1 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
3 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
4 ffi_closure_helper_SYSV): Add long double handling.
6 2007-02-02 Jakub Jelinek <jakub@redhat.com>
8 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
9 immediately after bctrl instruction.
11 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
13 * Makefile.am (all-recursive, install-recursive,
14 mostlyclean-recursive, clean-recursive, distclean-recursive,
15 maintainer-clean-recursive): Add missing targets.
16 * Makefile.in: Rebuilt.
18 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
20 * configure.ac: Add TARGET for x86_64-*-darwin*.
21 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
23 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
24 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
25 * configure: Regenerate.
26 * Makefile.in: Regenerate.
27 * include/Makefile.in: Regenerate.
28 * testsuite/Makefile.in: Regenerate.
29 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
32 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
34 * aclocal.m4: Regenerate with aclocal -I .. as written in the
37 2006-10-31 Geoffrey Keating <geoffk@apple.com>
39 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
40 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
42 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
43 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
45 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
46 Sandro Tolaini <tolaini@libero.it>
48 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
50 * configure: Regenerated.
51 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
52 (EXTRA_DIST): Add src/x86/darwin.S.
53 * Makefile.in: Regenerated.
54 * include/Makefile.in: Regenerated.
55 * testsuite/Makefile.in: Regenerated.
57 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
58 X86_WIN32, and additionally align stack to 16 bytes.
59 * src/x86/darwin.S: New, based on sysv.S.
60 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
62 2006-09-12 David Daney <ddaney@avtrex.com>
65 * include/Makefile.am: Install both ffi.h and ffitarget.h in
66 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
67 * aclocal.m4: Regenerated for automake 1.9.6.
68 * Makefile.in: Regenerated.
69 * include/Makefile.in: Regenerated.
70 * testsuite/Makefile.in: Regenerated.
72 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
74 * include/ffi_common.h (struct): Revert accidental commit.
76 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
78 * include/ffi_common.h: Remove lint directives.
79 * include/ffi.h.in: Likewise.
81 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
83 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
84 for 32-bit architectures.
85 * testsuite/libffi.call/return_ul.c: New test case.
87 2006-07-19 David Daney <ddaney@avtrex.com>
89 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
90 xfail remains for mips64.
92 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
94 * Makefile.am: Add install-html target. Add install-html to .PHONY
95 * Makefile.in: Regenerate.
96 * aclocal.m4: Regenerate.
97 * include/Makefile.in: Regenerate.
98 * testsuite/Makefile.in: Regenerate.
100 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
102 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
105 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
107 * README: Remove notice about 'Crazy Comments'.
108 * src/debug.c: Remove lint directives. Cleanup white spaces.
109 * src/java_raw_api.c: Likewise.
110 * src/prep_cif.c: Likewise.
111 * src/raw_api.c: Likewise.
112 * src/ffitest.c: Delete. No longer needed, all test cases migrated
114 * src/arm/ffi.c: Remove lint directives.
115 * src/m32r/ffi.c: Likewise.
116 * src/pa/ffi.c: Likewise.
117 * src/powerpc/ffi.c: Likewise.
118 * src/powerpc/ffi_darwin.c: Likewise.
119 * src/sh/ffi.c: Likewise.
120 * src/sh64/ffi.c: Likewise.
121 * src/x86/ffi.c: Likewise.
122 * testsuite/libffi.call/float2.c: Likewise.
123 * testsuite/libffi.call/promotion.c: Likewise.
124 * testsuite/libffi.call/struct1.c: Likewise.
126 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
128 * src/pa/hpux32.S: Correct unwind offset calculation for
130 * src/pa/linux.S: Likewise.
132 2006-04-12 James E Wilson <wilson@specifix.com>
135 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
136 (hfa_type_load): Call stf_spill.
137 (hfa_type_store): Call ldf_fill.
138 (ffi_call): Adjust calls to above routines. Add local temps for
141 2006-04-10 Matthias Klose <doko@debian.org>
143 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
144 directory names containing underscores.
146 2006-04-07 James E Wilson <wilson@specifix.com>
148 * testsuite/libffi.call/float4.c: New testcase.
150 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
151 Andreas Tobler <a.tobler@schweiz.ch>
153 * Makefile.am: Add PA_HPUX port.
154 * Makefile.in: Regenerate.
155 * include/Makefile.in: Likewise.
156 * testsuite/Makefile.in: Likewise.
157 * configure.ac: Add PA_HPUX rules.
158 * configure: Regenerate.
159 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
160 Add PA_HPUX and PA64_HPUX.
161 Rename FFI_LINUX ABI to FFI_PA32 ABI.
162 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
163 (FFI_TYPE_SMALL_STRUCT2): Define.
164 (FFI_TYPE_SMALL_STRUCT4): Likewise.
165 (FFI_TYPE_SMALL_STRUCT8): Likewise.
166 (FFI_TYPE_SMALL_STRUCT3): Redefine.
167 (FFI_TYPE_SMALL_STRUCT5): Likewise.
168 (FFI_TYPE_SMALL_STRUCT6): Likewise.
169 (FFI_TYPE_SMALL_STRUCT7): Likewise.
170 * src/pa/ffi.c (ROUND_DOWN): Delete.
171 (fldw, fstw, fldd, fstd): Use '__asm__'.
172 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
173 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
174 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
175 Simplify incrementing of stack slot variable. Change type of local
177 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
179 (ffi_prep_cif_machdep): Likewise.
180 (ffi_call): Likewise.
181 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
182 return type to ffi_status. Simplify incrementing of stack slot
183 variable. Only copy floating point argument registers when PA_LINUX
184 is true. Reformat debug statement.
185 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
186 FFI_TYPE_SMALL_STRUCT8.
187 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
189 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
190 Add nops to cache flush. Add trampoline for PA_HPUX.
191 * src/pa/hpux32.S: New file.
192 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
193 ffi_prep_args_LINUX to ffi_prep_args_pa32.
194 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
195 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
196 argument type checks so that common argument types appear first.
197 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
198 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
200 2006-03-24 Alan Modra <amodra@bigpond.net.au>
202 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
203 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
204 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
206 (ffi_prep_args64): Assert using IBM extended double.
207 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
208 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
209 (ffi_call): Handle FFI_LINUX.
210 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
211 gpr3 return pointer as for struct return. Handle FFI_LINUX
212 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
214 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
215 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
216 Don't use r6 as pointer to results, instead use sp offset. Don't
217 make a special call to load lr with case table address, instead
218 use offset from previous call.
219 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
220 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
223 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
225 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
226 passed with FP registers correctly.
227 (ffi_closure_helper_SYSV): Likewise.
228 * src/sh64/sysv.S: Likewise.
230 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
232 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
233 args and userdata unused.
234 (closure_test_fn1): Mark cif and userdata unused.
235 (main): Remove unused res.
237 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
239 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
240 -O2, -O3, -Os and the warning flags -W -Wall.
241 * testsuite/libffi.special/special.exp: Likewise.
242 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
243 unused parameter unused for gcc or else do nothing.
244 * testsuite/libffi.special/ffitestcxx.h: Likewise.
245 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
247 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
248 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
249 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
250 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
251 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
252 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
253 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
254 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
255 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
256 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
257 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
258 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
259 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
260 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
261 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
262 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
263 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
264 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
265 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
266 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
267 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
268 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
269 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
270 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
272 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
274 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
276 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
277 void* to avoid compiler warning.
279 (cls_struct_align_gn): Mark cif and userdata unused.
280 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
282 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
284 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
286 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
288 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
290 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
291 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
292 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
294 (main): Cast res_call to silence gcc.
295 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
297 (main): Cast res_call to silence gcc.
298 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
300 (main): Cast res_call to silence gcc.
301 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
303 (main): Cast res_call to silence gcc.
304 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
306 (main): Cast res_call to silence gcc.
307 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
309 (main): Cast res_call to silence gcc.
310 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
312 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
313 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
315 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
316 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
318 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
319 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
321 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
322 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
324 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
325 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
327 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
329 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
330 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
331 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
332 Cleanup white spaces.
333 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
334 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
335 cif and userdata unused.
336 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
338 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
339 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
340 formatters to silence gcc.
341 (B_gn): Mark cif and userdata unused.
342 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
344 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
345 (B_gn): Mark cif and userdata unused.
346 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
348 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
349 (B_gn): Mark cif and userdata unused.
350 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
352 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
353 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
354 * testsuite/libffi.call/problem1.c (stub): Likewise.
355 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
357 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
358 in the last commit for this test case in the test case itself.
359 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
361 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
362 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
363 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
364 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
365 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
366 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
368 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
370 * src/sh/sysv.S: Fix register numbers in the FDE for
373 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
375 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
376 declaration to avoid a false negative on ix86. See PR323.
378 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
380 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
381 and cast integer to void * if needed. Update the pointer to
382 the FP register saved area correctly.
384 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
386 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
388 * testsuite/libffi.call/nested_struct4.c: Likewise.
390 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
392 * testsuite/libffi.call/return_dbl.c: New test case.
393 * testsuite/libffi.call/return_dbl1.c: Likewise.
394 * testsuite/libffi.call/return_dbl2.c: Likewise.
395 * testsuite/libffi.call/return_fl.c: Likewise.
396 * testsuite/libffi.call/return_fl1.c: Likewise.
397 * testsuite/libffi.call/return_fl2.c: Likewise.
398 * testsuite/libffi.call/return_fl3.c: Likewise.
399 * testsuite/libffi.call/closure_fn6.c: Likewise.
401 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
403 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
404 here to be used by other test cases too.
