* Re: kvm PCI assignment & VFIO ramblings
From: Alexander Graf @ 2011-08-24 3:40 UTC (permalink / raw)
To: Benjamin Herrenschmidt
Cc: aafabbri, Alexey Kardashevskiy, kvm@vger.kernel.org list,
Paul Mackerras, linux-pci, qemu-devel Developers, iommu,
David Gibson, chrisw Wright, Alex Williamson, Avi Kivity,
Anthony Liguori, linuxppc-dev, benve
In-Reply-To: <1314143508.30478.72.camel@pasglop>
On 23.08.2011, at 18:51, Benjamin Herrenschmidt wrote:
>=20
>>> For us the most simple and logical approach (which is also what pHyp
>>> uses and what Linux handles well) is really to expose a given PCI =
host
>>> bridge per group to the guest. Believe it or not, it makes things
>>> easier :-)
>>=20
>> I'm all for easier. Why does exposing the bridge use less bus =
numbers
>> than emulating a bridge?
>=20
> Because a host bridge doesn't look like a PCI to PCI bridge at all for
> us. It's an entire separate domain with it's own bus number space
> (unlike most x86 setups).
>=20
> In fact we have some problems afaik in qemu today with the concept of
> PCI domains, for example, I think qemu has assumptions about a single
> shared IO space domain which isn't true for us (each PCI host bridge
> provides a distinct IO space domain starting at 0). We'll have to fix
> that, but it's not a huge deal.
>=20
> So for each "group" we'd expose in the guest an entire separate PCI
> domain space with its own IO, MMIO etc... spaces, handed off from a
> single device-tree "host bridge" which doesn't itself appear in the
> config space, doesn't need any emulation of any config space etc...
>=20
>> On x86, I want to maintain that our default assignment is at the =
device
>> level. A user should be able to pick single or multiple devices from
>> across several groups and have them all show up as individual,
>> hotpluggable devices on bus 0 in the guest. Not surprisingly, we've
>> also seen cases where users try to attach a bridge to the guest,
>> assuming they'll get all the devices below the bridge, so I'd be in
>> favor of making this "just work" if possible too, though we may have =
to
>> prevent hotplug of those.
>>=20
>> Given the device requirement on x86 and since everything is a PCI =
device
>> on x86, I'd like to keep a qemu command line something like -device
>> vfio,host=3D00:19.0. I assume that some of the iommu properties, =
such as
>> dma window size/address, will be query-able through an architecture
>> specific (or general if possible) ioctl on the vfio group fd. I hope
>> that will help the specification, but I don't fully understand what =
all
>> remains. Thanks,
>=20
> Well, for iommu there's a couple of different issues here but yes,
> basically on one side we'll have some kind of ioctl to know what =
segment
> of the device(s) DMA address space is assigned to the group and we'll
> need to represent that to the guest via a device-tree property in some
> kind of "parent" node of all the devices in that group.
>=20
> We -might- be able to implement some kind of hotplug of individual
> devices of a group under such a PHB (PCI Host Bridge), I don't know =
for
> sure yet, some of that PAPR stuff is pretty arcane, but basically, for
> all intend and purpose, we really want a group to be represented as a
> PHB in the guest.
>=20
> We cannot arbitrary have individual devices of separate groups be
> represented in the guest as siblings on a single simulated PCI bus.
So would it make sense for you to go the same route that we need to go =
on embedded power, with a separate VFIO style interface that simply =
exports memory ranges and irq bindings, but doesn't know anything about =
PCI? For e500, we'll be using something like that to pass through a full =
PCI bus into the system.
Alex
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: Alexander Graf @ 2011-08-24 3:36 UTC (permalink / raw)
To: Benjamin Herrenschmidt
Cc: chrisw, Alexey Kardashevskiy, kvm, Paul Mackerras,
linux-pci@vger.kernel.org, qemu-devel, iommu, David Gibson,
aafabbri, Alex Williamson, Avi Kivity, Anthony Liguori,
linuxppc-dev, benve
In-Reply-To: <1314142892.30478.64.camel@pasglop>
On 23.08.2011, at 18:41, Benjamin Herrenschmidt wrote:
> On Tue, 2011-08-23 at 10:23 -0600, Alex Williamson wrote:
>>=20
>> Yeah. Joerg's idea of binding groups internally (pass the fd of one
>> group to another via ioctl) is one option. The tricky part will be
>> implementing it to support hot unplug of any group from the
>> supergroup.
>> I believe Ben had a suggestion that supergroups could be created in
>> sysfs, but I don't know what the mechanism to do that looks like. It
>> would also be an extra management step to dynamically bind and unbind
>> groups to the supergroup around hotplug. Thanks,=20
>=20
> I don't really care that much what the method for creating them is, to
> be honest, I just prefer this concept of "meta groups" or "super =
groups"
> or "synthetic groups" (whatever you want to name them) to having a
> separate uiommu file descriptor.
>=20
> The one reason I have a slight preference for creating them =
"statically"
> using some kind of separate interface (again, I don't care whether =
it's
> sysfs, netlink, etc...) is that it means things like qemu don't have =
to
> care about them.
>=20
> In general, apps that want to use vfio can just get passed the path to
> such a group or the /dev/ path or the group number (whatever we chose =
as
> the way to identify a group), and don't need to know anything about
> "super groups", how to manipulate them, create them, possible
> constraints etc...
>=20
> Now, libvirt might want to know about that other API in order to =
provide
> control on the creation of these things, but that's a different issue.
>=20
> By "static" I mean they persist, they aren't tied to the lifetime of =
an
> fd.
>=20
> Now that's purely a preference on my side because I believe it will =
make
> life easier for actual programs wanting to use vfio to not have to =
care
> about those super-groups, but as I said earlier, I don't actually care
> that much :-)
Oh I think it's one of the building blocks we need for a sane user space =
device exposure API. If I want to pass user X a few devices that are all =
behind a single IOMMU, I just chown that device node to user X and be =
done with it.
The user space tool actually using the VFIO interface wouldn't be in =
configuration business then - and it really shouldn't. That's what =
system configuration is there for :).
But I'm fairly sure we managed to persuade Alex that this is the right =
path on the BOF :)
Alex
^ permalink raw reply
* Re: [PATCH v3] mtd/nand : workaround for Freescale FCM to support large-page Nand chip
From: LiuShuo @ 2011-08-24 2:48 UTC (permalink / raw)
To: Matthieu CASTET
Cc: Scott Wood, linuxppc-dev@ozlabs.org, dwmw2@infradead.org,
Li Yang-R58472, linux-mtd@lists.infradead.org
In-Reply-To: <4E537AC4.6000301@parrot.com>
=E4=BA=8E 2011=E5=B9=B408=E6=9C=8823=E6=97=A5 18:02, Matthieu CASTET =E5=86=
=99=E9=81=93:
> LiuShuo a =C3=A9crit :
>> =E4=BA=8E 2011=E5=B9=B408=E6=9C=8819=E6=97=A5 00:25, Scott Wood =E5=86=
=99=E9=81=93:
>>> On 08/17/2011 09:33 PM, b35362@freescale.com wrote:
>>>> From: Liu Shuo<b35362@freescale.com>
>>>>
>>>> Freescale FCM controller has a 2K size limitation of buffer RAM. In =
order
>>>> to support the Nand flash chip whose page size is larger than 2K byt=
es,
>>>> we divide a page into multi-2K pages for MTD layer driver. In that c=
ase,
>>>> we force to set the page size to 2K bytes. We convert the page addre=
ss of
>>>> MTD layer driver to a real page address in flash chips and a column =
index
>>>> in fsl_elbc driver. We can issue any column address by UA instructio=
n of
>>>> elbc controller.
>>>>
>>>> NOTE: Due to there is a limitation of 'Number of Partial Program Cyc=
les in
>>>> the Same Page (NOP)', the flash chip which is supported by this work=
around
>>>> have to meet below conditions.
>>>> 1. page size is not greater than 4KB
>>>> 2. 1) if main area and spare area have independent NOPs:
>>>> main area NOP :>=3D3
>>>> spare area NOP :>=3D2?
>>> How often are the NOPs split like this?
>>>
>>>> 2) if main area and spare area have a common NOP:
>>>> NOP :>=3D4
>>> This depends on how the flash is used. If you treat it as a NOP1 fla=
sh
>>> (e.g. run ubifs rather than jffs2), then you need NOP2 for a 4K chip =
and
>>> NOP4 for an 8K chip. OTOH, if you would be making full use of NOP4 o=
n a
>>> real 2K chip, you'll need NOP8 for a 4K chip.
>>>
>>> The NOP restrictions should be documented in the code itself, not jus=
t
>>> in the git changelog. Maybe print it to the console when this hack i=
s
>>> used, along with the NOP value read from the ID.
>> We can't read the NOP from the ID on any chip. Some chips don't
>> give this infomation.(e.g. Micron MT29F4G08BAC)
> Doesn't the micron chip provide it with onfi info ?
Sorry, there is something wrong with my expression.
We can get the NOP info from datasheet, but can't get it by READID=20
command in code.
-LiuShuo
> Matthieu
>
^ permalink raw reply
* Re: [Cbe-oss-dev] [PATCH part1 v2 4/9] Add region 1 memory early
From: Geoff Levand @ 2011-08-24 2:15 UTC (permalink / raw)
To: Antonio Ospite; +Cc: cbe-oss-dev, Andre Heider, linuxppc-dev, Hector Martin
In-Reply-To: <20110824003720.bc0589b8776467cc42e5d066@studenti.unina.it>
Hi Antonio,
On 08/23/2011 03:37 PM, Antonio Ospite wrote:
> Geoff Levand <geoff@infradead.org> wrote:
>> > + if (!map.r1.size) {
>> > + DBG("%s:%d: no region 1, not adding memory\n",
>> > + __func__, __LINE__);
>> > + } else {
>>
>> Remove brackets around a single line conditional.
>>
>> > + DBG("%s:%d: adding memory: start %llxh, size %llxh\n",
>> > + __func__, __LINE__, map.rm.size, map.r1.size);
>> > +
>> > + memblock_add(map.rm.size, map.r1.size);
>> > + memblock_analyze();
>> > + }
>> > +
>
> In Documentation/CodingStyle I read that if [only] one branch is a
> single statement then the parenthesis are OK (and even recommended) for
> both branches, I guess this is for style consistency. See Chapter 3,
> around line 169 on my copy. I guess the wording on that paragraph can
> be made more explicit, I'll try to fix that up.
Thanks for the comments. I don't think its such an important change,
mainly for consistency of style within the PS3 files.
-Geoff
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: Benjamin Herrenschmidt @ 2011-08-23 23:51 UTC (permalink / raw)
To: Alex Williamson
Cc: aafabbri, Alexey Kardashevskiy, kvm, Paul Mackerras,
linux-pci@vger.kernel.org, qemu-devel, David Gibson, chrisw,
iommu, Avi Kivity, Anthony Liguori, linuxppc-dev, benve
In-Reply-To: <1314127809.2859.121.camel@bling.home>
> > For us the most simple and logical approach (which is also what pHyp
> > uses and what Linux handles well) is really to expose a given PCI host
> > bridge per group to the guest. Believe it or not, it makes things
> > easier :-)
>
> I'm all for easier. Why does exposing the bridge use less bus numbers
> than emulating a bridge?
Because a host bridge doesn't look like a PCI to PCI bridge at all for
us. It's an entire separate domain with it's own bus number space
(unlike most x86 setups).
