ing freeing of wbinvd_dirty_mask.
Cc: sta...@vger.kernel.org
Signed-off-by: Ido Yariv <i...@wizery.com>
---
arch/x86/kvm/x86.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 6c633de..9ec8c1d 100644
--- a/arch/x86/kvm/x86.c
+++ b
ing freeing of wbinvd_dirty_mask.
Cc: sta...@vger.kernel.org
Signed-off-by: Ido Yariv
---
arch/x86/kvm/x86.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 6c633de..9ec8c1d 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x8
und this by setting the timeout to at least 2 jiffies on such
systems.
The 10ms figure was originally selected based on tests performed with
the current implementation and HZ = 1000. Thus, leave the behavior on
systems with HZ > 100 unchanged.
Signed-off-by: Ido Yariv
---
include/net/tcp.h
this by setting the timeout to at least 2 jiffies on such
systems.
The 10ms figure was originally selected based on tests performed with
the current implementation and HZ = 1000. Thus, leave the behavior on
systems with HZ 100 unchanged.
Signed-off-by: Ido Yariv idox.ya...@intel.com
---
include/net/tcp.h
Hi Eric,
On Wed, May 27, 2015 at 03:15:26PM -0400, Ido Yariv wrote:
> Hi Eric,
>
> On Wed, May 27, 2015 at 10:24:16AM -0700, Eric Dumazet wrote:
> > On Wed, 2015-05-27 at 12:54 -0400, Ido Yariv wrote:
> > > Hi Eric,
> >
> > > That's a nice optimi
Hi Eric,
On Wed, May 27, 2015 at 10:24:16AM -0700, Eric Dumazet wrote:
> On Wed, 2015-05-27 at 12:54 -0400, Ido Yariv wrote:
> > Hi Eric,
>
> > That's a nice optimization ;)
> >
> > However, I think that with Nicholas Mc Guire's recent changes to
> > msecs_t
Hi Eric,
On Wed, May 27, 2015 at 09:23:36AM -0700, Eric Dumazet wrote:
> On Wed, 2015-05-27 at 11:23 -0400, Ido Yariv wrote:
>
> > Signed-off-by: Ido Yariv
> > ---
> > include/net/tcp.h | 9 +
> > net/ipv4/tcp_output.c | 2 +-
> > 2 files c
Hi Eric,
On Wed, May 27, 2015 at 07:56:25AM -0700, Eric Dumazet wrote:
> On Wed, 2015-05-27 at 10:40 -0400, Ido Yariv wrote:
>
> > HZ=100 is used on some embedded platforms, so it's still something we
> > have to deal with unfortunately..
> >
> > Since
Hi Eric,
On Wed, May 27, 2015 at 06:41:17AM -0700, Eric Dumazet wrote:
> On Wed, 2015-05-27 at 11:36 +, David Laight wrote:
> > From: Of Ido Yariv
> > > Sent: 26 May 2015 21:17
> > > The Tail Loss Probe RFC specifies that the PTO value should be set to
> >
Hi Eric,
On Wed, May 27, 2015 at 10:24:16AM -0700, Eric Dumazet wrote:
On Wed, 2015-05-27 at 12:54 -0400, Ido Yariv wrote:
Hi Eric,
That's a nice optimization ;)
However, I think that with Nicholas Mc Guire's recent changes to
msecs_to_jiffies (http://marc.info/?l=linux-kernelm
Hi Eric,
On Wed, May 27, 2015 at 07:56:25AM -0700, Eric Dumazet wrote:
On Wed, 2015-05-27 at 10:40 -0400, Ido Yariv wrote:
HZ=100 is used on some embedded platforms, so it's still something we
have to deal with unfortunately..