406 * testsuite/libffi.call/nested_struct10.c: New test case.
407 * testsuite/libffi.call/nested_struct9.c: Likewise.
408 * testsuite/libffi.call/nested_struct8.c: Likewise.
409 * testsuite/libffi.call/nested_struct7.c: Likewise.
410 * testsuite/libffi.call/nested_struct6.c: Likewise.
411 * testsuite/libffi.call/nested_struct5.c: Likewise.
412 * testsuite/libffi.call/nested_struct4.c: Likewise.
414 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
416 * configure.ac: Enable libffi for sparc64-*-freebsd*.
417 * configure: Rebuilt.
419 2006-01-18 Jakub Jelinek <jakub@redhat.com>
421 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
422 instead do the shifting inline.
423 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
424 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
425 and 8 byte structs, for the remaining struct sizes don't call
426 __lshrdi3, instead do the shifting inline.
428 2005-12-07 Thiemo Seufer <ths@networkno.de>
430 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
432 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
433 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
434 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
435 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
436 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
438 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
440 (ffi_prep_closure): Improve trampoline instruction scheduling.
441 (ffi_closure_mips_inner_O32): Fix endianness bugs.
443 2005-12-03 Alan Modra <amodra@bigpond.net.au>
445 * src/powerpc/ffi.c: Formatting.
446 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
447 (ffi_prep_args64): Likewise.
449 2005-09-30 Geoffrey Keating <geoffk@apple.com>
451 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
452 darwin, use -shared-libgcc not -lgcc_s, and explain why.
454 2005-09-26 Tom Tromey <tromey@redhat.com>
456 * testsuite/libffi.call/float1.c (value_type): New typedef.
457 (CANARY): New define.
458 (main): Check for result buffer overflow.
459 * src/powerpc/linux64.S: Handle linux64 long double returns.
460 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
461 (ffi_prep_cif_machdep): Handle linux64 long double returns.
463 2005-08-25 Alan Modra <amodra@bigpond.net.au>
466 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
468 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
470 2005-08-11 Jakub Jelinek <jakub@redhat.com>
472 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
473 (AH_BOTTOM): Add FFI_HIDDEN definition.
474 * configure: Rebuilt.
475 * fficonfig.h.in: Rebuilt.
476 * src/powerpc/ffi.c (hidden): Remove.
477 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
478 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
479 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
480 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
481 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
482 add FFI_HIDDEN to its prototype.
483 (ffi_closure_SYSV_inner): New.
484 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
485 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
487 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
490 * configure: Rebuilt.
491 * configure.ac: Handle i*86-*-gnu*.
493 2005-08-09 Jakub Jelinek <jakub@redhat.com>
495 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
496 DW_CFA_offset_extended_sf rather than
497 DW_CFA_GNU_negative_offset_extended.
498 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
500 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
502 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
504 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
505 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
506 partially on register.
507 (ffi_closure_helper_SYSV): Likewise.
508 (ffi_prep_cif_machdep): Don't set too many cif->flags.
510 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
512 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
514 * src/sh64/ffi.c (simple_type): Remove.
515 (return_type): Handle small structures correctly.
516 (ffi_prep_args): Likewise.
517 (ffi_call): Likewise.
518 (ffi_closure_helper_SYSV): Likewise.
519 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
520 Emit position independent code if PIC and remove wrong datalabel
521 prefixes from EH data.
523 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
525 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
526 * Makefile.in: Regenerate.
527 * include/Makefile.in: Likewise.
528 * testsuite/Makefile.in: Likewise.
529 * configure.ac: Add POWERPC_FREEBSD rules.
530 * configure: Regenerate.
531 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
532 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
533 * src/powerpc/ffi.c: Add flags to handle small structure returns
535 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
537 (ffi_closure_helper_SYSV): Likewise.
538 * src/powerpc/ppc_closure.S: Add return types for small structures.
539 * src/powerpc/sysv.S: Add bits to handle small structures for
542 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
544 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
545 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
546 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
548 2005-07-05 Randolph Chung <tausq@debian.org>
550 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
551 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
552 structures. Kill compilation warnings.
553 (ffi_closure_inner_LINUX): Print return values as hex in debug
554 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
555 Properly handle 5-7 byte structure returns.
556 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
557 (FFI_TYPE_SMALL_STRUCT2): Remove.
558 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
559 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
560 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
561 (checksmst1, checksmst2): Remove.
562 (checksmst3): Optimize handling of 3-byte struct returns.
563 (checksmst567): Properly handle 5-7 byte struct returns.
565 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
568 * src/mips/n32.S: Enforce PIC code.
569 * src/mips/o32.S: Likewise.
571 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
573 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
574 * configure: Regenerate.
576 2005-06-01 Alan Modra <amodra@bigpond.net.au>
578 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
579 to call ffi_closure_helper_SYSV. Append @local instead.
580 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
582 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
584 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
585 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
586 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
587 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
588 include/Makefile.in, testsuite/Makefile.in: Regenerate.
590 2005-05-09 Mike Stump <mrs@apple.com>
592 * configure: Regenerate.
594 2005-05-08 Richard Henderson <rth@redhat.com>
597 * src/alpha/osf.S: Update unwind into to match code.
599 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
600 Richard Henderson <rth@redhat.com>
602 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
604 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
605 (ffi_prep_closure): Set carry bit if sse-used flag set.
606 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
607 Only load sse registers if ssecount non-zero.
608 (ffi_closure_unix64): Only save sse registers if carry set on entry.
610 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
612 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
613 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
614 * configure: Regenerate.
616 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
618 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
619 have Tcl8.3-compatible intermediate variable.
621 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
622 Hans-Peter Nilsson <hp@axis.com>
624 * Makefile.am: Add CRIS support.
625 * configure.ac: Likewise.
626 * Makefile.in, configure, testsuite/Makefile.in,
627 include/Makefile.in: Regenerate.
628 * src/cris: New directory.
629 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
630 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
632 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
633 \r?\n in output tests.
635 2005-04-12 Mike Stump <mrs@apple.com>
637 * configure: Regenerate.
639 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
641 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
643 2005-03-30 Steve Ellcey <sje@cup.hp.com>
645 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
647 * src/ia64/unix.S (ffi_closure_unix): Extend gp
648 to 64 bits in ILP32 mode.
649 Load 64 bits even for short data.
651 2005-03-23 Mike Stump <mrs@apple.com>
653 * src/powerpc/darwin.S: Update for -m64 multilib.
654 * src/powerpc/darwin_closure.S: Likewise.
656 2005-03-21 Zack Weinberg <zack@codesourcery.com>
658 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
659 Do not set tool_include_dir.
660 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
662 * include/Makefile.am: Set gcc_version and toollibffidir.
663 * include/Makefile.in: Regenerate.
665 2005-02-22 Andrew Haley <aph@redhat.com>
667 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
668 odd-numbered register pairs for 64-bit integer types.
670 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
673 * testsuite/libffi.call/return_ll1.c: New test case.
675 2005-02-11 Janis Johnson <janis187@us.ibm.com>
677 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
678 * testsuite/libffi.call/float.c: Ditto.
679 * testsuite/libffi.call/float2.c: Ditto.
680 * testsuite/libffi.call/float3.c: Ditto.
682 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
684 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
686 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
688 * testsuite/libffi.special/special.exp (cxx_options): Add
691 2004-12-31 Richard Henderson <rth@redhat.com>
693 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
694 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
695 offset parameters with a type parameter; deduce size and structure
696 alignment. Update all users.
698 2004-12-31 Richard Henderson <rth@redhat.com>
700 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
701 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
702 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
703 into ffi_prep_closure.
704 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
707 2004-12-27 Richard Henderson <rth@redhat.com>
709 * src/x86/unix64.S: Fix typo in unwind info.
711 2004-12-25 Richard Henderson <rth@redhat.com>
713 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
714 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
715 (merge_classes): Check for it.
717 (classify_argument): Pass byte_offset by value; perform all updates
719 (examine_argument): Add classes argument; handle
720 X86_64_COMPLEX_X87_CLASS.
721 (ffi_prep_args): Merge into ...
722 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
723 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
724 (ffi_fill_return_value): Remove.
725 (ffi_prep_closure): Remove dead assert.
726 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
727 Rewrite to use struct register_args instead of va_list. Create
728 flags for handling structure returns.
729 * src/x86/unix64.S: Remove dead strings.
730 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
731 stack frame with ffi_call. Handle structure returns properly.
732 (float2sse, floatfloat2sse, double2sse): Remove.
733 (sse2float, sse2double, sse2floatfloat): Remove.
734 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
735 to handle structure returns properly.
737 2004-12-08 David Edelsohn <edelsohn@gnu.org>
739 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
741 * Makefile.in: Regenerated.
743 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
745 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
746 * configure, aclocal.m4, Makefile.in: Regenerate.
747 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
749 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
751 * configure: Regenerate for libtool change.
753 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
755 * configure: Regenerate for libtool reversion.
757 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
759 * configure: Regenerate for libtool change.
761 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
763 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
765 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
767 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
768 of jal. Use an absolute encoding for the frame information.
770 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
772 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
773 * acinclude.m4: Delete logic for sincludes.
774 * aclocal.m4, Makefile.in, configure: Regenerate.
775 * include/Makefile: Likewise.
776 * testsuite/Makefile: Likewise.
778 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
780 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
781 on a 8-byte boundary.
782 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
784 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
786 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
787 long long values. Round stack allocation to a multiple of 8 bytes
788 for ATPCS compatibility.
789 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
790 names. Handle returning long long types. Add Thumb and interworking
791 support. Improve soft-float code.
