Mips-arc tests ever work?

Mips-arc tests ever work?

[Glenn Washburn]

Hi,

I'm looking into getting grub qemu testing working for the "mis-arc"
target. Is anyone running these tests successfully?  In grub-shell that
target passed the -M indy arguments to qemu.  qemu-system-mips64 is
saying that that is an invalid machine type.  The available machine
types are magnum, malta, mips, mipssim, none, and pica61.  The indy
machine type appears to not be available since at least qemu 2.5,
perhaps it was dropped some time ago?  According to wikipedia the SGI
Indy originally used R4k processors.  So would machine type mips be the
one to use since its description says "mips r4k platform"?  But then
why wouldn't it already use that machine type like the target
"mips-qemu_mips"?

So can anyone familiar with this tell me an appropriate machine type of
the above listed to use instead of indy?  Perhaps Vladimir who
committed that code can chime in?

Thanks,
Glenn

Re: Mips-arc tests ever work?

[John Paul Adrian Glaubitz]

Hello!

On 12/5/20 1:33 PM, Glenn Washburn wrote:
> I'm looking into getting grub qemu testing working for the "mis-arc"
> target. Is anyone running these tests successfully?

I'm running manual tests for all architectures before each release.

> In grub-shell that target passed the -M indy arguments to qemu.
> qemu-system-mips64 is saying that that is an invalid machine type.
> The available machine types are magnum, malta, mips, mipssim, none,
> and pica61. The indy machine type appears to not be available since
> at least qemu 2.5, perhaps it was dropped some time ago?

I could not find any reference to "indy" in the qemu git log. Are you
sure it was ever officially supported? I assume the tests here were
run on real hardware.

> According to wikipedia the SGI Indy originally used R4k processors.
> So would machine type mips be the one to use since its description
> says "mips r4k platform"?

No, that was a different, generic machine type, see [1].

> But then why wouldn't it already use that machine type like the
> target "mips-qemu_mips"?

It's possible the author of this change assumed that qemu supports
the indy target which apparently it does not.

> So can anyone familiar with this tell me an appropriate machine type of
> the above listed to use instead of indy?  Perhaps Vladimir who
> committed that code can chime in?

Adrian

> [1] https://git.qemu.org/?p=qemu.git;a=commit;h=f169413c27130ac9ebf96a1212100bba92f348ba

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - glaubitz@debian.org
`. `'   Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

Re: Mips-arc tests ever work?

[Glenn Washburn]

On Sat, 5 Dec 2020 14:47:40 +0100
John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> wrote:

> Hello!
> 
> On 12/5/20 1:33 PM, Glenn Washburn wrote:
> > I'm looking into getting grub qemu testing working for the "mis-arc"
> > target. Is anyone running these tests successfully?
> 
> I'm running manual tests for all architectures before each release.

What exactly are manual tests?  Is that running manually the tests for
that target? (ie "make check" for that target)

Also what version of qemu are you running to do the tests?

> > In grub-shell that target passed the -M indy arguments to qemu.
> > qemu-system-mips64 is saying that that is an invalid machine type.
> > The available machine types are magnum, malta, mips, mipssim, none,
> > and pica61. The indy machine type appears to not be available since
> > at least qemu 2.5, perhaps it was dropped some time ago?
> 
> I could not find any reference to "indy" in the qemu git log. Are you
> sure it was ever officially supported? I assume the tests here were
> run on real hardware.

Take a look in "tests/util/grub-shell.in" and you'll find the offending
test code in grub.  I have no clue either way whether it was ever
supported, but someone obviously did because its in the test code.  And
that someone might be Vladimir (phcoder) who committed the code.  This
is why I'm wondering if it ever worked.

> > According to wikipedia the SGI Indy originally used R4k processors.
> > So would machine type mips be the one to use since its description
> > says "mips r4k platform"?
> 
> No, that was a different, generic machine type, see [1].

I'm confused by this response.  My qemu doesn't list a "r4k" machine
type and its version 4.2.1.  The machine type "mips" says its the r4k
platform, but that doesn't look like what's being removed in that
commit.  Reviewing the hw/mips/Kconfig, it looks like the machine types
have changed a lot since 4.2.1, so maybe it was renamed to "r4k"
(haven't checked yet).  However if it was renamed and is being removed,
that's going to cause further problems (when/if we use qemu 5.2 for
testing) because grub-shell is using the "mips" machine type for
"mips-qemu_mips" and "mipsel-qemu_mips".

> > But then why wouldn't it already use that machine type like the
> > target "mips-qemu_mips"?
> 
> It's possible the author of this change assumed that qemu supports
> the indy target which apparently it does not.

Vladimir?

> > So can anyone familiar with this tell me an appropriate machine
> > type of the above listed to use instead of indy?  Perhaps Vladimir
> > who committed that code can chime in?

Ultimately, I just want to know what machine should I use if not indy
for the "mips-arc" target?  I know very little about mips and what an
appropriate machine type would be for emulating this target.

