view, from OFP
perspective.
BR,
Sorin Vultureanu
Software Engineer
Linux R&D
Email sorin.vulture...@enea.com
Phone +40 723.651.943
www.enea.com
This message, including attachments, is CONFIDENTIAL. It may also be privileged
or otherwise protected by law. If you received thisĀ email b
Agreed!
Thanks Bill!
From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
Sent: Thursday, September 01, 2016 4:57 PM
To: Sorin Vultureanu
Cc: lng-odp@lists.linaro.org; lng-odp-d...@lists.linaro.org; Mike Holmes
; Bogdan Pricope ; Dan
Lilliehorn
Subject: Re: [lng-odp-dpdk] option to drop
Hi,
There is a lock on odp_queue_deq(..)
I think there should be at least some option to disable the lock on
odp_queue_deq(). For example, when I am polling on a queue with a single core
it impacts the performance. Lockless design would be even better.
"
odp_buffer_hdr_t *queue_deq(queue_entry_
Hi,
Please increase the number of available pools for odp-dpdk and odp-linux as now
16 is on the small side.
ofp/scripts/odp-dpdk/platform/linux-dpdk/include/odp_config_internal.h:17:
#define ODP_CONFIG_POOLS 16
I suppose a platform will put here the real maximum number of pools, so there
is n
4__
-
#define ODP_CACHE_LINE_SIZE 64
-#endif
I have attached config.log that should contain all related information.
Kind Regards,
Sorin
From: Bill Fischofer [mailto:bill.fischo...@linaro.org]
Sent: Friday, June 10, 2016 5:00 PM
To: Sorin Vultureanu
Cc: lng-odp@lists.linaro.org; Bogdan Pricope
Subj
Hi,
I have a build error when building ODP 1.10 on i686 arch:
CC odp_buffer.lo
In file included from ./include/odp_pool_internal.h:23:0,
from odp_buffer.c:8:
./include/odp_align_internal.h:63:23: error: 'ODP_CACHE_LINE_SIZE' undeclared
here (not in a function)
ODP_ALIGN
Hi,
The BSD TCP/IP stack requires try operation for recursive write locking.
The OFP project has some malfunctioning implementation of recursive rwlocks
with this try operation.
1. Can ODP add try operation for recursive_rwlock and rwlock?
2. How can I implement this try operation over available
Hi,
I think this IP lookup fits better within OpenFastPath Project - OFP.
www.openfastpath.org
OFP has DIR 1688 (lockless, multicore with linear scalability, very low
overhead) and Radix Tree(read/write locks).
TBM algorithm and Front-End /Back-End architecture are interesting for OFP
project