793 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
795 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
796 (libffi_exit): New function.
797 (libffi_init): Build the testglue wrapper if needed.
799 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
802 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
804 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
806 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
808 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
810 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
811 * testsuite/libffi.call/float3.c: New test case.
813 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
815 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
816 the function returning a structure pointed with R2.
817 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
818 the structure return value if T bit set. Emit position
819 independent code and EH data if PIC.
821 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
823 * Makefile.am: Add m32r support.
824 * configure.ac: Likewise.
825 * Makefile.in: Regenerate.
826 * confiugre: Regenerate.
827 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
828 (uint64, sint64, double, longdouble)
829 * src/m32r: New directory.
830 * src/m32r/ffi.c: New file.
831 * src/m32r/sysv.S: Likewise.
832 * src/m32r/ffitarget.h: Likewise.
834 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
836 * testsuite/libffi.call/negint.c: New test case.
838 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
841 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
842 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
843 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
846 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
848 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
849 * testsuite/libffi.call/promotion.c: Likewise.
850 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
852 * testsuite/libffi.call/return_sc.c: Likewise.
853 * testsuite/libffi.call/return_uc.c: Likewise.
855 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
857 * src/powerpc/darwin.S: Fix comments and identation.
858 * src/powerpc/darwin_closure.S: Likewise.
860 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
862 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
863 (ffi_prep_args): Handle longdouble arguments.
864 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
866 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
867 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
869 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
870 * src/types.c: Defined longdouble size and alignment for darwin.
872 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
874 * src/powerpc/aix.S: Remove whitespaces.
875 * src/powerpc/aix_closure.S: Likewise.
876 * src/powerpc/asm.h: Likewise.
877 * src/powerpc/ffi.c: Likewise.
878 * src/powerpc/ffitarget.h: Likewise.
879 * src/powerpc/linux64.S: Likewise.
880 * src/powerpc/linux64_closure.S: Likewise.
881 * src/powerpc/ppc_closure.S: Likewise.
882 * src/powerpc/sysv.S: Likewise.
884 2004-08-30 Anthony Green <green@redhat.com>
886 * Makefile.am: Add frv support.
887 * Makefile.in, testsuite/Makefile.in: Rebuilt.
888 * configure.ac: Read configure.host.
889 * configure.in: Read configure.host.
890 * configure.host: New file. frv-elf needs libgloss.
891 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
892 alignment. This is needed to frv and shouldn't harm the others.
893 * include/ffi_common.h (ALIGN_DOWN): New macro.
894 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
896 2004-08-24 David Daney <daney@avtrex.com>
898 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
899 * testsuite/libffi.call/closure_fn1.c: Likewise.
900 * testsuite/libffi.call/closure_fn2.c Likewise.
901 * testsuite/libffi.call/closure_fn3.c: Likewise.
902 * testsuite/libffi.call/closure_fn4.c: Likewise.
903 * testsuite/libffi.call/closure_fn5.c: Likewise.
904 * testsuite/libffi.call/cls_18byte.c: Likewise.
905 * testsuite/libffi.call/cls_19byte.c: Likewise.
906 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
907 * testsuite/libffi.call/cls_20byte.c: Likewise.
908 * testsuite/libffi.call/cls_20byte1.c: Likewise.
909 * testsuite/libffi.call/cls_24byte.c: Likewise.
910 * testsuite/libffi.call/cls_2byte.c: Likewise.
911 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
912 * testsuite/libffi.call/cls_3byte1.c: Likewise.
913 * testsuite/libffi.call/cls_3byte2.c: Likewise.
914 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
915 * testsuite/libffi.call/cls_4byte.c: Likewise.
916 * testsuite/libffi.call/cls_64byte.c: Likewise.
917 * testsuite/libffi.call/cls_6byte.c: Likewise.
918 * testsuite/libffi.call/cls_7byte.c: Likewise.
919 * testsuite/libffi.call/cls_8byte.c: Likewise.
920 * testsuite/libffi.call/cls_9byte1.c: Likewise.
921 * testsuite/libffi.call/cls_9byte2.c: Likewise.
922 * testsuite/libffi.call/cls_align_double.c: Likewise.
923 * testsuite/libffi.call/cls_align_float.c: Likewise.
924 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
925 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
926 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
927 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
928 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
929 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
930 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
931 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
932 * testsuite/libffi.call/cls_double.c: Likewise.
933 * testsuite/libffi.call/cls_float.c: Likewise.
934 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
935 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
936 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
937 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
938 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
939 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
940 * testsuite/libffi.call/cls_schar.c: Likewise.
941 * testsuite/libffi.call/cls_sint.c: Likewise.
942 * testsuite/libffi.call/cls_sshort.c: Likewise.
943 * testsuite/libffi.call/cls_uchar.c: Likewise.
944 * testsuite/libffi.call/cls_uint.c: Likewise.
945 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
946 * testsuite/libffi.call/cls_ushort.c: Likewise.
947 * testsuite/libffi.call/nested_struct.c: Likewise.
948 * testsuite/libffi.call/nested_struct1.c: Likewise.
949 * testsuite/libffi.call/nested_struct2.c: Likewise.
950 * testsuite/libffi.call/nested_struct3.c: Likewise.
951 * testsuite/libffi.call/problem1.c: Likewise.
952 * testsuite/libffi.special/unwindtest.cc: Likewise.
953 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
955 * testsuite/libffi.call/cls_16byte.c: Likewise.
956 * testsuite/libffi.call/cls_5byte.c: Likewise.
958 2004-08-23 David Daney <daney@avtrex.com>
961 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
962 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
963 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
964 parameters and return types.
965 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
966 (ffi_prep_closure): Ditto.
967 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
968 alignment calculations.
969 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
970 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
972 2004-08-14 Casey Marshall <csm@gnu.org>
974 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
975 contain `FFI_TYPE_UINT64' as return type for any 64-bit
976 integer (O32 ABI only).
977 (ffi_prep_closure): new function.
978 (ffi_closure_mips_inner_O32): new function.
979 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
980 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
981 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
982 64 bit integers correctly.
983 (ffi_closure_O32): new function.
984 Added DWARF-2 unwind info for both functions.
986 2004-08-10 Andrew Haley <aph@redhat.com>
988 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
990 2004-08-01 Robert Millan <robertmh@gnu.org>
992 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
993 * configure: Regenerate.
995 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
997 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
998 and mmap() explicitly instead of relying on preset autoconf cache
1000 * aclocal.m4: Regenerate.
1001 * configure: Regenerate.
1003 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1005 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1006 (ffi_check_float_struct): Remove unused prototype.
1008 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1010 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1011 character on Darwin, use '\n\t' instead.
1013 2004-06-26 Matthias Klose <doko@debian.org>
1015 * libtool-version: Fix typo in revision/age.
1017 2004-06-17 Matthias Klose <doko@debian.org>
1019 * libtool-version: New.
1020 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1021 * Makefile.in: Regenerate.
1023 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1025 * Makefile.am: Remove useless multilib rules.
1026 * Makefile.in: Regenerate.
1027 * aclocal.m4: Regenerate with automake 1.8.5.
1028 * configure.ac: Remove useless multilib configury.
1029 * configure: Regenerate.
1031 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1033 * .cvsignore: New file.
1035 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1037 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1039 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1042 2004-06-06 Sean McNeil <sean@mcneil.com>
1044 * configure.ac: Add x86_64-*-freebsd* support.
1045 * configure: Regenerate.
1047 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1050 * configure.ac: Test for existence of mmap and sys/mman.h before
1051 checking blacklist. Fix suggested by Jim Wilson.
1052 * configure: Regenerate.
1054 2004-04-26 Matt Austern <austern@apple.com>
1056 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1058 * src/powerpc/darwin_closure.S: Likewise.
1060 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1062 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1063 error. Reported by Thomas Heller <theller@python.net>.
1064 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1065 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1067 2004-03-20 Matthias Klose <doko@debian.org>
1069 * src/pa/linux.S: Fix typo.
1071 2004-03-19 Matthias Klose <doko@debian.org>
1073 * Makefile.am: Update.
1074 * Makefile.in: Regenerate.
1075 * src/pa/ffi.h.in: Remove.
1076 * src/pa/ffitarget.h: New file.
1078 2004-02-10 Randolph Chung <tausq@debian.org>
1080 * Makefile.am: Add PA support.
1081 * Makefile.in: Regenerate.
1082 * include/Makefile.in: Regenerate.
1083 * configure.ac: Add PA target.
1084 * configure: Regenerate.
1085 * src/pa/ffi.c: New file.
1086 * src/pa/ffi.h.in: Add PA support.
1087 * src/pa/linux.S: New file.
1088 * prep_cif.c: Add PA support.
1090 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1092 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1094 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1095 with ecif->cif->flags.
1096 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1098 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1099 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1100 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1101 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1103 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1105 * configure.in: Rename file to ...
1106 * configure.ac: ... this.
1107 * fficonfig.h.in: Regenerate.
1108 * Makefile.in: Regenerate.
1109 * include/Makefile.in: Regenerate.
1110 * testsuite/Makefile.in: Regenerate.
1112 2004-03-12 Matt Austern <austern@apple.com>
1114 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1115 changes in EH format resulting from addition of linkonce support.
1116 * src/powerpc/darwin_closure.S: Likewise.
1118 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1119 Paolo Bonzini <bonzini@gnu.org>
1121 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1122 Remove VPATH. Remove rules for object files. Remove multilib support.
1123 (AM_CCASFLAGS): Add.
1124 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1125 (AC_PREREQ): Bump version to 2.59.
1126 (AC_INIT): Fill with version info and bug address.
