[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH 6/7] s390x/pci: move the memory region write fro

From: Pierre Morel
Subject: Re: [Qemu-devel] [PATCH 6/7] s390x/pci: move the memory region write from pcistg
Date: Mon, 13 Nov 2017 15:44:27 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.4.0

On 13/11/2017 12:54, Cornelia Huck wrote:
On Mon, 13 Nov 2017 10:39:50 +0100
Pierre Morel <address@hidden> wrote:

On 10/11/2017 10:51, Cornelia Huck wrote:
On Fri, 10 Nov 2017 17:40:12 +0800
Yi Min Zhao <address@hidden> wrote:
在 2017/11/10 上午3:23, Cornelia Huck 写道:
On Tue,  7 Nov 2017 18:24:38 +0100
Pierre Morel <address@hidden> wrote:
Let's move the memory region write from pcistg into a dedicated
This allows us to prepare a later patch searching for subregions
inside of the memory region.
OK, so here is the memory region write. Do we have any sleeping
endianness bugs in there for when we wire up tcg? I'm not sure how this
plays with the bswaps (see patch 1).

But maybe I've just gotten lost somewhere.
I think there's no error. For PCI bars' MRs, we got the little-endian data
that is exactly fit to the byte ordering of pcilg instruction. For PCI
space, the data has been swapped according to the cpu byte ordering.

Host or target cpu?
So we use zpci_swap_endian() to swap the data back to the little-endian

I do not see where we use the zpci_swap_endian() function in this patch
or in the zpci_write_bar() function.

So, is that swap function only ever used to convert BE register
contents to LE?

Yes that is what I understood.
The value in the register is read from memory somehow but it may be an immediate value setup by another instruction, I think the TCG should have already make sure it is big endian.

On the other side the PCI memory is always Little endian AFAIK.

So the swap function is always from BIG to little.

But as I said, we do not use the zpci_swap_endian in this function, so I write this answer again in the thread where the definition of the function is to continue the discussion on the register to PCI translation there.

That swap is unconditional. If we were running on a little-endian host,
it would be wrong, wouldn't it?

I think there is no problem here, we do not use the swap function but we
use the memory_region_dispatch_write() to access a subregion of the PCI
device which is defined as DEVICE_LITTLE_ENDIAN

AFAIU The memory access process the endianness correctly.

OK, if this is all going through generic memory mechanisms, we should
be fine. >

Signed-off-by: Pierre Morel <address@hidden>
Reviewed-by: Yi Min Zhao <address@hidden>
    hw/s390x/s390-pci-inst.c | 27 +++++++++++++++++----------
    1 file changed, 17 insertions(+), 10 deletions(-)

Pierre Morel
Linux/KVM/QEMU in Böblingen - Germany

reply via email to

[Prev in Thread] Current Thread [Next in Thread]