Remove "-L$(libdir) -rapth $(libdr)" from LDFLAGS of boinc libraries in
Makefile.am of api, lib, sched and zlib directories to be able to
cross-compile boinc.
Indeed, libdir is not equal to the path where BOINC will be installed
but to exec_prefix/lib. The full installation path is
$(DESTDIR)/$(libdir).
To cross-compile boinc, exec_prefix will be set to the target path (for
example: /usr) and DESTDIR will be set (during make install) to the
staging or target directory on the host (for example /home/xxx/target).
The issue of adding -L$(libdir) is that it is not allowed by the
compiler, the error "unsafe header/library path used in
cross-compilation: '-L/usr/lib'" will be raised.
As a result, remove "-L$(libdir) -rapth $(libdr)" from LDFLAGS, the
default library search paths are sufficient for "standard" compilation
or can be updated manually by passing the additional search path to
LDFLAGS during the configure call for cross-compilation.
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Lets SCHED_MSG_LOG be used including only -lboinc, not -lsched.
This is a tactical move; SETI@home back-end code
uses SCHED_MSG_LOG and I need to make this work without -lsched,
which pulls in MySQL libraries which I don't want.
Old: if the timer thread gets a <suspend> message while we're in
a critical section, it sets a "suspend_request" flag.
The timer then periodically (10X/sec) checks whether
suspend_request is set and we're no longer in a critical section;
if so it suspends the worker thread.
Problem (pointed out by Oliver): this doesn't work if the worker thread
is almost always in a critical section
(as is the case for GPU apps, which treat GPU kernels as critical sections).
The app never gets suspended.
New:
1) boinc_end_critical_section() checks suspend_request;
if set, it calls suspend_activities()
2) On Unix, if suspend_activities() is called from the worker thread,
it calls sleep() in a loop until the suspension is over.
(Note: pthreads has no suspend/resume).
3) Add a mutex to protect the data structures shared between
the timer and worker threads.
Oliver pointed out that
use strcat instead
- client: don't use get_mac_address() to create host CPIDs
(we have plenty of other info to make them unique)
- storage stuff
svn path=/trunk/boinc/; revision=25269
Note: XML_PARSER::parse_str() doesn't currently work right
for something like <foo><a>xx</a></foo>.
It should return "<a>xx</a>".
TODO.
svn path=/trunk/boinc/; revision=24325
proc_control: controlling processes
procinfo: enumerating and querying processes
run_app_windows: launching apps as other users on Win
svn path=/trunk/boinc/; revision=24120
otherwise non-ASCII characters in client_state.xml
make it invalid XML
- client: fix (I think) to scheduling logic.
a job is preemptable if it's finished its time slice and
Old: has checkpointed in last 10 sec
New: has checkpointed since the end of the time slice
svn path=/trunk/boinc/; revision=23551
for the things that BOINC actually needs
(fixes server compile problems)
- client: fix various compile errors in coproc_detect.cpp
svn path=/trunk/boinc/; revision=23310
Win: enumerate all descendants, and kill them all TerminateProcess().
Unix:
send the main process a SIGTERM.
Check once a sec for existence of descendants.
if none, done
If any still exist after 10 sec, kill all descendants
- wrapper fix bug in Win env var stuff
- scheduler: check for VBox version 3.2+ in app_plan()
svn path=/trunk/boinc/; revision=23085
see http://boinc.berkeley.edu/trac/wiki/CreditNew
Projects will need to update DB and recompile all back-end programs.
Summary:
- new way of computing credit
- "reliable host" mechanism is per app version
- "host punishment" mechanism is per app version
- adjustment of wu.rsc_fpops_est provides the
equivalent of per app version DCF
- max jobs in progress is now per app
- max jobs per RPC is now per app
TODO:
- reliable mechanism:
- populate and use host_app_version.error_rate
- populate host_app_version.turnaround
- host punishment:
- populate host_app_version.max_jobs_per_day
- populate host_app_version.n_jobs_today
- use app.max_jobs_per_day_init
- job limits:
- use app.max_jobs_in_progress, max_gpu_jobs_in_progress
- use app.max_jobs_per_rpc
- adjust wu.rsc_fpops_est
- remove old credit stuff
fpops_cumulative, credit_multiplier
credit computation in scheduler
- AVERAGE class: use the Knuth algorithm (Wikipedia)
svn path=/trunk/boinc/; revision=21021
Triggering the work generator is now done via the DB
instead of flat files.
Since only E@h uses locality scheduling,
I kept the DB changes in a separate file (db/schema_locality.sql).
There's a new field in the workunit table,
and that's a required update (in db_update.php)
- manager: compile fix
svn path=/trunk/boinc/; revision=20807
Fabrice Fontaine