1127 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1128 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1129 De-precious CC so that the right flags are passed down to multilibs.
1130 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1131 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1132 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1133 * configure: Rebuilt.
1134 * aclocal.m4: Likewise.
1135 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1136 * fficonfig.h.in: Likewise.
1138 2004-03-11 Andreas Schwab <schwab@suse.de>
1140 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1141 arguments from fp registers only for the first 8 parameter slots.
1142 Don't convert a float parameter when passed in memory.
1144 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1146 * configure: Regenerate for config/accross.m4 correction.
1148 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1150 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1151 ecif->cif->bytes to bytes.
1152 (ffi_prep_cif_machdep): Add braces around nested if statement.
1154 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1156 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1158 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1159 (ffi_closure_helper_LINUX64): Fix typo.
1160 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1162 * testsuite/libffi.call/float.c: Likewise.
1163 * testsuite/libffi.call/float2.c: Likewise.
1165 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1167 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1168 long double function return and long double arg handling.
1169 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1170 Use "end_pfr" instead of "nf". Correct long double handling.
1172 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1174 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1175 space for long double return value. Adjust stack frame and offsets.
1176 Load f2 long double return.
1178 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1180 * src/types.c: Use 16 byte long double for POWERPC64.
1182 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1184 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1185 when the structure return address is passed in %o0.
1186 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1187 (ffi_v9_layout_struct): Align the field following a nested structure
1188 on a word boundary. Use memmove instead of memcpy.
1189 (ffi_call): Update call to ffi_V9_return_struct.
1190 (ffi_prep_closure): Define 'ctx' only for V8.
1191 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1192 and ffi_closure_sparc_inner_v9.
1193 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1194 Always skip the structure return address. For structures and long
1195 doubles, copy the argument directly.
1196 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1197 if required. Shift the maximum floating-point slot accordingly. For
1198 big structures, copy the argument directly; otherwise, left-justify the
1199 argument and call ffi_v9_layout_struct to lay out the structure on
1201 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1202 (ffi_closure_v8): Pass the structure return address. Update call to
1203 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1204 Skip the 'unimp' insn when returning long doubles and structures.
1205 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1206 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1207 FFI_TYPE_INT handling. Load structures both in integers and
1208 floating-point registers on return.
1209 * README: Update status of the SPARC port.
1211 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1213 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1215 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1217 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1219 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1220 value as of type ffi_arg, not unsigned int.
1222 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1224 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1227 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1229 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1232 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1234 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1237 2003-12-10 Richard Henderson <rth@redhat.com>
1239 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1240 size_t instead of int.
1242 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1244 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1245 * testsuite/libffi.call/many_win32.c (main): Replace variable
1246 int i with unsigned long ul.
1248 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1249 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1250 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1251 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1252 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1253 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1254 * testsuite/libffi.call/cls_align_float.c: Likewise.
1255 * testsuite/libffi.call/cls_align_double.c: Likewise.
1256 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1257 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1259 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1262 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1263 Align arguments to 32 bits.
1265 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1268 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1269 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1270 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1271 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1272 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1273 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1275 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1277 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1279 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1280 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1282 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1284 * Makefile.in: Rebuilt.
1285 * configure: Likewise.
1286 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1289 2003-11-21 Andreas Jaeger <aj@suse.de>
1290 Andreas Tobler <a.tobler@schweiz.ch>
1292 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1293 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1294 * Makefile.in: Rebuilt.
1295 * aclocal.m4: Likewise.
1296 * configure: Likewise.
1297 * fficonfig.h.in: Likewise.
1298 * testsuite/lib/libffi-dg.exp: Add include dir.
1299 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1300 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1301 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1302 for ffi_closure if available.
1303 * testsuite/libffi.call/closure_fn1.c: Likewise.
1304 * testsuite/libffi.call/closure_fn2.c: Likewise.
1305 * testsuite/libffi.call/closure_fn3.c: Likewise.
1306 * testsuite/libffi.call/closure_fn4.c: Likewise.
1307 * testsuite/libffi.call/closure_fn5.c: Likewise.
1308 * testsuite/libffi.call/cls_12byte.c: Likewise.
1309 * testsuite/libffi.call/cls_16byte.c: Likewise.
1310 * testsuite/libffi.call/cls_18byte.c: Likewise.
1311 * testsuite/libffi.call/cls_19byte.c: Likewise.
1312 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1313 * testsuite/libffi.call/cls_20byte.c: Likewise.
1314 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1315 * testsuite/libffi.call/cls_24byte.c: Likewise.
1316 * testsuite/libffi.call/cls_2byte.c: Likewise.
1317 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1318 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1319 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1320 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1321 * testsuite/libffi.call/cls_4byte.c: Likewise.
1322 * testsuite/libffi.call/cls_5byte.c: Likewise.
1323 * testsuite/libffi.call/cls_64byte.c: Likewise.
1324 * testsuite/libffi.call/cls_6byte.c: Likewise.
1325 * testsuite/libffi.call/cls_7byte.c: Likewise.
1326 * testsuite/libffi.call/cls_8byte.c: Likewise.
1327 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1328 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1329 * testsuite/libffi.call/cls_double.c: Likewise.
1330 * testsuite/libffi.call/cls_float.c: Likewise.
1331 * testsuite/libffi.call/cls_schar.c: Likewise.
1332 * testsuite/libffi.call/cls_sint.c: Likewise.
1333 * testsuite/libffi.call/cls_sshort.c: Likewise.
1334 * testsuite/libffi.call/cls_uchar.c: Likewise.
1335 * testsuite/libffi.call/cls_uint.c: Likewise.
1336 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1337 * testsuite/libffi.call/cls_ushort.c: Likewise.
1338 * testsuite/libffi.call/nested_struct.c: Likewise.
1339 * testsuite/libffi.call/nested_struct1.c: Likewise.
1340 * testsuite/libffi.call/nested_struct2.c: Likewise.
1341 * testsuite/libffi.call/nested_struct3.c: Likewise.
1342 * testsuite/libffi.call/problem1.c: Likewise.
1343 * testsuite/libffi.special/unwindtest.cc: Likewise.
1345 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
1347 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
1349 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
1351 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
1352 Add -lgcc_s to additional flags.
1354 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
1356 * configure.in, include/Makefile.am: PR libgcj/11147, install
1357 the ffitarget.h header file in a gcc versioned and target
1359 * configure: Regenerated.
1360 * Makefile.in, include/Makefile.in: Likewise.
1361 * testsuite/Makefile.in: Likewise.
1363 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
1365 * testsuite/libffi.call/closure_fn0.c: Print result and check
1366 with dg-output to make debugging easier.
1367 * testsuite/libffi.call/closure_fn1.c: Likewise.
1368 * testsuite/libffi.call/closure_fn2.c: Likewise.
1369 * testsuite/libffi.call/closure_fn3.c: Likewise.
1370 * testsuite/libffi.call/closure_fn4.c: Likewise.
1371 * testsuite/libffi.call/closure_fn5.c: Likewise.
1372 * testsuite/libffi.call/cls_12byte.c: Likewise.
1373 * testsuite/libffi.call/cls_16byte.c: Likewise.
1374 * testsuite/libffi.call/cls_18byte.c: Likewise.
1375 * testsuite/libffi.call/cls_19byte.c: Likewise.
1376 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1377 * testsuite/libffi.call/cls_20byte.c: Likewise.
1378 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1379 * testsuite/libffi.call/cls_24byte.c: Likewise.
1380 * testsuite/libffi.call/cls_2byte.c: Likewise.
1381 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1382 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1383 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1384 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1385 * testsuite/libffi.call/cls_4byte.c: Likewise.
1386 * testsuite/libffi.call/cls_5byte.c: Likewise.
1387 * testsuite/libffi.call/cls_64byte.c: Likewise.
1388 * testsuite/libffi.call/cls_6byte.c: Likewise.
1389 * testsuite/libffi.call/cls_7byte.c: Likewise.
1390 * testsuite/libffi.call/cls_8byte.c: Likewise.
1391 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1392 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1393 * testsuite/libffi.call/cls_double.c: Likewise.
1394 * testsuite/libffi.call/cls_float.c: Likewise.
1395 * testsuite/libffi.call/cls_schar.c: Likewise.
1396 * testsuite/libffi.call/cls_sint.c: Likewise.
1397 * testsuite/libffi.call/cls_sshort.c: Likewise.
1398 * testsuite/libffi.call/cls_uchar.c: Likewise.
1399 * testsuite/libffi.call/cls_uint.c: Likewise.
1400 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1401 * testsuite/libffi.call/cls_ushort.c: Likewise.
1402 * testsuite/libffi.call/problem1.c: Likewise.
1404 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
1407 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
1409 * testsuite/libffi.call/cls_9byte2.c: New test case.
1410 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1411 * testsuite/libffi.call/cls_64byte.c: Likewise.
1412 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1413 * testsuite/libffi.call/cls_19byte.c: Likewise.
1414 * testsuite/libffi.call/cls_18byte.c: Likewise.
1415 * testsuite/libffi.call/closure_fn4.c: Likewise.
1416 * testsuite/libffi.call/closure_fn5.c: Likewise.
1417 * testsuite/libffi.call/cls_schar.c: Likewise.
1418 * testsuite/libffi.call/cls_sint.c: Likewise.
1419 * testsuite/libffi.call/cls_sshort.c: Likewise.
1420 * testsuite/libffi.call/nested_struct2.c: Likewise.
1421 * testsuite/libffi.call/nested_struct3.c: Likewise.
1423 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
1425 * testsuite/libffi.call/cls_double.c: Do a check on the result.
