Activity for Uffe Jakobsen
-
Uffe Jakobsen modified a comment on ticket #2065
Have you tried this: x64sc.exe --help 2>clihelp.txt AND/OR x64sc.exe --help >clihelp.txt 2>&1 AND/OR x64sc.exe --help 2>&1 | more
-
Have you tried this: x64sc.exe --help 2>clihelp.txt OR x64sc.exe --help >clihelp.txt 2>&1
-
Lorenz: nextdisk logic fixes
-
Yes I did test that - it still runs The old logic masks out bit 1 - that would be equivalent with $35 versus $37 to address $01
-
trap tests: use rom macros - this fixes the following trap tests for xplus4: trap 1, 2, 3, 9, 11. 13
-
not using the original kernal perhaps? No - I was using the wrong basic test program from bug #2052 :-D
-
Strange - pasting the basic program into x64sc gives me multiple crosses no matter what model I specify... I guess I must have missed some important detail here :-)
-
Thanks :-) Is there a reason that the corrections for subdir src/common is left out ? I guess that you could have committed all the src/*.sand forgot the src/common Anyway the corrections missing is attached to this entry mvh/Uffe :-)
-
It should already be rebased rev 45236 (the commit you made some hours ago) Let me know is there is something wrong :-)
-
Trim trailing whitespaces from Lorenz Test Suite sources
-
@gpz you are right the FIXME comments was added by you about 2 months ago in commit rev 45166 - so I guess they can be removed
-
Fix compile warning with define/macro TED_RASTER_CYCLE
-
yes but that is another patch - right now we are discussing a possible clean up of comments as a tail-fix of the already applied patch
-
Thanks :-) BTW: just above the TED_LINE_START_CLK in tedtypes.h there are two FIXME: notes - they refer to the two functions that my patch has added type casts into. But I have no idea if this is what the FIXME statement refers to - or something else ? In other words should these FIXME be deleted or ? /* FIXME: assigned to (CLOCK)ted.raster_irq_clk in ted-irq.c:ted_irq_set_raster_line() */ /* FIXME: assigned to (CLOCK)ted.raster_irq_clk in ted-mem.c:ted1c1d_store() */ #define TED_LINE_START_CLK(clk)...
-
Just for the record I did a build of latest trunk with clang - that is without this patch And the warnings you refer to are still there - they are not a result of this patch... I'll be happy to try and fix them - but I think that it should be in a separate patch
-
I have lots of warnings from other parts of the source - just not the ones that you have :-) This is the warnings I have with GCC-14.1.1 (on archlinux) https://pastebin.com/bJwGGGHx Now I switched to clang-18.1.8 (on archlinux) Now warnings that GCC gave are now gone - but I see the warnings that you reported. I'll look into them :-)
-
patch rev3 uploaded - let me know if anything is missing/wrong /Uffe :-)
-
Well, technically it is possible to copy the commit message and add it as a SVN message - but I got the point :-D
-
Well, technocally it is possible to copy the commit message and add it as a SVN message - but I got the point :-D
-
@Thomas Winkler: The easiest test is just to run this commandline: xplus4 -warp -keybuf "10 print ti: goto 10\nrun\n" If the TI output counter passes 144896 and continues without freeze - then the issue is solved :-)
-
Thanks for the feedback Attached is a revised patch - it may not be complete as I'm not sure if @gpz wants the comment/ explanation in the commit message (as it is now) - or if it should be duplicated four times in the patch - one for every change - or something else... Comments are welcome :-)
-
Thanks for the feedback Attached is a revised patch - if may not be complete as I'm not sure if @gpz wants the comment/ explanation in the commit message (as it is now) - or if it should be duplicated four times in the patch - one for every change - or something else... Comments are welcome :-)
-
did you try changing the type of cycles_per_line to int instead (less casts are better)? ted.cycles_per_line is already an int (declared in tedtypes.h) Implicit type conversion is really the issue here The issue is that the (existing) logic of the calculation of (from ted-irq.c) (line - current_line) needs to be able to become negative (as line may not be larger than current_line in all cases) Due to Implicit type conversion I cannot see a way without adding an explicit (int) cast That result - multiplied...
