Re: [sheepdog] any plan to use dpdk for network-io?

At Fri, 9 Jan 2015 13:46:04 +0800, =?utf-8?B?6ZmI5p2O57Ku?= wrote: > > [1 ] > [1.1 ] > any plan to use dpdk for network-io?‍ > http://www.dpdk.org/‍ > may improve io performance ? Currently, NTT doesn't have a plan related to DPDK. If you are interested in, feel free to work on it :) BTW, are

Re: [sheepdog] consistency problem when replicating to multi-replica

At Thu, 08 Jan 2015 16:25:08 +0800, guping wrote: > > I find a interesting subject at > http://copilotco.com/mail-archives/drbd.2009/msg00750.html , > > The 'concurrent local write' in DRBD is very similar to the problem in > my previous mail. > > Hope to be useful and we can come out a solution

[sheepdog] [PATCH 6/6] lib: add tracepoints related to event loop

Cc: Gu Ping Signed-off-by: Hitoshi Mitake --- lib/Makefile.am | 4 +-- lib/event.c | 9 +++ lib/tracepoint/event_tp.c | 13 ++ lib/tracepoint/event_tp.h | 63 +++ 4 files changed, 87 insertions(+), 2 deletions(-)

[sheepdog] [PATCH 5/6] lib: add tracepoints related to sockfd management

Cc: Gu Ping Signed-off-by: Hitoshi Mitake --- lib/Makefile.am | 4 +-- lib/sockfd_cache.c | 16 + lib/tracepoint/sockfd_cache_tp.c | 13 lib/tracepoint/sockfd_cache_tp.h | 70 4 files changed, 101 insertion

[sheepdog] [PATCH 0/6] employ LTTng for performance analysis and debugging

LTTng is the de facto standard of tracing framework. This patchset employes it for performance analysis and debugging of sheepdog. It would be useful for analyzing bottlenecks. Hitoshi Mitake (6): configure: enable LTTng with --enable-lttng-ust option sheep: use _exit(2) instead of exit(3) in

[sheepdog] [PATCH 1/6] configure: enable LTTng with --enable-lttng-ust option

LTTng is a tool for low overhead tracing of events. It is already employed by QEMU and friendly with tracing mechanism of linux kernel, so it is very useful for performance analysis and debugging of sheepdog. Below is a brief example of lttng tracing: $ lttng create sheepdog Session sheepdog crea

[sheepdog] [PATCH 4/6] lib: add tracepoints related to work queue

Cc: Gu Ping Signed-off-by: Hitoshi Mitake --- lib/Makefile.am | 8 + lib/tracepoint/work_tp.c | 13 +++ lib/tracepoint/work_tp.h | 93 lib/work.c | 10 ++ 4 files changed, 124 insertions(+) create mode 100644 l

[sheepdog] [PATCH 2/6] sheep: use _exit(2) instead of exit(3) in lock_and_daemon() for gcc's destructors

Because exit(3) calls registered destructors of gcc, it calling it in lock_and_daemon() conflicts with LTTng. This patch let the function use _exit(2) for avoiding this problem. Cc: Gu Ping Signed-off-by: Hitoshi Mitake --- sheep/sheep.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

[sheepdog] [PATCH 3/6] lib: add a header for a case of disabling LTTng

This is a preparation patch. The header file is required for disabling LTTng tracing. Cc: Gu Ping Signed-off-by: Hitoshi Mitake --- include/Makefile.am | 2 +- include/lttng_disable.h | 17 + 2 files changed, 18 insertions(+), 1 deletion(-) create mode 100644 include/lttng

[sheepdog] any plan to use dpdk for network-io?

any plan to use dpdk for network-io?‍ http://www.dpdk.org/‍ may improve io performance ?-- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

[sheepdog] Build failed in Jenkins: sheepdog-build #597

See -- [...truncated 51 lines...] checking for grep that handles long lines and -e... /bin/grep checking for egrep... /bin/grep -E checking for ANSI C header files... yes checking for sys/typ

Re: [sheepdog] consistency problem when replicating to multi-replica

I find a interesting subject at http://copilotco.com/mail-archives/drbd.2009/msg00750.html , The 'concurrent local write' in DRBD is very similar to the problem in my previous mail. Hope to be useful and we can come out a solution. 在 15/1/5 上午9:26, Hitoshi Mitake 写道: > At Mon, 05 Jan 2015 09:12: