Dor Laor wrote:
> Anthony Liguori wrote:
>
>> Hi Dor,
>>
>> I can't tell if you made any changes to this series. I have to look
>> back to my notes, but I believe it would be okay to pull this series
>> into KVM. I still need to spend some time updating the patchset to be
>> more friendly f
The next patch is a resend for patch #3 in the userspace series.
The previous one did not apply on kvm current head.
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http:/
From: Dor Laor <[EMAIL PROTECTED]>
This patch implements the backend support for the virtio block device. It's
designed to support in-order queueing of a virtually unlimited size so it will
be able to perform better than SCSI. Besides performance, the virtio block
interface passes through guest
Anthony Liguori wrote:
> Hi Dor,
>
> I can't tell if you made any changes to this series. I have to look
> back to my notes, but I believe it would be okay to pull this series
> into KVM. I still need to spend some time updating the patchset to be
> more friendly for QEMU though before resubmi
Hi Dor,
I can't tell if you made any changes to this series. I have to look
back to my notes, but I believe it would be okay to pull this series
into KVM. I still need to spend some time updating the patchset to be
more friendly for QEMU though before resubmitting to qemu-devel. Any
help on
From: Anthony Liguori <[EMAIL PROTECTED]>
Using unsigned int resulted in silent truncation of the upper 32-bit on x86_64
resulting in an OOPS since the ring was being initialized wrong.
Please reconsider my previous patch to just use PAGE_ALIGN(). Open coding this
sort of stuff, no matter how si
From: Anthony Liguori <[EMAIL PROTECTED]>
This is a PCI device that implements a transport for virtio. It allows virtio
devices to be used by QEMU based VMMs like KVM or Xen.
Signed-off-by: Anthony Liguori <[EMAIL PROTECTED]>
Signed-off-by: Rusty Russell <[EMAIL PROTECTED]>
---
drivers/virtio/K
From: Anthony Liguori <[EMAIL PROTECTED]>
This fixes a potential dangling xmit problem.
We also suppress refill interrupts until we need them.
Signed-off-by: Rusty Russell <[EMAIL PROTECTED]>
---
drivers/net/virtio_net.c | 69 -
1 files changed, 49
From: Dor Laor <[EMAIL PROTECTED]>
Both VIRTIO_CONFIG_S_DRIVER and VIRTIO_CONFIG_S_DRIVER_OK should
be cleaned upon driver unload.
Signed-off-by: Dor Laor <[EMAIL PROTECTED]>
---
drivers/virtio/virtio.c |3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/drivers/virtio/vir
From: Anthony Liguori <[EMAIL PROTECTED]>
Previously we used a type/len pair within the config space, but this
seems overkill. We now simply define a structure which represents the
layout in the config space: the config space can now only be extended
at the end.
The main driver-visible changes:
From: Dor Laor <[EMAIL PROTECTED]>
The current start_xmit sets 500us hrtimer to kick the host.
The problem is that if another xmit happens before the timer was fired then
the first xmit will have to wait additional 500us.
This patch does not re-arm the timer if there is existing one.
This will sho
From: Anthony Liguori <[EMAIL PROTECTED]>
This is needed for the virtio PCI device to be compiled as a module.
Signed-off-by: Anthony Liguori <[EMAIL PROTECTED]>
Signed-off-by: Rusty Russell <[EMAIL PROTECTED]>
---
drivers/virtio/virtio_ring.c |6 ++
1 files changed, 6 insertions(+), 0 d
From: Anthony Liguori <[EMAIL PROTECTED]>
Not sure why hrtimer's cb_mode only exists with CONFIG_HIGH_RES_TIMERS; that
seems like a bug to me.
Signed-off-by: Rusty Russell <[EMAIL PROTECTED]>
---
drivers/net/virtio_net.c | 41 -
1 files changed, 40 inser
From: Rusty Russell <[EMAIL PROTECTED]>
Signed-off-by: Rusty Russell <[EMAIL PROTECTED]>
---
Documentation/lguest/lguest.c | 53 +++
drivers/char/hw_random/Kconfig | 10 +++
drivers/char/hw_random/Makefile |1 +
drivers/char/hw_random/virtio-rng.c | 124 +++
From: Dor Laor <[EMAIL PROTECTED]>
Add the device release function.
Signed-off-by: Dor Laor <[EMAIL PROTECTED]>
---
drivers/virtio/virtio.c |7 +++
drivers/virtio/virtio_pci.c |7 +++
2 files changed, 14 insertions(+), 0 deletions(-)
diff --git a/drivers/virtio/virtio.c b/dr
From: Dor Laor <[EMAIL PROTECTED]>
Signed-off-by: Dor Laor <[EMAIL PROTECTED]>
---
drivers/net/virtio_net.c |5 -
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 0403d65..d35dbe7 100644
--- a/drivers/net/virtio_net.c
From: Anthony Liguori <[EMAIL PROTECTED]>
ifconfig down or remove the module to get the statistics dump.
---
drivers/net/virtio_net.c | 120 +++---
1 files changed, 103 insertions(+), 17 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/vir
From: Anthony Liguori <[EMAIL PROTECTED]>
This patch moves virtio under the virtualization menu and changes virtio
devices to not claim to only be for lguest.
Signed-off-by: Anthony Liguori <[EMAIL PROTECTED]>
Signed-off-by: Rusty Russell <[EMAIL PROTECTED]>
---
arch/x86/kvm/Kconfig |1 +
d
This patchset updates kvm repository with Rusty's/Anthony's virtio
implementation along with already sent tx performance bug fix and
new module reload fixes.
The (new) patches were accepted by Rusty, this is a resend to fix
email space mangling software.
This code together with the userspace patch
From: Dor Laor <[EMAIL PROTECTED]>
Call it when the driver sets the pfn to zero and
also on qemu_reset notifier list.
This fixes crashes when virtio driver unloads or when
rebooting the guest.
Signed-off-by: Dor Laor <[EMAIL PROTECTED]>
---
qemu/hw/virtio.c | 24 +---
1 fil
From: Dor Laor <[EMAIL PROTECTED]>
Stop using the old, unregsitered 0x5002, 0x2258 pair and use
Qumranet's donated 0x1AF4,0x10XX pairs.
Signed-off-by: Dor Laor <[EMAIL PROTECTED]>
---
qemu/hw/pc.c |2 +-
qemu/hw/virtio-blk.c |2 +-
2 files changed, 2 insertions(+), 2 deletions(-
From: Dor Laor <[EMAIL PROTECTED]>
This patch prevents qemu handlers from reading the tap and instead
it selects the tap descriptors for virtio devices.
This eliminates copies and also batch guest notifications (interrupts).
Using this patch the rx performance reaches 800Mbps.
-net user option re
From: Dor Laor <[EMAIL PROTECTED]>
This patch implements the basic infrastructure for virtio devices. These
devices are exposed to the guest as real PCI devices. The PCI vendor/device
IDs have been donated by Qumranet and the subsystem IDs are used to distinguish
the virtio device itself.
Virti
From: Dor Laor <[EMAIL PROTECTED]>
Allow tx only if VIRTIO_CONFIG_S_DRIVER_OK.
It can prevent an smp race of tx and driver unload.
Signed-off-by: Dor Laor <[EMAIL PROTECTED]>
---
qemu/hw/virtio-net.c |3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/qemu/hw/virtio-net.c
This patchset updates kvm repository with Anthony's virtio
implementation along with rx performance improvements and
guest reset handling. The original code was sent to qemu
devel list 2 weeks ago. V2 fixes Avi's comments for enabling
-net user option (although with old performance).
It contains
From: Dor Laor <[EMAIL PROTECTED]>
This patch implements the backend support for the virtio network device. The
device is optimized for virtualized environments by limiting the number of
guest=>host transitions per-packet. In the best case, the number of
transitions per-packet is < 1.