In fact we have some problems afaik in qemu today with the concept of
PCI domains, for example, I think qemu has assumptions about a single
shared IO space domain which isn't true for us (each PCI host bridge
provides a distinct IO space domain starting at 0). We'll have to fix
that, but it's not a huge deal.
So for each "group" we'd expose in the guest an entire separate PCI
domain space with its own IO, MMIO etc... spaces, handed off from a
single device-tree "host bridge" which doesn't itself appear in the
config space, doesn't need any emulation of any config space etc...
> On x86, I want to maintain that our default assignment is at the device
> level. A user should be able to pick single or multiple devices from
> across several groups and have them all show up as individual,
> hotpluggable devices on bus 0 in the guest. Not surprisingly, we've
> also seen cases where users try to attach a bridge to the guest,
> assuming they'll get all the devices below the bridge, so I'd be in
> favor of making this "just work" if possible too, though we may have to
> prevent hotplug of those.
>
> Given the device requirement on x86 and since everything is a PCI device
> on x86, I'd like to keep a qemu command line something like -device
> vfio,host=00:19.0. I assume that some of the iommu properties, such as
> dma window size/address, will be query-able through an architecture
> specific (or general if possible) ioctl on the vfio group fd. I hope
> that will help the specification, but I don't fully understand what all
> remains. Thanks,
Well, for iommu there's a couple of different issues here but yes,
basically on one side we'll have some kind of ioctl to know what segment
of the device(s) DMA address space is assigned to the group and we'll
need to represent that to the guest via a device-tree property in some
kind of "parent" node of all the devices in that group.
We -might- be able to implement some kind of hotplug of individual
devices of a group under such a PHB (PCI Host Bridge), I don't know for
sure yet, some of that PAPR stuff is pretty arcane, but basically, for
all intend and purpose, we really want a group to be represented as a
PHB in the guest.
We cannot arbitrary have individual devices of separate groups be
represented in the guest as siblings on a single simulated PCI bus.
Cheers,
Ben.
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: Benjamin Herrenschmidt @ 2011-08-23 23:41 UTC (permalink / raw)
To: Alex Williamson
Cc: chrisw, Alexey Kardashevskiy, kvm, Paul Mackerras,
linux-pci@vger.kernel.org, qemu-devel, David Gibson, aafabbri,
iommu, Avi Kivity, Anthony Liguori, linuxppc-dev, benve
In-Reply-To: <1314116600.2859.28.camel@bling.home>
On Tue, 2011-08-23 at 10:23 -0600, Alex Williamson wrote:
>
> Yeah. Joerg's idea of binding groups internally (pass the fd of one
> group to another via ioctl) is one option. The tricky part will be
> implementing it to support hot unplug of any group from the
> supergroup.
> I believe Ben had a suggestion that supergroups could be created in
> sysfs, but I don't know what the mechanism to do that looks like. It
> would also be an extra management step to dynamically bind and unbind
> groups to the supergroup around hotplug. Thanks,
I don't really care that much what the method for creating them is, to
be honest, I just prefer this concept of "meta groups" or "super groups"
or "synthetic groups" (whatever you want to name them) to having a
separate uiommu file descriptor.
The one reason I have a slight preference for creating them "statically"
using some kind of separate interface (again, I don't care whether it's
sysfs, netlink, etc...) is that it means things like qemu don't have to
care about them.
In general, apps that want to use vfio can just get passed the path to
such a group or the /dev/ path or the group number (whatever we chose as
the way to identify a group), and don't need to know anything about
"super groups", how to manipulate them, create them, possible
constraints etc...
Now, libvirt might want to know about that other API in order to provide
control on the creation of these things, but that's a different issue.
By "static" I mean they persist, they aren't tied to the lifetime of an
fd.
Now that's purely a preference on my side because I believe it will make
life easier for actual programs wanting to use vfio to not have to care
about those super-groups, but as I said earlier, I don't actually care
that much :-)
Cheers,
Ben.
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: Benjamin Herrenschmidt @ 2011-08-23 23:35 UTC (permalink / raw)
To: Roedel, Joerg
Cc: Alexey Kardashevskiy, kvm@vger.kernel.org, Paul Mackerras,
linux-pci@vger.kernel.org, qemu-devel, iommu, chrisw,
Alex Williamson, Avi Kivity, Anthony Liguori, linuxppc-dev,
benve@cisco.com
In-Reply-To: <20110823131819.GO2079@amd.com>
On Tue, 2011-08-23 at 15:18 +0200, Roedel, Joerg wrote:
> On Mon, Aug 22, 2011 at 05:03:53PM -0400, Benjamin Herrenschmidt wrote:
> >
> > > I am in favour of /dev/vfio/$GROUP. If multiple devices should be
> > > assigned to a guest, there can also be an ioctl to bind a group to an
> > > address-space of another group (certainly needs some care to not allow
> > > that both groups belong to different processes).
> > >
> > > Btw, a problem we havn't talked about yet entirely is
> > > driver-deassignment. User space can decide to de-assign the device from
> > > vfio while a fd is open on it. With PCI there is no way to let this fail
> > > (the .release function returns void last time i checked). Is this a
> > > problem, and yes, how we handle that?
> >
> > We can treat it as a hard unplug (like a cardbus gone away).
> >
> > IE. Dispose of the direct mappings (switch to MMIO emulation) and return
> > all ff's from reads (& ignore writes).
> >
> > Then send an unplug event via whatever mechanism the platform provides
> > (ACPI hotplug controller on x86 for example, we haven't quite sorted out
> > what to do on power for hotplug yet).
>
> Hmm, good idea. But as far as I know the hotplug-event needs to be in
> the guest _before_ the device is actually unplugged (so that the guest
> can unbind its driver first). That somehow brings back the sleep-idea
> and the timeout in the .release function.
That's for normal assisted hotplug, but don't we support hard hotplug ?
I mean, things like cardbus, thunderbolt (if we ever support that)
etc... will need it and some platforms do support hard hotplug of PCIe
devices.
(That's why drivers should never spin on MMIO waiting for a 1 bit to
clear without a timeout :-)
Cheers,
Ben.
^ permalink raw reply
* Re: [Cbe-oss-dev] [PATCH part1 v2 4/9] Add region 1 memory early
From: Antonio Ospite @ 2011-08-23 22:37 UTC (permalink / raw)
To: Geoff Levand; +Cc: cbe-oss-dev, Andre Heider, linuxppc-dev, Hector Martin
In-Reply-To: <4E54135A.7030509@infradead.org>
[-- Attachment #1: Type: text/plain, Size: 1095 bytes --]
On Tue, 23 Aug 2011 13:53:46 -0700
Geoff Levand <geoff@infradead.org> wrote:
[...]
> >
> > + if (!map.r1.size) {
> > + DBG("%s:%d: no region 1, not adding memory\n",
> > + __func__, __LINE__);
> > + } else {
>
> Remove brackets around a single line conditional.
>
> > + DBG("%s:%d: adding memory: start %llxh, size %llxh\n",
> > + __func__, __LINE__, map.rm.size, map.r1.size);
> > +
> > + memblock_add(map.rm.size, map.r1.size);
> > + memblock_analyze();
> > + }
> > +
In Documentation/CodingStyle I read that if [only] one branch is a
single statement then the parenthesis are OK (and even recommended) for
both branches, I guess this is for style consistency. See Chapter 3,
around line 169 on my copy. I guess the wording on that paragraph can
be made more explicit, I'll try to fix that up.
Regards,
Antonio
--
Antonio Ospite
http://ao2.it
PGP public key ID: 0x4553B001
A: Because it messes up the order in which people normally read text.
See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?
[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]
^ permalink raw reply
* Re: [PATCH part1 v2 8/9] ps3flash: Refuse to work in lpars other than OtherOS
From: Geoff Levand @ 2011-08-23 22:12 UTC (permalink / raw)
To: Andre Heider; +Cc: cbe-oss-dev, Hector Martin, linuxppc-dev
In-Reply-To: <1313091073-4572-9-git-send-email-a.heider@gmail.com>
On 08/11/2011 12:31 PM, Andre Heider wrote:
> The driver implements a character and misc device, meant for the
> axed OtherOS to exchange various settings with GameOS.
> Since Firmware 3.21 there is no support anymore to write these
> settings, so test if we're running in OtherOS, and refuse to load
> if that is not the case.
Please see my comments to the v1 patch regarding this text.
> Signed-off-by: Andre Heider <a.heider@gmail.com>
> ---
> arch/powerpc/platforms/ps3/Kconfig | 2 +-
> drivers/char/ps3flash.c | 7 +++++++
> 2 files changed, 8 insertions(+), 1 deletions(-)
>
> diff --git a/arch/powerpc/platforms/ps3/Kconfig b/arch/powerpc/platforms/ps3/Kconfig
> index 84df5c8..72fdecd 100644
> --- a/arch/powerpc/platforms/ps3/Kconfig
> +++ b/arch/powerpc/platforms/ps3/Kconfig
> @@ -121,7 +121,7 @@ config PS3_FLASH
>
> This support is required to access the PS3 FLASH ROM, which
> contains the boot loader and some boot options.
> - In general, all users will say Y or M.
> + In general, all PS3 OtherOS users will say Y or M.
>
> As this driver needs a fixed buffer of 256 KiB of memory, it can
> be disabled on the kernel command line using "ps3flash=off", to
> diff --git a/drivers/char/ps3flash.c b/drivers/char/ps3flash.c
> index d0c57c2..fc6d867 100644
> --- a/drivers/char/ps3flash.c
> +++ b/drivers/char/ps3flash.c
> @@ -25,6 +25,7 @@
>
> #include <asm/lv1call.h>
> #include <asm/ps3stor.h>
> +#include <asm/firmware.h>
>
>
> #define DEVICE_NAME "ps3flash"
> @@ -464,6 +465,12 @@ static struct ps3_system_bus_driver ps3flash = {
>
> static int __init ps3flash_init(void)
> {
> + if (!firmware_has_feature(FW_FEATURE_PS3_LV1))
> + return -ENODEV;
> +
> + if (ps3_get_ss_laid() != PS3_SS_LAID_OTHEROS)
> + return -ENODEV;
> +
> return ps3_system_bus_driver_register(&ps3flash);
> }
>
^ permalink raw reply
* Re: [PATCH part1 v2 6/9] ps3: Detect the current lpar
From: Geoff Levand @ 2011-08-23 22:08 UTC (permalink / raw)
To: Andre Heider; +Cc: cbe-oss-dev, Hector Martin, linuxppc-dev
In-Reply-To: <1313091073-4572-7-git-send-email-a.heider@gmail.com>
Hi,
On 08/11/2011 12:31 PM, Andre Heider wrote:
> Detect it by reading the ss laid repository node, and make it
> accessible via ps3_get_ss_laid().
I'm wondering now if we even need this. It is mainly used by your
later patch 8/9 that modifies ps3flash_init() to test if we should
call ps3_system_bus_driver_register(). If we don't use
ps3_get_ss_laid() and just allow ps3_system_bus_driver_register()
to be called, would the device probe fail and have the same result
as the test?
I would prefer to not have ps3_get_ss_laid() if possible.
-Geoff
^ permalink raw reply
* Re: [PATCH part1 v2 5/9] ps3: MEMORY_HOTPLUG is not a requirement anymore
From: Geoff Levand @ 2011-08-23 20:53 UTC (permalink / raw)
To: Andre Heider; +Cc: cbe-oss-dev, Hector Martin, linuxppc-dev
In-Reply-To: <1313091073-4572-6-git-send-email-a.heider@gmail.com>
On 08/11/2011 12:31 PM, Andre Heider wrote:
> Signed-off-by: Andre Heider <a.heider@gmail.com>
> ---
> arch/powerpc/platforms/ps3/Kconfig | 1 -
> 1 files changed, 0 insertions(+), 1 deletions(-)
>
> diff --git a/arch/powerpc/platforms/ps3/Kconfig b/arch/powerpc/platforms/ps3/Kconfig
> index 476d9d9..84df5c8 100644
> --- a/arch/powerpc/platforms/ps3/Kconfig
> +++ b/arch/powerpc/platforms/ps3/Kconfig
> @@ -7,7 +7,6 @@ config PPC_PS3
> select USB_OHCI_BIG_ENDIAN_MMIO
> select USB_ARCH_HAS_EHCI
> select USB_EHCI_BIG_ENDIAN_MMIO
> - select MEMORY_HOTPLUG
This one is OK, once the others are fixed up.
-Geoff
^ permalink raw reply
* Re: [PATCH part1 v2 4/9] Add region 1 memory early
From: Geoff Levand @ 2011-08-23 20:53 UTC (permalink / raw)
To: Andre Heider; +Cc: cbe-oss-dev, Hector Martin, linuxppc-dev
In-Reply-To: <1313091073-4572-5-git-send-email-a.heider@gmail.com>
On 08/11/2011 12:31 PM, Andre Heider wrote:
> From: Hector Martin <hector@marcansoft.com>
>
> Real mode memory can be limited and runs out quickly as memory is
> allocated during kernel startup.
> Having region1 available sooner fixes this.
>
> Signed-off-by: Hector Martin <hector@marcansoft.com>
> [a.heider: Various cleanups to make checkpatch.pl happy]
> Signed-off-by: Andre Heider <a.heider@gmail.com>
> ---
> arch/powerpc/platforms/ps3/mm.c | 75 +++++++--------------------------------
> 1 files changed, 13 insertions(+), 62 deletions(-)
>
> diff --git a/arch/powerpc/platforms/ps3/mm.c b/arch/powerpc/platforms/ps3/mm.c
> index 983b719..68b3879 100644
> --- a/arch/powerpc/platforms/ps3/mm.c
> +++ b/arch/powerpc/platforms/ps3/mm.c
> @@ -20,7 +20,6 @@
>
> #include <linux/kernel.h>
> #include <linux/module.h>
> -#include <linux/memory_hotplug.h>
> #include <linux/memblock.h>
> #include <linux/slab.h>
>
> @@ -94,10 +93,8 @@ struct mem_region {
> * @vas_id - HV virtual address space id
> * @htab_size: htab size in bytes
> *
> - * The HV virtual address space (vas) allows for hotplug memory regions.
> - * Memory regions can be created and destroyed in the vas at runtime.
This is still true, so we should keep these comments. We
are only changing the way we use the feature.
> * @rm: real mode (bootmem) region
> - * @r1: hotplug memory region(s)
> + * @r1: high memory region
high memory region(s)
> *
> * ps3 addresses
> * virt_addr: a cpu 'translated' effective address
> @@ -223,10 +220,6 @@ void ps3_mm_vas_destroy(void)
> }
> }
>
> -/*============================================================================*/
> -/* memory hotplug routines */
> -/*============================================================================*/
> -
> /**
> * ps3_mm_region_create - create a memory region in the vas
> * @r: pointer to a struct mem_region to accept initialized values
> @@ -319,57 +312,6 @@ zero_region:
> return result;
> }
>
> -/**
> - * ps3_mm_add_memory - hot add memory
> - */
> -
> -static int __init ps3_mm_add_memory(void)
> -{
> - int result;
> - unsigned long start_addr;
> - unsigned long start_pfn;
> - unsigned long nr_pages;
> -
> - if (!firmware_has_feature(FW_FEATURE_PS3_LV1))
> - return -ENODEV;
> -
> - BUG_ON(!mem_init_done);
> -
> - if (!map.r1.size) {
> - DBG("%s:%d: no region 1, not adding memory\n",
> - __func__, __LINE__);
> - return 0;
> - }
> -
> - start_addr = map.rm.size;
> - start_pfn = start_addr >> PAGE_SHIFT;
> - nr_pages = (map.r1.size + PAGE_SIZE - 1) >> PAGE_SHIFT;
> -
> - DBG("%s:%d: start_addr %lxh, start_pfn %lxh, nr_pages %lxh\n",
> - __func__, __LINE__, start_addr, start_pfn, nr_pages);
> -
> - result = add_memory(0, start_addr, map.r1.size);
> -
> - if (result) {
> - pr_err("%s:%d: add_memory failed: (%d)\n",
> - __func__, __LINE__, result);
> - return result;
> - }
> -
> - memblock_add(start_addr, map.r1.size);
> - memblock_analyze();
> -
> - result = online_pages(start_pfn, nr_pages);
> -
> - if (result)
> - pr_err("%s:%d: online_pages failed: (%d)\n",
> - __func__, __LINE__, result);
> -
> - return result;
> -}
> -
> -device_initcall(ps3_mm_add_memory);
> -
> /*============================================================================*/
> /* dma routines */
> /*============================================================================*/
> @@ -1256,14 +1198,23 @@ void __init ps3_mm_init(void)
> BUG_ON(!map.rm.size);
>
> /* check if we got the highmem region from an earlier boot step */
> - if (ps3_mm_get_repository_highmem(&map.r1)) {
> - /* arrange to do this in ps3_mm_add_memory */
> + if (ps3_mm_get_repository_highmem(&map.r1))
> ps3_mm_region_create(&map.r1, map.total - map.rm.size);
> - }
This should be folded into patch #3.
>
> /* correct map.total for the real total amount of memory we use */
> map.total = map.rm.size + map.r1.size;
>
> + if (!map.r1.size) {
> + DBG("%s:%d: no region 1, not adding memory\n",
> + __func__, __LINE__);
> + } else {
Remove brackets around a single line conditional.
> + DBG("%s:%d: adding memory: start %llxh, size %llxh\n",
> + __func__, __LINE__, map.rm.size, map.r1.size);
> +
> + memblock_add(map.rm.size, map.r1.size);
> + memblock_analyze();
> + }
> +
> DBG(" <- %s:%d\n", __func__, __LINE__);
> }
>
^ permalink raw reply
* Re: [PATCH part1 v2 3/9] ps3: Get lv1 high memory region from the repository
From: Geoff Levand @ 2011-08-23 20:53 UTC (permalink / raw)
To: Andre Heider; +Cc: cbe-oss-dev, Hector Martin, linuxppc-dev
In-Reply-To: <1313091073-4572-4-git-send-email-a.heider@gmail.com>
On 08/11/2011 12:31 PM, Andre Heider wrote:
> This lets the bootloader preallocate the high lv1 region and pass its
> location to the kernel through the repository. Thus, it can be used to
> hold the initrd. If the region info doesn't exist, the kernel retains
> the old behavior and attempts to allocate the region itself.
>
> Based on the patch
> "[PS3] Get lv1 high memory region from devtree"
> from Hector Martin <hector@marcansoft.com>
>
> Signed-off-by: Andre Heider <a.heider@gmail.com>
> ---
> arch/powerpc/platforms/ps3/mm.c | 46 ++++++++++++++++++++++++++++++++++++--
> 1 files changed, 43 insertions(+), 3 deletions(-)
>
> diff --git a/arch/powerpc/platforms/ps3/mm.c b/arch/powerpc/platforms/ps3/mm.c
> index c204588..983b719 100644
> --- a/arch/powerpc/platforms/ps3/mm.c
> +++ b/arch/powerpc/platforms/ps3/mm.c
> @@ -78,12 +78,14 @@ enum {
> * @base: base address
> * @size: size in bytes
> * @offset: difference between base and rm.size
> + * @destroy: flag if region should be destroyed upon shutdown
> */
>
> struct mem_region {
> u64 base;
> u64 size;
> unsigned long offset;
> + int destroy;
> };
>
> /**
> @@ -261,6 +263,7 @@ static int ps3_mm_region_create(struct mem_region *r, unsigned long size)
> goto zero_region;
> }
>
> + r->destroy = 1;
> r->offset = r->base - map.rm.size;
> return result;
>
> @@ -279,6 +282,12 @@ static void ps3_mm_region_destroy(struct mem_region *r)
> int result;
>
> DBG("%s:%d: r->base = %llxh\n", __func__, __LINE__, r->base);
> +
> + if (!r->destroy) {
> + DBG("%s:%d: not destroying region\n", __func__, __LINE__);
> + return;
> + }
> +
> if (r->base) {
> result = lv1_release_memory(r->base);
> BUG_ON(result);
> @@ -287,6 +296,29 @@ static void ps3_mm_region_destroy(struct mem_region *r)
> }
> }
>
> +static int ps3_mm_get_repository_highmem(struct mem_region *r)
> +{
> + int result = ps3_repository_read_highmem_info(&r->base, &r->size);
> +
> + if (result)
> + goto zero_region;
> +
> + if (!r->base || !r->size) {
> + result = -1;
> + goto zero_region;
> + }
> +
> + r->offset = r->base - map.rm.size;
> + DBG("%s:%d got high region from repository: %llxh %llxh\n",
> + __func__, __LINE__, r->base, r->size);
> + return 0;
> +
> +zero_region:
> + DBG("%s:%d no high region in repository...\n", __func__, __LINE__);
Three dots implies something more is on its way. I
don't think we need them.
> + r->size = r->base = r->offset = 0;
> + return result;
> +}
> +
> /**
> * ps3_mm_add_memory - hot add memory
> */
> @@ -303,6 +335,12 @@ static int __init ps3_mm_add_memory(void)
>
> BUG_ON(!mem_init_done);
>
> + if (!map.r1.size) {
> + DBG("%s:%d: no region 1, not adding memory\n",
> + __func__, __LINE__);
> + return 0;
> + }
> +
> start_addr = map.rm.size;
> start_pfn = start_addr >> PAGE_SHIFT;
> nr_pages = (map.r1.size + PAGE_SIZE - 1) >> PAGE_SHIFT;
> @@ -1217,9 +1255,11 @@ void __init ps3_mm_init(void)
> BUG_ON(map.rm.base);
> BUG_ON(!map.rm.size);
>
> -
> - /* arrange to do this in ps3_mm_add_memory */
> - ps3_mm_region_create(&map.r1, map.total - map.rm.size);
> + /* check if we got the highmem region from an earlier boot step */
> + if (ps3_mm_get_repository_highmem(&map.r1)) {
> + /* arrange to do this in ps3_mm_add_memory */
> + ps3_mm_region_create(&map.r1, map.total - map.rm.size);
> + }
>
> /* correct map.total for the real total amount of memory we use */
> map.total = map.rm.size + map.r1.size;
^ permalink raw reply
* Re: [PATCH part1 v2 2/9] ps3: Add helper functions to read highmem info from the repository
From: Geoff Levand @ 2011-08-23 20:53 UTC (permalink / raw)
To: Andre Heider; +Cc: cbe-oss-dev, Hector Martin, linuxppc-dev
In-Reply-To: <1313091073-4572-3-git-send-email-a.heider@gmail.com>
On 08/11/2011 12:31 PM, Andre Heider wrote:
> An earlier step in the boot chain can preallocate the highmem region.
> A boot loader doing so will place the region infos in the repository.
> Provide helper functions to read the required nodes.