Since the '2' here is a lower bound, and msecs_to_jiffies
Hi Eric,
On Wed, May 27, 2015 at 06:41:17AM -0700, Eric Dumazet wrote:
On Wed, 2015-05-27 at 11:36 +, David Laight wrote:
From: Of Ido Yariv
Sent: 26 May 2015 21:17
The Tail Loss Probe RFC specifies that the PTO value should be set to
max(2 * SRTT, 10ms), where SRTT
Hi Eric,
On Wed, May 27, 2015 at 09:23:36AM -0700, Eric Dumazet wrote:
On Wed, 2015-05-27 at 11:23 -0400, Ido Yariv wrote:
Signed-off-by: Ido Yariv idox.ya...@intel.com
---
include/net/tcp.h | 9 +
net/ipv4/tcp_output.c | 2 +-
2 files changed, 10 insertions(+), 1
Hi Eric,
On Wed, May 27, 2015 at 03:15:26PM -0400, Ido Yariv wrote:
Hi Eric,
On Wed, May 27, 2015 at 10:24:16AM -0700, Eric Dumazet wrote:
On Wed, 2015-05-27 at 12:54 -0400, Ido Yariv wrote:
Hi Eric,
That's a nice optimization ;)
However, I think that with Nicholas Mc
und this by setting the timeout to at least 2 jiffies on such
systems.
The 10ms figure was originally selected based on tests performed with
the current implementation and HZ = 1000. Thus, leave the behavior on
systems with HZ > 100 unchanged.
Signed-off-by: Ido Yariv
---
net/ipv4/tcp_output.c | 3
Hi Eric,
On Tue, May 26, 2015 at 11:25:21AM -0700, Eric Dumazet wrote:
> On Tue, 2015-05-26 at 13:55 -0400, Ido Yariv wrote:
>
> >
> > The platform this was tested on was an embedded platform with a wifi
> > module (11n, 20MHZ). The other end was a
Hi Eric,
On Tue, May 26, 2015 at 10:13:40AM -0700, Eric Dumazet wrote:
> On Tue, 2015-05-26 at 13:02 -0400, Ido Yariv wrote:
> > Hi Eric,
> >
> > On Tue, May 26, 2015 at 09:23:55AM -0700, Eric Dumazet wrote:
> > > Have you really hit an issue, or did
Hi Eric,
On Tue, May 26, 2015 at 09:23:55AM -0700, Eric Dumazet wrote:
> On Tue, 2015-05-26 at 10:25 -0400, Ido Yariv wrote:
> > The Tail Loss Probe RFC specifies that the PTO value should be set to
> > max(2 * SRTT, 10ms), where SRTT is the smoothed round-trip time.
> &g
this issue, increase the number of jiffies by one,
ensuring that the timeout won't expire in less than 10ms.
Signed-off-by: Ido Yariv
---
net/ipv4/tcp_output.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 534e5fd..6f57d3d 100644
Hi Eric,
On Tue, May 26, 2015 at 10:13:40AM -0700, Eric Dumazet wrote:
On Tue, 2015-05-26 at 13:02 -0400, Ido Yariv wrote:
Hi Eric,
On Tue, May 26, 2015 at 09:23:55AM -0700, Eric Dumazet wrote:
Have you really hit an issue, or did you send this patch after all these
msecs_to_jiffies
this by setting the timeout to at least 2 jiffies on such
systems.
The 10ms figure was originally selected based on tests performed with
the current implementation and HZ = 1000. Thus, leave the behavior on
systems with HZ 100 unchanged.
Signed-off-by: Ido Yariv idox.ya...@intel.com
---
net/ipv4
Hi Eric,
On Tue, May 26, 2015 at 11:25:21AM -0700, Eric Dumazet wrote:
On Tue, 2015-05-26 at 13:55 -0400, Ido Yariv wrote:
The platform this was tested on was an embedded platform with a wifi
module (11n, 20MHZ). The other end was a computer running Windows, and
the benchmarking
this issue, increase the number of jiffies by one,
ensuring that the timeout won't expire in less than 10ms.
Signed-off-by: Ido Yariv idox.ya...@intel.com
---
net/ipv4/tcp_output.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index
Hi Eric,
On Tue, May 26, 2015 at 09:23:55AM -0700, Eric Dumazet wrote:
On Tue, 2015-05-26 at 10:25 -0400, Ido Yariv wrote:
The Tail Loss Probe RFC specifies that the PTO value should be set to
max(2 * SRTT, 10ms), where SRTT is the smoothed round-trip time.
The PTO value is converted
Hi Sjur,
On Thu, Feb 21, 2013 at 06:15:34PM +0100, sjur.brandel...@stericsson.com wrote:
> From: Sjur Brændeland
>
> Add function find_rsc_table_va to firmware ops. This function
> returns the location of the resource table in shared memory
> after loading.
