mdacon driver updates
I've posted a small fix to mdacon.c at the end of last year (while Marcello was in vacation). It has'nt made it to the kernel yet, nor was there any feedback. I've thought, thats the way like open source works...
So, here we go again (against 2.4.17; my MDA is detected, but not initialized, see my last posting):
video/mdacon.c Sun Dec 30 21:36:50 2001 renamed from video/mdacon.c.orig Sun Dec 30 02:44:25 2001 +1 -0
@@ -24,6 +24,7 @@ * * Changelog: * Paul G. (03/2001) Fix mdacon= boot prompt to use __setup(). + * 20011230 [email protected] - consider non-Hercules MDA compatible */
#include <linux/types.h>
@@ -291,6 +292,10 @@
break;
}
}
+ else { /* consider non-Hercules as Hercules-compatible */
+ mda_type = TYPE_HERC;
+ mda_type_name = "Hercules compatible (hopefully)";+ }
return 1;
}
@@ -342,9 +347,8 @@
return NULL;
}- if (mda_type != TYPE_MDA) {
- mda_initialize();
- } + /* at this point, we found an MDA */ + mda_initialize();
/* cursor looks ugly during boot-up, so turn it off */
mda_set_cursor(mda_vram_len - 1);Quoted 3 lines · Jan Schubert
I've posted a small fix to
mdacon.cat the end of last year (while Marcello was in vacation). It has'nt made it to the kernel yet, nor was there any feedback. I've thought, thats the way like open source works...
James Simmons <[email protected]> volunteered to take in patches.
Regards,
Zwane Mwaikambo
Quoted 4 lines · Jan Schubert
I've posted a small fix to
mdacon.cat the end of last year (while Marcello was in vacation). It has'nt made it to the kernel yet, nor was there any feedback. I've thought, thats the way like open source works...
Well, you can't be too surprised when everybody doesn't jump to review driver changes for hardware that dates to the early 80's :)
I think your change may break true MDA cards (as in an original full length 8 bit text only card, clustered with enough TTL chips that could heat a small room...)
If I read the current code right, the p==q case means that TYPE_MDA remains set, and these cards don't get mda_initialize(). Only the TYPE_HERC* cards get mda_initialize().
Your change forces TYPE_HERC and forces mda_initialize() on all non Herc cards, which is a departure from existing behaviour.
So, there are two possibilities:
1) mda_initialize() should really be called hga_initialize() and your card is simply broken in some way :)
or
2) mda_initialize() is also needed for some non Herc cards too.
What kind of card did you say needed this change?
Paul.
Quoted 43 lines
So, here we go again (against 2.4.17; my MDA is detected, but not initialized, see my last posting):
video/mdacon.c Sun Dec 30 21:36:50 2001 renamed from video/mdacon.c.orig Sun Dec 30 02:44:25 2001 +7
-4@@ -24,6 +24,7 @@ * * Changelog: * Paul G. (03/2001) Fix mdacon= boot prompt to use __setup(). + * 20011230 [email protected] - consider non-Hercules MDA compatible */ #include <linux/types.h>@@ -291,6 +292,10 @@ break; } } + else { /* consider non-Hercules as Hercules-compatible */ + mda_type = TYPE_HERC; + mda_type_name = "Hercules compatible (hopefully)"; + } return 1; }@@ -342,9 +347,8 @@ return NULL; } - if (mda_type != TYPE_MDA) { - mda_initialize(); - } + /* at this point, we found an MDA */ + mda_initialize(); /* cursor looks ugly during boot-up, so turn it off */ mda_set_cursor(mda_vram_len - 1); -To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/