On Tue, Sep 23, 2014 at 03:43:59AM +, Saha, Avik (AWS) wrote:
> So with DPDK 1.7 there are 2 separate implementations - one is the rte_hash
> which does not support LRU (at least to my understanding - I could be wrong
> here) and then there is the librte_table library which has support for
On 9/22/2014 6:33 PM, Matthew Hall wrote:
> On Tue, Sep 23, 2014 at 01:08:21AM +, Saha, Avik (AWS) wrote:
>> I was wondering if there is way to use the rte_table_hash_lru without
>> building a pipeline - Basically using the same hash table like functionality
>> of add, delete and lookup
Provides a small sample application(guest_vm_power_mgr) to run on a VM.
The application is run by providing a core mask(-c) and number of memory
channels(-n). The core mask corresponds to the number of lcore channels to
attempt to open. A maximum of 64 channels per VM is allowed. The channels must
Signed-off-by: Alan Carew
---
app/test/Makefile | 1 +
app/test/autotest_data.py | 13 +++
app/test/test_power_vm.c | 215 ++
3 files changed, 229 insertions(+)
create mode 100644 app/test/test_power_vm.c
diff --git a/app/test/Makefile
Add CONFIG_RTE_LIBRTE_POWER_VM to config/common_linuxapp, default=n
As both host and guest side rely on the same API(librte_power) but different
implementations, it requires the following configurations:
Host: CONFIG_RTE_LIBRTE_POWER_VM=n and Add CONFIG_RTE_LIBRTE_POWER=y
Guest:
Provides a command packet format for host and guest.
Signed-off-by: Alan Carew
---
lib/librte_power_vm/channel_commands.h | 68 ++
1 file changed, 68 insertions(+)
create mode 100644 lib/librte_power_vm/channel_commands.h
diff --git
Re-using the host based librte_power API the alternate implementation uses
the guest channel API to forward request for frequency changes to the host
monitor.
A subset of the librte_power API is supported:
rte_power_init(unsigned lcore_id)
rte_power_exit(unsigned lcore_id)
Allows for the opening of Virtio-Serial devices on a VM, where a DPDK
application can send packets to the host based monitor. The packet formatted is
specified in channel_commands.h
Each device appears as a serial device in path
/dev/virtio-ports/virtio.serial.port.. where each lcore
in a DPDK
A wrapper around librte_power, providing locking around the non-threadsafe
library, allowing for frequency changes based on core masks and core numbers
from both the CLI thread and epoll monitor thread.
Signed-off-by: Alan Carew
---
examples/vm_power_manager/Makefile | 57 +++
A wrapper around librte_power, providing locking around the non-threadsafe
library, allowing for frequency changes based on core masks and core numbers
from both the CLI thread and epoll monitor thread.
Signed-off-by: Alan Carew
---
examples/vm_power_manager/power_manager.c | 234
The CLI is used for administrating the channel monitor and manager and
manually setting the CPU frequency on the host.
Supports the following commands:
add_vm [Mul-choice STRING]: add_vm|rm_vm , add a VM for subsequent
operations with the CLI or remove a previously added VM from the VM Power
The manager is responsible for adding communications channels to the Monitor
thread, tracking and reporting VM state and employs the libvirt API for
synchronization with the KVM Hypervisor. The manager interacts with the
Hypervisor to discover the mapping of virtual CPUS(vCPUs) to the host
The following patches add two DPDK sample applications and an alternate
implementation of librte_power for use in virtualized environments.
The idea is to provide librte_power functionality from within a VM to address
the lack of MSRs to facilitate frequency changes from within a VM.
It is ideally
On Tue, Sep 23, 2014 at 01:08:21AM +, Saha, Avik (AWS) wrote:
> I was wondering if there is way to use the rte_table_hash_lru without
> building a pipeline - Basically using the same hash table like functionality
> of add, delete and lookup without setting up a pipeline and connect it to
>
Hi Ivan
(2014/09/22 16:46), Ivan Boule wrote:
>
> This is a "nice to have" PMD.
>
> Just a minor comment.
> I guess it should be NTT instead of Intel in the following comment
> "# * Neither the name of Intel Corporation nor the names of its".
>
I appreciate your comment. I will rewrite and
On Mon, Sep 22, 2014 at 12:23:36PM -0700, Matthew Hall wrote:
> I fixed some of the clang errors a few weeks ago. But some of my patches got
> sent back due to issues seen by others and I didn't have time to fix them yet.
Can you elaborate on the specific issue here?
Neil
On Mon, Sep 22, 2014 at 2:22 PM, Neil Horman wrote:
>
> just a note here - This usage is great reflection of deprecation policy, we
> deprecated this back in the 1.6 time frame, left it around for the 1.7
> release,
> and remove it for the 1.8 release. I'd love to add some deprecation
>
On Mon, Sep 22, 2014 at 07:34:35PM +0100, Alan Carew wrote:
> Re-using the host based librte_power API the alternate implementation uses
> the guest channel API to forward request for frequency changes to the host
> monitor.
> A subset of the librte_power API is supported:
>
On Mon, Sep 22, 2014 at 04:05:29PM -0400, Neil Horman wrote:
> On Mon, Sep 22, 2014 at 12:23:36PM -0700, Matthew Hall wrote:
> > I fixed some of the clang errors a few weeks ago. But some of my patches
> > got sent back due to issues seen by others and I didn't have time to fix
> > them yet.
>
On Tue, Sep 16, 2014 at 3:05 PM, David Marchand
wrote:
> On Tue, Aug 26, 2014 at 4:12 PM, David Marchand
> wrote:
>
>> This patchset reworks the bonding pmd so that we don't need to modify the
>> eal
>> for this pmd to work.
>>
>> Basically, the arguments parsed at bond_init are stored in the
I fixed some of the clang errors a few weeks ago. But some of my patches got
sent back due to issues seen by others and I didn't have time to fix them yet.
--
Sent from my mobile device.
On September 22, 2014 6:18:51 AM PDT, Neil Horman
wrote:
>On Mon, Sep 22, 2014 at 09:36:33AM +,
Hi,
I am able to access the link now.
Regards,
Mukesh
On Mon, Sep 22, 2014 at 10:14 AM, Mukesh Dua wrote:
> Hi,
>
> I am not able to access the link. Seems the site is offline.
> Can you please share some alternate link.
>
> Regards,
> Mukesh
>
> On Sat, Sep 20, 2014 at 1:45 AM, Aaro Koskinen
Indent files modified in previous commit.
Signed-off-by: David Marchand
---
lib/librte_eal/common/eal_common_options.c | 48 +++
lib/librte_eal/linuxapp/eal/eal.c | 58 ++--
2 files changed, 53 insertions(+), 53 deletions(-)
diff --git
Identify all options through the getopt_long return value.
This way, we only need a big switch/case.
Indentation is broken to ease commit review (fixed in next commit).
Suggested-by: Neil Horman
Signed-off-by: David Marchand
---
lib/librte_eal/bsdapp/eal/eal.c | 21 +++-
All common options are now in a single file.
Common usage() has been moved as well.
Signed-off-by: David Marchand
---
lib/librte_eal/bsdapp/eal/Makefile |1 +
lib/librte_eal/bsdapp/eal/eal.c | 340 +++-
lib/librte_eal/common/eal_common_options.c |
Signed-off-by: David Marchand
---
lib/librte_eal/bsdapp/eal/eal.c | 21 +++
lib/librte_eal/linuxapp/eal/eal.c | 42 +
2 files changed, 21 insertions(+), 42 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal.c
We can handle both short and long options for those in the same case.
Signed-off-by: David Marchand
---
lib/librte_eal/bsdapp/eal/eal.c | 18 ++
lib/librte_eal/linuxapp/eal/eal.c | 18 ++
2 files changed, 4 insertions(+), 32 deletions(-)
diff --git
Signed-off-by: David Marchand
---
lib/librte_eal/bsdapp/eal/eal.c | 21 ++---
lib/librte_eal/linuxapp/eal/eal.c |6 ++
2 files changed, 12 insertions(+), 15 deletions(-)
diff --git a/lib/librte_eal/bsdapp/eal/eal.c b/lib/librte_eal/bsdapp/eal/eal.c
index
Following commit cac6d08c8bde2fdb57806c49038187cdb54219a8 and
4bf3fe634a4d9dfce90c4167f3a47d0e2ddf1e64, this option is not available anymore.