1426 * testsuite/libffi.call/cls_uchar.c: Likewise.
1427 * testsuite/libffi.call/cls_uint.c: Likewise.
1428 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1429 * testsuite/libffi.call/cls_ushort.c: Likewise.
1430 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
1432 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
1434 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
1437 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
1439 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
1440 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
1442 2003-10-22 David Daney <ddaney@avtrex.com>
1444 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
1445 __attribute__((__mode__(__SI__))) and friends.
1447 2003-10-22 Andreas Schwab <schwab@suse.de>
1449 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
1451 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
1453 * configure.in: AC_LINK_FILES(ffitarget.h).
1454 * configure: Regenerate.
1455 * Makefile.in: Likewise.
1456 * include/Makefile.in: Likewise.
1457 * testsuite/Makefile.in: Likewise.
1458 * fficonfig.h.in: Likewise.
1460 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
1461 Richard Henderson <rth@redhat.com>
1463 Avoid that ffi.h includes fficonfig.h.
1465 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
1466 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
1467 (TARGET_SRC_MIPS_SGI): Removed.
1468 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
1469 (MIPS_SGI): Removed.
1470 (CLEANFILES): Removed.
1471 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
1473 * acconfig.h: Removed.
1474 * configure.in: Compute sizeofs only for double and long double.
1475 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
1476 into AC_DEFINE instead of using acconfig.h. Create
1477 include/ffitarget.h instead of include/fficonfig.h. Rename
1478 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
1479 AC_DEFINE EH_FRAME_FLAGS.
1480 * include/Makefile.am (DISTCLEANFILES): New automake macro.
1481 (hack_DATA): Add ffitarget.h.
1482 * include/ffi.h.in: Remove all system specific definitions.
1483 Declare raw API even if it is not installed, why bother?
1484 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
1485 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
1486 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
1487 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
1488 * include/ffi_common.h (bool): Do not define.
1489 (ffi_assert): Accept failed assertion.
1490 (ffi_type_test): Return void and accept file/line.
1491 (FFI_ASSERT): Pass stringized failed assertion.
1492 (FFI_ASSERT_AT): New macro.
1493 (FFI_ASSERT_VALID_TYPE): New macro.
1494 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
1495 UINT64, SINT64): Define here with gcc's __attribute__ macro
1497 (FLOAT32, ALIGN): Define here instead of in ffi.h
1498 * include/ffi-mips.h: Removed. Its content moved to
1499 src/mips/ffitarget.h after separating assembly and C sections.
1500 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
1501 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
1502 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
1503 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
1504 SIZEOF_ARG -> FFI_SIZEOF_ARG.
1505 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
1506 * src/debug.c (ffi_assert): Accept stringized failed assertion.
1507 (ffi_type_test): Rewritten.
1508 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
1509 FFI_ASSERT_VALID_TYPE.
1510 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
1511 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
1512 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
1513 src/s390/ffitarget.h, src/sh/ffitarget.h,
1514 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
1515 src/x86/ffitarget.h: New files.
1516 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
1517 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
1518 src/powerpc/aix.S, src/powerpc/darwin.S,
1519 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
1520 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
1521 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
1522 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
1523 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
1526 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1528 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
1529 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
1531 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
1533 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
1534 Used when FFI_DEBUG = 1.
1536 2003-10-14 Alan Modra <amodra@bigpond.net.au>
1538 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
1541 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1543 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
1544 FFI_MIPS_O32 for O32 ABI.
1546 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
1548 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
1549 SPARC64. Cleanup whitespaces.
1551 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
1553 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
1554 strongarm, xscale. Cleanup whitespaces.
1555 * testsuite/libffi.call/closure_fn1.c: Likewise.
1556 * testsuite/libffi.call/closure_fn2.c: Likewise.
1557 * testsuite/libffi.call/closure_fn3.c: Likewise.
1558 * testsuite/libffi.call/cls_12byte.c: Likewise.
1559 * testsuite/libffi.call/cls_16byte.c: Likewise.
1560 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1561 * testsuite/libffi.call/cls_20byte.c: Likewise.
1562 * testsuite/libffi.call/cls_24byte.c: Likewise.
1563 * testsuite/libffi.call/cls_2byte.c: Likewise.
1564 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1565 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1566 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1567 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1568 * testsuite/libffi.call/cls_4byte.c: Likewise.
1569 * testsuite/libffi.call/cls_5byte.c: Likewise.
1570 * testsuite/libffi.call/cls_6byte.c: Likewise.
1571 * testsuite/libffi.call/cls_7byte.c: Likewise.
1572 * testsuite/libffi.call/cls_8byte.c: Likewise.
1573 * testsuite/libffi.call/cls_double.c: Likewise.
1574 * testsuite/libffi.call/cls_float.c: Likewise.
1575 * testsuite/libffi.call/cls_uchar.c: Likewise.
1576 * testsuite/libffi.call/cls_uint.c: Likewise.
1577 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1578 * testsuite/libffi.call/cls_ushort.c: Likewise.
1579 * testsuite/libffi.call/nested_struct.c: Likewise.
1580 * testsuite/libffi.call/nested_struct1.c: Likewise.
1581 * testsuite/libffi.call/problem1.c: Likewise.
1582 * testsuite/libffi.special/unwindtest.cc: Likewise.
1583 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
1585 2003-09-18 David Edelsohn <edelsohn@gnu.org>
1587 * src/powerpc/aix.S: Cleanup whitespaces.
1588 * src/powerpc/aix_closure.S: Likewise.
1590 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
1592 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
1593 * src/powerpc/darwin_closure.S: Likewise.
1594 * src/powerpc/ffi_darwin.c: Likewise.
1596 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
1597 David Edelsohn <edelsohn@gnu.org>
1599 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
1600 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
1602 * src/powerpc/darwin_closure.S: Likewise.
1603 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
1604 according to the Darwin/AIX ABI.
1605 (ffi_prep_cif_machdep): Likewise.
1606 (ffi_closure_helper_DARWIN): Likewise.
1607 Remove the outgoing parameter stack logic. Simplify the evaluation
1608 of the different CASE types.
1609 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
1610 statement in the trampoline code.
1612 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
1614 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
1615 for the register size.
1616 (ffi_closure_helper_SYSV): Handle the structure return value
1618 (ffi_closure_helper_SYSV): Return the appropriate type when
1619 the registers are used for the structure return value.
1620 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
1621 the 64-bit return value. Update copyright years.
1623 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1625 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
1626 srcdir for ffi_mips.h.
1628 2003-09-12 Alan Modra <amodra@bigpond.net.au>
1630 * src/prep_cif.c (initialize_aggregate): Include tail padding in
1632 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
1633 placement of float result.
1634 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
1635 cast of "resp" for big-endian 64 bit machines.
1637 2003-09-11 Alan Modra <amodra@bigpond.net.au>
1639 * src/types.c (double, longdouble): Merge identical SH and ARM
1640 typedefs, and add POWERPC64.
1641 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
1642 struct split over gpr and rest.
1643 (ffi_prep_cif_machdep): Correct intarg_count for structures.
1644 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
1646 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
1648 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
1651 2003-09-09 Alan Modra <amodra@bigpond.net.au>
1653 * configure: Regenerate.
1655 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
1657 * Makefile.am: Remove build rules for ffitest.
1658 * Makefile.in: Rebuilt.
1660 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
1662 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
1663 about implicit declaration of abort().
1665 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
1667 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
1668 * Makefile.in: Rebuilt.
1669 * configure.in: Add dejagnu test framework.
1670 * configure: Rebuilt.
1672 * testsuite/Makefile.am: New file.
1673 * testsuite/Makefile.in: Built
1674 * testsuite/lib/libffi-dg.exp: New file.
1675 * testsuite/config/default.exp: Likewise.
1676 * testsuite/libffi.call/call.exp: Likewise.
1677 * testsuite/libffi.call/ffitest.h: Likewise.
1678 * testsuite/libffi.call/closure_fn0.c: Likewise.
1679 * testsuite/libffi.call/closure_fn1.c: Likewise.
1680 * testsuite/libffi.call/closure_fn2.c: Likewise.
1681 * testsuite/libffi.call/closure_fn3.c: Likewise.
1682 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1683 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1684 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1685 * testsuite/libffi.call/cls_2byte.c: Likewise.
1686 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1687 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1688 * testsuite/libffi.call/cls_4byte.c: Likewise.
1689 * testsuite/libffi.call/cls_5byte.c: Likewise.
1690 * testsuite/libffi.call/cls_6byte.c: Likewise.
1691 * testsuite/libffi.call/cls_7byte.c: Likewise.
1692 * testsuite/libffi.call/cls_8byte.c: Likewise.
1693 * testsuite/libffi.call/cls_12byte.c: Likewise.
1694 * testsuite/libffi.call/cls_16byte.c: Likewise.
1695 * testsuite/libffi.call/cls_20byte.c: Likewise.
1696 * testsuite/libffi.call/cls_24byte.c: Likewise.
1697 * testsuite/libffi.call/cls_double.c: Likewise.
1698 * testsuite/libffi.call/cls_float.c: Likewise.
1699 * testsuite/libffi.call/cls_uchar.c: Likewise.
1700 * testsuite/libffi.call/cls_uint.c: Likewise.
1701 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1702 * testsuite/libffi.call/cls_ushort.c: Likewise.
1703 * testsuite/libffi.call/float.c: Likewise.
1704 * testsuite/libffi.call/float1.c: Likewise.
1705 * testsuite/libffi.call/float2.c: Likewise.
