Geert Uytterhoeven writes:
> Perhaps we can write some asm macros for simple FPU instructions,
> calling into the existing emulation code for simple FPU instructions, and
> just include e.g. arch/m68k/fpsp040/slogn.S?
fpsp040 depends a lot on the internals of the unimplemented fpu insn
exception
On Mon, Nov 18, 2013 at 10:35 AM, Andreas Schwab wrote:
> Finn Thain writes:
>> On Sun, 20 Oct 2013, Andreas Schwab wrote:
>>> Michael Schmitz writes:
>>> > Even for 030, I cannot recall specific instructions that are
>>> > unimplemented or dodgy.
>>>
>>> None of the trigonometric or logarithmic
Finn Thain writes:
> On Sun, 20 Oct 2013, Andreas Schwab wrote:
>
>> Michael Schmitz writes:
>>
>> > Even for 030, I cannot recall specific instructions that are
>> > unimplemented or dodgy.
>>
>> None of the trigonometric or logarithmic insns are implemented (grep for
>> uprint).
>
> If any
On Sun, 20 Oct 2013, Andreas Schwab wrote:
> Michael Schmitz writes:
>
> > Even for 030, I cannot recall specific instructions that are
> > unimplemented or dodgy.
>
> None of the trigonometric or logarithmic insns are implemented (grep for
> uprint).
If anyone wants to get them working, th
On Tue, Nov 12, 2013 at 1:21 AM, Finn Thain wrote:
>> > Hence I was trying to find an explanation for the screen shot posted,
>> > which says "console [tty0] enabled" even though the macfb console did
>> > not appear.
>>
>> No, without fb/console support you wouldn't even see those.
>
> "console [
On Mon, 11 Nov 2013, Geert Uytterhoeven wrote:
> On Mon, Nov 11, 2013 at 6:16 AM, Finn Thain
> wrote:
> >> > ! CONFIG_FB_SYS_COPYAREA : m n
> >> > ! CONFIG_FB_SYS_FILLRECT : m n
> >> > ! CONFIG_FB_SYS_FOPS : m n
> >> > ! CONFIG_FB_SYS_IM
Geert Uytterhoeven dixit:
>On Mon, Nov 11, 2013 at 6:16 AM, Finn Thain wrote:
>>> > ! CONFIG_FB_SYS_COPYAREA : m n
>>> > ! CONFIG_FB_SYS_FILLRECT : m n
>>> > ! CONFIG_FB_SYS_FOPS : m n
>>> > ! CONFIG_FB_SYS_IMAGEBLIT : m n
On Mon, Nov 11, 2013 at 6:16 AM, Finn Thain wrote:
>> > ! CONFIG_FB_SYS_COPYAREA : m n
>> > ! CONFIG_FB_SYS_FILLRECT : m n
>> > ! CONFIG_FB_SYS_FOPS : m n
>> > ! CONFIG_FB_SYS_IMAGEBLIT : m n
>> >
>> >Would this prevent the
On Mon, 11 Nov 2013, Thorsten Glaser wrote:
> Finn Thain dixit:
>
> > ! multi_defconfig -+
> > ! config-3.10-2-m68k + |
> > !| |
> >
> > ...
> >
> > ! CONFIG_FB_SYS_COPYAREA : m n
Finn Thain dixit:
>Would this prevent the macfb framebuffer console from working in the
>absence of kernel modules?
With the switch to initrd, absence of modules is simply no longer
supported at all, period. I am not happy about this myself, but,
realistically, the monoliths cannot be kept worki
On Sun, 20 Oct 2013, Geert Uytterhoeven wrote:
> This is the difference between my multi_defconfig and config-3.10-2-m68k.
...
>
> ! multi_defconfig -+
> ! config-3.10-2-m68k + |
> !| |
...
> ! CO
On Sun, 20 Oct 2013, Geert Uytterhoeven wrote:
> This is the difference between my multi_defconfig and
> config-3.10-2-m68k.
Nice!
> Big differences are due to many drivers being modular in the initrd for
> the Debian kernel, which I want to avoid for my defconfigs, for
> hassle-free testing
Andreas Schwab dixit:
>That's just double rounding. But saying that it doesn't handle it at
>all is stupid.
It doesn’t handle the documented way of switching the FPU into
a mode that doesn’t do that.
bye,
//mirabilos
--
Beware of ritual lest you forget the meaning behind it.
yeah but it mean
Thorsten Glaser writes:
>>> On the other hand, I cannot help but wonder how well it compares
>>> to whatever ARAnyM passes as FPU. I know ARAnyM doesn’t handle the
>>> 80/64bit precision switch at all
>>
>>It doesn't?
>
> In my tests, it didn’t, and I later got a mail from someone saying
> that i
Andreas Schwab dixit:
>None of the trigonometric or logarithmic insns are implemented (grep for
OK, decide on whether this should let it be disabled or not.
Ingo J�rgensmann dixit:
>Hmmm... if Aranym is the problematic part in this story, then Aranym
No, it’s not, that’s a totally different s
Thorsten Glaser writes:
> On the other hand, I cannot help but wonder how well it compares
> to whatever ARAnyM passes as FPU. I know ARAnyM doesn’t handle the
> 80/64bit precision switch at all
It doesn't?
Andreas.