Signed-off-by: David Marchand
---
lib/librte_eal/bsdapp/eal/eal.c |7 ---
lib/librte_eal/linuxapp/eal/eal.c |7 ---
2 files changed, 14
Following Neil comments, here is a patchset to rework the eal options parsing.
I tried to have everything common to linux and bsd in a single file.
I ran a little make test on linux, it looks fine (at least I have as many fails
as before my changes).
There is still work in this part, but I want
Hi,
I am not able to access the link. Seems the site is offline.
Can you please share some alternate link.
Regards,
Mukesh
On Sat, Sep 20, 2014 at 1:45 AM, Aaro Koskinen wrote:
> Hi,
>
> On Fri, Sep 19, 2014 at 03:57:52PM +0530, Mukesh Dua wrote:
> > Did someone tried porting DPDK to ARM
On 09/19/2014 02:27 PM, mukawa at igel.co.jp wrote:
> From: Tetsuya Mukawa
>
> 'null PMD' is a virtual device driver particulary designed to measure
> performance of DPDK applications and DPDK PMDs. When an application call rx,
> null PMD just allocate mbufs and return those. Also tx, the PMD
Hi all,
just looking to see if anyone has any suggestions to help me debug an issue I'm
seeing here. Basically, the clang target is no longer working for me on Fedora
20 -due to errors when compiling up the kernel modules. The interesting thing
is that the gcc target works fine, while the
On Mon, Sep 22, 2014 at 09:36:33AM +, Richardson, Bruce wrote:
> Hi all,
>
> just looking to see if anyone has any suggestions to help me debug an issue
> I'm seeing here. Basically, the clang target is no longer working for me on
> Fedora 20 -due to errors when compiling up the kernel
On Mon, Sep 22, 2014 at 11:59:33AM +, Jastrzebski, MichalX K wrote:
> Hi Neil,
> I agree with you that the most important is to release a code that works best
> without errors and this is what we all are working on. Pawel's answer "no
> time" doesn't sounds good here (he meant something
On Mon, Sep 22, 2014 at 01:02:08PM +0200, David Marchand wrote:
> On Tue, Sep 16, 2014 at 3:05 PM, David Marchand
> wrote:
>
> > On Tue, Aug 26, 2014 at 4:12 PM, David Marchand > 6wind.com>
> > wrote:
> >
> >> This patchset reworks the bonding pmd so that we don't need to modify the
> >> eal
>
On Mon, Sep 22, 2014 at 10:37:54AM +0200, David Marchand wrote:
> Following Neil comments, here is a patchset to rework the eal options parsing.
> I tried to have everything common to linux and bsd in a single file.
>
> I ran a little make test on linux, it looks fine (at least I have as many
>
On Mon, Sep 22, 2014 at 10:38:00AM +0200, David Marchand wrote:
> Identify all options through the getopt_long return value.
> This way, we only need a big switch/case.
>
> Indentation is broken to ease commit review (fixed in next commit).
>
> Suggested-by: Neil Horman
> Signed-off-by: David
On Mon, Sep 22, 2014 at 10:37:55AM +0200, David Marchand wrote:
> Following commit cac6d08c8bde2fdb57806c49038187cdb54219a8 and
> 4bf3fe634a4d9dfce90c4167f3a47d0e2ddf1e64, this option is not available
> anymore.
>
> Signed-off-by: David Marchand
> ---
> lib/librte_eal/bsdapp/eal/eal.c |7
> I think that will work, but I believe you're making it more complicated (and
> less reusable) than it needs to be. What I think you really need to do is
> create a new rte api call, rte_eal_alarm_cancel_sync (something like the
> equivalent of del_timer_sync in linux, that wraps up the
>
On Mon, Sep 22, 2014 at 06:26:21AM +, Wodkowski, PawelX wrote:
> > I think that will work, but I believe you're making it more complicated (and
> > less reusable) than it needs to be. What I think you really need to do is
> > create a new rte api call, rte_eal_alarm_cancel_sync (something
41 matches
Mail list logo
