glusterfs.git/libglusterfs/src/event.c, branch v3.8rc0

core: use syscall wrappers instead of direct syscalls - libglusterfs

2015-11-16T14:18:10+00:00

various xlators and other components are invoking system calls
directly instead of using the libglusterfs/syscall.[ch] wrappers.

If not using the system call wrappers there should be a comment
in the source explaining why the wrapper isn't used.

Change-Id: Ieeca2d36adbc884e4cfa0026dba40df70310d40b
BUG: 1267967
Signed-off-by: Kaleb S. KEITHLEY 
Reviewed-on: http://review.gluster.org/12275
Tested-by: NetBSD Build System 
Tested-by: Gluster Build System 
Reviewed-by: Jeff Darcy

event: remove vestigial mutex_unlock from event_dispatch

2015-08-29T16:56:26+00:00

This was missed erroneously by self in http://review.gluster.org/12004

BUG: 1242421
Change-Id: Id6cb156fd584704486e49bc4c4a7bdd44a98679f
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/12044
Reviewed-by: Atin Mukherjee 
Reviewed-by: Raghavendra Talur 
Tested-by: Gluster Build System 
Tested-by: NetBSD Build System

event-epoll: Use pollers[] to check if event_pool_dispatch was called

2015-08-28T17:14:01+00:00

BUG: 1242421
Change-Id: I1a0044653f15d33f89ffe16edc5baba40393dec3
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/12004
Tested-by: NetBSD Build System 
Reviewed-by: Shyamsundar Ranganathan 
Reviewed-by: Raghavendra Talur

event: add dispatched flag to know if event_dispatch was called

2015-08-21T12:03:06+00:00

This is important for glusterfs processes that choose to reconfigure
no. of event-threads (a.k.a epoll worker-threads) before they call
event_dispatch on the event_pool. glusterd needs this today.

Change-Id: Ia8df3c958545324472262c555ed84b71797f002e
BUG: 1242421
Signed-off-by: Krishnan Parthasarathi 
Reviewed-on: http://review.gluster.org/11911
Tested-by: NetBSD Build System 
Reviewed-by: Niels de Vos 
Tested-by: Gluster Build System

event,parse-utils,quota-common-utils,rbthash/libglusterfs : Porting to new logging framework

2015-06-25T06:29:28+00:00

Change-Id: I13ae9a97fe442c06cf50fe77c63718ea0858dc4d
BUG: 1194640
Signed-off-by: Mohamed Ashiq 
Reviewed-on: http://review.gluster.org/10823
Tested-by: NetBSD Build System 
Reviewed-by: Pranith Kumar Karampuri

build: do not #include "config.h" in each file

2015-05-29T12:43:17+00:00

Instead of including config.h in each file, and have the additional
config.h included from the compiler commandline (-include option).

When a .c file tests for a certain #define, and config.h was not
included, incorrect assumtions were made. With this change, it can not
happen again.

BUG: 1222319
Change-Id: I4f9097b8740b81ecfe8b218d52ca50361f74cb64
Signed-off-by: Niels de Vos 
Reviewed-on: http://review.gluster.org/10808
Tested-by: Gluster Build System 
Tested-by: NetBSD Build System
Reviewed-by: Kaleb KEITHLEY 
Reviewed-by: Pranith Kumar Karampuri

libglusterfs: Replace pipe2 with pipe.

2015-03-11T02:20:52+00:00

pipe2() doesn't works on Linux kernel version < 2.6.27 and 
glibc < version 2.9. Hence replacing it with pipe(),
so that the build will not fail on Centos5.

Change-Id: If17aed0d51466cd7528cf8dde0edfa28b68139e5
BUG: 1200255
Signed-off-by: Poornima G 
Reviewed-on: http://review.gluster.org/9844
Reviewed-by: Raghavendra Talur 
Tested-by: Gluster Build System 
Reviewed-by: Krishnan Parthasarathi

glfsheal: Avoid infinite loop on exit

2015-03-09T17:42:02+00:00

Make sure we do not get stuck looping forever in event_dispatch_destroy()
by limiting the retries when waiting for other threads, and by giving up
when writing to other thread fails.

This fixes regression tests hanging forever on NetBSD.

BUG: 1129939
Change-Id: I4459cfb1ab7294e8c15a21b592e0154c22abae07
Signed-off-by: Emmanuel Dreyfus 
Reviewed-on: http://review.gluster.org/9825
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos 
Reviewed-by: Shyamsundar Ranganathan 
Reviewed-by: Jeff Darcy 
Reviewed-by: Vijay Bellur

event_pool: Add the code to destroy the poller threads and event pool gracefully.

2015-03-03T14:45:21+00:00

Change-Id: I49b6ceebb45773620c318fb5d20b81623db75ab6
BUG: 1093594
Signed-off-by: Poornima G 
Reviewed-on: http://review.gluster.org/9691
Reviewed-by: Krishnan Parthasarathi 
Tested-by: Gluster Build System 
Reviewed-by: Shyamsundar Ranganathan

epoll: Adding the ability to configure epoll threads

2015-02-07T21:23:03+00:00

Add the ability to configure the number of event threads
for various gluster services.

Currently with the multi thread epoll patch, it is possible
to have more than one thread waiting on socket activity and
processing the same. This thread count is currently static,
which this commit makes dynamic.

The current services which use IO path, i.e brick processes,
any client process (nfs, FUSE, gfapi, heal,
rebalance, etc.a), gain 2 set parameters to control the number
of threads that are processing events. These settings are,
  - client.event-threads 
  - server.event-threads 

The client setting affects the client graph consumers, and the
server setting affects the brick processes. These are processed
and inited/reconfigured using the client/server protocol xlators.

Other services (say glusterd) would need to extend similar
configuration settings to take advantage of multi threaded event
processing.

At present glusterd is not enabled with this commit, as it does not
stand to gain from this multi-threading (as I understand it).

Change-Id: Id8422fc57a9f95a135158eb6477ccf9d3c9ea4d9
BUG: 1104462
Signed-off-by: Shyam 
Reviewed-on: http://review.gluster.org/9488
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur