARWIN): Likewise.
2174 Remove the outgoing parameter stack logic. Simplify the evaluation
2175 of the different CASE types.
2176 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2177 statement in the trampoline code.
2179 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2181 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2182 for the register size.
2183 (ffi_closure_helper_SYSV): Handle the structure return value
2185 (ffi_closure_helper_SYSV): Return the appropriate type when
2186 the registers are used for the structure return value.
2187 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2188 the 64-bit return value. Update copyright years.
2190 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2192 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2193 srcdir for ffi_mips.h.
2195 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2197 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2199 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2200 placement of float result.
2201 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2202 cast of "resp" for big-endian 64 bit machines.
2204 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2206 * src/types.c (double, longdouble): Merge identical SH and ARM
2207 typedefs, and add POWERPC64.
2208 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2209 struct split over gpr and rest.
2210 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2211 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2213 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2215 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2218 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2220 * configure: Regenerate.
2222 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2224 * Makefile.am: Remove build rules for ffitest.
2225 * Makefile.in: Rebuilt.
2227 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2229 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2230 about implicit declaration of abort().
2232 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2234 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2235 * Makefile.in: Rebuilt.
2236 * configure.in: Add dejagnu test framework.
2237 * configure: Rebuilt.
2239 * testsuite/Makefile.am: New file.
2240 * testsuite/Makefile.in: Built
2241 * testsuite/lib/libffi-dg.exp: New file.
2242 * testsuite/config/default.exp: Likewise.
2243 * testsuite/libffi.call/call.exp: Likewise.
2244 * testsuite/libffi.call/ffitest.h: Likewise.
2245 * testsuite/libffi.call/closure_fn0.c: Likewise.
2246 * testsuite/libffi.call/closure_fn1.c: Likewise.
2247 * testsuite/libffi.call/closure_fn2.c: Likewise.
2248 * testsuite/libffi.call/closure_fn3.c: Likewise.
2249 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2250 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2251 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2252 * testsuite/libffi.call/cls_2byte.c: Likewise.
2253 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2254 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2255 * testsuite/libffi.call/cls_4byte.c: Likewise.
2256 * testsuite/libffi.call/cls_5byte.c: Likewise.
2257 * testsuite/libffi.call/cls_6byte.c: Likewise.
2258 * testsuite/libffi.call/cls_7byte.c: Likewise.
2259 * testsuite/libffi.call/cls_8byte.c: Likewise.
2260 * testsuite/libffi.call/cls_12byte.c: Likewise.
2261 * testsuite/libffi.call/cls_16byte.c: Likewise.
2262 * testsuite/libffi.call/cls_20byte.c: Likewise.
2263 * testsuite/libffi.call/cls_24byte.c: Likewise.
2264 * testsuite/libffi.call/cls_double.c: Likewise.
2265 * testsuite/libffi.call/cls_float.c: Likewise.
2266 * testsuite/libffi.call/cls_uchar.c: Likewise.
2267 * testsuite/libffi.call/cls_uint.c: Likewise.
2268 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2269 * testsuite/libffi.call/cls_ushort.c: Likewise.
2270 * testsuite/libffi.call/float.c: Likewise.
2271 * testsuite/libffi.call/float1.c: Likewise.
2272 * testsuite/libffi.call/float2.c: Likewise.
2273 * testsuite/libffi.call/many.c: Likewise.
2274 * testsuite/libffi.call/many_win32.c: Likewise.
2275 * testsuite/libffi.call/nested_struct.c: Likewise.
2276 * testsuite/libffi.call/nested_struct1.c: Likewise.
2277 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2278 * testsuite/libffi.call/problem1.c: Likewise.
2279 * testsuite/libffi.call/promotion.c: Likewise.
2280 * testsuite/libffi.call/return_ll.c: Likewise.
2281 * testsuite/libffi.call/return_sc.c: Likewise.