--
Andreas Schwab, sch...@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53
On Sun, 20 Oct 2013, Michael Schmitz wrote:
> The docs are quite clearly out of date...
ISTR those docs are older than the most recent work Roman did on the FPU
emulation code...
> Certain mask revisions of the LC040 have a firmware bug that prevents
> floating point exceptions to properly wo
On 10/20/2013 01:25 AM, Michael Schmitz wrote:
> The docs are quite clearly out of date. From memory, there may be the
> odd FPU instruction that does not work perfectly, but that would be
> limited to 030 type emulation. 040 and 060 use code from the Motorola
> floating-point support package which
On Sun, Oct 20, 2013 at 11:25 AM, Laurent Vivier wrote:
> Le 20/10/2013 10:52, Geert Uytterhoeven a écrit :
>>
>> On Sat, Oct 19, 2013 at 6:13 PM, Geert Uytterhoeven
>> wrote:
>> [...]
>>
>>
>> Apart from that, (probably both) configs need more stripping in the common
>> part. I want to have boot
Le 20/10/2013 10:52, Geert Uytterhoeven a écrit :
On Sat, Oct 19, 2013 at 6:13 PM, Geert Uytterhoeven
wrote:
[...]
Apart from that, (probably both) configs need more stripping in the common
part. I want to have bootable (i.e. < 4 MiB) defconfig kernels.
Moreover, a compressed kernel fitting on
On Sat, Oct 19, 2013 at 6:13 PM, Geert Uytterhoeven
wrote:
> Doh, I really should have an in-depth look at your .config, and compare it to
> multi_defconfig (which does need more stripping!), to catch these things.
This is the difference between my multi_defconfig and config-3.10-2-m68k.
Big diff
On Sun, Oct 20, 2013 at 10:04 AM, Ingo Jürgensmann
wrote:
> Am 20.10.2013 um 01:46 schrieb Thorsten Glaser :
>> Geert Uytterhoeven dixit:
>>> FPU emulation should be fine, AFAIK (never used it myself, though).
>> Michael Schmitz dixit:
>>> The docs are quite clearly out of date. From memory, there
Am 20.10.2013 um 01:46 schrieb Thorsten Glaser :
> Geert Uytterhoeven dixit:
>> FPU emulation should be fine, AFAIK (never used it myself, though).
> Michael Schmitz dixit:
>> The docs are quite clearly out of date. From memory, there may be the odd FPU
> […]
>> Seeing as this code is crucial, I'd
Thorsten,
Michael Schmitz dixit:
The docs are quite clearly out of date. From memory, there may be the
odd FPU
[…]
Seeing as this code is crucial, I'd opt for putting it back in, and
finding
These two statements are “it should be fine but I don’t really know”
and “this is important so pu
Michael Schmitz writes:
> Even for 030, I cannot recall specific instructions that are
> unimplemented or dodgy.
None of the trigonometric or logarithmic insns are implemented (grep for
uprint).
Andreas.
--
Andreas Schwab, sch...@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756
Geert Uytterhoeven dixit:
>Hmm, yes, it does look scary... But as usual, there's even less time to
>update the docs than to update the code ;-)
Thanks for volunteering to do that, though ;-)
Geert Uytterhoeven dixit:
>FPU emulation should be fine, AFAIK (never used it myself, though).
Michael
Thorsten,
Any reasoning why FPU emulation has been disabled?
The docs say it’s incomplete, bogus and experimental and generally
not-working so I removed it for space reasons.
The docs most likely are talking about LC040 FPU trap emulation.
The docs are quite clearly out of date. From memor
Any reasoning why FPU emulation has been disabled?
The docs say it’s incomplete, bogus and experimental and generally
not-working so I removed it for space reasons.
The docs most likely are talking about LC040 FPU trap emulation.
John
On Sat, Oct 19, 2013 at 5:53 PM, Thorsten Glaser wrote:
>>Please put it back in!
>
> Not without a statement from the m68k Linux kernel devs that
> it indeed does its job correctly.
Doh, I really should have an in-depth look at your .config, and compare it to
multi_defconfig (which does need more
John Paul Adrian Glaubitz dixit:
>Please put it back in!
Not without a statement from the m68k Linux kernel devs that
it indeed does its job correctly.
bye,
//mirabilos
--
Beware of ritual lest you forget the meaning behind it.
yeah but it means if you really care about something, don't
r
On 10/19/2013 05:13 PM, Thorsten Glaser wrote:
> The docs say it’s incomplete, bogus and experimental and generally
> not-working so I removed it for space reasons.
It worked like a charm for me on my A1200 which doesn't have
an FPU. Many 68030 accelerators don't have an FPU and people
with these
John Paul Adrian Glaubitz dixit:
>Any reasoning why FPU emulation has been disabled?
The docs say it’s incomplete, bogus and experimental and generally
not-working so I removed it for space reasons.
bye,
//mirabilos
--
Beware of ritual lest you forget the meaning behind it.
yeah but it means
Hi folks!
I was trying to upgrade the kernel on my Amiga 1200 to 3.10-2-m68k
today and was unfortunately greeted by this:
[0.21] *** YOU DO NOT HAVE A FLOATING POINT UNIT, WHICH IS
REQUIRED BY LINUX/M68K ***
[0.22] Upgrade your hardware or join the FPU emulation project
[0.230
33 matches
Mail list logo