1706 * testsuite/libffi.call/many.c: Likewise.
1707 * testsuite/libffi.call/many_win32.c: Likewise.
1708 * testsuite/libffi.call/nested_struct.c: Likewise.
1709 * testsuite/libffi.call/nested_struct1.c: Likewise.
1710 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
1711 * testsuite/libffi.call/problem1.c: Likewise.
1712 * testsuite/libffi.call/promotion.c: Likewise.
1713 * testsuite/libffi.call/return_ll.c: Likewise.
1714 * testsuite/libffi.call/return_sc.c: Likewise.
1715 * testsuite/libffi.call/return_uc.c: Likewise.
1716 * testsuite/libffi.call/strlen.c: Likewise.
1717 * testsuite/libffi.call/strlen_win32.c: Likewise.
1718 * testsuite/libffi.call/struct1.c: Likewise.
1719 * testsuite/libffi.call/struct2.c: Likewise.
1720 * testsuite/libffi.call/struct3.c: Likewise.
1721 * testsuite/libffi.call/struct4.c: Likewise.
1722 * testsuite/libffi.call/struct5.c: Likewise.
1723 * testsuite/libffi.call/struct6.c: Likewise.
1724 * testsuite/libffi.call/struct7.c: Likewise.
1725 * testsuite/libffi.call/struct8.c: Likewise.
1726 * testsuite/libffi.call/struct9.c: Likewise.
1727 * testsuite/libffi.special/special.exp: New file.
1728 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1729 * testsuite/libffi.special/unwindtest.cc: Likewise.
1732 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
1734 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
1737 2003-08-02 Alan Modra <amodra@bigpond.net.au>
1739 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
1741 (ffi_closure_helper_LINUX64): Likewise.
1742 * src/powerpc/linux64.S: Remove code writing to parm save area.
1743 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
1744 address in lr from ffi_closure_helper_LINUX64 call to calculate
1745 table address. Optimize function tail.
1747 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
1749 * src/sparc/ffi.c: Handle all floating point registers.
1750 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
1752 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
1754 * README: Note that libffi is not part of GCC. Update the project
1757 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
1759 * src/powerpc/ppc_closure.S: Include ffi.h.
1761 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1763 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
1764 Use C style comments.
1766 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
1768 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
1769 * Makefile.in: Regenerate.
1770 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
1771 * configure: Regenerate.
1772 * include/ffi.h.in: Add SHmedia support.
1773 * src/sh64/ffi.c: New file.
1774 * src/sh64/sysv.S: New file.
1776 2003-05-16 Jakub Jelinek <jakub@redhat.com>
1778 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
1779 should be read-only.
1780 * configure: Rebuilt.
1781 * fficonfig.h.in: Rebuilt.
1782 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
1783 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
1784 * src/powerpc/linux64.S: Likewise.
1785 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
1786 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
1787 if -fpic/-fPIC/-mrelocatable.
1788 * src/powerpc/powerpc_closure.S: Likewise.
1789 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
1790 #write in .eh_frame flags.
1791 * src/sparc/v9.S: Likewise.
1792 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
1793 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
1794 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
1796 2003-05-07 Jeff Sturm <jsturm@one-point.com>
1798 Fixes PR bootstrap/10656
1799 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
1800 support for .register pseudo-op.
1801 * src/sparc/v8.S: Use it.
1802 * fficonfig.h.in: Rebuilt.
1803 * configure: Rebuilt.
1805 2003-04-18 Jakub Jelinek <jakub@redhat.com>
1807 * include/ffi.h.in (POWERPC64): Define if 64-bit.
1808 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
1809 Make it the default on POWERPC64.
1810 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
1811 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
1812 * configure: Rebuilt.
1813 * src/powerpc/ffi.c (hidden): Define.
1814 (ffi_prep_args_SYSV): Renamed from
1815 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
1816 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
1817 ASM_NEEDS_REGISTERS64): New.
1818 (ffi_prep_args64): New function.
1819 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
1820 (ffi_call): Likewise.
1821 (ffi_prep_closure): Likewise.
1822 (flush_icache): Surround by #ifndef POWERPC64.
1823 (ffi_dblfl): New union type.
1824 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
1825 (ffi_closure_helper_LINUX64): New function.
1826 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
1828 * src/powerpc/sysv.S: Likewise.
1829 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
1830 * src/powerpc/linux64.S: New file.
1831 * src/powerpc/linux64_closure.S: New file.
1832 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
1833 src/powerpc/linux64_closure.S.
1834 (TARGET_SRC_POWERPC): Likewise.
1836 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
1837 closure_test_fn3): Fix result printing on big-endian 64-bit
1839 (main): Print tst2_arg instead of uninitialized tst2_result.
1841 * src/ffitest.c (main): Hide what closure pointer really points to
1844 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
1846 * configure.in (arm-*-netbsdelf*): Add configuration.
1847 (configure): Regenerated.
1849 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
1851 * include/Makefile.in: Regenerate.
1853 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
1855 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
1857 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
1858 Receive closure pointer through parameter, read args using
1859 __builtin_dwarf_cfa.
1860 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
1862 2003-03-12 Andreas Schwab <schwab@suse.de>
1864 * configure.in: Avoid trailing /. in toolexeclibdir.
1865 * configure: Rebuilt.
1867 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
1869 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
1871 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
1873 * libffi/src/powerpc/darwin_closure.S:
1874 Fix alignement bug, allocate 8 bytes for the result.
1875 * libffi/src/powerpc/aix_closure.S:
1877 * libffi/src/powerpc/ffi_darwin.c:
1878 Update stackframe description for aix/darwin_closure.S.
1880 2003-02-06 Jakub Jelinek <jakub@redhat.com>
1882 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
1885 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
1886 Andreas Schwab <schwab@suse.de>
1888 * configure.in: Adjust command to source config-ml.in to account
1889 for changes to the libffi_basedir definition.
1890 (libffi_basedir): Remove ${srcdir} from value and include trailing
1893 * configure: Regenerate.
1895 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
1897 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
1899 2003-01-28 Andrew Haley <aph@redhat.com>
1901 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
1902 * src/x86/ffi64.c (ffi_prep_closure): New.
1903 (ffi_closure_UNIX64_inner): New.
1904 * src/x86/unix64.S (ffi_closure_UNIX64): New.
1906 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
1908 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
1909 Remove USE_LIBDIR conditional.
1910 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
1911 * Makefile.in, configure: Rebuilt.
1913 2003-01027 David Edelsohn <edelsohn@gnu.org>
1915 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
1916 * Makefile.in: Regenerate.
1918 2003-01-22 Andrew Haley <aph@redhat.com>
1920 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
1923 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1925 * src/powerpc/darwin.S: Add unwind info.
1926 * src/powerpc/darwin_closure.S: Likewise.
1928 2003-01-14 Andrew Haley <aph@redhat.com>
1930 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
1931 (ffi_prep_cif_machdep): Likewise.
1932 * src/x86/unix64.S: Add unwind info.
1934 2003-01-14 Andreas Jaeger <aj@suse.de>
1936 * src/ffitest.c (main): Only use ffi_closures if those are
1939 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
1941 * libffi/src/ffitest.c
1942 add closure testcases
1944 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
1946 * libffi/src/powerpc/ffi.c
1947 fix alignment bug for float (4 byte aligned iso 8 byte)
1949 2003-01-09 Geoffrey Keating <geoffk@apple.com>
1951 * src/powerpc/ffi_darwin.c: Remove RCS version string.
1952 * src/powerpc/darwin.S: Remove RCS version string.
1954 2003-01-03 Jeff Sturm <jsturm@one-point.com>
1956 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
1957 * src/ffitest.c (main): Use static storage for closure.
1958 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
1959 * src/sparc/v8.S (ffi_closure_v8): New.
1960 * src/sparc/v9.S (ffi_closure_v9): New.
1962 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
1964 * include/ffi.h.in: Added FFI_STDCALL ffi_type
1965 enumeration for X86_WIN32.
1966 * src/x86/win32.S: Added ffi_call_STDCALL function
1968 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
1969 switch cases for recognising FFI_STDCALL and
1970 calling ffi_call_STDCALL if target is X86_WIN32.
1971 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
1972 stdcall versions of the "my_strlen" and "many"
1973 test functions (for X86_WIN32).
1974 Added test cases to test stdcall invocation using
1977 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
1979 * src/sh/sysv.S: Add DWARF2 unwind info.
1981 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
1983 * src/s390/sysv.S (.eh_frame section): Make section read-only.
1985 2002-11-26 Jim Wilson <wilson@redhat.com>
1987 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
1989 2002-11-23 H.J. Lu <hjl@gnu.org>
1991 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
1992 Include ../config/accross.m4.
1993 * aclocal.m4; Rebuild.
1994 * configure: Likewise.
1996 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
1998 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2000 2002-11-11 DJ Delorie <dj@redhat.com>
2002 * configure.in: Look for common files in the right place.
2004 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2006 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2007 raw data as _Jv_word values, not ffi_raw.
2008 (ffi_java_ptrarray_to_raw): Likewise.
2009 (ffi_java_rvalue_to_raw): New function.
2010 (ffi_java_raw_call): Call it.
2011 (ffi_java_raw_to_rvalue): New function.
2012 (ffi_java_translate_args): Call it.
2013 * src/ffitest.c (closure_test_fn): Interpret return value
2014 as ffi_arg, not int.
2015 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2016 FFI_TYPE_POINTER case.
2017 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2018 values extended to word size.
2020 2002-10-02 Andreas Jaeger <aj@suse.de>
2022 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2024 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2026 * include/ffi.h.in: Fix i386 win32 compilation.
