Lauri Kasanen
05b73276c2
Update test description
2017-05-20 09:53:30 +03:00
Lauri Kasanen
e8a735492d
Correct comment style
2017-05-19 15:38:50 +03:00
Lauri Kasanen
901ac80026
Wrapper functions use all registers
2017-05-19 15:25:50 +03:00
Lauri Kasanen
80e6afd335
Add two new trampoline test cases
2017-05-19 14:34:10 +03:00
Oliver Schmidt
11b01b908d
Merge pull request #424 from clbr/trampoline
...
Trampoline support
2017-05-18 16:29:34 +02:00
Lauri Kasanen
219b239a34
Adjustments in response to latest comments
2017-05-18 17:14:26 +03:00
Lauri Kasanen
4c05c46cd1
Remove RAM-trampoline array support
2017-05-18 16:11:39 +03:00
Lauri Kasanen
f6fa74a636
Rename trampoline to wrappedcall everywhere
2017-05-18 16:00:04 +03:00
Lauri Kasanen
7feced8862
More tabs to spaces
2017-05-18 15:54:47 +03:00
Lauri Kasanen
a590b99f5b
Align with spaces instead of tabs here
2017-05-18 14:21:43 +03:00
Oliver Schmidt
390ee4e22f
Merge pull request #429 from pfusik/opt-stdfunc
...
Optimize the inlined strlen/memcpy/memset
2017-05-17 22:58:54 +02:00
Oliver Schmidt
73f1c0e11d
Migrate 'encrypted variables' variables to 'repository settings'.
...
The GitHub token used for GitHub Pages deployment was revoked (see https://blog.travis-ci.com/2017-05-08-security-advisory ) so I took the opportunity to make use of the "new" repository settings feature instead of fiddling with variable encryption again.
2017-05-17 20:56:21 +02:00
Oliver Schmidt
f151142e6a
Keep GCC from complaining about 'List' may be used uninitialized in this function.
2017-05-17 19:39:02 +02:00
Oliver Schmidt
298dda0e71
Increased GCC optimization level.
2017-05-17 18:30:35 +02:00
Piotr Fusik
675dd3c963
Optimize the inlined memcpy/memset for the sizes of 128/129.
2017-05-17 10:08:08 +02:00
Piotr Fusik
6011bdb2f6
Optimize the inlined strlen.
2017-05-17 10:00:35 +02:00
Lauri Kasanen
ac7a0e21ac
Correct missed doc label
2017-05-16 20:14:19 +03:00
Lauri Kasanen
ca41af41c2
Rename pragma trampoline to wrapped-call, value to identifier
2017-05-16 20:10:24 +03:00
Oliver Schmidt
0291c92ffb
Merge pull request #425 from clbr/arrayaccess
...
Add fast paths for char postinc/dec
2017-05-09 19:26:41 +02:00
Lauri Kasanen
077b14ccfa
test/val/postincdec: Test for the final val value
2017-05-08 10:19:26 +03:00
Lauri Kasanen
7adcde1707
Add explicit postinc/dec testcase
2017-05-07 20:35:49 +03:00
Lauri Kasanen
a2f61c667a
Update comments and location checks according to Greg
2017-05-07 20:31:41 +03:00
Lauri Kasanen
ead0338c0d
Add fast path for char postdec
2017-05-06 13:35:39 +03:00
Lauri Kasanen
767f093ff8
Add fast path for char postinc
2017-05-06 13:35:39 +03:00
Lauri Kasanen
2890b3a810
Pragma trampoline
2017-05-02 20:51:53 +03:00
Lauri Kasanen
9c86c03a96
Add test case
2017-05-02 20:43:41 +03:00
Lauri Kasanen
d091a57e91
Trampoline stack
2017-05-02 20:43:41 +03:00
Lauri Kasanen
679bfb0ae9
Add IntPtrStack
2017-05-01 21:08:04 +03:00
Lauri Kasanen
a84463a3df
callptr4
2017-05-01 21:08:03 +03:00
Lauri Kasanen
82e62209d6
Document pragma trampoline
2017-05-01 21:08:03 +03:00
Oliver Schmidt
18b0aaf9f2
Removed help for non-existent option.