>
> Signed-off-by: Andre Heider <a.heider@gmail.com>
> ---
> arch/powerpc/platforms/ps3/platform.h | 3 ++
> arch/powerpc/platforms/ps3/repository.c | 36 +++++++++++++++++++++++++++++++
> 2 files changed, 39 insertions(+), 0 deletions(-)
>
> diff --git a/arch/powerpc/platforms/ps3/platform.h b/arch/powerpc/platforms/ps3/platform.h
> index 9a196a8..d9b4ec0 100644
> --- a/arch/powerpc/platforms/ps3/platform.h
> +++ b/arch/powerpc/platforms/ps3/platform.h
> @@ -187,6 +187,9 @@ int ps3_repository_read_rm_size(unsigned int ppe_id, u64 *rm_size);
> int ps3_repository_read_region_total(u64 *region_total);
> int ps3_repository_read_mm_info(u64 *rm_base, u64 *rm_size,
> u64 *region_total);
> +int ps3_repository_read_highmem_base(u64 *highmem_base);
> +int ps3_repository_read_highmem_size(u64 *highmem_size);
> +int ps3_repository_read_highmem_info(u64 *highmem_base, u64 *highmem_size);
In the general case we could have multiple regions. If we
add a region arg here we can handle that if needed.
region_index would be {1..}. ps3_repository_read_highmem_info
could hold how many regions, so:
int ps3_repository_read_highmem_base(unsigned int region_index, u64 *highmem_base);
int ps3_repository_read_highmem_size(unsigned int region_index, u64 *highmem_size);
int ps3_repository_read_highmem_region(unsigned int region_index, u64 *highmem_base, u64 *highmem_size);
>
> /* repository pme info */
>
> diff --git a/arch/powerpc/platforms/ps3/repository.c b/arch/powerpc/platforms/ps3/repository.c
> index 5e304c2..9908d61 100644
> --- a/arch/powerpc/platforms/ps3/repository.c
> +++ b/arch/powerpc/platforms/ps3/repository.c
> @@ -778,6 +778,42 @@ int ps3_repository_read_mm_info(u64 *rm_base, u64 *rm_size, u64 *region_total)
> : ps3_repository_read_region_total(region_total);
> }
>
> +int ps3_repository_read_highmem_base(u64 *highmem_base)
> +{
> + return read_node(PS3_LPAR_ID_CURRENT,
> + make_first_field("bi", 0),
> + make_field("highmem", 0),
> + make_field("base", 0),
> + 0,
> + highmem_base, NULL);
> +}
I think something like this seems better:
int ps3_repository_read_highmem_base(unsigned int region_index, u64 *highmem_base)
{
return read_node(PS3_LPAR_ID_CURRENT,
make_first_field("highmem", 0),
make_field("region", region_index),
make_field("base", 0),
0,
highmem_base, NULL);
}
> +
> +int ps3_repository_read_highmem_size(u64 *highmem_size)
> +{
> + return read_node(PS3_LPAR_ID_CURRENT,
> + make_first_field("bi", 0),
> + make_field("highmem", 0),
> + make_field("size", 0),
> + 0,
> + highmem_size, NULL);
> +}
> +
> +/**
> + * ps3_repository_read_highmem_info - Read high memory info
> + * @highmem_base: High memory base address.
> + * @highmem_size: High mode memory size.
> + */
> +
> +int ps3_repository_read_highmem_info(u64 *highmem_base, u64 *highmem_size)
> +{
> + int result;
> +
> + *highmem_base = 0;
> + result = ps3_repository_read_highmem_base(highmem_base);
> + return result ? result
> + : ps3_repository_read_highmem_size(highmem_size);
ps3_repository_read_highmem_base(1, highmem_base);
...
> +}
> +
> /**
> * ps3_repository_read_num_spu_reserved - Number of physical spus reserved.
> * @num_spu: Number of physical spus.
-Geoff
^ permalink raw reply
* Re: [PATCH part1 v2 1/9] Add udbg driver using the PS3 gelic Ethernet device
From: Geoff Levand @ 2011-08-23 20:53 UTC (permalink / raw)
To: Andre Heider; +Cc: cbe-oss-dev, Hector Martin, linuxppc-dev
In-Reply-To: <1313091073-4572-2-git-send-email-a.heider@gmail.com>
Hi,
We had some questions as to why we have this totally separate driver
from the gelic driver, so I think it worthwhile to have an
explanation of why in the commit log. Otherwise, the code looks
OK.
-Geoff
> From: Hector Martin <hector@marcansoft.com>
>
> Signed-off-by: Hector Martin <hector@marcansoft.com>
> [a.heider: Various cleanups to make checkpatch.pl happy]
> Signed-off-by: Andre Heider <a.heider@gmail.com>
> ---
> arch/powerpc/Kconfig.debug | 8 +
> arch/powerpc/include/asm/udbg.h | 1 +
> arch/powerpc/kernel/udbg.c | 2 +
> arch/powerpc/platforms/ps3/Kconfig | 12 ++
> arch/powerpc/platforms/ps3/Makefile | 1 +
> arch/powerpc/platforms/ps3/gelic_udbg.c | 273 +++++++++++++++++++++++++++++++
> drivers/net/ps3_gelic_net.c | 3 +
> drivers/net/ps3_gelic_net.h | 6 +
> 8 files changed, 306 insertions(+), 0 deletions(-)
> create mode 100644 arch/powerpc/platforms/ps3/gelic_udbg.c
>
> diff --git a/arch/powerpc/Kconfig.debug b/arch/powerpc/Kconfig.debug
> index 067cb84..ab2335f 100644
> --- a/arch/powerpc/Kconfig.debug
> +++ b/arch/powerpc/Kconfig.debug
> @@ -258,6 +258,14 @@ config PPC_EARLY_DEBUG_WSP
> depends on PPC_WSP
> select PPC_UDBG_16550
>
> +config PPC_EARLY_DEBUG_PS3GELIC
> + bool "Early debugging through the PS3 Ethernet port"
> + depends on PPC_PS3
> + select PS3GELIC_UDBG
> + help
> + Select this to enable early debugging for the PlayStation3 via
> + UDP broadcasts sent out through the Ethernet port.
> +
> endchoice
>
> config PPC_EARLY_DEBUG_HVSI_VTERMNO
> diff --git a/arch/powerpc/include/asm/udbg.h b/arch/powerpc/include/asm/udbg.h
> index 93e05d1..7cf796f 100644
> --- a/arch/powerpc/include/asm/udbg.h
> +++ b/arch/powerpc/include/asm/udbg.h
> @@ -54,6 +54,7 @@ extern void __init udbg_init_40x_realmode(void);
> extern void __init udbg_init_cpm(void);
> extern void __init udbg_init_usbgecko(void);
> extern void __init udbg_init_wsp(void);
> +extern void __init udbg_init_ps3gelic(void);
>
> #endif /* __KERNEL__ */
> #endif /* _ASM_POWERPC_UDBG_H */
> diff --git a/arch/powerpc/kernel/udbg.c b/arch/powerpc/kernel/udbg.c
> index faa82c1..5b3e98e 100644
> --- a/arch/powerpc/kernel/udbg.c
> +++ b/arch/powerpc/kernel/udbg.c
> @@ -67,6 +67,8 @@ void __init udbg_early_init(void)
> udbg_init_usbgecko();
> #elif defined(CONFIG_PPC_EARLY_DEBUG_WSP)
> udbg_init_wsp();
> +#elif defined(CONFIG_PPC_EARLY_DEBUG_PS3GELIC)
> + udbg_init_ps3gelic();
> #endif
>
> #ifdef CONFIG_PPC_EARLY_DEBUG
> diff --git a/arch/powerpc/platforms/ps3/Kconfig b/arch/powerpc/platforms/ps3/Kconfig
> index dfe316b..476d9d9 100644
> --- a/arch/powerpc/platforms/ps3/Kconfig
> +++ b/arch/powerpc/platforms/ps3/Kconfig
> @@ -148,4 +148,16 @@ config PS3_LPM
> profiling support of the Cell processor with programs like
> oprofile and perfmon2, then say Y or M, otherwise say N.
>
> +config PS3GELIC_UDBG
> + bool "PS3 udbg output via UDP broadcasts on Ethernet"
> + depends on PPC_PS3
> + help
> + Enables udbg early debugging output by sending broadcast UDP
> + via the Ethernet port (UDP port number 18194).
> +
> + This driver uses a trivial implementation and is independent
> + from the main network driver.
> +
> + If in doubt, say N here.
> +
> endmenu
> diff --git a/arch/powerpc/platforms/ps3/Makefile b/arch/powerpc/platforms/ps3/Makefile
> index ac1bdf8..02b9e63 100644
> --- a/arch/powerpc/platforms/ps3/Makefile
> +++ b/arch/powerpc/platforms/ps3/Makefile
> @@ -2,6 +2,7 @@ obj-y += setup.o mm.o time.o hvcall.o htab.o repository.o
> obj-y += interrupt.o exports.o os-area.o
> obj-y += system-bus.o
>
> +obj-$(CONFIG_PS3GELIC_UDBG) += gelic_udbg.o
> obj-$(CONFIG_SMP) += smp.o
> obj-$(CONFIG_SPU_BASE) += spu.o
> obj-y += device-init.o
> diff --git a/arch/powerpc/platforms/ps3/gelic_udbg.c b/arch/powerpc/platforms/ps3/gelic_udbg.c
> new file mode 100644
> index 0000000..20b46a1
> --- /dev/null
> +++ b/arch/powerpc/platforms/ps3/gelic_udbg.c
> @@ -0,0 +1,273 @@
> +/*
> + * udbg debug output routine via GELIC UDP broadcasts
> + *
> + * Copyright (C) 2007 Sony Computer Entertainment Inc.
> + * Copyright 2006, 2007 Sony Corporation
> + * Copyright (C) 2010 Hector Martin <hector@marcansoft.com>
> + * Copyright (C) 2011 Andre Heider <a.heider@gmail.com>
> + *
> + * This program is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU General Public License
> + * as published by the Free Software Foundation; either version 2
> + * of the License, or (at your option) any later version.
