Re: [Qemu-devel] [PATCH 2/4] Allow ISA bus to be configured out

["Andreas Färber" <afaerber@suse.de>]

[-- Attachment #1: Type: text/plain, Size: 3156 bytes --]

Am 30.06.2015 um 06:48 schrieb David Gibson:
> On Tue, Jun 30, 2015 at 01:44:10PM +1000, David Gibson wrote:
>> On Mon, Jun 29, 2015 at 12:26:50PM +0200, Andreas Färber wrote:
>>> Am 29.06.2015 um 07:36 schrieb David Gibson:
>>>> Currently, the code to handle the legacy ISA bus is always included in
>>>> qemu.  However there are lots of platforms that don't include ISA legacy
>>>> devies, and quite a few that have never used ISA legacy devices at all.
>>>>
>>>> This patch allows the ISA bus code to be disabled in the configuration for
>>>> platforms where it doesn't make sense.  For now, the default configs are
>>>> adjusted to include ISA on all platforms including PCI (since
>>>> CONFIG_IDE_CORE which is in pci.mak requires ISA support) and also several
>>>> others which include ISA devices.  We may want to pare this down in future.
>>>
>>> PCI also allows to have a PCI-ISA bridge such as the i82378, which means
>>> that all PCI platforms can potentially obtain an actual ISA bus. So at
>>> least the commit message could use a makeover to avoid someone touching
>>> IDE to blindly disable the dependency.
>>
>> Sorry, I'm not quite following what you're getting at here.
>>
>>> You'd also need to think about the qtests then:
>>>
>>> $ git grep i82378 -- tests/
>>> tests/endianness-test.c:    { "ppc", "g3beige", 0xfe000000, .bswap =
>>> true, .superio = "i82378" },
>>> tests/endianness-test.c:    { "ppc", "bamboo", 0xe8000000, .bswap =
>>> true, .superio = "i82378" },
>>> tests/endianness-test.c:    { "ppc64", "mac99", 0xf2000000, .bswap =
>>> true, .superio = "i82378" },
>>> tests/endianness-test.c:      .bswap = true, .superio = "i82378" },
>>> tests/endianness-test.c:    { "sh4", "r2d", 0xfe240000, .superio =
>>> "i82378" },
>>> tests/endianness-test.c:    { "sh4eb", "r2d", 0xfe240000, .bswap = true,
>>> .superio = "i82378" },
>>
>> Ah, good point.  I'll need to rework for that,
> 
> Actually.. on second thoughts..
> 
> CONFIG_I82378 already exists, and will break those tests of disabled.
> 
> So while making the tests more robust against config changes would be
> a good thing in general, I don't think it's in the scope of what I'm
> trying to do here - making ISA configurable won't make these tests any
> more broken with nonstandard configs than they already are.

I am mainly saying that your commit message is misleading.

CONFIG_IDE_CORE is not the only reason that pci.mak needs
CONFIG_ISA_BUS=y, as you make it sound. The other problem is that we
don't have Kconfig yet, so we have no way of modeling that I82378
depends on ISA_BUS, therefore either pci.mak (as done here) or
ppc-softmmu.mak and ppc64-softmmu.mak need CONFIG_ISA_BUS=y. And due to
the qtest either sh4/sh4eb need that same dependency too, or the test
needs to be changed.

So please either revise the commit message when you resend, or you'll
need to repeat CONFIG_ISA_BUS=y for ppc, ppc64, sh4, sh4eb.

Regards,
Andreas

-- 
SUSE Linux GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Felix Imendörffer, Jane Smithard, Dilip Upmanyu, Graham Norton; HRB
21284 (AG Nürnberg)


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]