With some
From: Dor Laor <[EMAIL PROTECTED]>
This patch implements the backend support for the virtio block device. It's
designed to support in-order queueing of a virtually unlimited size so it will
be able to perform better than SCSI. Besides performance, the virtio block
interface passes through guest
From: Dor Laor <[EMAIL PROTECTED]>
It allows picking other NIC vendors than rtl8139.
Use --nictype=xxx, the ddefault is rtl8139.
Signed-off-by: Dor Laor <[EMAIL PROTECTED]>
---
kvm |8 +++-
1 files changed, 7 insertions(+), 1 deletions(-)
diff --git a/kvm b/kvm
index 0374fbc..2a7dc85 10
Avi Kivity wrote:
> Avi Kivity wrote:
>> From: Marcelo Tosatti <[EMAIL PROTECTED]>
>>
>> Do not hold kvm->lock mutex across the entire pagefault code,
>> only acquire it in places where it is necessary, such as mmu
>> hash list, active list, rmap and parent pte handling.
>>
>> Allow concurrent gues
Avi Kivity wrote:
> From: Marcelo Tosatti <[EMAIL PROTECTED]>
>
> Do not hold kvm->lock mutex across the entire pagefault code,
> only acquire it in places where it is necessary, such as mmu
> hash list, active list, rmap and parent pte handling.
>
> Allow concurrent guest walkers by switching walk
Carlo Marcelo Arenas Belon wrote:
> This reverts commit a2c1905ff7c6faf9253d4185952a5b85036114f7.
>
> It was made obsolete by 3f5daaafad51a2b75171c2a93a28cc55ce205cae
> and would redefine MAX_IO_APIC to a smaller value in amd64 that
> the one used by for non SMP kernels > 2.6.21
>
Applied, than
Carlo Marcelo Arenas Belon wrote:
> This patch complement 6eabe2ba30a673c4280df75b28520e0e1f4a2171 so that
> all callbacks used for MMIO read/write use the ABI exported in libkvm.h.
>
> This fixes the following warnings was reported at compile time :
>
> main.c:392: warning: initialization from i
Carlo Marcelo Arenas Belon wrote:
> This patch corrects 03f1b5e137e7e4c8dd51dbaea6779be853f4fde0, that
> modified the logic behind kvm_arch_run because of a mangled endif
> which accidentally included the default entry for the switch.
>
> kvm_arch_run as defined in 4f19bdea03330641fd48514ea84d1ed1b
From: Marcelo Tosatti <[EMAIL PROTECTED]>
Do not hold kvm->lock mutex across the entire pagefault code,
only acquire it in places where it is necessary, such as mmu
hash list, active list, rmap and parent pte handling.
Allow concurrent guest walkers by switching walk_addr() to use
mmap_sem in rea
From: Marcelo Tosatti <[EMAIL PROTECTED]>
Convert the synchronization of the shadow handling to a separate mmu_lock
spinlock.
Also guard fetch() by mmap_sem in read-mode to protect against alias
and memslot changes.
Signed-off-by: Marcelo Tosatti <[EMAIL PROTECTED]>
Signed-off-by: Avi Kivity <[E
The following patchset, based on Marcelo's original mmu scaling patch,
allows the kvm guest walker to run concurrently. Shadow pagetable
manipulation is still single threaded.
Handling pte writes is similar to an early version of the patchset: the page
is derived from the guest pte being written.
Since gfn_to_page() is a sleeping function, and we want to make the core mmu
spinlocked, we need to pass the page from the walker context (which can sleep)
to the shadow context (which cannot).
Signed-off-by: Avi Kivity <[EMAIL PROTECTED]>
---
arch/x86/kvm/mmu.c | 58 +++
From: Marcelo Tosatti <[EMAIL PROTECTED]>
In preparation for a mmu spinlock, add kvm_read_guest_atomic()
and use it in fetch() and prefetch_page().
Signed-off-by: Marcelo Tosatti <[EMAIL PROTECTED]>
Signed-off-by: Avi Kivity <[EMAIL PROTECTED]>
---
arch/x86/kvm/paging_tmpl.h | 28 +
38 matches
Mail list logo