2282 * testsuite/libffi.call/return_uc.c: Likewise.
2283 * testsuite/libffi.call/strlen.c: Likewise.
2284 * testsuite/libffi.call/strlen_win32.c: Likewise.
2285 * testsuite/libffi.call/struct1.c: Likewise.
2286 * testsuite/libffi.call/struct2.c: Likewise.
2287 * testsuite/libffi.call/struct3.c: Likewise.
2288 * testsuite/libffi.call/struct4.c: Likewise.
2289 * testsuite/libffi.call/struct5.c: Likewise.
2290 * testsuite/libffi.call/struct6.c: Likewise.
2291 * testsuite/libffi.call/struct7.c: Likewise.
2292 * testsuite/libffi.call/struct8.c: Likewise.
2293 * testsuite/libffi.call/struct9.c: Likewise.
2294 * testsuite/libffi.special/special.exp: New file.
2295 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2296 * testsuite/libffi.special/unwindtest.cc: Likewise.
2299 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2301 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2304 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2306 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2308 (ffi_closure_helper_LINUX64): Likewise.
2309 * src/powerpc/linux64.S: Remove code writing to parm save area.
2310 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2311 address in lr from ffi_closure_helper_LINUX64 call to calculate
2312 table address. Optimize function tail.
2314 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2316 * src/sparc/ffi.c: Handle all floating point registers.
2317 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2319 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2321 * README: Note that libffi is not part of GCC. Update the project
2324 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2326 * src/powerpc/ppc_closure.S: Include ffi.h.
2328 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2330 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2331 Use C style comments.
2333 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2335 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2336 * Makefile.in: Regenerate.
2337 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2338 * configure: Regenerate.
2339 * include/ffi.h.in: Add SHmedia support.
2340 * src/sh64/ffi.c: New file.
2341 * src/sh64/sysv.S: New file.
2343 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2345 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2346 should be read-only.
2347 * configure: Rebuilt.
2348 * fficonfig.h.in: Rebuilt.
2349 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2350 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2351 * src/powerpc/linux64.S: Likewise.
2352 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2353 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2354 if -fpic/-fPIC/-mrelocatable.
2355 * src/powerpc/powerpc_closure.S: Likewise.
2356 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2357 #write in .eh_frame flags.
2358 * src/sparc/v9.S: Likewise.
2359 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2360 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2361 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2363 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2365 Fixes PR bootstrap/10656
2366 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2367 support for .register pseudo-op.
2368 * src/sparc/v8.S: Use it.
2369 * fficonfig.h.in: Rebuilt.
2370 * configure: Rebuilt.
2372 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2374 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2375 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2376 Make it the default on POWERPC64.
2377 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2378 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2379 * configure: Rebuilt.
2380 * src/powerpc/ffi.c (hidden): Define.
2381 (ffi_prep_args_SYSV): Renamed from
2382 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2383 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2384 ASM_NEEDS_REGISTERS64): New.
2385 (ffi_prep_args64): New function.
2386 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2387 (ffi_call): Likewise.
2388 (ffi_prep_closure): Likewise.
2389 (flush_icache): Surround by #ifndef POWERPC64.
2390 (ffi_dblfl): New union type.
2391 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2392 (ffi_closure_helper_LINUX64): New function.
2393 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2395 * src/powerpc/sysv.S: Likewise.
2396 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2397 * src/powerpc/linux64.S: New file.
2398 * src/powerpc/linux64_closure.S: New file.
2399 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2400 src/powerpc/linux64_closure.S.
2401 (TARGET_SRC_POWERPC): Likewise.
2403 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2404 closure_test_fn3): Fix result printing on big-endian 64-bit
2406 (main): Print tst2_arg instead of uninitialized tst2_result.
2408 * src/ffitest.c (main): Hide what closure pointer really points to
2411 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2413 * configure.in (arm-*-netbsdelf*): Add configuration.
2414 (configure): Regenerated.
