similar to: [PATCH 1/4] Support accelerated network plugin modules

There would seem to be a potential deadlock in the netfront accelerator plugin support. When the configured accelerator changes in xenstore, netfront tries to load the new plugin using request_module(). It does this from a workqueue work item. request_module() will invoke modprobe which in some circumstances (I''m not sure exactly what - I''ve not managed to reproduce it myself)

Signed-off-by: Kieran Mansley <kmansley@solarflare.com> Cleanup vif_state on device removal diff -r f992a517aba5 drivers/xen/netfront/accel.c --- a/drivers/xen/netfront/accel.c Wed Oct 03 13:41:33 2007 +0100 +++ b/drivers/xen/netfront/accel.c Wed Oct 03 13:45:55 2007 +0100 @@ -500,6 +500,30 @@ EXPORT_SYMBOL_GPL(netfront_accelerator_r /* + * Remove the hooks from a single vif state.

This patch against linux-2.6.18-xen.hg changest 974:aa133dee7792 Update sfc_netback driver to match sfc_resource 3.0.2.2074 Add support for direct guest access and acceleration of SFC9000 series NICs. Improve handling of NIC reset in sfc_netback Remove nic_index state and replace with if_index from struct net_device Remove duplication of header files with sfc_resource driver Signed-off-by:

Signed-off-by: Kieran Mansley <kmansley@solarflare.com> Allow iomem permissions to be set up through grant table ops diff -r 749b60ccc177 xen/arch/x86/mm.c --- a/xen/arch/x86/mm.c Wed Jul 25 14:03:08 2007 +0100 +++ b/xen/arch/x86/mm.c Wed Jul 25 14:03:12 2007 +0100 @@ -594,6 +594,14 @@ get_##level##_linear_pagetable( return 1; \ } + +int iomem_page_test(unsigned long mfn,

I''m having problems starting more than three domains. It sometimes works fine, but more often than not the 4th domain''s root block device times out and so the domU kernel panics as there''s no /dev/root: XENBUS: Timeout connecting to device: device/vbd/2057 (state 6) XENBUS: Timeout connecting to device: device/vif/0 (state 6) XENBUS: Timeout connecting to device:

While writing a driver for a device doing lots of DMA I''ve hit an "swiotlb_full()" problem. This surprised me somewhat as I wouldn''t have expected to need the use of the software TLB - it''s a 64 bit capable device on a server with only 2 GB of RAM, and so I''d have expected to be using a hardware TLB. Is this a peculiarity of Xen, or should I be right

cleanup accelerators list on netfront unload Signed-off-by <kmansley@solarflare.com> diff -r fdfd708d2ec4 drivers/xen/netfront/accel.c --- a/drivers/xen/netfront/accel.c Wed Oct 24 15:38:35 2007 +0100 +++ b/drivers/xen/netfront/accel.c Wed Oct 24 16:25:57 2007 +0100 @@ -75,6 +75,24 @@ void netif_init_accel(void) spin_lock_init(&accelerators_lock); } +void netif_exit_accel(void) +{

This set of patches adds the support for acceleration plugins to the netfront/netback drivers. These plugins are intended to support virtualisable network hardware that can be directly accessed from the guest, bypassing dom0. This is in response to the RFC we posted to xen-devel with an outline of our approach at the end of September. To follow will be another set of patches to provide our

hi, I want to know some details about this network accelerated modules. 1)what''s the main purpose of this plugin module? 2)any requirement for harware nic? or other hardware platform? 3)I''ve not found the code in Xen3.1 code set. so was it added to xen source code yet? Thanks in advance _______________________________________________ Xen-devel mailing list

On Tue, Jun 9, 2020 at 8:04 PM Kieran Bingham <kieran.bingham+renesas at ideasonboard.com> wrote: > > The word 'descriptor' is misspelled throughout the tree. I think, *throughout the tree* is not appropriate* here. This patch has fixed it within the file. With that, Acked-by: Souptick Joarder <jrdr.linux at gmail.com> > > Fix it up accordingly: >