2028 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2030 * configure.in: Add s390x-*-linux-* target.
2031 * configure: Regenerate.
2032 * include/ffi.h.in: Define S390X for s390x targets.
2033 (FFI_CLOSURES): Define for s390/s390x.
2034 (FFI_TRAMPOLINE_SIZE): Likewise.
2035 (FFI_NATIVE_RAW_API): Likewise.
2036 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2037 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2038 * src/s390/ffi.c: Major rework of existing code. Add support for
2039 s390x targets. Add closure support.
2040 * src/s390/sysv.S: Likewise.
2042 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2044 * src/arm/sysv.S: Fix typo.
2046 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2048 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2049 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2050 (ffi_call_SYSV): Handle soft-float.
2052 2002-09-27 Bo Thorsen <bo@suse.de>
2054 * include/ffi.h.in: Fix multilib x86-64 support.
2056 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2058 * Makefile.am (all-multi): Fix multilib parallel build.
2060 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2062 * configure.in (sh[34]*-*-linux*): Add brackets.
2063 * configure: Regenerate.
2065 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2067 * Makefile.am: Add SH support.
2068 * Makefile.in: Regenerate.
2069 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2070 * configure: Regenerate.
2071 * include/ffi.h.in: Add SH support.
2072 * src/sh/ffi.c: New file.
2073 * src/sh/sysv.S: New file.
2074 * src/types.c: Add SH support.
2076 2002-07-16 Bo Thorsen <bo@suse.de>
2078 * src/x86/ffi64.c: New file that adds x86-64 support.
2079 * src/x86/unix64.S: New file that handles argument setup for
2081 * src/x86/sysv.S: Don't use this on x86-64.
2082 * src/x86/ffi.c: Don't use this on x86-64.
2084 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2086 * src/ffitest.c (struct6): New test that tests a special case in
2088 (struct7): Likewise.
2089 (struct8): Likewise.
2090 (struct9): Likewise.
2091 (closure_test_fn): Silence warning about this when it's not used.
2092 (main): Add the new tests.
2093 (main): Fix a couple of wrong casts and silence some compiler warnings.
2094 * include/ffi.h.in: Add x86-64 ABI definition.
2095 * fficonfig.h.in: Regenerate.
2096 * Makefile.am: Add x86-64 support.
2097 * configure.in: Likewise.
2098 * Makefile.in: Regenerate.
2099 * configure: Likewise.
2101 2002-06-24 Bo Thorsen <bo@suse.de>
2103 * src/types.c: Merge settings for similar architectures.
2104 Add x86-64 sizes and alignments.
2106 2002-06-23 Bo Thorsen <bo@suse.de>
2108 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2109 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2110 * src/mips/ffi.c (ffi_prep_args): Likewise.
2111 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2113 2002-07-18 H.J. Lu (hjl@gnu.org)
2115 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2116 (libffi_la_SOURCES): Support MIPS_LINUX.
2117 (libffi_convenience_la_SOURCES): Likewise.
2118 * Makefile.in: Regenerated.
2120 * configure.in (mips64*-*): Skip.
2121 (mips*-*-linux*): New.
2122 * configure: Regenerated.
2124 * src/mips/ffi.c: Include <sgidefs.h>.
2126 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2128 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2130 2002-05-27 Roger Sayle <roger@eyesopen.com>
2132 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2134 2002-05-27 Bo Thorsen <bo@suse.de>
2136 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2139 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2141 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2142 beginning of function (for older apple cc).
2144 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2146 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2147 script entry, and set LD to it when configuring multilibs.
2148 * configure: Rebuilt.
2150 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2152 * configure.in (sparc64-*-netbsd*): Add target.
2153 (sparc-*-netbsdelf*): Likewise.
2154 * configure: Regenerate.
2156 2002-04-28 David S. Miller <davem@redhat.com>
2158 * configure.in, configure: Fix SPARC test in previous change.
2160 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2162 * Makefile.am: Add Linux for S/390 support.
2163 * Makefile.in: Regenerate.
2164 * configure.in: Add Linux for S/390 support.
2165 * configure: Regenerate.
2166 * include/ffi.h.in: Add Linux for S/390 support.
2167 * src/s390/ffi.c: New file from libffi CVS tree.
2168 * src/s390/sysv.S: New file from libffi CVS tree.
2170 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2172 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2174 * src/sparc/v8.S: Use it.
2175 * src/sparc/v9.S: Likewise.
2176 * fficonfig.h.in: Rebuilt.
2177 * configure: Rebuilt.
2179 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2181 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2183 * src/ia64/unix.S: Add unwind information. Fix comments.
2184 Save sp in a way that's compatible with unwind info.
2185 (ffi_call_unix): Correctly restore sp in all cases.
2186 * src/ia64/ffi.c: Add, fix comments.
2188 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2190 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2192 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2194 * configure.in (alpha*-*-netbsd*): Add target.
2195 * configure: Regenerate.
2197 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2199 * src/sparc/v8.S: Add unwind info.
2200 * src/sparc/v9.S: Likewise.
2202 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2204 * configure.in: Enable i*86-*-netbsdelf*.
2205 * configure: Rebuilt.
2207 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2210 * src/mips/n32.s: Delete
2211 * src/mips/o32.s: Delete
2213 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2215 * configure.in: Enable alpha*-*-freebsd*.
2216 * configure: Rebuilt.
2218 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2220 * Makefile.am: libfficonvenience -> libffi_convenience.
2221 * Makefile.in: Rebuilt.
2223 * Makefile.am: Define ffitest_OBJECTS.
2224 * Makefile.in: Rebuilt.
2226 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2227 David Edelsohn <edelsohn@gnu.org>
2229 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2230 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2231 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2232 * Makefile.in: Regenerate.
2233 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2234 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2235 (flush_icache, flush_range): New functions.
2236 (ffi_closure_helper_DARWIN): New function.
2237 * src/powerpc/aix_closure.S: New file.
2238 * src/powerpc/darwin_closure.S: New file.
2240 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2242 * include/ffi.h.in: Add typedef for ffi_arg.
2243 * src/ffitest.c (main): Declare rint with ffi_arg.
2245 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2247 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2248 number of GPRs for floating-point arguments.
2250 2002-01-31 Anthony Green <green@redhat.com>
2252 * configure: Rebuilt.
2253 * configure.in: Replace CHECK_SIZEOF and endian tests with
2254 cross-compiler friendly macros.
2255 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2258 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2260 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2261 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2263 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2265 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2266 (TARGET_SRC_POWERPC_AIX): New.
2267 (POWERPC_AIX): New stanza.
2268 * Makefile.in: Regenerate.
2269 * configure.in: Add AIX case.
2270 * configure: Regenerate.
2271 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2272 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2273 size. Fix "long double" support.
2274 (ffi_call): Add FFI_AIX case.
2275 * src/powerpc/aix.S: New.
2277 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2279 Implement Darwin PowerPC ABI.
2280 * configure.in: Handle powerpc-*-darwin*.
2281 * Makefile.am: Set source files for POWERPC_DARWIN.
2282 * configure: Rebuilt.
2283 * Makefile.in: Rebuilt.
2284 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2286 * src/powerpc/darwin.S: New file.
2287 * src/powerpc/ffi_darwin.c: New file.
2289 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2291 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2293 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2295 * src/x86/sysv.S: Avoid gas-only .balign directive.
2296 Use C style comments.
2298 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2300 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2301 Fixes PR bootstrap/3563.
2303 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2305 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2307 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2309 * configure.in: Recognize sparc*-sun-* host.
2310 * configure: Regenerate.
2312 2001-06-06 Andrew Haley <aph@redhat.com>
2314 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
2316 2001-06-03 Andrew Haley <aph@redhat.com>
2318 * src/alpha/osf.S: Add unwind info.
2319 * src/powerpc/sysv.S: Add unwind info.
2320 * src/powerpc/ppc_closure.S: Likewise.
2322 2000-05-31 Jeff Sturm <jsturm@one-point.com>
2324 * configure.in: Fix AC_ARG_ENABLE usage.
2325 * configure: Rebuilt.
2327 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2329 * configure.in: Remove warning about beta code.
2330 * configure: Rebuilt.
2332 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
2334 * src/ia64/unix.S: Restore stack pointer when returning from
2336 * src/ia64/ffi.c: Fix typo in comment.
2338 2001-04-18 Jim Wilson <wilson@redhat.com>
2340 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
2341 to eliminate RAW DV.
2343 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
2345 * Makefile.am: Make a libtool convenience library.
2346 * Makefile.in: Rebuilt.
2348 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
2350 * configure.in: Use different syntax for subdirectory creation.
2351 * configure: Rebuilt.
2353 2001-03-27 Jon Beniston <jon@beniston.com>
2355 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
2356 * configure: Rebuilt.
2357 * Makefile.am: Added X86_WIN32 target support.
2358 * Makefile.in: Rebuilt.
2360 * include/ffi.h.in: Added X86_WIN32 target support.
2362 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
2363 * src/types.c: Added X86_WIN32 target support.
2365 * src/x86/win32.S: New file. Based on sysv.S, but with EH
2366 stuff removed and made to work with CygWin's gas.
2368 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
2370 * configure.in: Make target subdirectory in build dir.
2371 * Makefile.am: Override suffix based rules to specify correct output
2373 * Makefile.in: Rebuilt.
2374 * configure: Rebuilt.
2376 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
2378 * src/powerpc/ppc_closure.S: New file.
2379 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
2380 involving long long and register pairs.
2381 (ffi_prep_closure): New function.
2382 (flush_icache): Likewise.