>
> Signed-off-by: Sjur Brændeland
Hi Sjur,
On Thu, Feb 21, 2013 at 06:15:34PM +0100, sjur.brandel...@stericsson.com wrote:
From: Sjur Brændeland sjur.brandel...@stericsson.com
Add function find_rsc_table_va to firmware ops. This function
returns the location of the resource table in shared memory
after loading.
Hi Sjur,
On Sun, Feb 10, 2013 at 12:39:12PM +0100, sjur.brandel...@stericsson.com wrote:
> From: Sjur Brændeland
>
> Copy resource table from first to second firmware loading.
> After firmware is loaded to memory, update the vdevs resource
> pointer to the resource table kept in device memory.
Hi Sjur,
On Sun, Feb 10, 2013 at 12:39:11PM +0100, sjur.brandel...@stericsson.com wrote:
> From: Sjur Brændeland
>
> Simplify handling of max_notifyid by simply counting the
> number of vrings.
>
> Signed-off-by: Sjur Brændeland
> ---
> drivers/remoteproc/remoteproc_core.c | 34
>
Hi Sjur,
On Sun, Feb 10, 2013 at 12:39:09PM +0100, sjur.brandel...@stericsson.com wrote:
> From: Sjur Brændeland
>
> Support virtio configuration space and device status and
> feature negotiation with remote device. This virtio device
> can now access the resource table in shared memory.
>
>
Hi Sjur,
Sorry for the (very) late reply.
On Sun, Feb 10, 2013 at 12:39:05PM +0100, sjur.brandel...@stericsson.com wrote:
> From: Sjur Brændeland
>
> Refactor rproc_elf_find_rsc_table and split out the scanning
> for the section header named resource table. This is done to
> prepare for
Hi Sjur,
On Sun, Feb 10, 2013 at 12:39:06PM +0100, sjur.brandel...@stericsson.com wrote:
> From: Sjur Brændeland
>
> Add function find_rsc_table_va to firmware ops. This function
> returns the location of the resource table in shared memory
> after loading.
>
> Signed-off-by: Sjur Brændeland
Hi Sjur,
Sorry for the (very) late reply.
On Sun, Feb 10, 2013 at 12:39:05PM +0100, sjur.brandel...@stericsson.com wrote:
From: Sjur Brændeland sjur.brandel...@stericsson.com
Refactor rproc_elf_find_rsc_table and split out the scanning
for the section header named resource table. This is
Hi Sjur,
On Sun, Feb 10, 2013 at 12:39:06PM +0100, sjur.brandel...@stericsson.com wrote:
From: Sjur Brændeland sjur.brandel...@stericsson.com
Add function find_rsc_table_va to firmware ops. This function
returns the location of the resource table in shared memory
after loading.
Hi Sjur,
On Sun, Feb 10, 2013 at 12:39:09PM +0100, sjur.brandel...@stericsson.com wrote:
From: Sjur Brændeland sjur.brandel...@stericsson.com
Support virtio configuration space and device status and
feature negotiation with remote device. This virtio device
can now access the resource table
Hi Sjur,
On Sun, Feb 10, 2013 at 12:39:11PM +0100, sjur.brandel...@stericsson.com wrote:
From: Sjur Brændeland sjur.brandel...@stericsson.com
Simplify handling of max_notifyid by simply counting the
number of vrings.
Signed-off-by: Sjur Brændeland sjur.brandel...@stericsson.com
---
Hi Sjur,
On Sun, Feb 10, 2013 at 12:39:12PM +0100, sjur.brandel...@stericsson.com wrote:
From: Sjur Brændeland sjur.brandel...@stericsson.com
Copy resource table from first to second firmware loading.
After firmware is loaded to memory, update the vdevs resource
pointer to the resource
Hi Steve,
On Sun, Dec 30, 2012 at 10:13:08PM +0800, steve.zhan wrote:
> Hi,
> Acked-by: Steve zhan zhanzhe...@gmail.com
>
> "I'd rather not test the spinlocks after they are registering as they
> might already be in use by then."
>
> Why? I think user must use it after
Hi Steve,
On Sat, Dec 29, 2012 at 05:19:08PM +0800, steve.zhan wrote:
> Hi,
>
> It is good idea add this feature.
>
> 1: Can we let the "ret = hwspin_lock_tests(ops, hwlock);" add after
> hwspin_lock_register_single have return
> succeed, that can avoid test duplicated Or error lockid. Of
Hi Steve,
On Sat, Dec 29, 2012 at 05:19:08PM +0800, steve.zhan wrote:
Hi,
It is good idea add this feature.