2017-04-15 23:12:19 +02:00
Oliver Schmidt
9b243ff68c
Merge pull request #418 from greg-king5/while-nothing
...
Fix the handling of the C construct "while (0) { }".
2017-04-07 16:07:38 +02:00
Greg King
1aab287189
Fixed the handling of "while (0) {}".
...
It's a corner case; but, conditional macroes might create it -- better safe than sorry.
2017-04-07 09:26:58 -04:00
Oliver Schmidt
e43dbe1c24
Just some white space fixes of recently contributed code.
2017-04-06 17:53:57 +02:00
Oliver Schmidt
be772c0198
Bumped version.
...
As suggested the incompatibility resulting from 02daf9f8b5
2017-04-06 12:22:39 +02:00
Oliver Schmidt
5d129f6e8a
Merge pull request #417 from greg-king5/sim-interrupts
...
sim65: Fix the simulations of the stack pointer and the "break" and decimal flags.
2017-04-05 10:57:47 +02:00
Greg King
dce5783dc9
Fixed the simulations of the stack pointer and the "break" and decimal-mode flags.
...
* The pointer wraps around the stack page.
* The break flag exists on only the stack, and only after an interrupt.
* 65C02 interrupts clear the decimal-mode flag.
2017-04-04 13:23:44 -04:00
Oliver Schmidt
e806515541
Removed workaround.
2017-04-04 15:53:14 +02:00
Oliver Schmidt
93193727be
Removed workaround.
2017-04-04 15:52:37 +02:00
Oliver Schmidt
cd27fef6a8
Removed workaround.
2017-04-04 15:52:01 +02:00
Oliver Schmidt
3789174162
Removed workaround.
2017-04-04 15:51:00 +02:00
Oliver Schmidt
14268510f5
Removed workaround.
2017-04-04 15:45:24 +02:00
Oliver Schmidt
275b27647e
Removed superfluous final linefeed.
2017-04-04 15:43:56 +02:00
Oliver Schmidt
233dd5062a
Merge pull request #389 from IrgendwerA8/stringimprovements
...
Optimization of string functions (size & speed).
2017-04-04 15:39:37 +02:00
Oliver Schmidt
4e856546c1
Minor language update.
2017-04-04 13:42:57 +02:00
Greg King
3b17c3a411
Fixed 2 typoes.
...
Warning messages from nsgmls usually mean a typing mistake a few lines above the first line that's named by nsgmls.
2017-04-04 07:23:05 -04:00
Oliver Schmidt
9b34509956
Use the more explicit tag style for recently added internal links.
2017-04-04 00:41:40 +02:00
Oliver Schmidt
02daf9f8b5
So far the built-in inlining of several known standard function was always (!) enabled and the option -Os enabled additional, potentially unsafe inlining of some of those functions.
...
There were two aspects of this behavior that were considered undesirable:
- Although the safe inlining is in general desirable it should only be enabled if asked for it - like any other optimization.
- The option name -Os implies that it is a safe option, the potentially unsafe inlining should have a more explicit name.
So now:
- The option -Os enables the safe inlining.
- The new option --eagerly-inline-funcs enables the potentially unsafe inlining (including the safe inlining).
Additionally was added:
- The option --inline-stdfuncs that does like -Os enable the safe inlining but doesn't enable optimizations.
- The pragma inline-stdfuncs that works identical to --inline-stdfuncs.
- The pragma allow-eager-inline that enables the potentially unsafe inlining but doesn't include the safe inlining. That means that by itself it only marks code as safe for potentially unsafe inlining but doesn't actually enable any inlining.
2017-04-03 23:20:26 +02:00
Oliver Schmidt
cbb2833c01
Merge pull request #414 from IrgendwerA8/SED_Fix
...
Fixed sim65 SED
2017-03-30 12:26:37 +02:00
IrgendwerA8
170d96898f
Fixed SED
2017-03-30 12:17:29 +02:00