2383 (ffi_closure_helper_SYSV): Likewise.
2384 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
2385 (FFI_TRAMPOLINE_SIZE): Likewise.
2386 (FFI_NATIVE_RAW_API): Likewise.
2387 * Makefile.in: Rebuilt.
2388 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
2389 (TARGET_SRC_POWERPC): Likewise.
2391 2001-03-19 Tom Tromey <tromey@redhat.com>
2393 * Makefile.in: Rebuilt.
2394 * Makefile.am (ffitest_LDFLAGS): New macro.
2396 2001-03-02 Nick Clifton <nickc@redhat.com>
2398 * include/ffi.h.in: Remove RCS ident string.
2399 * include/ffi_mips.h: Remove RCS ident string.
2400 * src/debug.c: Remove RCS ident string.
2401 * src/ffitest.c: Remove RCS ident string.
2402 * src/prep_cif.c: Remove RCS ident string.
2403 * src/types.c: Remove RCS ident string.
2404 * src/alpha/ffi.c: Remove RCS ident string.
2405 * src/alpha/osf.S: Remove RCS ident string.
2406 * src/arm/ffi.c: Remove RCS ident string.
2407 * src/arm/sysv.S: Remove RCS ident string.
2408 * src/mips/ffi.c: Remove RCS ident string.
2409 * src/mips/n32.S: Remove RCS ident string.
2410 * src/mips/o32.S: Remove RCS ident string.
2411 * src/sparc/ffi.c: Remove RCS ident string.
2412 * src/sparc/v8.S: Remove RCS ident string.
2413 * src/sparc/v9.S: Remove RCS ident string.
2414 * src/x86/ffi.c: Remove RCS ident string.
2415 * src/x86/sysv.S: Remove RCS ident string.
2417 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
2419 * include/ffi.h.in: Change sourceware.cygnus.com references to
2422 2000-12-09 Richard Henderson <rth@redhat.com>
2424 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
2425 (ffi_closure_osf_inner): Index rather than increment avalue
2426 and arg_types. Give ffi_closure_osf the raw return value type.
2427 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
2430 2000-12-07 Richard Henderson <rth@redhat.com>
2432 * src/raw_api.c (ffi_translate_args): Fix typo.
2433 (ffi_prep_closure): Likewise.
2435 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
2436 FFI_TRAMPOLINE_SIZE.
2437 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
2438 cif->bytes for new ffi_call_osf implementation.
2439 (ffi_prep_args): Absorb into ...
2440 (ffi_call): ... here. Do all stack allocation here and
2441 avoid a callback function.
2442 (ffi_prep_closure, ffi_closure_osf_inner): New.
2443 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
2444 (ffi_closure_osf): New.
2446 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
2448 * config.guess, config.sub, install-sh: Removed.
2449 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
2450 * Makefile.in: Rebuilt.
2452 * acinclude.m4: Include libtool macros from the top level.
2453 * aclocal.m4, configure: Rebuilt.
2455 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
2457 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
2458 * configure: Rebuilt.
2460 2000-05-11 Scott Bambrough <scottb@netwinder.org>
2462 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
2463 memory correctly. Use conditional instructions, not branches where
2466 2000-05-04 Tom Tromey <tromey@cygnus.com>
2468 * configure: Rebuilt.
2469 * configure.in: Match `arm*-*-linux-*'.
2470 From Chris Dornan <cdornan@arm.com>.
2472 2000-04-28 Jakub Jelinek <jakub@redhat.com>
2474 * Makefile.am (SUBDIRS): Define.
2475 (AM_MAKEFLAGS): Likewise.
2476 (Multilib support.): Add section.
2477 * Makefile.in: Rebuilt.
2478 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
2479 New variables. Set for gcc using -print-multi-lib. Export them
2481 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
2482 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
2484 (extra_compiler_flags_value, extra_compiler_flags): Check these
2485 for extra compiler options which need to be passed down in
2488 2000-04-16 Anthony Green <green@redhat.com>
2490 * configure: Rebuilt.
2491 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
2493 2000-04-14 Jakub Jelinek <jakub@redhat.com>
2495 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
2496 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
2497 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
2498 Replace all void * sizeofs with sizeof(int).
2499 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
2500 different than DOUBLE.
2501 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
2502 (ffi_prep_args_v9): New function.
2503 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
2504 (ffi_V9_return_struct): New function.
2505 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
2506 32bit code (not yet cross-arch calls).
2507 * src/sparc/v8.S: Add struct return delay nop.
2509 * src/sparc/v9.S: New file.
2510 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
2511 is used on sparc64 only for structures larger than 32 bytes.
2512 Pass by reference for structures is done for structure arguments
2513 larger than 16 bytes.
2514 * src/ffitest.c (main): Use 64bit rint on sparc64.
2515 Run long long tests on sparc.
2516 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
2518 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
2520 * configure.in (sparc-*-linux*): New supported target.
2521 (sparc64-*-linux*): Likewise.
2522 * configure: Rebuilt.
2523 * Makefile.am: Add v9.S to SPARC files.
2524 * Makefile.in: Likewise.
2525 (LINK): Surround $(CCLD) into double quotes, so that multilib
2526 compiles work correctly.
2528 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
2530 * configure: Rebuilt.
2531 * configure.in: (i*86-*-solaris*): New libffi target. Patch
2532 proposed by Bryce McKinlay.
2534 2000-03-20 Tom Tromey <tromey@cygnus.com>
2536 * Makefile.in: Hand edit for java_raw_api.lo.
2538 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
2540 * config.guess, config.sub: Update from the gcc tree.
2541 Fix for PR libgcj/168.
2543 2000-03-03 Tom Tromey <tromey@cygnus.com>
2545 * Makefile.in: Fixed ia64 by hand.
2547 * configure: Rebuilt.
2548 * configure.in (--enable-multilib): New option.
2549 (libffi_basedir): New subst.
2550 (AC_OUTPUT): Added multilib code.
2552 2000-03-02 Tom Tromey <tromey@cygnus.com>
2554 * Makefile.in: Rebuilt.
2555 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
2558 2000-02-25 Hans Boehm <boehm@acm.org>
2560 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
2562 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
2564 (ffi_prep_raw_closure): Use ffi_translate_args, not
2565 ffi_closure_translate.
2566 * src/java_raw_api.c: New file.
2567 * src/ffitest.c (closure_test_fn): New function.
2568 (main): Define `rint' as long long on IA64. Added new test when
2569 FFI_CLOSURES is defined.
2570 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
2571 (ffi_abi): Recognize IA64.
2572 (ffi_raw): Added `flt' field.
2573 Added "Java raw API" code.
2574 * configure.in: Recognize ia64.
2575 * Makefile.am (TARGET_SRC_IA64): New macro.
2576 (libffi_la_common_SOURCES): Added java_raw_api.c.
2577 (libffi_la_SOURCES): Define in IA64 case.
2579 2000-01-04 Tom Tromey <tromey@cygnus.com>
2581 * Makefile.in: Rebuilt with newer automake.
2583 1999-12-31 Tom Tromey <tromey@cygnus.com>
2585 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
2587 1999-09-01 Tom Tromey <tromey@cygnus.com>
2589 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
2591 * fficonfig.h.in: Rebuilt.
2592 * configure: Rebuilt.
2593 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
2594 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
2595 * acconfig.h: Don't #undef PACKAGE or VERSION.
2597 1999-08-09 Anthony Green <green@cygnus.com>
2599 * include/ffi.h.in: Try to work around messy header problem
2600 with PACKAGE and VERSION.
2602 * configure: Rebuilt.
2603 * configure.in: Change version to 2.00-beta.
2605 * fficonfig.h.in: Rebuilt.
2606 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
2608 * src/x86/ffi.c (ffi_raw_call): Rename.
2610 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
2612 * src/x86/ffi.c (ffi_closure_SYSV): New function.
2613 (ffi_prep_incoming_args_SYSV): Ditto.
2614 (ffi_prep_closure): Ditto.
2615 (ffi_closure_raw_SYSV): Ditto.
2616 (ffi_prep_raw_closure): More ditto.
2617 (ffi_call_raw): Final ditto.
2619 * include/ffi.h.in: Add definitions for closure and raw API.
2621 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
2624 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
2626 * src/raw_api.c: New file.
2628 * include/ffi.h.in (ffi_raw): New type.
2629 (UINT_ARG, SINT_ARG): New defines.
2630 (ffi_closure, ffi_raw_closure): New types.
2631 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
2633 * configure.in: Add check for endianness and sizeof void*.
2635 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
2636 instead of directly.
2638 * configure: Rebuilt.
2640 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
2642 * configure.in: Add x86 and powerpc BeOS configurations.
2643 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
2645 1999-05-09 Anthony Green <green@cygnus.com>
2647 * configure.in: Add warning about this being beta code.
2648 Remove src/Makefile.am from the picture.
2649 * configure: Rebuilt.
2651 * Makefile.am: Move logic from src/Makefile.am. Add changes
2652 to support libffi as a target library.
2653 * Makefile.in: Rebuilt.
2655 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
2656 Upgraded to new autoconf, automake, libtool.
2660 * LICENSE: Update copyright date.
2662 * src/Makefile.am, src/Makefile.in: Removed.
2664 1998-11-29 Anthony Green <green@cygnus.com>
2666 * include/ChangeLog: Removed.
2667 * src/ChangeLog: Removed.
2668 * src/mips/ChangeLog: Removed.
2669 * src/sparc/ChangeLog: Remboved.
2670 * src/x86/ChangeLog: Removed.
2672 * ChangeLog.v1: Created.