Mercurial > pub > dyncall > dyncall
annotate ToDo @ 364:3bdd326dc269
- Todo entries
- some clarifications in code about GNU thiscalls (which are effectively idential to cdecl calls)
* took away own mode identifier, as aliasing to cdecl now
- cleanups
author | Tassilo Philipp |
---|---|
date | Tue, 14 Apr 2020 19:35:30 +0200 |
parents | 5ac53b2d5f1e |
children | ad5f9803f52f |
rev | line source |
---|---|
355 | 1 1.2: |
0 | 2 ---- |
364 | 3 - test atpcs thumb on arm and verify if ellipsis and C++ thiscalls work, if so adapt doc appendix |
4 - test eabi thumb on arm and verify if ellipsis and C++ thiscalls work, if so adapt doc appendix | |
5 - new platform matrix | |
6 * check dark green cells and validate in general | |
355 | 7 - openbsd-4.0-* test/resolve_self fails, but unsure why. dlopen correctly returns RTLD_DEFAULT, which |
8 should lookup via dlsym the function in the object itself, but cannot. So... some compiler flag might | |
9 be missing to keep symbol names, or some prefixing happens | |
10 - dyncall/dyncall_call*h could probably be all removed, as forward declarations in corresponding .c | |
11 files would be enough; no headers are needed b/c those functions declared in there are of no public use | |
12 - test/thunk win/x64 doesn't output anything decent for stack test (neither sigsegv nor output), which can create a problem for test log file as it'll mess with the subsequent test's output) | |
13 - remove test/gen-masm alltogether | |
329 | 14 - test code: add sparc tests for when out of register windows, meaning multiple layers deep in the call stack |
286 | 15 - ./configure on macOS does not pass CFLAGS etc.. to Makefile.config |
256
7520e2260097
- added to configure some detection on how to compile .s/.S with --noexecstack (or similar), as compilers' defaults are often insane
Tassilo Philipp
parents:
250
diff
changeset
|
16 - support platforms without shared libraries so they can be built by just typing `./configure;make` (e.g. minix < 3.2.1) |
7520e2260097
- added to configure some detection on how to compile .s/.S with --noexecstack (or similar), as compilers' defaults are often insane
Tassilo Philipp
parents:
250
diff
changeset
|
17 * also make sure minix with shared library support builds and works in general (>= 3.2.1) as well as ELF on Minix (>= 3.2.0) to begin with (as that's the base, and we only test on 3.1.8 so far) |
201 | 18 - dyncall supports calls to functions where prototype exists (with and without varargs); however not when no |
19 prototype exist (e.g. just calling some pointer), without any declaration... even possible in C, as one has | |
20 to cast to the function type? need to support? | |
21 | |
364 | 22 - dyncall sparc v7/v8: -xO3 on SunPro segfaults in tests (observed years ago) |
165
572aff021627
- file name/layout cleanup, removed "-att" from x64 .S filenames, as unnecessary and also misleading
cslag
parents:
159
diff
changeset
|
23 |
155 | 24 - dyncallback's "stack_cleanup" can be removed from non x86 impls... test if they still work |
165
572aff021627
- file name/layout cleanup, removed "-att" from x64 .S filenames, as unnecessary and also misleading
cslag
parents:
159
diff
changeset
|
25 - target a minimum c version, for use with -pedantic, etc.? Right now we use e.g. extensions |
329 | 26 for anon structs, etc. -> we do now target c99 |
86 | 27 - the CallVM-free functions are per VTable, however the latter can be changed out |
28 at runtime... which basically changes potentially the deallocator... | |
29 same goes for functions setting the mode. think about how to make this safer - currently | |
30 we have a higher flexibility and we need to potentially clean up from the mode we are in, | |
31 however, this is open to misuse | |
78 | 32 - check all vararg calling conv implementations if they promote correctly |
33 (e.g. float to double); we used to not care as it was the user's job to | |
34 think of that, but we allow users to use ArgFloat, etc. on "..." now, | |
35 taking care of the promotion; so it should be consistent all across the | |
36 implementations -> also write testcode specifically for that | |
0 | 37 - autovar was never really pushed all the way through and still plays second |
9 | 38 fiddle to dyncall_macros.h, clean this all up, eventually (Tassilo: for now, I |
39 removed out of sync files to avoid confusion; but still probably duped code) | |
69 | 40 - add portable "make install" |
41 * make install should also install manpages | |
0 | 42 - mailing list announcements (html email?) |
43 - fix for NDS: | |
69 | 44 * ARM: |
104
dbca6763f2be
- complete, working mips o32 callback (using hardware fp); fixes error from last commit, which ignored first 2 float args
cslag
parents:
100
diff
changeset
|
45 - callback_plain retval wrong (not only platform), area on stack where |
dbca6763f2be
- complete, working mips o32 callback (using hardware fp); fixes error from last commit, which ignored first 2 float args
cslag
parents:
100
diff
changeset
|
46 it's stored is probably not zeroed, so half-words, etc., come with garbage |
69 | 47 * THUMB: |
48 - ellipsis (might be test itself, not respecting fixed part of args) | |
104
dbca6763f2be
- complete, working mips o32 callback (using hardware fp); fixes error from last commit, which ignored first 2 float args
cslag
parents:
100
diff
changeset
|
49 - callback_plain retval, see above under ARM |
69 | 50 - callf |
24 | 51 - support Solaris11 CFLAGS with -D_FILE_OFFSET_BITS=64 |
256
7520e2260097
- added to configure some detection on how to compile .s/.S with --noexecstack (or similar), as compilers' defaults are often insane
Tassilo Philipp
parents:
250
diff
changeset
|
52 - make sure selinux works (esp. regarding NX bits) |
54 | 53 Daniel has a patch for it, if needed, implementing allocwx with two memory blocks, one for w^x, mmaped to one for r|x |
68 | 54 - support /SAFESEH on cl/win32 |
83
54930a037e8a
- PPC64 single-precision float fixes for more than 13 float args (thanks Masanori!)