2416 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2418 * include/Makefile.in: Regenerate.
2420 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2422 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2424 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2425 Receive closure pointer through parameter, read args using
2426 __builtin_dwarf_cfa.
2427 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2429 2003-03-12 Andreas Schwab <schwab@suse.de>
2431 * configure.in: Avoid trailing /. in toolexeclibdir.
2432 * configure: Rebuilt.
2434 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2436 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2438 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2440 * libffi/src/powerpc/darwin_closure.S:
2441 Fix alignement bug, allocate 8 bytes for the result.
2442 * libffi/src/powerpc/aix_closure.S:
2444 * libffi/src/powerpc/ffi_darwin.c:
2445 Update stackframe description for aix/darwin_closure.S.
2447 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2449 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2452 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2453 Andreas Schwab <schwab@suse.de>
2455 * configure.in: Adjust command to source config-ml.in to account
2456 for changes to the libffi_basedir definition.
2457 (libffi_basedir): Remove ${srcdir} from value and include trailing
2460 * configure: Regenerate.
2462 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2464 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2466 2003-01-28 Andrew Haley <aph@redhat.com>
2468 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2469 * src/x86/ffi64.c (ffi_prep_closure): New.
2470 (ffi_closure_UNIX64_inner): New.
2471 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2473 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2475 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2476 Remove USE_LIBDIR conditional.
2477 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2478 * Makefile.in, configure: Rebuilt.
2480 2003-01027 David Edelsohn <edelsohn@gnu.org>
2482 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2483 * Makefile.in: Regenerate.
2485 2003-01-22 Andrew Haley <aph@redhat.com>
2487 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2490 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2492 * src/powerpc/darwin.S: Add unwind info.
2493 * src/powerpc/darwin_closure.S: Likewise.
2495 2003-01-14 Andrew Haley <aph@redhat.com>
2497 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2498 (ffi_prep_cif_machdep): Likewise.
2499 * src/x86/unix64.S: Add unwind info.
2501 2003-01-14 Andreas Jaeger <aj@suse.de>
2503 * src/ffitest.c (main): Only use ffi_closures if those are
2506 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2508 * libffi/src/ffitest.c
2509 add closure testcases
2511 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2513 * libffi/src/powerpc/ffi.c
2514 fix alignment bug for float (4 byte aligned iso 8 byte)
2516 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2518 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2519 * src/powerpc/darwin.S: Remove RCS version string.
2521 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2523 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2524 * src/ffitest.c (main): Use static storage for closure.
2525 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2526 * src/sparc/v8.S (ffi_closure_v8): New.
2527 * src/sparc/v9.S (ffi_closure_v9): New.
2529 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2531 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2532 enumeration for X86_WIN32.
2533 * src/x86/win32.S: Added ffi_call_STDCALL function
2535 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2536 switch cases for recognising FFI_STDCALL and
2537 calling ffi_call_STDCALL if target is X86_WIN32.
2538 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2539 stdcall versions of the "my_strlen" and "many"
2540 test functions (for X86_WIN32).
2541 Added test cases to test stdcall invocation using
2544 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2546 * src/sh/sysv.S: Add DWARF2 unwind info.
2548 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2550 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2552 2002-11-26 Jim Wilson <wilson@redhat.com>
2554 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2556 2002-11-23 H.J. Lu <hjl@gnu.org>
2558 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2559 Include ../config/accross.m4.
2560 * aclocal.m4; Rebuild.
2561 * configure: Likewise.
2563 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2565 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2567 2002-11-11 DJ Delorie <dj@redhat.com>
2569 * configure.in: Look for common files in the right place.
2571 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2573 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2574 raw data as _Jv_word values, not ffi_raw.
2575 (ffi_java_ptrarray_to_raw): Likewise.
2576 (ffi_java_rvalue_to_raw): New function.
2577 (ffi_java_raw_call): Call it.
2578 (ffi_java_raw_to_rvalue): New function.