> + *
> + */
> +
> +#include <asm/io.h>
> +#include <asm/udbg.h>
> +#include <asm/lv1call.h>
> +
> +#define GELIC_BUS_ID 1
> +#define GELIC_DEVICE_ID 0
> +#define GELIC_DEBUG_PORT 18194
> +#define GELIC_MAX_MESSAGE_SIZE 1000
> +
> +#define GELIC_LV1_GET_MAC_ADDRESS 1
> +#define GELIC_LV1_GET_VLAN_ID 4
> +#define GELIC_LV1_VLAN_TX_ETHERNET_0 2
> +
> +#define GELIC_DESCR_DMA_STAT_MASK 0xf0000000
> +#define GELIC_DESCR_DMA_CARDOWNED 0xa0000000
> +
> +#define GELIC_DESCR_TX_DMA_IKE 0x00080000
> +#define GELIC_DESCR_TX_DMA_NO_CHKSUM 0x00000000
> +#define GELIC_DESCR_TX_DMA_FRAME_TAIL 0x00040000
> +
> +#define GELIC_DESCR_DMA_CMD_NO_CHKSUM (GELIC_DESCR_DMA_CARDOWNED | \
> + GELIC_DESCR_TX_DMA_IKE | \
> + GELIC_DESCR_TX_DMA_NO_CHKSUM)
> +
> +static u64 bus_addr;
> +
> +struct gelic_descr {
> + /* as defined by the hardware */
> + __be32 buf_addr;
> + __be32 buf_size;
> + __be32 next_descr_addr;
> + __be32 dmac_cmd_status;
> + __be32 result_size;
> + __be32 valid_size; /* all zeroes for tx */
> + __be32 data_status;
> + __be32 data_error; /* all zeroes for tx */
> +} __attribute__((aligned(32)));
> +
> +struct debug_block {
> + struct gelic_descr descr;
> + u8 pkt[1520];
> +} __packed;
> +
> +struct ethhdr {
> + u8 dest[6];
> + u8 src[6];
> + u16 type;
> +} __packed;
> +
> +struct vlantag {
> + u16 vlan;
> + u16 subtype;
> +} __packed;
> +
> +struct iphdr {
> + u8 ver_len;
> + u8 dscp_ecn;
> + u16 total_length;
> + u16 ident;
> + u16 frag_off_flags;
> + u8 ttl;
> + u8 proto;
> + u16 checksum;
> + u32 src;
> + u32 dest;
> +} __packed;
> +
> +struct udphdr {
> + u16 src;
> + u16 dest;
> + u16 len;
> + u16 checksum;
> +} __packed;
> +
> +static __iomem struct ethhdr *h_eth;
> +static __iomem struct vlantag *h_vlan;
> +static __iomem struct iphdr *h_ip;
> +static __iomem struct udphdr *h_udp;
> +
> +static __iomem char *pmsg;
> +static __iomem char *pmsgc;
> +
> +static __iomem struct debug_block dbg __attribute__((aligned(32)));
> +
> +static int header_size;
> +
> +static void map_dma_mem(int bus_id, int dev_id, void *start, size_t len,
> + u64 *real_bus_addr)
> +{
> + s64 result;
> + u64 real_addr = ((u64)start) & 0x0fffffffffffffffUL;
> + u64 real_end = real_addr + len;
> + u64 map_start = real_addr & ~0xfff;
> + u64 map_end = (real_end + 0xfff) & ~0xfff;
> + u64 bus_addr = 0;
> +
> + u64 flags = 0xf800000000000000UL;
> +
> + result = lv1_allocate_device_dma_region(bus_id, dev_id,
> + map_end - map_start, 12, 0,
> + &bus_addr);
> + if (result)
> + lv1_panic(0);
> +
> + result = lv1_map_device_dma_region(bus_id, dev_id, map_start,
> + bus_addr, map_end - map_start,
> + flags);
> + if (result)
> + lv1_panic(0);
> +
> + *real_bus_addr = bus_addr + real_addr - map_start;
> +}
> +
> +static int unmap_dma_mem(int bus_id, int dev_id, u64 bus_addr, size_t len)
> +{
> + s64 result;
> + u64 real_bus_addr;
> +
> + real_bus_addr = bus_addr & ~0xfff;
> + len += bus_addr - real_bus_addr;
> + len = (len + 0xfff) & ~0xfff;
> +
> + result = lv1_unmap_device_dma_region(bus_id, dev_id, real_bus_addr,
> + len);
> + if (result)
> + return result;
> +
> + return lv1_free_device_dma_region(bus_id, dev_id, real_bus_addr);
> +}
> +
> +static void gelic_debug_init(void)
> +{
> + s64 result;
> + u64 v2;
> + u64 mac;
> + u64 vlan_id;
> +
> + result = lv1_open_device(GELIC_BUS_ID, GELIC_DEVICE_ID, 0);
> + if (result)
> + lv1_panic(0);
> +
> + map_dma_mem(GELIC_BUS_ID, GELIC_DEVICE_ID, &dbg, sizeof(dbg),
> + &bus_addr);
> +
> + memset(&dbg, 0, sizeof(dbg));
> +
> + dbg.descr.buf_addr = bus_addr + offsetof(struct debug_block, pkt);
> +
> + wmb();
> +
> + result = lv1_net_control(GELIC_BUS_ID, GELIC_DEVICE_ID,
> + GELIC_LV1_GET_MAC_ADDRESS, 0, 0, 0,
> + &mac, &v2);
> + if (result)
> + lv1_panic(0);
> +
> + mac <<= 16;
> +
> + h_eth = (struct ethhdr *)dbg.pkt;
> +
> + memset(&h_eth->dest, 0xff, 6);
> + memcpy(&h_eth->src, &mac, 6);
> +
> + header_size = sizeof(struct ethhdr);
> +
> + result = lv1_net_control(GELIC_BUS_ID, GELIC_DEVICE_ID,
> + GELIC_LV1_GET_VLAN_ID,
> + GELIC_LV1_VLAN_TX_ETHERNET_0, 0, 0,
> + &vlan_id, &v2);
> + if (!result) {
> + h_eth->type = 0x8100;
> +
> + header_size += sizeof(struct vlantag);
> + h_vlan = (struct vlantag *)(h_eth + 1);
> + h_vlan->vlan = vlan_id;
> + h_vlan->subtype = 0x0800;
> + h_ip = (struct iphdr *)(h_vlan + 1);
> + } else {
> + h_eth->type = 0x0800;
> + h_ip = (struct iphdr *)(h_eth + 1);
> + }
> +
> + header_size += sizeof(struct iphdr);
> + h_ip->ver_len = 0x45;
> + h_ip->ttl = 10;
> + h_ip->proto = 0x11;
> + h_ip->src = 0x00000000;
> + h_ip->dest = 0xffffffff;
> +
> + header_size += sizeof(struct udphdr);
> + h_udp = (struct udphdr *)(h_ip + 1);
> + h_udp->src = GELIC_DEBUG_PORT;
> + h_udp->dest = GELIC_DEBUG_PORT;
> +
> + pmsgc = pmsg = (char *)(h_udp + 1);
> +}
> +
> +static void gelic_debug_shutdown(void)
> +{
> + if (bus_addr)
> + unmap_dma_mem(GELIC_BUS_ID, GELIC_DEVICE_ID,
> + bus_addr, sizeof(dbg));
> + lv1_close_device(GELIC_BUS_ID, GELIC_DEVICE_ID);
> +}
> +
> +static void gelic_sendbuf(int msgsize)
> +{
> + u16 *p;
> + u32 sum;
> + int i;
> +
> + dbg.descr.buf_size = header_size + msgsize;
> + h_ip->total_length = msgsize + sizeof(struct udphdr) +
> + sizeof(struct iphdr);
> + h_udp->len = msgsize + sizeof(struct udphdr);
> +
> + h_ip->checksum = 0;
> + sum = 0;
> + p = (u16 *)h_ip;
> + for (i = 0; i < 5; i++)
> + sum += *p++;
> + h_ip->checksum = ~(sum + (sum >> 16));
> +
> + dbg.descr.dmac_cmd_status = GELIC_DESCR_DMA_CMD_NO_CHKSUM |
> + GELIC_DESCR_TX_DMA_FRAME_TAIL;
> + dbg.descr.result_size = 0;
> + dbg.descr.data_status = 0;
> +
> + wmb();
> +
> + lv1_net_start_tx_dma(GELIC_BUS_ID, GELIC_DEVICE_ID, bus_addr, 0);
> +
> + while ((dbg.descr.dmac_cmd_status & GELIC_DESCR_DMA_STAT_MASK) ==
> + GELIC_DESCR_DMA_CARDOWNED)
> + cpu_relax();
> +}
> +
> +static void ps3gelic_udbg_putc(char ch)
> +{
> + *pmsgc++ = ch;
> + if (ch == '\n' || (pmsgc-pmsg) >= GELIC_MAX_MESSAGE_SIZE) {
> + gelic_sendbuf(pmsgc-pmsg);
> + pmsgc = pmsg;
> + }
> +}
> +
> +void __init udbg_init_ps3gelic(void)
> +{
> + gelic_debug_init();
> + udbg_putc = ps3gelic_udbg_putc;
> +}
> +
> +void udbg_shutdown_ps3gelic(void)
> +{
> + udbg_putc = NULL;
> + gelic_debug_shutdown();
> +}
> +EXPORT_SYMBOL(udbg_shutdown_ps3gelic);
> diff --git a/drivers/net/ps3_gelic_net.c b/drivers/net/ps3_gelic_net.c
> index d82a82d..e743c94 100644
> --- a/drivers/net/ps3_gelic_net.c
> +++ b/drivers/net/ps3_gelic_net.c
> @@ -1674,6 +1674,9 @@ static int __devinit ps3_gelic_driver_probe(struct ps3_system_bus_device *dev)
> int result;
>
> pr_debug("%s: called\n", __func__);
> +
> + udbg_shutdown_ps3gelic();
> +
> result = ps3_open_hv_device(dev);
>
> if (result) {
> diff --git a/drivers/net/ps3_gelic_net.h b/drivers/net/ps3_gelic_net.h
> index d3fadfb..a93df6a 100644
> --- a/drivers/net/ps3_gelic_net.h
> +++ b/drivers/net/ps3_gelic_net.h
> @@ -359,6 +359,12 @@ static inline void *port_priv(struct gelic_port *port)
> return port->priv;
> }
>
> +#ifdef CONFIG_PPC_EARLY_DEBUG_PS3GELIC
> +extern void udbg_shutdown_ps3gelic(void);
> +#else
> +static inline void udbg_shutdown_ps3gelic(void) {}
> +#endif
> +
> extern int gelic_card_set_irq_mask(struct gelic_card *card, u64 mask);
> /* shared netdev ops */
> extern void gelic_card_up(struct gelic_card *card);
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: Alex Williamson @ 2011-08-23 19:30 UTC (permalink / raw)
To: Benjamin Herrenschmidt
Cc: aafabbri, Alexey Kardashevskiy, kvm, Paul Mackerras,
linux-pci@vger.kernel.org, qemu-devel, David Gibson, chrisw,
iommu, Avi Kivity, Anthony Liguori, linuxppc-dev, benve
In-Reply-To: <1314046904.7662.37.camel@pasglop>
On Tue, 2011-08-23 at 07:01 +1000, Benjamin Herrenschmidt wrote:
> On Mon, 2011-08-22 at 09:45 -0600, Alex Williamson wrote:
>
> > Yes, that's the idea. An open question I have towards the configuration
> > side is whether we might add iommu driver specific options to the
> > groups. For instance on x86 where we typically have B:D.F granularity,
> > should we have an option not to trust multi-function devices and use a
> > B:D granularity for grouping?
>
> Or even B or range of busses... if you want to enforce strict isolation
> you really can't trust anything below a bus level :-)
>
> > Right, we can also combine models. Binding a device to vfio
> > creates /dev/vfio$GROUP, which only allows a subset of ioctls and no
> > device access until all the group devices are also bound. I think
> > the /dev/vfio/$GROUP might help provide an enumeration interface as well
> > though, which could be useful.
>
> Could be tho in what form ? returning sysfs pathes ?
I'm at a loss there, please suggest. I think we need an ioctl that
returns some kind of array of devices within the group and another that
maybe takes an index from that array and returns an fd for that device.
A sysfs path string might be a reasonable array element, but it sounds
like a pain to work with.
> > 1:1 group<->process is probably too strong. Not allowing concurrent
> > open()s on the group file enforces a single userspace entity is
> > responsible for that group. Device fds can be passed to other
> > processes, but only retrieved via the group fd. I suppose we could even
> > branch off the dma interface into a different fd, but it seems like we
> > would logically want to serialize dma mappings at each iommu group
> > anyway. I'm open to alternatives, this just seemed an easy way to do
> > it. Restricting on UID implies that we require isolated qemu instances
> > to run as different UIDs. I know that's a goal, but I don't know if we
> > want to make it an assumption in the group security model.