Glenn

> 
> Adrian
> 
> > [1]
> > https://git.qemu.org/?p=qemu.git;a=commit;h=f169413c27130ac9ebf96a1212100bba92f348ba
> 

Re: Mips-arc tests ever work?

[John Paul Adrian Glaubitz]

On 12/5/20 8:10 PM, Glenn Washburn wrote:
>> I'm running manual tests for all architectures before each release.
> 
> What exactly are manual tests?  Is that running manually the tests for
> that target? (ie "make check" for that target)

Yes. I'm test-compiling GRUB natively, run "make check" and also perform
test installations in some cases.

> Also what version of qemu are you running to do the tests?

I'm performing all tests on real hardware. I have access to all architectures
supported by GRUB either through my personal hardware zoo, the Debian project
of which I am member of or the GCC compile farm for which I am also providing
hardware.

You are free to register for the GCC compile farm and use it yourself [1].

>> I could not find any reference to "indy" in the qemu git log. Are you
>> sure it was ever officially supported? I assume the tests here were
>> run on real hardware.
> 
> Take a look in "tests/util/grub-shell.in" and you'll find the offending
> test code in grub.  I have no clue either way whether it was ever
> supported, but someone obviously did because its in the test code.  And
> that someone might be Vladimir (phcoder) who committed the code.  This
> is why I'm wondering if it ever worked.

Maybe he used a fork of qemu for these tests.

>>> According to wikipedia the SGI Indy originally used R4k processors.
>>> So would machine type mips be the one to use since its description
>>> says "mips r4k platform"?
>>
>> No, that was a different, generic machine type, see [1].
> 
> I'm confused by this response.  My qemu doesn't list a "r4k" machine
> type and its version 4.2.1.  The machine type "mips" says its the r4k
> platform, but that doesn't look like what's being removed in that
> commit.  Reviewing the hw/mips/Kconfig, it looks like the machine types
> have changed a lot since 4.2.1, so maybe it was renamed to "r4k"
> (haven't checked yet).  However if it was renamed and is being removed,
> that's going to cause further problems (when/if we use qemu 5.2 for
> testing) because grub-shell is using the "mips" machine type for
> "mips-qemu_mips" and "mipsel-qemu_mips".

Well, either way, the r4k machine no longer exists in QEMU 5.2.

>>> So can anyone familiar with this tell me an appropriate machine
>>> type of the above listed to use instead of indy?  Perhaps Vladimir
>>> who committed that code can chime in?
> 
> Ultimately, I just want to know what machine should I use if not indy
> for the "mips-arc" target?  I know very little about mips and what an
> appropriate machine type would be for emulating this target.

I guess "malta" would work, too.

Adrian

> [1] https://gcc.gnu.org/wiki/CompileFarm

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - glaubitz@debian.org
`. `'   Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

Re: Mips-arc tests ever work?

[Glenn Washburn]

On Sat, 5 Dec 2020 21:28:05 +0100
John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> wrote:

> On 12/5/20 8:10 PM, Glenn Washburn wrote:
> >> I'm running manual tests for all architectures before each release.
> > 
> > What exactly are manual tests?  Is that running manually the tests
> > for that target? (ie "make check" for that target)
> 
> Yes. I'm test-compiling GRUB natively, run "make check" and also
> perform test installations in some cases.

Ok so I'm clear, since you're running "make check", your qemu tests for
mips-arc are failing due to indy not being a machine type, correct?

> > Also what version of qemu are you running to do the tests?
> 
> I'm performing all tests on real hardware. I have access to all
> architectures supported by GRUB either through my personal hardware
> zoo, the Debian project of which I am member of or the GCC compile
> farm for which I am also providing hardware.
> 
> You are free to register for the GCC compile farm and use it yourself
> [1].

Cool project, thanks for the tip.

> >> I could not find any reference to "indy" in the qemu git log. Are
> >> you sure it was ever officially supported? I assume the tests here
> >> were run on real hardware.
> > 
> > Take a look in "tests/util/grub-shell.in" and you'll find the
> > offending test code in grub.  I have no clue either way whether it
> > was ever supported, but someone obviously did because its in the
> > test code.  And that someone might be Vladimir (phcoder) who
> > committed the code.  This is why I'm wondering if it ever worked.
> 
> Maybe he used a fork of qemu for these tests.
> 
> >>> According to wikipedia the SGI Indy originally used R4k
> >>> processors. So would machine type mips be the one to use since
> >>> its description says "mips r4k platform"?
> >>
> >> No, that was a different, generic machine type, see [1].
> > 
> > I'm confused by this response.  My qemu doesn't list a "r4k" machine
> > type and its version 4.2.1.  The machine type "mips" says its the
> > r4k platform, but that doesn't look like what's being removed in
> > that commit.  Reviewing the hw/mips/Kconfig, it looks like the
> > machine types have changed a lot since 4.2.1, so maybe it was
> > renamed to "r4k" (haven't checked yet).  However if it was renamed
> > and is being removed, that's going to cause further problems
> > (when/if we use qemu 5.2 for testing) because grub-shell is using
> > the "mips" machine type for "mips-qemu_mips" and "mipsel-qemu_mips".
> 
> Well, either way, the r4k machine no longer exists in QEMU 5.2.
> 
> >>> So can anyone familiar with this tell me an appropriate machine
> >>> type of the above listed to use instead of indy?  Perhaps Vladimir
> >>> who committed that code can chime in?
> > 
> > Ultimately, I just want to know what machine should I use if not
> > indy for the "mips-arc" target?  I know very little about mips and
> > what an appropriate machine type would be for emulating this target.
> 
> I guess "malta" would work, too.

Ok, see if that works.

> Adrian
> 
> > [1] https://gcc.gnu.org/wiki/CompileFarm
> 

Re: Mips-arc tests ever work?

[Vladimir 'phcoder' Serbinenko]

[-- Attachment #1: Type: text/plain, Size: 1252 bytes --]

I have added -M indy in my own branch but never released it

сб, 5 дек. 2020 г., 13:34 Glenn Washburn <development@efficientek.com>:

> Hi,
>
> I'm looking into getting grub qemu testing working for the "mis-arc"
> target. Is anyone running these tests successfully?  In grub-shell that
> target passed the -M indy arguments to qemu.  qemu-system-mips64 is
> saying that that is an invalid machine type.  The available machine
> types are magnum, malta, mips, mipssim, none, and pica61.  The indy
> machine type appears to not be available since at least qemu 2.5,
> perhaps it was dropped some time ago?  According to wikipedia the SGI
> Indy originally used R4k processors.  So would machine type mips be the
> one to use since its description says "mips r4k platform"?  But then
> why wouldn't it already use that machine type like the target
> "mips-qemu_mips"?
>
> So can anyone familiar with this tell me an appropriate machine type of
> the above listed to use instead of indy?  Perhaps Vladimir who
> committed that code can chime in?
>
> Thanks,
> Glenn
>
> _______________________________________________
> Grub-devel mailing list
> Grub-devel@gnu.org
> https://lists.gnu.org/mailman/listinfo/grub-devel
>

[-- Attachment #2: Type: text/html, Size: 1795 bytes --]

Re: Mips-arc tests ever work?

[Glenn Washburn]

On Wed, 9 Dec 2020 10:46:45 +0100
"Vladimir 'phcoder' Serbinenko" <phcoder@gmail.com> wrote:

> I have added -M indy in my own branch but never released it

So I understand correctly, you have a -M indy in your branch of Qemu?
If so, is that what we should be using to get the mips-arc target?  Or
what do you recommend for getting the mips-arc tests working with Qemu?

> сб, 5 дек. 2020 г., 13:34 Glenn Washburn
> <development@efficientek.com>:
> 
> > Hi,
> >
> > I'm looking into getting grub qemu testing working for the "mis-arc"
> > target. Is anyone running these tests successfully?  In grub-shell
> > that target passed the -M indy arguments to qemu.
> > qemu-system-mips64 is saying that that is an invalid machine type.
> > The available machine types are magnum, malta, mips, mipssim, none,
> > and pica61.  The indy machine type appears to not be available
> > since at least qemu 2.5, perhaps it was dropped some time ago?
> > According to wikipedia the SGI Indy originally used R4k processors.
> >  So would machine type mips be the one to use since its description
> > says "mips r4k platform"?  But then why wouldn't it already use
> > that machine type like the target "mips-qemu_mips"?
> >
> > So can anyone familiar with this tell me an appropriate machine
> > type of the above listed to use instead of indy?  Perhaps Vladimir
> > who committed that code can chime in?
> >
> > Thanks,
> > Glenn
> >
> > _______________________________________________
> > Grub-devel mailing list
> > Grub-devel@gnu.org
> > https://lists.gnu.org/mailman/listinfo/grub-devel
> >

Re: Mips-arc tests ever work?

[Glenn Washburn]

On Sat, 5 Dec 2020 21:28:05 +0100
John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> wrote:

> On 12/5/20 8:10 PM, Glenn Washburn wrote:
> >>> So can anyone familiar with this tell me an appropriate machine
> >>> type of the above listed to use instead of indy?  Perhaps Vladimir
> >>> who committed that code can chime in?
> > 
> > Ultimately, I just want to know what machine should I use if not
> > indy for the "mips-arc" target?  I know very little about mips and
> > what an appropriate machine type would be for emulating this target.
> 
> I guess "malta" would work, too.

Ok, I've gotten around to testing with the "malta" machine type and I
get the following error:

$ qemu-system-mips64 -M malta -nographic -monitor file:/dev/null -fw_cfg
name=etc/sercon-port,string=0 -serial file:/dev/stdout -cdrom
/tmp/test-mips-arc-75f96ab1/./file_filter_test/tmp.THAl4u4RAX -boot d
qemu-system-mips64: Could not load MIPS bios 'mips_bios.bin', and no
-kernel argument was specified

I suspect the -kernel should be needed, but I imagine I do need a bios.
Do you know where I can get a mips bios file? I also get similar
errors for mips-qemu_mips, mipsel-arc, and mipsel-qemu_mips.

Glenn