cslag
parents:
78
diff
changeset
|
55 - test code functions should be in .so files, optionally, so the suites can directly |
54930a037e8a
- PPC64 single-precision float fixes for more than 13 float args (thanks Masanori!)
cslag
parents:
78
diff
changeset
|
56 be used to test dycnall bindings |
54930a037e8a
- PPC64 single-precision float fixes for more than 13 float args (thanks Masanori!)
cslag
parents:
78
diff
changeset
|
57 - pkg-config support? |
90 | 58 - quadmath support (long double) |
329 | 59 - add tests using alloca(), as this might depend on the frame pointer being correctly set (we don't |
60 check for this, currently, at all - but might not affect dyncall's stubs, anyways) - verify | |
0 | 61 |
62 portasm: | |
63 -------- | |
64 - add solaris support for x64 | |
65 | |
66 dyncall: | |
67 -------- | |
364 | 68 - callbacks and use of CC signature characters (only x86 has any support for different CCs for callbacks): |
69 * support for arm (arm/thumb on same platform) | |
70 * check if ellipsis support is needed (is there any need for calling back into an ellipse function?) | |
71 * check if syscalls support is needed (is there any need for calling back into a syscall?) | |
72 - think about using 'inline' for all/most dyncall_api.c functions? | |
73 - support for where syscalls originate from on OpenBSD: http://undeadly.org/cgi?action=article;sid=20191202105849 | |
288 | 74 - merge arm32_thumb_{gas/apple} and include thumb support for portasm |
75 - x64 verification: return values are passed via RAX and RDX, and XMM0 and XMM1. | |
146
0ab08c1541f0
- fixed armhf ellipsis calls, were ignoring spill area
cslag
parents:
145
diff
changeset
|
76 - find big endian arm box (also hardfloat), e.g. cubietruck: http://blog.netbsd.org/tnf/entry/working_arm_multiprocessor_support |
0ab08c1541f0
- fixed armhf ellipsis calls, were ignoring spill area
cslag
parents:
145
diff
changeset
|
77 - test if armhf/thumb is actually working... pulls in arm mode code in dyncall_callvm.c |
0ab08c1541f0
- fixed armhf ellipsis calls, were ignoring spill area
cslag
parents:
145
diff
changeset
|
78 - dyncall_call_mips_n32.h and dyncall_call_mips_n64.h are pretty much the same, share code |
0 | 79 - support for return values: aggregate return values |
80 - support for argument values: structures, half-precision, vector types, long double | |
288 | 81 * make sure that struct support for ellipsis calls are not forgotten (copy everything by value) |
0 | 82 - consider automatic type-promotion for arguments passed through DC_CALL_C_ELLIPSIS_VARARGS; |
83 this would make it easier to just pass arguments instead of having to know about the C | |
278 | 84 type promotions (this is partly done, see todo-item above under 1.1) |
0 | 85 - other syscalls |
135 | 86 - test MIPS32 eabi big endian (current port works on psp, which is little endian) |
132
b0cca984c740
- manual update, started splitting platform overview in EP and EL for some platforms
cslag
parents:
129
diff
changeset
|
87 - implement MIPS64 N32 (gcc -mabi=n32); both, little and big-endian (looks like NetNBSD on and |
b0cca984c740
- manual update, started splitting platform overview in EP and EL for some platforms
cslag
parents:
129
diff
changeset
|
88 EdgeRouter lite uses this: https://blog.netbsd.org/tnf/entry/hands_on_experience_with_edgerouter) |
306
9f2aff560fa7
- todo update reflecting mips softfloat status more precisely (as only lacking for mips64/n32 callconv)
Tassilo Philipp
parents:
295
diff
changeset
|
89 - MIPS64 n32 softfloat support (-msoft-float) |
129 | 90 - implement MIPS64 EABI (gcc -mabi=eabi); both, little and big-endian |
141 | 91 - support for Minix/arm |
146
0ab08c1541f0
- fixed armhf ellipsis calls, were ignoring spill area
cslag
parents:
145
diff
changeset
|
92 - improve ellipsis test: |
0ab08c1541f0
- fixed armhf ellipsis calls, were ignoring spill area
cslag
parents:
145
diff
changeset
|
93 * test not just with one fixed arg, to also simulate alignment problems |
0ab08c1541f0
