[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-ppc] [PATCH v2] spapr_vio/spapr_iommu: Move VIO bypass where i
From: |
David Gibson |
Subject: |
Re: [Qemu-ppc] [PATCH v2] spapr_vio/spapr_iommu: Move VIO bypass where it belongs |
Date: |
Thu, 29 Jan 2015 11:31:09 +1100 |
User-agent: |
Mutt/1.5.23 (2014-03-12) |
On Wed, Jan 28, 2015 at 07:49:48PM +1100, Alexey Kardashevskiy wrote:
> On 01/28/2015 08:21 AM, Alexey Kardashevskiy wrote:
> > On 01/27/2015 05:13 PM, Alexey Kardashevskiy wrote:
[snip]
> >> diff --git a/include/hw/ppc/spapr_vio.h b/include/hw/ppc/spapr_vio.h
> >> index 46edc2a..6ad55d1 100644
> >> --- a/include/hw/ppc/spapr_vio.h
> >> +++ b/include/hw/ppc/spapr_vio.h
> >> @@ -64,6 +64,8 @@ struct VIOsPAPRDevice {
> >> target_ulong signal_state;
> >> VIOsPAPR_CRQ crq;
> >> AddressSpace as;
> >> + MemoryRegion mrroot;
> >> + MemoryRegion mrbypass;
> >> sPAPRTCETable *tcet;
> >> };
> >>
> >>
> >
> >
>
>
> I believe doing something like this is way too disguising because of
> tobj->parent?
It's kinda ugly, but it's the best way I can think of.
So, I think this is a reasonable approach, but it does need a comment
saying why the hack is necessary.
>
>
>
> --- a/hw/ppc/spapr_iommu.c
> +++ b/hw/ppc/spapr_iommu.c
> @@ -25,6 +25,7 @@
> #include "trace.h"
>
> #include "hw/ppc/spapr.h"
> +#include "hw/ppc/spapr_vio.h"
>
> #include <libfdt.h>
>
> @@ -88,10 +89,26 @@ static IOMMUTLBEntry
> spapr_tce_translate_iommu(MemoryRegion *iommu, hwaddr addr,
> return ret;
> }
>
> +static int spapr_tce_table_post_load(void *opaque, int version_id)
> +{
> + sPAPRTCETable *tcet = SPAPR_TCE_TABLE(opaque);
> + Object *tobj = OBJECT(tcet);
> + Object *vobj = object_dynamic_cast(tobj->parent, TYPE_VIO_SPAPR_DEVICE);
> +
> + if (!vobj) {
> + return 0;
> + }
> +
> + spapr_vio_set_bypass((VIOsPAPRDevice *) vobj, tcet->bypass);
> +
> + return 0;
> +}
> +
> static const VMStateDescription vmstate_spapr_tce_table = {
> .name = "spapr_iommu",
> .version_id = 2,
> .minimum_version_id = 2,
> + .post_load = spapr_tce_table_post_load,
> .fields = (VMStateField []) {
> /* Sanity check */
> VMSTATE_UINT32_EQUAL(liobn, sPAPRTCETable),
>
>
>
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
pgpkz1sZn5U1a.pgp
Description: PGP signature
Re: [Qemu-ppc] [PATCH v2] spapr_vio/spapr_iommu: Move VIO bypass where it belongs, David Gibson, 2015/01/28