2579 (ffi_java_translate_args): Call it.
2580 * src/ffitest.c (closure_test_fn): Interpret return value
2581 as ffi_arg, not int.
2582 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2583 FFI_TYPE_POINTER case.
2584 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2585 values extended to word size.
2587 2002-10-02 Andreas Jaeger <aj@suse.de>
2589 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2591 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2593 * include/ffi.h.in: Fix i386 win32 compilation.
2595 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2597 * configure.in: Add s390x-*-linux-* target.
2598 * configure: Regenerate.
2599 * include/ffi.h.in: Define S390X for s390x targets.
2600 (FFI_CLOSURES): Define for s390/s390x.
2601 (FFI_TRAMPOLINE_SIZE): Likewise.
2602 (FFI_NATIVE_RAW_API): Likewise.
2603 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2604 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2605 * src/s390/ffi.c: Major rework of existing code. Add support for
2606 s390x targets. Add closure support.
2607 * src/s390/sysv.S: Likewise.
2609 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2611 * src/arm/sysv.S: Fix typo.
2613 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2615 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2616 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2617 (ffi_call_SYSV): Handle soft-float.
2619 2002-09-27 Bo Thorsen <bo@suse.de>
2621 * include/ffi.h.in: Fix multilib x86-64 support.
2623 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2625 * Makefile.am (all-multi): Fix multilib parallel build.
2627 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2629 * configure.in (sh[34]*-*-linux*): Add brackets.
2630 * configure: Regenerate.
2632 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2634 * Makefile.am: Add SH support.
2635 * Makefile.in: Regenerate.
2636 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2637 * configure: Regenerate.
2638 * include/ffi.h.in: Add SH support.
2639 * src/sh/ffi.c: New file.
2640 * src/sh/sysv.S: New file.
2641 * src/types.c: Add SH support.
2643 2002-07-16 Bo Thorsen <bo@suse.de>
2645 * src/x86/ffi64.c: New file that adds x86-64 support.
2646 * src/x86/unix64.S: New file that handles argument setup for
2648 * src/x86/sysv.S: Don't use this on x86-64.
2649 * src/x86/ffi.c: Don't use this on x86-64.
2651 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2653 * src/ffitest.c (struct6): New test that tests a special case in
2655 (struct7): Likewise.
2656 (struct8): Likewise.
2657 (struct9): Likewise.
2658 (closure_test_fn): Silence warning about this when it's not used.
2659 (main): Add the new tests.
2660 (main): Fix a couple of wrong casts and silence some compiler warnings.
2661 * include/ffi.h.in: Add x86-64 ABI definition.
2662 * fficonfig.h.in: Regenerate.
2663 * Makefile.am: Add x86-64 support.
2664 * configure.in: Likewise.
2665 * Makefile.in: Regenerate.
2666 * configure: Likewise.
2668 2002-06-24 Bo Thorsen <bo@suse.de>
2670 * src/types.c: Merge settings for similar architectures.
2671 Add x86-64 sizes and alignments.
2673 2002-06-23 Bo Thorsen <bo@suse.de>
2675 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2676 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2677 * src/mips/ffi.c (ffi_prep_args): Likewise.
2678 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2680 2002-07-18 H.J. Lu (hjl@gnu.org)
2682 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2683 (libffi_la_SOURCES): Support MIPS_LINUX.
2684 (libffi_convenience_la_SOURCES): Likewise.
2685 * Makefile.in: Regenerated.
2687 * configure.in (mips64*-*): Skip.
2688 (mips*-*-linux*): New.
2689 * configure: Regenerated.
2691 * src/mips/ffi.c: Include <sgidefs.h>.
2693 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2695 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2697 2002-05-27 Roger Sayle <roger@eyesopen.com>
2699 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2701 2002-05-27 Bo Thorsen <bo@suse.de>
2703 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2706 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2708 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2709 beginning of function (for older apple cc).