>
> 1:1 process has the advantage of linking to an -mm which makes the whole
> mmu notifier business doable. How do you want to track down mappings and
> do the second level translation in the case of explicit map/unmap (like
> on power) if you are not tied to an mm_struct ?
Right, I threw away the mmu notifier code that was originally part of
vfio because we can't do anything useful with it yet on x86. I
definitely don't want to prevent it where it makes sense though. Maybe
we just record current->mm on open and restrict subsequent opens to the
same.
> > Yes. I'm not sure there's a good ROI to prioritize that model. We have
> > to assume >1 device per guest is a typical model and that the iotlb is
> > large enough that we might improve thrashing to see both a resource and
> > performance benefit from it. I'm open to suggestions for how we could
> > include it though.
>
> Sharing may or may not be possible depending on setups so yes, it's a
> bit tricky.
>
> My preference is to have a static interface (and that's actually where
> your pet netlink might make some sense :-) to create "synthetic" groups
> made of other groups if the arch allows it. But that might not be the
> best approach. In another email I also proposed an option for a group to
> "capture" another one...
I already made some comments on this in a different thread, so I won't
repeat here.
> > > If that's
> > > not what you're saying, how would the domains - now made up of a
> > > user's selection of groups, rather than individual devices - be
> > > configured?
> > >
> > > > Hope that captures it, feel free to jump in with corrections and
> > > > suggestions. Thanks,
> > >
>
> Another aspect I don't see discussed is how we represent these things to
> the guest.
>
> On Power for example, I have a requirement that a given iommu domain is
> represented by a single dma window property in the device-tree. What
> that means is that that property needs to be either in the node of the
> device itself if there's only one device in the group or in a parent
> node (ie a bridge or host bridge) if there are multiple devices.
>
> Now I do -not- want to go down the path of simulating P2P bridges,
> besides we'll quickly run out of bus numbers if we go there.
>
> For us the most simple and logical approach (which is also what pHyp
> uses and what Linux handles well) is really to expose a given PCI host
> bridge per group to the guest. Believe it or not, it makes things
> easier :-)
I'm all for easier. Why does exposing the bridge use less bus numbers
than emulating a bridge?
On x86, I want to maintain that our default assignment is at the device
level. A user should be able to pick single or multiple devices from
across several groups and have them all show up as individual,
hotpluggable devices on bus 0 in the guest. Not surprisingly, we've
also seen cases where users try to attach a bridge to the guest,
assuming they'll get all the devices below the bridge, so I'd be in
favor of making this "just work" if possible too, though we may have to
prevent hotplug of those.
Given the device requirement on x86 and since everything is a PCI device
on x86, I'd like to keep a qemu command line something like -device
vfio,host=00:19.0. I assume that some of the iommu properties, such as
dma window size/address, will be query-able through an architecture
specific (or general if possible) ioctl on the vfio group fd. I hope
that will help the specification, but I don't fully understand what all
remains. Thanks,
Alex
^ permalink raw reply
* Re: [PATCH 8/9] arch/powerpc/sysdev/ehv_pic.c: add missing kfree
From: Timur Tabi @ 2011-08-23 19:04 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Kumar Gala
Cc: kernel-janitors, Julia Lawall, linuxppc-dev, Paul Mackerras,
linux-kernel
In-Reply-To: <CAOZdJXX_BS7GMtmyqiS91QByhrjANtKpbqrstR8g-L80r_N8qg@mail.gmail.com>
Ben, Kumar, can one of you take a look at my question and help me out?
wrote:
> On Mon, Aug 8, 2011 at 7:18 AM, Julia Lawall <julia@diku.dk> wrote:
>
>> diff --git a/arch/powerpc/sysdev/ehv_pic.c b/arch/powerpc/sysdev/ehv_pic.c
>> index af1a5df..b6731e4 100644
>> --- a/arch/powerpc/sysdev/ehv_pic.c
>> +++ b/arch/powerpc/sysdev/ehv_pic.c
>> @@ -280,6 +280,7 @@ void __init ehv_pic_init(void)
>>
>> if (!ehv_pic->irqhost) {
>> of_node_put(np);
>> + kfree(ehv_pic);
>> return;
>> }
>
> Although the fix is correct, I think there is another bug in this
> function. 'np' is not released when the function finishes
> successfully. I've looked at other functions that use
> irq_alloc_host(), and most of them do the same thing: they don't call
> of_node_put() on the device node pointer. The only exception I've
> found is mpc5121_ads_cpld_pic_init().
>
> Ben, Kumar: am I missing something? irq_alloc_host() calls of_node_get():
>
> host->of_node = of_node_get(of_node);
>
> so doesn't that mean that the caller of irq_alloc_host() should
> release the device node pointer?
>
--
Timur Tabi
Linux kernel developer at Freescale
^ permalink raw reply
* Re: [PATCH 8/9] arch/powerpc/sysdev/ehv_pic.c: add missing kfree
From: Timur Tabi @ 2011-08-23 18:38 UTC (permalink / raw)
To: Julia Lawall; +Cc: kernel-janitors, linux-kernel, Paul Mackerras, linuxppc-dev
In-Reply-To: <1312802283-9107-8-git-send-email-julia@diku.dk>
Julia Lawall wrote:
> At this point, ehv_pic has been allocated but not stored anywhere, so it
> should be freed before leaving the function.
Acked-by: Timur Tabi <timur@freescale.com>
FYI, Ashish is no longer with Freescale, so I've taken over maintainership of
ehv_pic.
--
Timur Tabi
Linux kernel developer at Freescale
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: Alex Williamson @ 2011-08-23 18:01 UTC (permalink / raw)
To: Aaron Fabbri
Cc: Alexey Kardashevskiy, kvm, Paul Mackerras, qemu-devel, chrisw,
iommu, Avi Kivity, Anthony Liguori, linux-pci@vger.kernel.org,
linuxppc-dev, benve
In-Reply-To: <CA79326A.FB97%aafabbri@cisco.com>
On Tue, 2011-08-23 at 10:33 -0700, Aaron Fabbri wrote:
>
>
> On 8/23/11 10:01 AM, "Alex Williamson" <alex.williamson@redhat.com> wrote:
>
> > On Tue, 2011-08-23 at 16:54 +1000, Benjamin Herrenschmidt wrote:
> >> On Mon, 2011-08-22 at 17:52 -0700, aafabbri wrote:
> >>
> >>> I'm not following you.
> >>>
> >>> You have to enforce group/iommu domain assignment whether you have the
> >>> existing uiommu API, or if you change it to your proposed
> >>> ioctl(inherit_iommu) API.
> >>>
> >>> The only change needed to VFIO here should be to make uiommu fd assignment
> >>> happen on the groups instead of on device fds. That operation fails or
> >>> succeeds according to the group semantics (all-or-none assignment/same
> >>> uiommu).
> >>
> >> Ok, so I missed that part where you change uiommu to operate on group
> >> fd's rather than device fd's, my apologies if you actually wrote that
> >> down :-) It might be obvious ... bare with me I just flew back from the
> >> US and I am badly jet lagged ...
> >
> > I missed it too, the model I'm proposing entirely removes the uiommu
> > concept.
> >
> >> So I see what you mean, however...
> >>
> >>> I think the question is: do we force 1:1 iommu/group mapping, or do we allow
> >>> arbitrary mapping (satisfying group constraints) as we do today.
> >>>
> >>> I'm saying I'm an existing user who wants the arbitrary iommu/group mapping
> >>> ability and definitely think the uiommu approach is cleaner than the
> >>> ioctl(inherit_iommu) approach. We considered that approach before but it
> >>> seemed less clean so we went with the explicit uiommu context.
> >>
> >> Possibly, the question that interest me the most is what interface will
> >> KVM end up using. I'm also not terribly fan with the (perceived)
> >> discrepancy between using uiommu to create groups but using the group fd
> >> to actually do the mappings, at least if that is still the plan.
> >
> > Current code: uiommu creates the domain, we bind a vfio device to that
> > domain via a SET_UIOMMU_DOMAIN ioctl on the vfio device, then do
> > mappings via MAP_DMA on the vfio device (affecting all the vfio devices
> > bound to the domain)
> >
> > My current proposal: "groups" are predefined. groups ~= iommu domain.
>
> This is my main objection. I'd rather not lose the ability to have multiple
> devices (which are all predefined as singleton groups on x86 w/o PCI
> bridges) share IOMMU resources. Otherwise, 20 devices sharing buffers would
> require 20x the IOMMU/ioTLB resources. KVM doesn't care about this case?
We do care, I just wasn't prioritizing it as heavily since I think the
typical model is probably closer to 1 device per guest.
> > The iommu domain would probably be allocated when the first device is
> > bound to vfio. As each device is bound, it gets attached to the group.
> > DMAs are done via an ioctl on the group.
> >
> > I think group + uiommu leads to effectively reliving most of the
> > problems with the current code. The only benefit is the group
> > assignment to enforce hardware restrictions. We still have the problem
> > that uiommu open() = iommu_domain_alloc(), whose properties are
> > meaningless without attached devices (groups). Which I think leads to
> > the same awkward model of attaching groups to define the domain, then we
> > end up doing mappings via the group to enforce ordering.
>
> Is there a better way to allow groups to share an IOMMU domain?
>
> Maybe, instead of having an ioctl to allow a group A to inherit the same
> iommu domain as group B, we could have an ioctl to fully merge two groups
> (could be what Ben was thinking):
>
> A.ioctl(MERGE_TO_GROUP, B)
>
> The group A now goes away and its devices join group B. If A ever had an
> iommu domain assigned (and buffers mapped?) we fail.
>
> Groups cannot get smaller (they are defined as minimum granularity of an
> IOMMU, initially). They can get bigger if you want to share IOMMU
> resources, though.
>
> Any downsides to this approach?
That's sort of the way I'm picturing it. When groups are bound
together, they effectively form a pool, where all the groups are peers.
When the MERGE/BIND ioctl is called on group A and passed the group B
fd, A can check compatibility of the domain associated with B, unbind
devices from the B domain and attach them to the A domain. The B domain
would then be freed and it would bump the refcnt on the A domain. If we
need to remove A from the pool, we call UNMERGE/UNBIND on B with the A
fd, it will remove the A devices from the shared object, disassociate A
with the shared object, re-alloc a domain for A and rebind A devices to
that domain.
This is where it seems like it might be helpful to make a GET_IOMMU_FD
ioctl so that an iommu object is ubiquitous and persistent across the
pool. Operations on any group fd work on the pool as a whole. Thanks,
Alex
^ permalink raw reply
* Re: [PATCH] powerpc: fixup QE_General4 errata
From: Timur Tabi @ 2011-08-23 17:48 UTC (permalink / raw)
To: Joakim Tjernlund; +Cc: Kumar Gala, linuxppc-dev
In-Reply-To: <1314102605-9718-1-git-send-email-Joakim.Tjernlund@transmode.se>
Joakim Tjernlund wrote:
> QE_General4 should only round up the divisor iff divisor is > 3.
> Rounding up lower divisors makes the error too big, causing USB
> on MPC832x to fail.
>
> Signed-off-by: Joakim Tjernlund <Joakim.Tjernlund@transmode.se>
Acked-by: Timur Tabi <timur@freescale.com>
--
Timur Tabi
Linux kernel developer at Freescale
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: Aaron Fabbri @ 2011-08-23 17:33 UTC (permalink / raw)
To: Alex Williamson, Benjamin Herrenschmidt
Cc: Alexey Kardashevskiy, kvm, Paul Mackerras,
linux-pci@vger.kernel.org, qemu-devel, chrisw, iommu, Avi Kivity,
Anthony Liguori, linuxppc-dev, benve
In-Reply-To: <1314118861.2859.51.camel@bling.home>
On 8/23/11 10:01 AM, "Alex Williamson" <alex.williamson@redhat.com> wrote:
> On Tue, 2011-08-23 at 16:54 +1000, Benjamin Herrenschmidt wrote:
>> On Mon, 2011-08-22 at 17:52 -0700, aafabbri wrote:
>>
>>> I'm not following you.
>>>
>>> You have to enforce group/iommu domain assignment whether you have the
>>> existing uiommu API, or if you change it to your proposed
>>> ioctl(inherit_iommu) API.
>>>
>>> The only change needed to VFIO here should be to make uiommu fd assignment
>>> happen on the groups instead of on device fds. That operation fails or
>>> succeeds according to the group semantics (all-or-none assignment/same
>>> uiommu).
>>
>> Ok, so I missed that part where you change uiommu to operate on group
>> fd's rather than device fd's, my apologies if you actually wrote that
>> down :-) It might be obvious ... bare with me I just flew back from the
>> US and I am badly jet lagged ...
>
> I missed it too, the model I'm proposing entirely removes the uiommu
> concept.
>
>> So I see what you mean, however...
>>
>>> I think the question is: do we force 1:1 iommu/group mapping, or do we allow
>>> arbitrary mapping (satisfying group constraints) as we do today.
>>>
>>> I'm saying I'm an existing user who wants the arbitrary iommu/group mapping
>>> ability and definitely think the uiommu approach is cleaner than the
>>> ioctl(inherit_iommu) approach. We considered that approach before but it
>>> seemed less clean so we went with the explicit uiommu context.
>>
>> Possibly, the question that interest me the most is what interface will
>> KVM end up using. I'm also not terribly fan with the (perceived)
>> discrepancy between using uiommu to create groups but using the group fd
>> to actually do the mappings, at least if that is still the plan.
>
> Current code: uiommu creates the domain, we bind a vfio device to that
> domain via a SET_UIOMMU_DOMAIN ioctl on the vfio device, then do
> mappings via MAP_DMA on the vfio device (affecting all the vfio devices
> bound to the domain)
>
> My current proposal: "groups" are predefined. groups ~= iommu domain.
This is my main objection. I'd rather not lose the ability to have multiple
devices (which are all predefined as singleton groups on x86 w/o PCI
bridges) share IOMMU resources. Otherwise, 20 devices sharing buffers would
require 20x the IOMMU/ioTLB resources. KVM doesn't care about this case?
> The iommu domain would probably be allocated when the first device is
> bound to vfio. As each device is bound, it gets attached to the group.
> DMAs are done via an ioctl on the group.
>
> I think group + uiommu leads to effectively reliving most of the
> problems with the current code. The only benefit is the group
> assignment to enforce hardware restrictions. We still have the problem
> that uiommu open() = iommu_domain_alloc(), whose properties are
> meaningless without attached devices (groups). Which I think leads to
> the same awkward model of attaching groups to define the domain, then we
> end up doing mappings via the group to enforce ordering.
Is there a better way to allow groups to share an IOMMU domain?
Maybe, instead of having an ioctl to allow a group A to inherit the same
iommu domain as group B, we could have an ioctl to fully merge two groups
(could be what Ben was thinking):
A.ioctl(MERGE_TO_GROUP, B)
The group A now goes away and its devices join group B. If A ever had an
iommu domain assigned (and buffers mapped?) we fail.
Groups cannot get smaller (they are defined as minimum granularity of an
IOMMU, initially). They can get bigger if you want to share IOMMU
resources, though.
Any downsides to this approach?
-AF
>
>> If the separate uiommu interface is kept, then anything that wants to be
>> able to benefit from the ability to put multiple devices (or existing
>> groups) into such a "meta group" would need to be explicitly modified to
>> deal with the uiommu APIs.
>>
>> I tend to prefer such "meta groups" as being something you create
>> statically using a configuration interface, either via sysfs, netlink or
>> ioctl's to a "control" vfio device driven by a simple command line tool
>> (which can have the configuration stored in /etc and re-apply it at
>> boot).
>
> I cringe anytime there's a mention of "static". IMHO, we have to
> support hotplug. That means "meta groups" change dynamically. Maybe
> this supports the idea that we should be able to retrieve a new fd from
> the group to do mappings. Any groups bound together will return the
> same fd and the fd will persist so long as any member of the group is
> open.
>
>> That way, any program capable of exploiting VFIO "groups" will
>> automatically be able to exploit those "meta groups" (or groups of
>> groups) as well as long as they are supported on the system.
>>
>> If we ever have system specific constraints as to how such groups can be
>> created, then it can all be handled at the level of that configuration
>> tool without impact on whatever programs know how to exploit them via
>> the VFIO interfaces.
>
> I'd prefer to have the constraints be represented in the ioctl to bind
> groups. It works or not and the platform gets to define what it
> considers compatible. Thanks,
>
> Alex
>
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: Alex Williamson @ 2011-08-23 17:08 UTC (permalink / raw)
To: Roedel, Joerg
Cc: Alexey Kardashevskiy, kvm@vger.kernel.org, Paul Mackerras,
qemu-devel, chrisw, iommu, Avi Kivity, Anthony Liguori,
linux-pci@vger.kernel.org, linuxppc-dev, benve@cisco.com
In-Reply-To: <20110823131441.GN2079@amd.com>
On Tue, 2011-08-23 at 15:14 +0200, Roedel, Joerg wrote:
> On Mon, Aug 22, 2011 at 03:17:00PM -0400, Alex Williamson wrote:
> > On Mon, 2011-08-22 at 19:25 +0200, Joerg Roedel wrote:
>
> > > I am in favour of /dev/vfio/$GROUP. If multiple devices should be
> > > assigned to a guest, there can also be an ioctl to bind a group to an
> > > address-space of another group (certainly needs some care to not allow
> > > that both groups belong to different processes).
> >
> > That's an interesting idea. Maybe an interface similar to the current
> > uiommu interface, where you open() the 2nd group fd and pass the fd via
> > ioctl to the primary group. IOMMUs that don't support this would fail
> > the attach device callback, which would fail the ioctl to bind them. It
> > will need to be designed so any group can be removed from the super-set
> > and the remaining group(s) still works. This feels like something that
> > can be added after we get an initial implementation.
>
> Handling it through fds is a good idea. This makes sure that everything
> belongs to one process. I am not really sure yet if we go the way to
> just bind plain groups together or if we create meta-groups. The
> meta-groups thing seems somewhat cleaner, though.
I'm leaning towards binding because we need to make it dynamic, but I
don't really have a good picture of the lifecycle of a meta-group.
> > > Btw, a problem we havn't talked about yet entirely is
> > > driver-deassignment. User space can decide to de-assign the device from
> > > vfio while a fd is open on it. With PCI there is no way to let this fail
> > > (the .release function returns void last time i checked). Is this a
> > > problem, and yes, how we handle that?
> >
> > The current vfio has the same problem, we can't unbind a device from
> > vfio while it's attached to a guest. I think we'd use the same solution
> > too; send out a netlink packet for a device removal and have the .remove
> > call sleep on a wait_event(, refcnt == 0). We could also set a timeout
> > and SIGBUS the PIDs holding the device if they don't return it
> > willingly. Thanks,
>
> Putting the process to sleep (which would be uninterruptible) seems bad.
> The process would sleep until the guest releases the device-group, which
> can take days or months.
> The best thing (and the most intrusive :-) ) is to change PCI core to
> allow unbindings to fail, I think. But this probably further complicates
> the way to upstream VFIO...
Yes, it's not ideal but I think it's sufficient for now and if we later
get support for returning an error from release, we can set a timeout
after notifying the user to make use of that. Thanks,
Alex
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: Alex Williamson @ 2011-08-23 17:01 UTC (permalink / raw)
To: Benjamin Herrenschmidt
Cc: chrisw, Alexey Kardashevskiy, kvm, Paul Mackerras,
linux-pci@vger.kernel.org, qemu-devel, aafabbri, iommu,
Avi Kivity, Anthony Liguori, linuxppc-dev, benve
In-Reply-To: <1314082483.30478.43.camel@pasglop>
On Tue, 2011-08-23 at 16:54 +1000, Benjamin Herrenschmidt wrote:
> On Mon, 2011-08-22 at 17:52 -0700, aafabbri wrote:
>
> > I'm not following you.
> >
> > You have to enforce group/iommu domain assignment whether you have the
> > existing uiommu API, or if you change it to your proposed
> > ioctl(inherit_iommu) API.
> >
> > The only change needed to VFIO here should be to make uiommu fd assignment
> > happen on the groups instead of on device fds. That operation fails or
> > succeeds according to the group semantics (all-or-none assignment/same
> > uiommu).
>
> Ok, so I missed that part where you change uiommu to operate on group
> fd's rather than device fd's, my apologies if you actually wrote that
> down :-) It might be obvious ... bare with me I just flew back from the
> US and I am badly jet lagged ...
I missed it too, the model I'm proposing entirely removes the uiommu
concept.
> So I see what you mean, however...
>
> > I think the question is: do we force 1:1 iommu/group mapping, or do we allow
> > arbitrary mapping (satisfying group constraints) as we do today.
> >
> > I'm saying I'm an existing user who wants the arbitrary iommu/group mapping
> > ability and definitely think the uiommu approach is cleaner than the
> > ioctl(inherit_iommu) approach. We considered that approach before but it
> > seemed less clean so we went with the explicit uiommu context.
>
> Possibly, the question that interest me the most is what interface will
> KVM end up using. I'm also not terribly fan with the (perceived)
> discrepancy between using uiommu to create groups but using the group fd
> to actually do the mappings, at least if that is still the plan.
Current code: uiommu creates the domain, we bind a vfio device to that
domain via a SET_UIOMMU_DOMAIN ioctl on the vfio device, then do
mappings via MAP_DMA on the vfio device (affecting all the vfio devices
bound to the domain)
My current proposal: "groups" are predefined. groups ~= iommu domain.
The iommu domain would probably be allocated when the first device is
bound to vfio. As each device is bound, it gets attached to the group.
DMAs are done via an ioctl on the group.
I think group + uiommu leads to effectively reliving most of the
problems with the current code. The only benefit is the group
assignment to enforce hardware restrictions. We still have the problem
that uiommu open() = iommu_domain_alloc(), whose properties are
meaningless without attached devices (groups). Which I think leads to
the same awkward model of attaching groups to define the domain, then we
end up doing mappings via the group to enforce ordering.
> If the separate uiommu interface is kept, then anything that wants to be
> able to benefit from the ability to put multiple devices (or existing
> groups) into such a "meta group" would need to be explicitly modified to
> deal with the uiommu APIs.
>
> I tend to prefer such "meta groups" as being something you create
> statically using a configuration interface, either via sysfs, netlink or
> ioctl's to a "control" vfio device driven by a simple command line tool
> (which can have the configuration stored in /etc and re-apply it at
> boot).
I cringe anytime there's a mention of "static". IMHO, we have to
support hotplug. That means "meta groups" change dynamically. Maybe
this supports the idea that we should be able to retrieve a new fd from
the group to do mappings. Any groups bound together will return the
same fd and the fd will persist so long as any member of the group is
open.
> That way, any program capable of exploiting VFIO "groups" will
> automatically be able to exploit those "meta groups" (or groups of
> groups) as well as long as they are supported on the system.
>
> If we ever have system specific constraints as to how such groups can be
> created, then it can all be handled at the level of that configuration
> tool without impact on whatever programs know how to exploit them via
> the VFIO interfaces.
I'd prefer to have the constraints be represented in the ioctl to bind
groups. It works or not and the platform gets to define what it
considers compatible. Thanks,
Alex
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: aafabbri @ 2011-08-23 16:54 UTC (permalink / raw)
To: Joerg Roedel
Cc: Alexey Kardashevskiy, kvm@vger.kernel.org, Paul Mackerras,
linux-pci@vger.kernel.org, qemu-devel, chrisw, iommu, Avi Kivity,
Anthony Liguori, linuxppc-dev, benve@cisco.com
In-Reply-To: <20110823110431.GK2079@amd.com>
On 8/23/11 4:04 AM, "Joerg Roedel" <joerg.roedel@amd.com> wrote:
> On Mon, Aug 22, 2011 at 08:52:18PM -0400, aafabbri wrote:
>> You have to enforce group/iommu domain assignment whether you have the
>> existing uiommu API, or if you change it to your proposed
>> ioctl(inherit_iommu) API.
>>
>> The only change needed to VFIO here should be to make uiommu fd assignment
>> happen on the groups instead of on device fds. That operation fails or
>> succeeds according to the group semantics (all-or-none assignment/same
>> uiommu).
>
> That is makes uiommu basically the same as the meta-groups, right?
Yes, functionality seems the same, thus my suggestion to keep uiommu
explicit. Is there some need for group-groups besides defining sets of
groups which share IOMMU resources?
I do all this stuff (bringing up sets of devices which may share IOMMU
domain) dynamically from C applications. I don't really want some static
(boot-time or sysfs fiddling) supergroup config unless there is a good
reason KVM/power needs it.
As you say in your next email, doing it all from ioctls is very easy,
programmatically.
-Aaron Fabbri
^ permalink raw reply
* Re: kvm PCI assignment & VFIO ramblings
From: Alex Williamson @ 2011-08-23 16:23 UTC (permalink / raw)
To: David Gibson
Cc: chrisw, Alexey Kardashevskiy, kvm, Paul Mackerras,
linux-pci@vger.kernel.org, qemu-devel, aafabbri, iommu,
Avi Kivity, Anthony Liguori, linuxppc-dev, benve
In-Reply-To: <20110823023822.GO30097@yookeroo.fritz.box>
On Tue, 2011-08-23 at 12:38 +1000, David Gibson wrote:
> On Mon, Aug 22, 2011 at 09:45:48AM -0600, Alex Williamson wrote:
> > On Mon, 2011-08-22 at 15:55 +1000, David Gibson wrote:
> > > On Sat, Aug 20, 2011 at 09:51:39AM -0700, Alex Williamson wrote:
> > > > We had an extremely productive VFIO BoF on Monday. Here's my attempt to
> > > > capture the plan that I think we agreed to:
> > > >
> > > > We need to address both the description and enforcement of device
> > > > groups. Groups are formed any time the iommu does not have resolution
> > > > between a set of devices. On x86, this typically happens when a
> > > > PCI-to-PCI bridge exists between the set of devices and the iommu. For
> > > > Power, partitionable endpoints define a group. Grouping information
> > > > needs to be exposed for both userspace and kernel internal usage. This
> > > > will be a sysfs attribute setup by the iommu drivers. Perhaps:
> > > >
> > > > # cat /sys/devices/pci0000:00/0000:00:19.0/iommu_group
> > > > 42
> > > >
> > > > (I use a PCI example here, but attribute should not be PCI specific)
> > >
> > > Ok. Am I correct in thinking these group IDs are representing the
> > > minimum granularity, and are therefore always static, defined only by
> > > the connected hardware, not by configuration?
> >
> > Yes, that's the idea. An open question I have towards the configuration
> > side is whether we might add iommu driver specific options to the
> > groups. For instance on x86 where we typically have B:D.F granularity,
> > should we have an option not to trust multi-function devices and use a
> > B:D granularity for grouping?
>
> Right. And likewise I can see a place for configuration parameters
> like the present 'allow_unsafe_irqs'. But these would be more-or-less
> global options which affected the overall granularity, rather than
> detailed configuration such as explicitly binding some devices into a
> group, yes?
Yes, currently the interrupt remapping support is a global iommu
capability. I suppose it's possible that this could be an iommu option,
where the iommu driver would not advertise a group if the interrupt
remapping constraint isn't met.
> > > > >From there we have a few options. In the BoF we discussed a model where
> > > > binding a device to vfio creates a /dev/vfio$GROUP character device
> > > > file. This "group" fd provides provides dma mapping ioctls as well as
> > > > ioctls to enumerate and return a "device" fd for each attached member of
> > > > the group (similar to KVM_CREATE_VCPU). We enforce grouping by
> > > > returning an error on open() of the group fd if there are members of the
> > > > group not bound to the vfio driver. Each device fd would then support a
> > > > similar set of ioctls and mapping (mmio/pio/config) interface as current
> > > > vfio, except for the obvious domain and dma ioctls superseded by the
> > > > group fd.
> > >
> > > It seems a slightly strange distinction that the group device appears
> > > when any device in the group is bound to vfio, but only becomes usable
> > > when all devices are bound.
> > >
> > > > Another valid model might be that /dev/vfio/$GROUP is created for all
> > > > groups when the vfio module is loaded. The group fd would allow open()
> > > > and some set of iommu querying and device enumeration ioctls, but would
> > > > error on dma mapping and retrieving device fds until all of the group
> > > > devices are bound to the vfio driver.
> > >
> > > Which is why I marginally prefer this model, although it's not a big
> > > deal.
> >
> > Right, we can also combine models. Binding a device to vfio
> > creates /dev/vfio$GROUP, which only allows a subset of ioctls and no
> > device access until all the group devices are also bound. I think
> > the /dev/vfio/$GROUP might help provide an enumeration interface as well
> > though, which could be useful.
>
> I'm not entirely sure what you mean here. But, that's now several
> weak votes in favour of the always-present group devices, and none in
> favour of the created-when-first-device-bound model, so I suggest we
> take the /dev/vfio/$GROUP as our tentative approach.
Yep
> > > > In either case, the uiommu interface is removed entirely since dma
> > > > mapping is done via the group fd. As necessary in the future, we can
> > > > define a more high performance dma mapping interface for streaming dma
> > > > via the group fd. I expect we'll also include architecture specific
> > > > group ioctls to describe features and capabilities of the iommu. The
> > > > group fd will need to prevent concurrent open()s to maintain a 1:1 group
> > > > to userspace process ownership model.
> > >
> > > A 1:1 group<->process correspondance seems wrong to me. But there are
> > > many ways you could legitimately write the userspace side of the code,
> > > many of them involving some sort of concurrency. Implementing that
> > > concurrency as multiple processes (using explicit shared memory and/or
> > > other IPC mechanisms to co-ordinate) seems a valid choice that we
> > > shouldn't arbitrarily prohibit.
> > >
> > > Obviously, only one UID may be permitted to have the group open at a
> > > time, and I think that's enough to prevent them doing any worse than
> > > shooting themselves in the foot.
> >
> > 1:1 group<->process is probably too strong. Not allowing concurrent
> > open()s on the group file enforces a single userspace entity is
> > responsible for that group. Device fds can be passed to other
> > processes, but only retrieved via the group fd. I suppose we could even
> > branch off the dma interface into a different fd, but it seems like we
> > would logically want to serialize dma mappings at each iommu group
> > anyway. I'm open to alternatives, this just seemed an easy way to do
> > it. Restricting on UID implies that we require isolated qemu instances
> > to run as different UIDs.
>
> Well.. yes and know. It means guests which need to be isolated from
> malicious interference with each other need different UIDs, but given
> that if they have the same UID one qemu can kill() or ptrace() the
> other, they're not isolated in that sense anyway.
>
> It seems to me that running as the same UIDs with different device
> groups assigned, the guests are still pretty well isolated from
> accidental interference with each other.
If our only restriction is UID, what prevents a non-clueful user from
trying to create separate qemu instances making use of different devices
within the same group (or even the same device)? If we restrict
concurrent opens, it's just the subsequent instances get a -EBUSY.
> > I know that's a goal, but I don't know if we
> > want to make it an assumption in the group security model.
> >
> > > > Also on the table is supporting non-PCI devices with vfio. To do this,
> > > > we need to generalize the read/write/mmap and irq eventfd interfaces.
> > > > We could keep the same model of segmenting the device fd address space,
> > > > perhaps adding ioctls to define the segment offset bit position or we
> > > > could split each region into it's own fd (VFIO_GET_PCI_BAR_FD(0),
> > > > VFIO_GET_PCI_CONFIG_FD(), VFIO_GET_MMIO_FD(3)), though we're already
> > > > suffering some degree of fd bloat (group fd, device fd(s), interrupt
> > > > event fd(s), per resource fd, etc). For interrupts we can overload
> > > > VFIO_SET_IRQ_EVENTFD to be either PCI INTx or non-PCI irq
> > >
> > > Sounds reasonable.
> > >
> > > > (do non-PCI
> > > > devices support MSI?).
> > >
> > > They can. Obviously they might not have exactly the same semantics as
> > > PCI MSIs, but I know we have SoC systems with (non-PCI) on-die devices
> > > whose interrupts are treated by the (also on-die) root interrupt
> > > controller in the same way as PCI MSIs.
> >
> > Ok, I suppose we can define ioctls to enable these as we go. We also
> > need to figure out how non-PCI resources, interrupts, and iommu mapping
> > restrictions are described via vfio.
>
> Yeah. On device tree platforms we'd want it to be bound to the device
> tree representation in some way.
>
> For platform devices, at least, could we have the index into the array
> of resources take the place of BAR number for PCI?
That's what I was thinking, but we need some way to describe the set of
valid indexes and type and size for each as well. We already have the
BAR_LEN helper ioctl, we could make that generic (RANGE_LEN?) and add
NUM_RANGES and RANGE_TYPE. For PCI there would always be 7 ranges (6
BARs + ROM).
> > > > For qemu, these changes imply we'd only support a model where we have a
> > > > 1:1 group to iommu domain. The current vfio driver could probably
> > > > become vfio-pci as we might end up with more target specific vfio
> > > > drivers for non-pci. PCI should be able to maintain a simple -device
> > > > vfio-pci,host=bb:dd.f to enable hotplug of individual devices. We'll
> > > > need to come up with extra options when we need to expose groups to
> > > > guest for pvdma.
> > >
> > > Are you saying that you'd no longer support the current x86 usage of
> > > putting all of one guest's devices into a single domain?
> >
> > Yes. I'm not sure there's a good ROI to prioritize that model. We have
> > to assume >1 device per guest is a typical model and that the iotlb is
> > large enough that we might improve thrashing to see both a resource and
> > performance benefit from it. I'm open to suggestions for how we could
> > include it though.
>
> Creating supergroups of some sort seems to be what we need, but I'm
> not sure what's the best interface for doing that.
Yeah. Joerg's idea of binding groups internally (pass the fd of one
group to another via ioctl) is one option. The tricky part will be
implementing it to support hot unplug of any group from the supergroup.
I believe Ben had a suggestion that supergroups could be created in
sysfs, but I don't know what the mechanism to do that looks like. It
would also be an extra management step to dynamically bind and unbind
groups to the supergroup around hotplug. Thanks,
Alex
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox