... Woow! Lot's of patches! ... A few (cosmetic) comments only. I'm not in a position to really challenge Mac hackers :-) ... I see no definition for...
Geert Uytterhoeven
geert@...
Oct 4, 2004 7:30 pm
7055
... I may be wrong for pointing this out, but according to discussion elsewhere of some bugs brought about by GCC 3.4.x compilers, "inline static" is invalid...
Eric Shattow
shadow@...
Oct 4, 2004 9:56 pm
7056
I've got a mystic mac. This happens when you take the chassis of a Color Classic I and swap into it the logic board from an LC575. There are plans to do a...
Eric Shattow
shadow@...
Oct 4, 2004 10:36 pm
7057
... The code for enable/disable of nubus interrupt sources is pretty hackish for VIA style nubus interrupt routing. Last I knew, it wasn't guaranteed to work...
Brad Boyer
flar@...
Oct 5, 2004 12:58 am
7058
... Yes... the SONIC patch could probably be shorter, but I got carried away once I compared the existing driver with the chip's actual capabilities, and with...
Finn Thain
ft01@...
Oct 5, 2004 5:48 am
7059
Le mar 05/10/2004 à 00:36, Eric Shattow a écrit : [...] ... Do you think we need some modifications in EMILE ? What is really the problem: why linux cannot...
Laurent Vivier
LaurentVivier@...
Oct 5, 2004 6:26 am
7060
... I'm assuming that "inline static" is the same as "static inline"... gcc 3.4.1 compiles without complaint whether I use "static inline" or "extern inline". ...
Finn Thain
ft01@...
Oct 5, 2004 6:44 am
7061
... What bothered me about the "hackish" enable/disable of nubus interrupt sources is that wiring outputs together assumes that these pins on the VIA or...
Finn Thain
ft01@...
Oct 5, 2004 7:36 am
7062
... You mean dma_alloc_coherent()? Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@... In personal...
Geert Uytterhoeven
geert@...
Oct 5, 2004 12:47 pm
7063
... I agree that it's a mess. I seem to recall that the reports went this way: One type of behavior was observed on boxes with a genuine VIA 6522 chip for...
Brad Boyer
flar@...
Oct 6, 2004 6:10 am
7064
EAT WHAT YOU WANT! LOSE WEIGHT SAFE AND FAST! NO EXERCISE - NO DANGEROUS DIETS - NO DRUGS! 100% Natural and Maximum Strength Fat Binder!!! Fatblast X-treme is...
40KilosLighter
jenna.russell3464@...
Oct 22, 2004 5:57 am
7065
Hey guys! I don't want to be a nuisance, but I have to tell you that I finally got my webcam repaired! So now I am on full view again like usually! Also the...
TureAngel
jenna.russell3464@...
Oct 24, 2004 11:03 am
7066
... That's not necessary true anymore. Try EMILE (@sourceforge.net). It's a floppy based bootloader for Linux/m68k. Not all machine's are supported yet, but...
Erik C.J. Laan
elaan@...
Oct 26, 2004 8:00 am
7067
... I guess the CPU in my LC475 is free of this bug, since the in-kernel FPU emulation works fine with 2.6.8.1 (though it seems a lot slower than the full...
Finn Thain
ft01@...
Oct 26, 2004 11:17 am
7068
... Indeed. Then the kernel can abort and print a message if it's incompatible with buggy LC040s. Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's...
Geert Uytterhoeven
geert@...
Oct 26, 2004 11:21 am
7069
... Actually, only a few models have a real DMA engine. The AV models have the PSC doing DMA for the NCR53c94 chip, which has been partially documented, but...
Brad Boyer
flar@...
Oct 26, 2004 10:11 pm
7070
... Well, I don't know enough about the memory management code to write it, but a basic outline of a test that should work would be as follows: create a page...
Brad Boyer
flar@...
Oct 26, 2004 10:26 pm
7071
... For working around this bug, the page fault handler should check whether (a) the page fault happened during an instruction fetch for the first word of the ...
Geert Uytterhoeven
geert@...
Oct 27, 2004 8:41 am
7072
... Yes, that is my understanding. Perhaps this is enough for someone who knows that part of the code to actually put in a workaround for the bug. I don't know...
Brad Boyer
flar@...
Oct 27, 2004 4:33 pm
7073
... Ah, that's it, sorry I wasn't specific enough. I meant PDMA (Pseudo-DMA right?). This makes the 5380 based Mac's much more enjoyable compared to the 53c9x...
Erik C.J. Laan
elaan@...
Oct 29, 2004 2:47 pm
7074
... Yes, that's true. However, not all 5380 models have support in the current driver. In particular, I know the IIfx is still restricted to PIO mode. The real...
Brad Boyer
flar@...
Oct 29, 2004 4:42 pm
7075
... Which takes us back to the start of the thread again, where Rick Genter asked (though on debian-68k only, so you might have missed that): ".... I'd like to...
Erik C.J. Laan
elaan@...
Oct 29, 2004 6:13 pm
7076
... Not at all. ... - what's lacking in LC040 support (supporting a page fault the fetch of a floating point instruction that itself will generate an...
Rick Genter
rgenter@...
Oct 29, 2004 8:05 pm
7077
... Fixing the handling of real FPU instructions would be better. That way you don't crash stuff if you accidentally run a regular 68k binary on your box....
Brad Boyer
flar@...
Oct 29, 2004 8:58 pm
7078
... Has anybody thought of "fixing" the compiler with an optimization option to get arround the bug? From what I understand, all you would need to do is see if...
root
root@...
Oct 29, 2004 9:20 pm
7079
... I wonder if there is a performance benefit to either solution. My suspicion is there is not, assuming the emulation code is equal. (Though the linux kernel...
Finn Thain
ft01@...
Oct 30, 2004 12:58 am
7080
... That might fix the problem without a performance hit, but it's not as simple as it sounds. It would have to be implemented in the linker since that's where...
Brad Boyer
flar@...
Oct 30, 2004 6:31 am
7081
... If the emulation code is exactly the same quality, the soft float will beat out the trapped instructions every time. An unimplimented instruction trap is...
Brad Boyer
flar@...
Oct 30, 2004 6:41 am
7082
... Indeed. It executes FPU instructions to save/restore the FPU state. Please add CONFIG_FPU (and set it to N) to arch/m68k/Kconfig and try ...
Geert Uytterhoeven
geert@...
Oct 30, 2004 9:03 am
7083
... Doesn't glibc always initialize the FPU state, unless compiled with -msoft-float? Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's lots of Linux...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