-
FYI: I just created a patch ticket containing a fix for this issue Patch #377 Fix for bug ticket #2030: Plus/4 emulator hangs after a while https://sourceforge.net/p/vice-emu/patches/377/
-
Fix for bug ticket #2030: Plus/4 emulator hangs after a while
-
As far as I can see it is closer to 2^31 - indicating a signed int (int) - but that is all details - still it is damn hard to find... :-)
-
I know - but now the bisection is done ;-)
-
I'm using this command below for quick reproduction of the problem It always freezes on TI=144898 (and a few times on TI=144896) - with timer ticks around every 1/60 sec - that is a little over 40 minuttes (so it is "nice" that the problem can be reproduced in warp-mode) xplus4 -warp -keybuf "10 print ti: goto 10\nrun\n" It seems that commit rev 40432 has introduced the problem: https://sourceforge.net/p/vice-emu/code/40432/ CLOCK has changed from uint32_t to uint64_t. Clkguard has been removed,...
-
Damn - I messed up the formatting and forgot to set the category to "bugfix - Sorry
-
Lorenz TestSuite: fix "brkn" test for Plus/4 and C16
-
Improve Lorenz TestSuite Makefiles
-
Hi Spiro, Thanks for your reply. To lower the confusion I think that this PR should be closed I'll try to write Peter again BR/Uffe :-)
-
This PR is now found on github https://github.com/OpenCBM/OpenCBM/pull/113 I believe that we can close this issue/case
-
Hi Spiro, Thanks for your fast reply. I've just made a github PR for this fix. And thanks for explaining - I actually thought that it was the other way around... Because "nibtools" on github was read-only I also believed that "opencbm" was also read-only on github - and that sf.net was the origin of development :-) BR/Uffe :-)
-
Compile fix for modern header check/guards - nibtools will not compile without defining _DEFAULT_SOURCE
-
Compile fix for: grep: warning: stray \ before #
-
c1541 report 35 tracks when given a 40 tracks d64 file
-
Now x64 works as expected. The newly changed code is also more logical compared to the old approach Thanks :-)
-
x64 initializes (clears) memory on first manual reset
-
New patch - handle sterling symbol - replaces previous patch
-
Improve Danish GTK3 symbolic keymap for C64
-
Hahaha :-D thanks a lot for your time :-) Yes that is right - the patch fixes various problems around bindist - I think that you've mentioned the two most important ones :-) /Uffe :-)
-
Any news on accepting/applying this patch ? It is a simple oneliner change that fixes the number of arguments to a script :-)
-
Fix make-bindist_win32.sh arguments for win32/win64 SDL bindistzip target
-
Patch for dollar mapping on Danish GTK symbolic keymaps
-
SDL symbolic keymap for danish keybard
-
Hmm For now it seems that this change is also needed to prevent an undefined ... error As you propably can guess I'm building without network I would have That should have been taken care of at another level... I'll have to look at this later diff --git src/arch/msdos/archdep.c src/arch/msdos/archdep.c index f35c05604..de5bf7cfc 100644 --- src/arch/msdos/archdep.c +++ src/arch/msdos/archdep.c @@ -412,7 +412,9 @@ int archdep_rename(const char oldpath, const char newpath) void archdep_shutdown(void)...
-
Hi, Thanks for your reply. This typo seems to be the only real change needed diff --git src/arch/msdos/uisampler.c src/arch/msdos/uisampler.c index abb6cf564..0166798dc 100644 --- src/arch/msdos/uisampler.c +++ src/arch/msdos/uisampler.c @@ -27,7 +27,7 @@ #include "vice.h" #include <stdio.h> -#include <string.H> +#include <string.h> #include "lib.h" #include "mouse.h"
-
Patch for djgpp/dos cross compile builds on linux
-
Sorry - patch was empty - my bad
-
Patch for out-of-tree cross compile builds (gentranslate_h.sh)
-
Linux: Fix libimgcopy build warnings
-
Linux: newer usb.h does not have USB_LE16_TO_CPU() macro anymore
-
PATCH: x64 (SDL) prevent coredump when pressing menu key in -console mode
-
PATCH: xcbm2 (SDL) core dump when entering "Settings management" menu
-
ioutil.c functions does not handle filesystem links right
-
Patch for remote monitor binary memdump command decoding error
-
I've come across this problem too - seen on both FreeBSD and Linux - both 64bit I've...
-
Damn, I believe that I managed to get the milestone wrong - should have been v2.4.x...
-
Patch for vice (x64) coredumping when displaying drive registers in monitor
1