similar to: [NET] front: Fix features on resume when csum is off

Hi Herbert, I wonder if you''ve got a chance to look at netfront in light of the new stuff in davem''s network tree (the stuff that''s in http://git.kernel.org/?p=linux/kernel/git/davem/net-2.6.22.git). In particular, struct sk_buff has been changed so that "nh" has gone, and the replacement can be just an offset rather than a full pointer. This breaks the

Hi: Here are the patches I used for testing KVM with virtio-net using TSO. There are three patches for the tun device which are basically Rusty's patches with the mmap turned into copying (for correctness). Two patches are for the virtio-net frontend, one required to support receiving SG/TSO, and the other useful for testing SG per se. The other patch is to the KVM backend to make all this

Hi: Here are the patches I used for testing KVM with virtio-net using TSO. There are three patches for the tun device which are basically Rusty's patches with the mmap turned into copying (for correctness). Two patches are for the virtio-net frontend, one required to support receiving SG/TSO, and the other useful for testing SG per se. The other patch is to the KVM backend to make all this

Hi Keir: These two patches remove the need for netloop by performing the copying in netback and only if it is necessary. The rationale is that most packets will be processed without delay allowing them to be freed without copying at all. So instead of copying every packet destined to dom0 we''ll only copy those that linger longer than a specified amount of time (currently 0.5s). As it

An embedded and charset-unspecified text was scrubbed... Name: not available Url: http://lists.linux-foundation.org/pipermail/virtualization/attachments/20080714/a884a437/attachment.txt

An embedded and charset-unspecified text was scrubbed... Name: not available Url: http://lists.linux-foundation.org/pipermail/virtualization/attachments/20080714/a884a437/attachment.txt

Hi Keir: [LINUX] netfront: Cleanup and fix TSO/GSO/CHECKSUM conditionals This patch tries to minimise the amount of code that is conditionally compiled. This is desirable (and the Linux way) as it helps to prevent people breaking code unwittingly since conditionals may hide compile problems. It also adds a missing conditional around the TSO ethtool operations. This also helps the building of

Looks like the GSO is involved? I got this while running Dom0 only (no guests), with a BOINC/Rosetta@home application running on all 4 cores. changeset: 10649:8e55c5c11475 Build: x86_32p (pae). ------------[ cut here ]------------ kernel BUG at net/core/dev.c:1133! invalid opcode: 0000 [#1] SMP CPU: 0 EIP: 0061:[<c04dceb0>] Not tainted VLI EFLAGS: 00210297 (2.6.16.13-xen

Herbert tells me that returning NETDEV_TX_BUSY from hard_start_xmit is seen as a poor thing to do; we should cache the packet and stop the queue. Signed-off-by: Rusty Russell <rusty at rustcorp.com.au> --- drivers/net/virtio_net.c | 58 ++++++++++++++++++++++++++++++++--------------- 1 file changed, 40 insertions(+), 18 deletions(-) diff -r 26ba2dd67f46 drivers/net/virtio_net.c ---

Herbert tells me that returning NETDEV_TX_BUSY from hard_start_xmit is seen as a poor thing to do; we should cache the packet and stop the queue. Signed-off-by: Rusty Russell <rusty at rustcorp.com.au> --- drivers/net/virtio_net.c | 58 ++++++++++++++++++++++++++++++++--------------- 1 file changed, 40 insertions(+), 18 deletions(-) diff -r 26ba2dd67f46 drivers/net/virtio_net.c ---

On Thu, Oct 27, 2005 at 10:22:08PM -0700, H. Peter Anvin wrote: > > >So it looks like as long as dash can link with klibc then: OK, I've got it to build with klibc using klcc: $ size src/dash text data bss dec hex filename 63333 356 5940 69629 10ffd src/dash Unfortunately there seems to be something fishy in parameter expansion as it fails three

Hi: I noticed a bug in the realloc error checking code in the QEMU backend for RealTek8139. However, what''s worse is that there is no cap on the total size of the transmit buffer at all. So a guest can keep extending it until memory runs out. CP_TX_BUFFER_SIZE is already 64K. So it seems to me that we don''t need the while loop to extend the buffer at all since no transmitted

Hi: The following patch replaces a bash-specific substitution with equivalent POSIX /bin/sh substitutions. This allows a Xen-patched Linux to build when /bin/sh is a POSIX shell other than bash. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Thanks, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au> Home Page:

Hi Keir: It has been brought to my attention that modular oprofile is broken because it now calls get_vm_area which is not an exported function. This appears to be the result of the changeset that removed vm_map_xen_pages which was only used by oprofile. So we either need to export get_vm_area in Xen, or put that function back. What do you think about this? Thanks, -- Visit Openswan at

On Thursday 09 October 2008 06:34:59 Mark McLoughlin wrote: > From: Herbert Xu <herbert.xu at redhat.com> > > If segmentation offload is enabled by the host, we currently allocate > maximum sized packet buffers and pass them to the host. This uses up > 20 ring entries, allowing us to supply only 20 packet buffers to the > host with a 256 entry ring. This is a huge overhead

On Thursday 09 October 2008 06:34:59 Mark McLoughlin wrote: > From: Herbert Xu <herbert.xu at redhat.com> > > If segmentation offload is enabled by the host, we currently allocate > maximum sized packet buffers and pass them to the host. This uses up > 20 ring entries, allowing us to supply only 20 packet buffers to the > host with a 256 entry ring. This is a huge overhead

fix subject and added warning to patch mail, thanks for review. ----- Forwarded message from "H. Peter Anvin" <hpa at zytor.com> ----- Date: Mon, 29 Mar 2010 15:07:01 -0700 From: "H. Peter Anvin" <hpa at zytor.com> To: maximilian attems <max at stro.at> Cc: Colin Watson <cjwatson at ubuntu.com>, klibc at zytor.com, Herbert Xu <herbert at

Hello! The rcu_assign_pointer() primitive currently unconditionally executes a memory barrier, even when a NULL pointer is being assigned. This has lead some to avoid using rcu_assign_pointer() for NULL pointers, which loses the self-documenting advantages of rcu_assign_pointer() This patch uses __builtin_const_p() to omit needless memory barriers for NULL-pointer assignments at compile time

On Fri, Jun 04, 2010 at 10:28:56AM +0930, Rusty Russell wrote: > This patch is a subset of an already upstream patch, but this portion > is useful in earlier releases. > > Please consider for the 2.6.32 and 2.6.33 stable trees. > > If the add_buf operation fails, indicate failure to the caller. > > Signed-off-by: Bruce Rogers <brogers at novell.com> >

On Fri, Jun 04, 2010 at 10:28:56AM +0930, Rusty Russell wrote: > This patch is a subset of an already upstream patch, but this portion > is useful in earlier releases. > > Please consider for the 2.6.32 and 2.6.33 stable trees. > > If the add_buf operation fails, indicate failure to the caller. > > Signed-off-by: Bruce Rogers <brogers at novell.com> >