Jes Sorensen wrote:
Zhang, Xiantao wrote:
Hi Jes, Good news! On my machine, without the patch, smp guest can't
build one whole kernel at all. So if you can build it without errors
and use it to boot up the guest, I think it should work well. Xiantao
Yep, compiles, boots, the works ...
Zhang, Xiantao wrote:
Hi Jes,
Good news! On my machine, without the patch, smp guest can't build one whole kernel at all. So if you can build it without errors and use it to boot up the guest, I think it should work well.
Xiantao
Yep, compiles, boots, the works ...
Avi, I think we can
Zhang, Xiantao wrote:
Hi, Jes
Have you verified whether it works for you ? You may run kernel build in
the guest with 4 vcpus, if it can be done successfully without any error, it
should be Okay I think, otherwise, we may need to investigate it further. :)
Xiantao
Hi Xiantao,
I was
Jes Sorensen wrote:
Zhang, Xiantao wrote:
Hi, Jes
Have you verified whether it works for you ? You may run kernel
build in the guest with 4 vcpus, if it can be done successfully
without any error, it should be Okay I think, otherwise, we may need
to investigate it further. :) Xiantao
Avi Kivity wrote:
Zhang, Xiantao wrote:
Fine to me. But seems the change in qemu_iovec_from_buffer is lost in
this patch or that change is also not unnecessary ?
I think the fixed unmap handles that case. Can you test to make sure?
Avi and I went through the code and verified that it was
: Jes Sorensen [mailto:j...@sgi.com]
Sent: Tuesday, June 02, 2009 6:57 PM
To: Avi Kivity
Cc: Zhang, Xiantao; kvm@vger.kernel.org; kvm-i...@vger.kernel.org; Hollis
Blanchard
Subject: Re: [PATCH] qemu-kvm: Flush icache after dma operations for ia64
Avi Kivity wrote:
Zhang, Xiantao wrote:
Fine
Zhang, Xiantao wrote:
Avi Kivity wrote:
Jes Sorensen wrote:
Ok,
Trying once more. After spending a couple of hours trying to follow
the QEMU dma codeflow, I have convinced myself Avi is right and those
two functions don't need to do the flushing as they all end up
calling
Avi Kivity wrote:
Jes Sorensen wrote:
Ok,
Trying once more. After spending a couple of hours trying to follow
the QEMU dma codeflow, I have convinced myself Avi is right and those
two functions don't need to do the flushing as they all end up
calling dma_bdrv_cb() which calls
Jes Sorensen wrote:
Ok,
Trying once more. After spending a couple of hours trying to follow
the QEMU dma codeflow, I have convinced myself Avi is right and those
two functions don't need to do the flushing as they all end up calling
dma_bdrv_cb() which calls dma_brdv_unmap(). I have added a
Hi,
Looking at Avi's comments, I came up with a slightly modified version
of this patch, which avoids the #ifdef __ia64__ problem by introducing
dma_flush_range().
I have made it a noop on non ia64, but it may need to be changed for
PPC? Hollis would you check if you guys need this too?
It
ARGH!
I think the attachment was lost :-( Trying again.
Jes
Jes == Jes Sorensen j...@sgi.com writes:
Jes Hi, Looking at Avi's comments, I came up with a slightly modified
Jes version of this patch, which avoids the #ifdef __ia64__ problem
Jes by introducing dma_flush_range().
Jes I have
Jes Sorensen wrote:
+/*
+ * Is this correct for PPC?
+ */
+#define dma_flush_range(start, end) \
+do { (void) (start); (void) (end); } while (0)
Correct or not, should be a function/
+#define dma_flush_range(start, end) \
+do { (void) (start);
Ok,
Trying once more. After spending a couple of hours trying to follow
the QEMU dma codeflow, I have convinced myself Avi is right and those
two functions don't need to do the flushing as they all end up calling
dma_bdrv_cb() which calls dma_brdv_unmap(). I have added a couple
comments to the
: [PATCH] qemu-kvm: Flush icache after dma operations for ia64
ia64 system depends on that platform issues snoop cycle to flush
icache for memory touched by DMA write operations, but virtual DMA
operations is emulated by memcpy, so use explict instrustions to flush
the related icache, otherwise, guest
:15 +0800
Subject: [PATCH] qemu-kvm: Flush icache after dma operations for ia64
ia64 system depends on that platform issues snoop cycle to flush
icache for memory touched by DMA write operations, but virtual DMA
operations is emulated by memcpy, so use explict instrustions to flush
the related
15 matches
Mail list logo