- fixed armhf ellipsis calls, were ignoring spill area
cslag
parents:
145
diff
changeset
|
94 * test return values |
0ab08c1541f0
- fixed armhf ellipsis calls, were ignoring spill area
cslag
parents:
145
diff
changeset
|
95 * convert from c++ to c, to have it also on plan9, and more portability |
0 | 96 |
97 dynload: | |
98 -------- | |
244 | 99 - dlGetLibraryPath: dl_iterate_phdr() used on OpenBSD, but not available on version < 3.7 |
0 | 100 - bug: test/nm crashes on qemu mips linux o32 debian etch |
101 - rename SymsInit to InitSyms and similar for cleanup function to reflect naming | |
102 in dyncall | |
103 - get rid of Dl_info and friends (_GNU_SOURCE) in dynload/dynload_syms_elf.c | |
104 for implementation of dlSymsNameFromValue(..) | |
105 - check if PSP can support dynload (missing elf.h, dlfcn.h, etc.); fixup readme if successful | |
84 | 106 - some functionality like java's System.mapLibraryName('name') might be helpful? |
288 | 107 - Failed On Open Solaris 11: |
108 In file included from dynload_syms.c:28: | |
109 dynload_syms_elf.c:56: error: syntax error before "Elf_Dyn" | |
110 dynload_syms_elf.c:56: warning: data definition has no type or storage class | |
111 dynload_syms_elf.c: In function `dlSymsInit': | |
112 dynload_syms_elf.c:155: error: `pDyn' undeclared (first use in this function) | |
113 dynload_syms_elf.c:155: error: (Each undeclared identifier is reported only once | |
114 dynload_syms_elf.c:155: error: for each function it appears in.) | |
115 dynload_syms_elf.c:155: error: syntax error before ')' token | |
116 dynload_syms_elf.c:160: error: `DT_NULL' undeclared (first use in this function) | |
117 dynload_syms_elf.c:163: error: `DT_STRTAB' undeclared (first use in this function) | |
118 dynload_syms_elf.c:164: error: `DT_SYMTAB' undeclared (first use in this function) | |
119 dynload_syms_elf.c:165: error: `DT_HASH' undeclared (first use in this function) | |
120 make: *** [dynload_syms.o] Error 1 | |
0 | 121 |
122 dyncallback: | |
123 ------------ | |
104
dbca6763f2be
- complete, working mips o32 callback (using hardware fp); fixes error from last commit, which ignored first 2 float args
cslag
parents:
100
diff
changeset
|
124 - callback_plain's return value not correct anymore on NDS (maybe just broken testcode?), |
278 | 125 see above under 1.1 items |
104
dbca6763f2be
- complete, working mips o32 callback (using hardware fp); fixes error from last commit, which ignored first 2 float args
cslag
parents:
100
diff
changeset
|
126 * check other platforms also, if asm stub initializes retval space, correctly |
135 | 127 - test MIPS32 eabi big endian (current port works on psp, which is little endian) |
132
b0cca984c740
- manual update, started splitting platform overview in EP and EL for some platforms
cslag
parents:
129
diff
changeset
|
128 - implement MIPS64 N32 (gcc -mabi=n32); both, little and big-endian (looks like NetNBSD on and |
b0cca984c740
- manual update, started splitting platform overview in EP and EL for some platforms
cslag
parents:
129
diff
changeset
|
129 EdgeRouter lite uses this: https://blog.netbsd.org/tnf/entry/hands_on_experience_with_edgerouter) |
306
9f2aff560fa7
- todo update reflecting mips softfloat status more precisely (as only lacking for mips64/n32 callconv)
Tassilo Philipp
parents:
295
diff
changeset
|
130 - MIPS64 n32 softfloat support (-msoft-float) |
129 | 131 - implement MIPS64 EABI (gcc -mabi=eabi); both, little and big-endian |
141 | 132 - support for Minix/arm |
0 | 133 |
134 bindings: | |
135 --------- | |
136 - release bindings as standalone packages (already done for rbdc as a gem and rdyncall on cran) | |
137 * add note to documentation, where bindings are and that some bindings are svn-only ATM | |
138 - add rdoc documentation to ruby gem | |
139 - add pydoc for python | |
140 - add manpage for shdc | |
141 - expose callf-convenience functions to go | |
142 - add godoc for godc | |
143 - add javadoc to java | |
144 - add more bindings | |
145 - add callback support for erlang, go, java, lua, python and ruby (and shell? how?) | |
146 - add call-conv-mode-support for rbdc through signature | |
147 | |
148 manual: | |
149 ------- | |
201 | 150 - introduce doc overview tables and common vocab for calling convention specifics, following example of table on page 13 of https://www.linux-mips.org/pub/linux/mips/doc/ABI/MIPS-N32-ABI-Handbook.pdf; also as inspiration, e.g. user 'preserved' instead of 'permanent', etc. https://developer.apple.com/library/content/documentation/Xcode/Conceptual/iPhoneOSABIReference/Articles/ARMv6FunctionCallingConventions.html#//apple_ref/doc/uid/TP40009021-SW4 |
329 | 151 * also, get inspiration for naming for GCC predefs as talked about in https://stffrdhrn.github.io/software/embedded/openrisc/2018/06/08/gcc_stack_frames.html |
201 | 152 - make stack layout diagrams better, e.g. add grow direction (example: https://developer.apple.com/library/content/documentation/Xcode/Conceptual/iPhoneOSABIReference/art/arm_stack.jpg on https://developer.apple.com/library/content/documentation/Xcode/Conceptual/iPhoneOSABIReference/Articles/ARMv6FunctionCallingConventions.html#//apple_ref/doc/uid/TP40009021-SW1) |
332 | 153 * another example, chapter 3 of: ftp://www.sourceware.org/pub/binutils/ppc-docs/ppc-poweropen/ |
0 | 154 - provide more examples on how to use dyncall, dynload and dyncallback |
155 * e.g. enhance manual with a couple of examples (e.g. calling MessageBoxA on windows, etc.) | |
156 - manuals for bindings (as stated in bindings-section, above) | |
117 | 157 - under "dyncall support", in each calling convention appendix, list crucial details of e.g. big/little endian support, etc. |
158 - specify integer models in use, use common ILP32, LLP64, etc. terms | |
0 | 159 |
88 | 160 regressions: |
161 ------------ | |
0 | 162 - Haiku/gmake build doesn't build dynload due to the fact, that elf.h is not found |
163 on default installations (one has to install the system headers) | |
164 * add to doc | |
165 - Solaris/gmake build doesn't build dynload and tests anymore (maybe not even dyncallback) | |
166 - look over code and fix TODO or @@@ marks | |
167 - add ppc-support to doc (both, 32 and 64, as well as little and big endian remarks) | |
168 | |
169 nice to have: | |
170 ------------- | |
171 - update: microsoft visual c++ build files - integrate all test suites in solution file | |
172 - consistency: on x64 platforms: make both ABIs available ('win64','sysv') for all x64 OS ports | |
173 - microsoft build enhancement: autodetect assembly via C preprocessor (like with .S gcc files) | |
174 - test/thunk: some platforms do not allow for allocation of executable code on heap and stack -- | |
329 | 175 currently this test will segfault (handled), maybe we should collect information which platforms allow and which not. |
176 - PSP | |
177 - automate ppsspp stuff, seems to print to stdout for callback_suite? but not for call_suite.. (?) | |
178 - fix link with dummy versions of the following for test-bins: | |
179 void __stack_chk_fail(){} | |
180 void __stack_chk_guard(){} | |
0 | 181 |
182 bugs: | |
183 ----- | |
88 | 184 - setting callconv, specifically for platforms that have only one (default)) excepts on some |
6 | 185 platforms, eg. setting X64_sysv on amd64/Linux - keeping cdecl which is the same in |
88 | 186 this case, works (this should work now...) |
0 | 187 - vararg function calls don't work for ARM THUMB mode (Daniel's new interface for modesetting for varag |
188 args should be used); adapt test/ellipsis and/or retest | |
189 | |
190 missing implementations (descending priority): | |
191 ---------------------------------------------- | |
192 - riscv | |
193 - itanium | |
88 | 194 - superh/j-core |
0 | 195 - 68k, 88k, 6502, etc. |
88 | 196 - arc |
0 | 197 - alpha |
198 - cell | |
199 - blackfin | |
88 | 200 - seaforth |
21 | 201 - s/390 |
0 | 202 - z80 |
203 - hppa | |
204 - vax | |
205 - atmel avr | |
206 - propeller | |
207 - amd29k | |
208 |