On tis, 2019-07-30 at 21:21 +0200, David Marchand wrote:
>
> We are really close to -rc3.
> Can you send a review or test tag?
>
Tested-by: Johan Källström
Reviewed-by: Johan Källström
I might be to late for that. Tested the patch on the v19.08-rc3 tag.
BR
Johan
Hi, for the online check I referred to the check of "default_set" via the
initial thread affinity.
I see that pthread_getaffinity_np returns an already and:ed mask, was under the
impression that pthread_getaffinity_np would return the same mask as was set
using pthread_setaffinity_np.
Looking
On Tue, Jul 30, 2019 at 6:32 PM Johan Källström
wrote:
>
> Hi, for the online check I referred to the check of "default_set" via the
> initial thread affinity.
>
> I see that pthread_getaffinity_np returns an already and:ed mask, was under
> the impression that pthread_getaffinity_np would retur
On Tue, Jul 30, 2019 at 1:38 PM Johan Källström
wrote:
> The CPU failsafe is nice to have as you could set the thread affinity to
> offline cpus.
Created a "dpdk" cpuset and put cpus 4-7 into it (my system is mono
numa with 8 cpus)
# cd /sys/fs/cgroup/cpuset/
# mkdir dpdk
# cd dpdk
# echo 4-7 >
See inline comments about not using cpuset for "thread affinity" and possible
online cpu failsafe to detect if the thread affinity mask is not a subset of
online cpus.
This feature was present before your suggested change.
The CPU failsafe is nice to have as you could set the thread affinity to
l.com;
> > olivier.m...@6wind.com; sta...@dpdk.org
> > Subject: [dpdk-dev] [PATCH] eal: fix ctrl thread affinity with --lcores
> >
> > When using -l/-c options, each lcore is mapped to a physical cpu in a
> > 1:1 fashion.
> > On the contrary, when using --lcore
> -Original Message-
> From: dev On Behalf Of David Marchand
> Sent: Tuesday, July 30, 2019 3:06 PM
> To: dev@dpdk.org
> Cc: johan.kallst...@ericsson.com; anatoly.bura...@intel.com;
> olivier.m...@6wind.com; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] eal: fix ctrl
When using -l/-c options, each lcore is mapped to a physical cpu in a
1:1 fashion.
On the contrary, when using --lcores, each lcore has its own cpuset on
which the associated EAL thread runs.
To handle those two situations, rely on the per lcore cpuset.
Introduced macros to manipulate cpusets in
8 matches
Mail list logo
