Re: [PATCH 03/34] AMD IOMMU: add defines and structures for ACPI scanning code

From: Joerg Roedel
Date: Thu Jul 10 2008 - 08:18:51 EST


On Wed, Jul 09, 2008 at 06:41:41PM -0700, Andrew Morton wrote:
> On Thu, 26 Jun 2008 21:27:39 +0200 Joerg Roedel <joerg.roedel@xxxxxxx> wrote:
>
> > This patch adds the required data structures and constants required to parse
> > the ACPI table for the AMD IOMMU.
> >
> > Signed-off-by: Joerg Roedel <joerg.roedel@xxxxxxx>
> > ---
> > arch/x86/kernel/amd_iommu_init.c | 101 ++++++++++++++++++++++++++++++++++++++
> > 1 files changed, 101 insertions(+), 0 deletions(-)
> > create mode 100644 arch/x86/kernel/amd_iommu_init.c
> >
> > diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
> > new file mode 100644
> > index 0000000..6fce5ab
> > --- /dev/null
> > +++ b/arch/x86/kernel/amd_iommu_init.c
> > @@ -0,0 +1,101 @@
> > +/*
> > + * Copyright (C) 2007-2008 Advanced Micro Devices, Inc.
> > + * Author: Joerg Roedel <joerg.roedel@xxxxxxx>
> > + * Leo Duran <leo.duran@xxxxxxx>
> > + *
> > + * This program is free software; you can redistribute it and/or modify it
> > + * under the terms of the GNU General Public License version 2 as published
> > + * by the Free Software Foundation.
> > + *
> > + * This program is distributed in the hope that it will be useful,
> > + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> > + * GNU General Public License for more details.
> > + *
> > + * You should have received a copy of the GNU General Public License
> > + * along with this program; if not, write to the Free Software
> > + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
> > + */
> > +
> > +#include <linux/pci.h>
> > +#include <linux/acpi.h>
> > +#include <linux/gfp.h>
> > +#include <linux/list.h>
> > +#include <asm/pci-direct.h>
> > +#include <asm/amd_iommu_types.h>
> > +#include <asm/gart.h>
> > +
> > +/*
> > + * definitions for the ACPI scanning code
> > + */
> > +#define UPDATE_LAST_BDF(x) do {\
> > + if ((x) > amd_iommu_last_bdf) \
> > + amd_iommu_last_bdf = (x); \
> > + } while (0);
>
> Does this need to exist?
>
> If so, it could and should be implemented as a C function, not as a macro.

Yes, this macro is usefull because it makes the code look cleaner. But I
will convert that into a function.

> > +#define DEVID(bus, devfn) (((bus) << 8) | (devfn))
>
> I suspect that a helper function for this already exists.

I have not found one. Macros exist for slot/function of PCI devices. But
nothing which also includes the bus.

> > +#define PCI_BUS(x) (((x) >> 8) & 0xff)
>
> And this. If it doesn't exist, probably it should?

Ok, I move that one to linux/pci.h.

>
> > +#define IVRS_HEADER_LENGTH 48
> > +#define TBL_SIZE(x) (1 << (PAGE_SHIFT + get_order(amd_iommu_last_bdf * (x))))
>
> Can be implemented in C.

Right. I will convert that into a function too.

>
> > +#define ACPI_IVHD_TYPE 0x10
> > +#define ACPI_IVMD_TYPE_ALL 0x20
> > +#define ACPI_IVMD_TYPE 0x21
> > +#define ACPI_IVMD_TYPE_RANGE 0x22
> > +
> > +#define IVHD_DEV_ALL 0x01
> > +#define IVHD_DEV_SELECT 0x02
> > +#define IVHD_DEV_SELECT_RANGE_START 0x03
> > +#define IVHD_DEV_RANGE_END 0x04
> > +#define IVHD_DEV_ALIAS 0x42
> > +#define IVHD_DEV_ALIAS_RANGE 0x43
> > +#define IVHD_DEV_EXT_SELECT 0x46
> > +#define IVHD_DEV_EXT_SELECT_RANGE 0x47
> > +
> > +#define IVHD_FLAG_HT_TUN_EN 0x00
> > +#define IVHD_FLAG_PASSPW_EN 0x01
> > +#define IVHD_FLAG_RESPASSPW_EN 0x02
> > +#define IVHD_FLAG_ISOC_EN 0x03
> > +
> > +#define IVMD_FLAG_EXCL_RANGE 0x08
> > +#define IVMD_FLAG_UNITY_MAP 0x01
> > +
> > +#define ACPI_DEVFLAG_INITPASS 0x01
> > +#define ACPI_DEVFLAG_EXTINT 0x02
> > +#define ACPI_DEVFLAG_NMI 0x04
> > +#define ACPI_DEVFLAG_SYSMGT1 0x10
> > +#define ACPI_DEVFLAG_SYSMGT2 0x20
> > +#define ACPI_DEVFLAG_LINT0 0x40
> > +#define ACPI_DEVFLAG_LINT1 0x80
> > +#define ACPI_DEVFLAG_ATSDIS 0x10000000
> > +
> > +struct ivhd_header {
> > + u8 type;
> > + u8 flags;
> > + u16 length;
> > + u16 devid;
> > + u16 cap_ptr;
> > + u64 mmio_phys;
> > + u16 pci_seg;
> > + u16 info;
> > + u32 reserved;
> > +} __attribute__((packed));
> > +
> > +struct ivhd_entry {
> > + u8 type;
> > + u16 devid;
> > + u8 flags;
> > + u32 ext;
> > +} __attribute__((packed));
> > +
> > +struct ivmd_header {
> > + u8 type;
> > + u8 flags;
> > + u16 length;
> > + u16 devid;
> > + u16 aux;
> > + u64 resv;
> > + u64 range_start;
> > + u64 range_length;
> > +} __attribute__((packed));
>
> I can make a guess as to why these are "packed", but a code comment would
> prevent guesswork.

I am currently preparing patches to add a lot of comments to the code. I
will also add a comment to these structures.

Joerg

--
| AMD Saxony Limited Liability Company & Co. KG
Operating | Wilschdorfer Landstr. 101, 01109 Dresden, Germany
System | Register Court Dresden: HRA 4896
Research | General Partner authorized to represent:
Center | AMD Saxony LLC (Wilmington, Delaware, US)
| General Manager of AMD Saxony LLC: Dr. Hans-R. Deppe, Thomas McCoy

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/