2711 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2713 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2714 script entry, and set LD to it when configuring multilibs.
2715 * configure: Rebuilt.
2717 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2719 * configure.in (sparc64-*-netbsd*): Add target.
2720 (sparc-*-netbsdelf*): Likewise.
2721 * configure: Regenerate.
2723 2002-04-28 David S. Miller <davem@redhat.com>
2725 * configure.in, configure: Fix SPARC test in previous change.
2727 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2729 * Makefile.am: Add Linux for S/390 support.
2730 * Makefile.in: Regenerate.
2731 * configure.in: Add Linux for S/390 support.
2732 * configure: Regenerate.
2733 * include/ffi.h.in: Add Linux for S/390 support.
2734 * src/s390/ffi.c: New file from libffi CVS tree.
2735 * src/s390/sysv.S: New file from libffi CVS tree.
2737 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2739 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2741 * src/sparc/v8.S: Use it.
2742 * src/sparc/v9.S: Likewise.
2743 * fficonfig.h.in: Rebuilt.
2744 * configure: Rebuilt.
2746 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2748 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2750 * src/ia64/unix.S: Add unwind information. Fix comments.
2751 Save sp in a way that's compatible with unwind info.
2752 (ffi_call_unix): Correctly restore sp in all cases.
2753 * src/ia64/ffi.c: Add, fix comments.
2755 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2757 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2759 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2761 * configure.in (alpha*-*-netbsd*): Add target.
2762 * configure: Regenerate.
2764 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2766 * src/sparc/v8.S: Add unwind info.
2767 * src/sparc/v9.S: Likewise.
2769 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2771 * configure.in: Enable i*86-*-netbsdelf*.
2772 * configure: Rebuilt.
2774 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2777 * src/mips/n32.s: Delete
2778 * src/mips/o32.s: Delete
2780 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2782 * configure.in: Enable alpha*-*-freebsd*.
2783 * configure: Rebuilt.
2785 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2787 * Makefile.am: libfficonvenience -> libffi_convenience.
2788 * Makefile.in: Rebuilt.
2790 * Makefile.am: Define ffitest_OBJECTS.
2791 * Makefile.in: Rebuilt.
2793 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2794 David Edelsohn <edelsohn@gnu.org>
2796 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2797 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2798 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2799 * Makefile.in: Regenerate.
2800 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2801 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2802 (flush_icache, flush_range): New functions.
2803 (ffi_closure_helper_DARWIN): New function.
2804 * src/powerpc/aix_closure.S: New file.
2805 * src/powerpc/darwin_closure.S: New file.
2807 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2809 * include/ffi.h.in: Add typedef for ffi_arg.
2810 * src/ffitest.c (main): Declare rint with ffi_arg.
2812 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2814 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2815 number of GPRs for floating-point arguments.
2817 2002-01-31 Anthony Green <green@redhat.com>
2819 * configure: Rebuilt.
2820 * configure.in: Replace CHECK_SIZEOF and endian tests with
2821 cross-compiler friendly macros.
2822 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2825 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2827 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2828 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2830 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2832 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2833 (TARGET_SRC_POWERPC_AIX): New.
2834 (POWERPC_AIX): New stanza.
2835 * Makefile.in: Regenerate.
2836 * configure.in: Add AIX case.
2837 * configure: Regenerate.
2838 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2839 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2840 size. Fix "long double" support.
2841 (ffi_call): Add FFI_AIX case.
2842 * src/powerpc/aix.S: New.
2844 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2846 Implement Darwin PowerPC ABI.
2847 * configure.in: Handle powerpc-*-darwin*.
2848 * Makefile.am: Set source files for POWERPC_DARWIN.
2849 * configure: Rebuilt.
2850 * Makefile.in: Rebuilt.
2851 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2853 * src/powerpc/darwin.S: New file.
2854 * src/powerpc/ffi_darwin.c: New file.