On Thu, 2014-01-16 at 11:07 -0800, Michael Dalton wrote: > On Jan 16, 2014 at 10:57 AM, Ben Hutchings <bhutchings at solarflare.com> wrote: > > Why write a loop when you can do: > > i = queue - dev->_rx; > Good point, the loop approach was done in get_netdev_queue_index -- > I agree your fix is faster and simpler. I'll fix in next patchset. It's

On Thu, 2014-01-16 at 11:07 -0800, Michael Dalton wrote: > On Jan 16, 2014 at 10:57 AM, Ben Hutchings <bhutchings at solarflare.com> wrote: > > Why write a loop when you can do: > > i = queue - dev->_rx; > Good point, the loop approach was done in get_netdev_queue_index -- > I agree your fix is faster and simpler. I'll fix in next patchset. It's

hi guys I wonder if any of you might be using SFN6122F-R7 SFP+ (SFC9000, same firmware everywhere, Centos 7.5 too. I'm trying poor man's setup to get the servers onto 10GbE network. Setup is such that three Dell R815 are connected to each other, each has one Solarflare(SFP ports) and each Solarflare is set as net-team(both ports on a card are net-team device) with runner in broadcast

Hi All, I''m trying to choose between 10Gb Ethernet cards for Xen. The trouble is I''m getting quite confused with all the different terms different companies use to describe NIC''s virtualization support. Any ideas what the differences are between these brands? I am trying to sort a comparison table but will need some help: Feature Intel ET Series Mellanox

lejeczek via CentOS wrote: > hi guys > > I wonder if any of you might be using SFN6122F-R7 SFP+ (SFC9000, same > firmware everywhere, Centos 7.5 too. > > I'm trying poor man's setup to get the servers onto 10GbE network. > > > Setup is such that three Dell R815 are connected to each other, each has > one Solarflare(SFP ports) and each Solarflare is set as

On Thu, 2014-01-16 at 01:38 -0800, Michael Dalton wrote: [...] > --- a/include/linux/netdevice.h > +++ b/include/linux/netdevice.h [...] > @@ -2401,6 +2416,23 @@ static inline int netif_copy_real_num_queues(struct net_device *to_dev, > #endif > } > > +#ifdef CONFIG_SYSFS > +static inline unsigned int get_netdev_rx_queue_index( > + struct netdev_rx_queue *queue) >

On Thu, 2014-01-16 at 01:38 -0800, Michael Dalton wrote: [...] > --- a/include/linux/netdevice.h > +++ b/include/linux/netdevice.h [...] > @@ -2401,6 +2416,23 @@ static inline int netif_copy_real_num_queues(struct net_device *to_dev, > #endif > } > > +#ifdef CONFIG_SYSFS > +static inline unsigned int get_netdev_rx_queue_index( > + struct netdev_rx_queue *queue) >

On Mon, 2013-06-10 at 10:07 +0300, Michael S. Tsirkin wrote: > On Thu, Feb 07, 2013 at 01:14:20PM -0500, David Miller wrote: > > From: Ben Hutchings <bhutchings at solarflare.com> > > Date: Thu, 7 Feb 2013 16:20:46 +0000 > > > > > If the consensus is still that we must preserve packets exactly (aside > > > from the usual modifications by IP routers)

Hello again -- inevitably, I discovered the (documented) answer 30 seconds after sending my question. Using gsub(...,extended=FALSE) is the way to avoid this issue. Cheers, Kieran -- Kieran Healy, Assistant Professor, Sociology Dept, University of Arizona. kjhealy at u.arizona.edu :: (520) 621-3480 http://www.u.arizona.edu/~kjhealy -------------- next part -------------- An embedded message

On 09.06.20 14:46, Kieran Bingham wrote: > The word 'descriptor' is misspelled throughout the tree. > "mm/balloon_compaction:" is a better subject prefix Reviewed-by: David Hildenbrand <david at redhat.com> > Fix it up accordingly: > decriptors -> descriptors > > Signed-off-by: Kieran Bingham <kieran.bingham+renesas at ideasonboard.com>