1: Can we let the ret = hwspin_lock_tests(ops, hwlock); add after
hwspin_lock_register_single have return
succeed, that can avoid test duplicated Or error lockid. Of course, If
Hi Steve,
On Sun, Dec 30, 2012 at 10:13:08PM +0800, steve.zhan wrote:
Hi,
Acked-by: Steve zhan zhanzhe...@gmail.com
I'd rather not test the spinlocks after they are registering as they
might already be in use by then.
Why? I think user must use it after hwspin_lock_register
Hi Sjur,
On Fri, Dec 21, 2012 at 01:20:02PM +0100, Sjur BRENDELAND wrote:
> Hi Ido,
>
> > From: Ido Yariv [mailto:i...@wizery.com]
> > > Postpone the registration of virtio devices until all
> > > vritio ring resource has been allocated.
> > > This fixe
Hi Sjur,
On Fri, Dec 21, 2012 at 01:02:02PM +0100, Sjur BRENDELAND wrote:
> Hi Ido and Ohad,
>
> > In case the remote processor is added, booted and then rebooted, the
> > program sections wont be reinitialized. This can be a bit problematic,
> > since the firmware might assume values are
Hi Sjur,
On Fri, Dec 21, 2012 at 01:02:02PM +0100, Sjur BRENDELAND wrote:
Hi Ido and Ohad,
In case the remote processor is added, booted and then rebooted, the
program sections wont be reinitialized. This can be a bit problematic,
since the firmware might assume values are initialized in
Hi Sjur,
On Fri, Dec 21, 2012 at 01:20:02PM +0100, Sjur BRENDELAND wrote:
Hi Ido,
From: Ido Yariv [mailto:i...@wizery.com]
Postpone the registration of virtio devices until all
vritio ring resource has been allocated.
This fixes the following bug: The driver's start callback
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:58PM +0100, Sjur Brændeland wrote:
> Add function find_rsc_table_va to firmware ops. This function
> returns the location of the resource table in shared memory
> after loading.
>
> Signed-off-by: Sjur Brændeland
> ---
>
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:57PM +0100, Sjur Brændeland wrote:
> Refatcor rproc_elf_find_rsc_table and split out the scanning
Small typo there.
> for the section header named resource table. This is done to
> prepare for loading firmware once.
>
> Signed-off-by: Sjur Brændeland
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:59PM +0100, Sjur Brændeland wrote:
> Call rproc_get_rsctab_addr() to get the address
> of the resource table in shared memory.
>
> Signed-off-by: Sjur Brændeland
This patch should probably be squashed with the next one, since it
doesn't do much on its
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:56PM +0100, Sjur Brændeland wrote:
> Postpone the registration of virtio devices until all
> vritio ring resource has been allocated.
> This fixes the following bug: The driver's start callback
> is called before all vring notify ids are allocated and
>
Hi Sjur,
On Fri, Dec 14, 2012 at 05:07:00PM +0100, Sjur Brændeland wrote:
> Support virtio configuration space and device status and
> feature negotiation with remote device. This virtio device
> can now access the resource table in shared memory.
>
> Signed-off-by: Sjur Brændeland
> ---
...
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:54PM +0100, Sjur Brændeland wrote:
> Load the firmware once and pave the way for two way
> communication of virtio configuration and status bits.
> The virtio ring device address is now stored in the resource
> table.
>
> NOTE: This patch requires device
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:53PM +0100, Sjur Brændeland wrote:
> Add state RPROC_LOADED as firmware loading and startup will be
> different states.
>
> Signed-off-by: Sjur Brændeland
The rproc_state holds the state of the remote processor and it will
still be offline even when the
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:52PM +0100, Sjur Brændeland wrote:
> Set the vring addresses in the resource table so that
> the remote device can read the actual addresses used.
>
> Signed-off-by: Sjur Brændeland
Since this new member will only be referenced in the last patch, perhaps
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:52PM +0100, Sjur Brændeland wrote:
Set the vring addresses in the resource table so that
the remote device can read the actual addresses used.
Signed-off-by: Sjur Brændeland sjur.brandel...@stericsson.com
Since this new member will only be referenced
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:53PM +0100, Sjur Brændeland wrote:
Add state RPROC_LOADED as firmware loading and startup will be
different states.
Signed-off-by: Sjur Brændeland sjur.brandel...@steicsson.com
The rproc_state holds the state of the remote processor and it will
still
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:54PM +0100, Sjur Brændeland wrote:
Load the firmware once and pave the way for two way
communication of virtio configuration and status bits.
The virtio ring device address is now stored in the resource
table.
NOTE: This patch requires device
Hi Sjur,
On Fri, Dec 14, 2012 at 05:07:00PM +0100, Sjur Brændeland wrote:
Support virtio configuration space and device status and
feature negotiation with remote device. This virtio device
can now access the resource table in shared memory.
Signed-off-by: Sjur Brændeland
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:56PM +0100, Sjur Brændeland wrote:
Postpone the registration of virtio devices until all
vritio ring resource has been allocated.
This fixes the following bug: The driver's start callback
is called before all vring notify ids are allocated and
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:59PM +0100, Sjur Brændeland wrote:
Call rproc_get_rsctab_addr() to get the address
of the resource table in shared memory.
Signed-off-by: Sjur Brændeland sjur.brandel...@stericsson.com
This patch should probably be squashed with the next one, since it
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:57PM +0100, Sjur Brændeland wrote:
Refatcor rproc_elf_find_rsc_table and split out the scanning
Small typo there.
for the section header named resource table. This is done to
prepare for loading firmware once.
Signed-off-by: Sjur Brændeland
Hi Sjur,
On Fri, Dec 14, 2012 at 05:06:58PM +0100, Sjur Brændeland wrote:
Add function find_rsc_table_va to firmware ops. This function
returns the location of the resource table in shared memory
after loading.
Signed-off-by: Sjur Brændeland sjur.brandel...@stericsson.com
---
In case pm_runtime_get_sync() fails, __hwspin_lock_request will exit
without calling module_put. As a result, the module could never be
removed. Fix this.
Signed-off-by: Ido Yariv
---
drivers/hwspinlock/hwspinlock_core.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git
Add testing capabilities for verifying correctness of the underlying
hwspinlock layers. This can be handy especially during development.
These tests are performed only once as part of the hwspinlock
registration.
Signed-off-by: Ido Yariv
---
drivers/hwspinlock/Kconfig |9
Add testing capabilities for verifying correctness of the underlying
hwspinlock layers. This can be handy especially during development.
These tests are performed only once as part of the hwspinlock
registration.
Signed-off-by: Ido Yariv i...@wizery.com
---
drivers/hwspinlock/Kconfig
In case pm_runtime_get_sync() fails, __hwspin_lock_request will exit
without calling module_put. As a result, the module could never be
removed. Fix this.
Signed-off-by: Ido Yariv i...@wizery.com
---
drivers/hwspinlock/hwspinlock_core.c |1 +
1 files changed, 1 insertions(+), 0 deletions
Hi Fengguang,
On Thu, Aug 23, 2012 at 08:55:15AM +0800, Fengguang Wu wrote:
> Hi Ido,
>
> The below warning should be introduced by this commit at tip/master:
>
> commit c00679af441c5be76e195deed2c76fef65a5d102
> Author: Ido Yariv
> CommitDate: Wed Aug 22 10:48:05 20
Hi Fengguang,
On Thu, Aug 23, 2012 at 08:55:15AM +0800, Fengguang Wu wrote:
Hi Ido,
The below warning should be introduced by this commit at tip/master:
commit c00679af441c5be76e195deed2c76fef65a5d102
Author: Ido Yariv i...@wizery.com
CommitDate: Wed Aug 22 10:48:05 2012 +0200
vSMP Foundation detection and initialization could be done by the
hypervisor layer, so set it up that way.
Signed-off-by: Ido Yariv
Acked-by: Shai Fultheim
---
Changes from v1:
- Fixed build errors on 32bit kernels
arch/x86/include/asm/hypervisor.h |1 +
arch/x86/include/asm/setup.h
Hi Ingo,
On Wed, Aug 22, 2012 at 10:47:02AM +0200, Ingo Molnar wrote:
>
> * Ido Yariv wrote:
>
> > vSMP Foundation does not require to serialize CPA by guaranteeing that
> > the most recent TLB entry will always be used.
> >
> > To avoid needless contentio
Hi Ingo,
On Wed, Aug 22, 2012 at 11:11:41AM +0200, Ingo Molnar wrote:
>
> * tip-bot for Ido Yariv wrote:
>
> > Commit-ID: c00679af441c5be76e195deed2c76fef65a5d102
> > Gitweb:
> > http://git.kernel.org/tip/c00679af441c5be76e195deed2c76fef65a5d102
> > Auth
Commit-ID: c00679af441c5be76e195deed2c76fef65a5d102
Gitweb: http://git.kernel.org/tip/c00679af441c5be76e195deed2c76fef65a5d102
Author: Ido Yariv
AuthorDate: Wed, 8 Aug 2012 20:01:38 +0300
Committer: Ingo Molnar
CommitDate: Wed, 22 Aug 2012 10:48:05 +0200
x86/vsmp: Use hypervisor layer
Commit-ID: c00679af441c5be76e195deed2c76fef65a5d102
Gitweb: http://git.kernel.org/tip/c00679af441c5be76e195deed2c76fef65a5d102
Author: Ido Yariv i...@wizery.com
AuthorDate: Wed, 8 Aug 2012 20:01:38 +0300
Committer: Ingo Molnar mi...@kernel.org
CommitDate: Wed, 22 Aug 2012 10:48:05 +0200
Hi Ingo,
On Wed, Aug 22, 2012 at 11:11:41AM +0200, Ingo Molnar wrote:
* tip-bot for Ido Yariv i...@wizery.com wrote:
Commit-ID: c00679af441c5be76e195deed2c76fef65a5d102
Gitweb:
http://git.kernel.org/tip/c00679af441c5be76e195deed2c76fef65a5d102
Author: Ido Yariv i
Hi Ingo,
On Wed, Aug 22, 2012 at 10:47:02AM +0200, Ingo Molnar wrote:
* Ido Yariv i...@wizery.com wrote:
vSMP Foundation does not require to serialize CPA by guaranteeing that
the most recent TLB entry will always be used.
To avoid needless contention on cpa_lock, do not lock/unlock
vSMP Foundation detection and initialization could be done by the
hypervisor layer, so set it up that way.
Signed-off-by: Ido Yariv i...@wizery.com
Acked-by: Shai Fultheim s...@scalemp.com
---
Changes from v1:
- Fixed build errors on 32bit kernels
arch/x86/include/asm/hypervisor.h |1
On Wed, Aug 08, 2012 at 08:01:38PM +0300, Ido Yariv wrote:
> vSMP Foundation detection and initialization could be done by the
> hypervisor layer, so set it up that way.
>
> Signed-off-by: Ido Yariv
> Acked-by: Shai Fultheim
> ---
Ping?
Thanks,
Ido.
--
To unsubscribe fr
On Wed, Aug 08, 2012 at 08:01:38PM +0300, Ido Yariv wrote:
vSMP Foundation detection and initialization could be done by the
hypervisor layer, so set it up that way.
Signed-off-by: Ido Yariv i...@wizery.com
Acked-by: Shai Fultheim s...@scalemp.com
---
Ping?
Thanks,
Ido.
--
To unsubscribe
vSMP Foundation detection and initialization could be done by the
hypervisor layer, so set it up that way.
Signed-off-by: Ido Yariv
Acked-by: Shai Fultheim
---
arch/x86/include/asm/hypervisor.h |1 +
arch/x86/include/asm/setup.h |7 ---
arch/x86/kernel/cpu/hypervisor.c |1
vSMP Foundation does not require to serialize CPA by guaranteeing that
the most recent TLB entry will always be used.
To avoid needless contention on cpa_lock, do not lock/unlock it if it
isn't necessary.
Based on work by Shai Fultheim .
Signed-off-by: Ido Yariv
Acked-by: Shai Fultheim
vSMP Foundation does not require to serialize CPA by guaranteeing that
the most recent TLB entry will always be used.
To avoid needless contention on cpa_lock, do not lock/unlock it if it
isn't necessary.
Based on work by Shai Fultheim s...@scalemp.com.
Signed-off-by: Ido Yariv i...@wizery.com
vSMP Foundation detection and initialization could be done by the
hypervisor layer, so set it up that way.
Signed-off-by: Ido Yariv i...@wizery.com
Acked-by: Shai Fultheim s...@scalemp.com
---
arch/x86/include/asm/hypervisor.h |1 +
arch/x86/include/asm/setup.h |7 ---
arch/x86
80 matches
Mail list logo