an app version will run on a particular host.
- scheduler: fix memory leak: BEST_APP_VERSIONs weren't being freed
svn path=/trunk/boinc/; revision=21148
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
change various std::string to char[n], to improve performance.
NOTE: std::string should ONLY be used in cases where there is
no a priori limit on string length.
If there's a limit (e.g. because the string originates in
a fixed-size database field) always use char[n].
svn path=/trunk/boinc/; revision=20877
Removed my changes of 19 Jan 2010, which didn't work.
Added new mechanism: keep track of whether a job J has ever run in EDF.
If so, and if another job of the same project and resource type as J
is marked as deadline miss, then mark J as deadline miss,
so that it won't get preempted.
- web: change "result" to "task" in server status page
- admin web: show server stable SVN revision, not trunk
svn path=/trunk/boinc/; revision=20805
- scheduler: when calculate scheduler runtime,
don't include the part reading request msg from client.
That can be misleadingly long
svn path=/trunk/boinc/; revision=20781
or an auth_ops() function (not necessarily both).
New projects are given an .htaccess file with no .htpasswd,
so they must either create a .htpasswd file
or define auth_ops() in order to use admin web pages.
svn path=/trunk/boinc/; revision=20671
to restart app for 10 minutes.
Hopefully what will happen is:
- another instance of app is running in slot dir
(shouldn't happen, but sometimes does)
- that app will eventually finish, and will write
a checkpoint file saying so.
It will call boinc_finish(0), but the client won't notice
that it has exited.
- the next time the client starts the app,
it will acquire lock, see that it's done,
and call boinc_finish(0).
This time the client will notice,
and the job will be reported as correct.
The downside to all this is that the client won't know
that the CPU is in use, and will schedule NCPUS jobs.
svn path=/trunk/boinc/; revision=20128
- a project overestimates job FLOP counts
- the client starts jobs in EDF mode
- as job progresses and fraction done increases,
its completion time estimate decreases until
it's no longer a deadline miss.
- job gets preempted by other job from that project;
you end up with lots of partly completed jobs.
Solution (I hope): if an app version has running jobs,
compute a "temp DCF" for the app version,
which is the min of dynamic/static estimates for its jobs.
Apply this scaling factor to completion time estimates
for unstarted jobs in RR simulation
- client: the estimation of remaining time of running jobs was wrong
(how did this bug survive so long?)
svn path=/trunk/boinc/; revision=20077
the BOINC stylesheet (main.css) was included.
This is no good if we're exporting the HTML.
Add an option to generate generic HTML.
- web: add options to the forum RSS feed:
1) threads_only: just show threads (i.e. 1st post in each thread)
2) truncate: truncate posts to 256 chars.
If this is not set, convert post from BBcode to generic HTML,
and put this (XML-encoded) in item.description
This is preparation for using the forum code for project news,
and for displaying forum RSS feeds in the manager.
svn path=/trunk/boinc/; revision=19915
can increase or decrease at N times real time.
My checkin of 7 Dec reflects this by changing
the STD limits to +- N*MAX_STD.
This looks like a bug to users.
Instead, scale that rate of STD change by 1/N,
and keep the old limits of +- MAX_STD
svn path=/trunk/boinc/; revision=19851
only if the offset is positive.
- client: some cmdline args set members of config.
However, config was being cleared after cmdline args were parsed,
so these args had no effect.
Instead, clear config before parsing cmdline
svn path=/trunk/boinc/; revision=19776
Make them both peak FLOPS,
according to the formula supplied by the manufacturer.
The impact on the client is minor:
- the startup message describing the GPU
- the weight of the resource type in computing long-term debt
On the server, I changed the example app_plan() function
to assume that app FLOPS is 20% of peak FLOPS
(that's about what it is for SETI@home)
svn path=/trunk/boinc/; revision=19310
for certain periods (e.g. when Remote Desktop is used on Win).
- add is_usable() member function to COPROC.
Currently this just calls the respective (CUDA or CAL)
initialization function.
We need to check whether this works and/or causes problems.
- in enforce_schedule(), check whether usability has changed
for each GPU type.
If we've gone from usable to unusable,
flag all jobs for that GPU as coproc_missing
(so they won't get run, and will quit if they're running).
If we've gone from unusable to usable, clear the flag.
This should deal with all cases except where
the client is started up with GPUs unusable.
- scheduler: more query optimizations for locality scheduling
(from Oliver Bock)
svn path=/trunk/boinc/; revision=19301
- default: script can be run from command line;
to run from web you need to be logged in as a particular user.
(in addition to .htaccess protection).
- if include $cli_only at start of script:
can only be run from the command line.
- if include $skip_auth_ops at start of script:
can run from web without being logged in
(this is set ONLY in login_action.php)
- admin web: added $cli_only = true to a bunch of scripts
svn path=/trunk/boinc/; revision=19230
use job name as tiebreaker if received times are the same.
This makes the order deterministic.
- web: show elapsed time and CPU time,
and show app version info, in result display
svn path=/trunk/boinc/; revision=19008
FILE_INFO::url
HOST_INFO::p_vendor, p_model, p_features, os_name, os_version
The following fields are already entity-encoded:
PROJECT::user_name, team_name
APP_INIT_DATA::user_name, team_name
PROXY_INFO:: user names and passwds
These are (as far as I know) the only fields that can
contain special chars.
This is now documented in:
http://boinc.berkeley.edu/trac/wiki/XmlNotes
- client: XML_PARSER::parse_str() now does xml_unescape(),
same as ::parse_str().
svn path=/trunk/boinc/; revision=18915
make sure they're the same (i.e., enforce file immutability).
TODO: do this for input files too
- client: check return values from CUDA device detection functions
svn path=/trunk/boinc/; revision=18903
don't modify user preferences or CPID.
- client: fix bug that shows ATI version incorrectly
- database: host.posts has been repurposed as a salt (or seqno)
for a new type of weak authenticator that won't depend on password
- web code:
modify forum_preferences.posts instead of host.posts.
(actually, the former isn't used either, we just do a select count(*);
should fix this at some point).
svn path=/trunk/boinc/; revision=18865
don't skip the reference site check.
Instead, defer the reference site check until 30 secs after startup.
Also, if during this interval (when the check has been requested
but not started) return net status as LOOKUP_PENDING
svn path=/trunk/boinc/; revision=18760
to the executable (../../projects/x/y) into argv[0],
not just the executable filename.
Apparently the new NVIDIA drivers have a bug that cause
CUDA apps to crash unless this is done.
- Scheduler: in no-host-ID case, don't mark results as "detached"
if request contains any in-progress results
svn path=/trunk/boinc/; revision=18754
- client: (refinement to previous checkin)
don't skip file size check if file has multiple upload URLs.
We might have uploaded different amounts on different servers.
svn path=/trunk/boinc/; revision=18606
make_project now generates PROJECT/bin/boinc_path_config.py
to ensure that the interpreter will find the modules.
From Gabor Gombas.
svn path=/trunk/boinc/; revision=18374
limits the # of completed results handled per scheduler RPC.
This may be needed to avoid crashes due to memory allocation
failure (each reported result uses about 128KB memory).
- web: In showing result lists,
include "Validate error" results in the "Invalid" category.
(Previously they didn't appear in any category)
svn path=/trunk/boinc/; revision=18104
Rom Walton