2856 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2858 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2860 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2862 * src/x86/sysv.S: Avoid gas-only .balign directive.
2863 Use C style comments.
2865 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2867 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2868 Fixes PR bootstrap/3563.
2870 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2872 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2874 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2876 * configure.in: Recognize sparc*-sun-* host.
2877 * configure: Regenerate.
2879 2001-06-06 Andrew Haley <aph@redhat.com>
2881 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
2883 2001-06-03 Andrew Haley <aph@redhat.com>
2885 * src/alpha/osf.S: Add unwind info.
2886 * src/powerpc/sysv.S: Add unwind info.
2887 * src/powerpc/ppc_closure.S: Likewise.
2889 2000-05-31 Jeff Sturm <jsturm@one-point.com>
2891 * configure.in: Fix AC_ARG_ENABLE usage.
2892 * configure: Rebuilt.
2894 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2896 * configure.in: Remove warning about beta code.
2897 * configure: Rebuilt.
2899 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
2901 * src/ia64/unix.S: Restore stack pointer when returning from
2903 * src/ia64/ffi.c: Fix typo in comment.
2905 2001-04-18 Jim Wilson <wilson@redhat.com>
2907 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
2908 to eliminate RAW DV.
2910 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
2912 * Makefile.am: Make a libtool convenience library.
2913 * Makefile.in: Rebuilt.
2915 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
2917 * configure.in: Use different syntax for subdirectory creation.
2918 * configure: Rebuilt.
2920 2001-03-27 Jon Beniston <jon@beniston.com>
2922 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
2923 * configure: Rebuilt.
2924 * Makefile.am: Added X86_WIN32 target support.
2925 * Makefile.in: Rebuilt.
2927 * include/ffi.h.in: Added X86_WIN32 target support.
2929 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
2930 * src/types.c: Added X86_WIN32 target support.
2932 * src/x86/win32.S: New file. Based on sysv.S, but with EH
2933 stuff removed and made to work with CygWin's gas.
2935 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
2937 * configure.in: Make target subdirectory in build dir.
2938 * Makefile.am: Override suffix based rules to specify correct output
2940 * Makefile.in: Rebuilt.
2941 * configure: Rebuilt.
2943 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
2945 * src/powerpc/ppc_closure.S: New file.
2946 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
2947 involving long long and register pairs.
2948 (ffi_prep_closure): New function.
2949 (flush_icache): Likewise.
2950 (ffi_closure_helper_SYSV): Likewise.
2951 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
2952 (FFI_TRAMPOLINE_SIZE): Likewise.
2953 (FFI_NATIVE_RAW_API): Likewise.
2954 * Makefile.in: Rebuilt.
2955 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
2956 (TARGET_SRC_POWERPC): Likewise.
2958 2001-03-19 Tom Tromey <tromey@redhat.com>
2960 * Makefile.in: Rebuilt.
2961 * Makefile.am (ffitest_LDFLAGS): New macro.
2963 2001-03-02 Nick Clifton <nickc@redhat.com>
2965 * include/ffi.h.in: Remove RCS ident string.
2966 * include/ffi_mips.h: Remove RCS ident string.
2967 * src/debug.c: Remove RCS ident string.
2968 * src/ffitest.c: Remove RCS ident string.
2969 * src/prep_cif.c: Remove RCS ident string.
2970 * src/types.c: Remove RCS ident string.
2971 * src/alpha/ffi.c: Remove RCS ident string.
2972 * src/alpha/osf.S: Remove RCS ident string.
2973 * src/arm/ffi.c: Remove RCS ident string.
2974 * src/arm/sysv.S: Remove RCS ident string.
2975 * src/mips/ffi.c: Remove RCS ident string.
2976 * src/mips/n32.S: Remove RCS ident string.
2977 * src/mips/o32.S: Remove RCS ident string.
2978 * src/sparc/ffi.c: Remove RCS ident string.
2979 * src/sparc/v8.S: Remove RCS ident string.
2980 * src/sparc/v9.S: Remove RCS ident string.
2981 * src/x86/ffi.c: Remove RCS ident string.
2982 * src/x86/sysv.S: Remove RCS ident string.
2984 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
2986 * include/ffi.h.in: Change sourceware.cygnus.com references to
2989 2000-12-09 Richard Henderson <rth@redhat.com>
2991 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
2992 (ffi_closure_osf_inner): Index rather than increment avalue
2993 and arg_types. Give ffi_closure_osf the raw return value type.
2994 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
2997 2000-12-07 Richard Henderson <rth@redhat.com>
2999 * src/raw_api.c (ffi_translate_args): Fix typo.
3000 (ffi_prep_closure): Likewise.
3002 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3003 FFI_TRAMPOLINE_SIZE.
3004 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3005 cif->bytes for new ffi_call_osf implementation.
3006 (ffi_prep_args): Absorb into ...
3007 (ffi_call): ... here. Do all stack allocation here and
3008 avoid a callback function.
3009 (ffi_prep_closure, ffi_closure_osf_inner): New.
3010 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3011 (ffi_closure_osf): New.
3013 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3015 * config.guess, config.sub, install-sh: Removed.
3016 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3017 * Makefile.in: Rebuilt.
3019 * acinclude.m4: Include libtool macros from the top level.
3020 * aclocal.m4, configure: Rebuilt.
3022 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3024 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3025 * configure: Rebuilt.
3027 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3029 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3030 memory correctly. Use conditional instructions, not branches where
3033 2000-05-04 Tom Tromey <tromey@cygnus.com>
3035 * configure: Rebuilt.
3036 * configure.in: Match `arm*-*-linux-*'.
3037 From Chris Dornan <cdornan@arm.com>.
3039 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3041 * Makefile.am (SUBDIRS): Define.
3042 (AM_MAKEFLAGS): Likewise.
3043 (Multilib support.): Add section.
3044 * Makefile.in: Rebuilt.
3045 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3046 New variables. Set for gcc using -print-multi-lib. Export them
3048 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3049 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3051 (extra_compiler_flags_value, extra_compiler_flags): Check these
3052 for extra compiler options which need to be passed down in
3055 2000-04-16 Anthony Green <green@redhat.com>
3057 * configure: Rebuilt.
3058 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3060 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3062 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3063 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3064 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3065 Replace all void * sizeofs with sizeof(int).
3066 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3067 different than DOUBLE.
3068 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3069 (ffi_prep_args_v9): New function.
3070 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3071 (ffi_V9_return_struct): New function.
3072 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3073 32bit code (not yet cross-arch calls).
3074 * src/sparc/v8.S: Add struct return delay nop.
3076 * src/sparc/v9.S: New file.
3077 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3078 is used on sparc64 only for structures larger than 32 bytes.
3079 Pass by reference for structures is done for structure arguments
3080 larger than 16 bytes.
3081 * src/ffitest.c (main): Use 64bit rint on sparc64.
3082 Run long long tests on sparc.
3083 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3085 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3087 * configure.in (sparc-*-linux*): New supported target.
3088 (sparc64-*-linux*): Likewise.
3089 * configure: Rebuilt.
3090 * Makefile.am: Add v9.S to SPARC files.
3091 * Makefile.in: Likewise.
3092 (LINK): Surround $(CCLD) into double quotes, so that multilib
3093 compiles work correctly.
3095 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3097 * configure: Rebuilt.
3098 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3099 proposed by Bryce McKinlay.
3101 2000-03-20 Tom Tromey <tromey@cygnus.com>
3103 * Makefile.in: Hand edit for java_raw_api.lo.
3105 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3107 * config.guess, config.sub: Update from the gcc tree.
3108 Fix for PR libgcj/168.
3110 2000-03-03 Tom Tromey <tromey@cygnus.com>
3112 * Makefile.in: Fixed ia64 by hand.
3114 * configure: Rebuilt.
3115 * configure.in (--enable-multilib): New option.
3116 (libffi_basedir): New subst.
3117 (AC_OUTPUT): Added multilib code.
3119 2000-03-02 Tom Tromey <tromey@cygnus.com>
3121 * Makefile.in: Rebuilt.
3122 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3125 2000-02-25 Hans Boehm <boehm@acm.org>
3127 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3129 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3131 (ffi_prep_raw_closure): Use ffi_translate_args, not
3132 ffi_closure_translate.
3133 * src/java_raw_api.c: New file.
3134 * src/ffitest.c (closure_test_fn): New function.
3135 (main): Define `rint' as long long on IA64. Added new test when
3136 FFI_CLOSURES is defined.
3137 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3138 (ffi_abi): Recognize IA64.
3139 (ffi_raw): Added `flt' field.
3140 Added "Java raw API" code.
3141 * configure.in: Recognize ia64.
3142 * Makefile.am (TARGET_SRC_IA64): New macro.
3143 (libffi_la_common_SOURCES): Added java_raw_api.c.
3144 (libffi_la_SOURCES): Define in IA64 case.
3146 2000-01-04 Tom Tromey <tromey@cygnus.com>
3148 * Makefile.in: Rebuilt with newer automake.
3150 1999-12-31 Tom Tromey <tromey@cygnus.com>
3152 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3154 1999-09-01 Tom Tromey <tromey@cygnus.com>
3156 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3158 * fficonfig.h.in: Rebuilt.
3159 * configure: Rebuilt.
3160 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3161 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3162 * acconfig.h: Don't #undef PACKAGE or VERSION.
3164 1999-08-09 Anthony Green <green@cygnus.com>
3166 * include/ffi.h.in: Try to work around messy header problem
3167 with PACKAGE and VERSION.
3169 * configure: Rebuilt.
3170 * configure.in: Change version to 2.00-beta.
3172 * fficonfig.h.in: Rebuilt.
3173 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3175 * src/x86/ffi.c (ffi_raw_call): Rename.
3177 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3179 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3180 (ffi_prep_incoming_args_SYSV): Ditto.
3181 (ffi_prep_closure): Ditto.
3182 (ffi_closure_raw_SYSV): Ditto.
3183 (ffi_prep_raw_closure): More ditto.
3184 (ffi_call_raw): Final ditto.
3186 * include/ffi.h.in: Add definitions for closure and raw API.
3188 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3191 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3193 * src/raw_api.c: New file.
3195 * include/ffi.h.in (ffi_raw): New type.
3196 (UINT_ARG, SINT_ARG): New defines.
3197 (ffi_closure, ffi_raw_closure): New types.
3198 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3200 * configure.in: Add check for endianness and sizeof void*.
3202 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3203 instead of directly.
3205 * configure: Rebuilt.
3207 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3209 * configure.in: Add x86 and powerpc BeOS configurations.
3210 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3212 1999-05-09 Anthony Green <green@cygnus.com>
3214 * configure.in: Add warning about this being beta code.
3215 Remove src/Makefile.am from the picture.
3216 * configure: Rebuilt.
3218 * Makefile.am: Move logic from src/Makefile.am. Add changes
3219 to support libffi as a target library.
3220 * Makefile.in: Rebuilt.
3222 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3223 Upgraded to new autoconf, automake, libtool.
3227 * LICENSE: Update copyright date.
3229 * src/Makefile.am, src/Makefile.in: Removed.
3231 1998-11-29 Anthony Green <green@cygnus.com>
3233 * include/ChangeLog: Removed.
3234 * src/ChangeLog: Removed.
3235 * src/mips/ChangeLog: Removed.
3236 * src/sparc/ChangeLog: Remboved.
3237 * src/x86/ChangeLog: Removed.
3239 * ChangeLog.v1: Created.