Rooba
/
boinc
mirror of https://github.com/BOINC/boinc.git
1
0
Fork 0
Code Issues 1 Packages Projects Releases Wiki Activity

*** empty log message *** 

svn path=/trunk/boinc/; revision=2420
This commit is contained in:
David Anderson 2003-10-07 04:35:56 +00:00
parent bf5c01f624
commit 4947c97be0
4 changed files with 91 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
api/graphics_api.h
View File

@ -35,6 +35,8 @@
#include <OpenGL/gl.h>
#elif defined(HAVE_MESAGL_GL_H)
#include <MesaGL/gl.h>
#else
#error No gl.h
#endif
#ifdef HAVE_GLU_H
#include <glu.h>

22
checkin_notes
View File

@ -6471,6 +6471,17 @@ Eric Oct 1 2003 11:15 am
graphics_api.h
m4/sah_grx.m4
David Oct 2 2003
- Added code to limit frame rate in either or both of two ways:
1) max frames per second
2) upper bound on rendering CPU time as a fraction or real time
NOTE: currently these limits are hardware in the code.
Need to put them in preferences.
api/
graphics_api.C,h
windows_opengl.C
Eric Oct 2 2003
-moved xml_indent() from seti_boinc/db/sqlrow.[cpp,h] to i
@ -6563,3 +6574,14 @@ Karl 2003/10/06
sched/
transitioner.C
David Oct 6 2003
- Don't send two results from the same WU in a single RPC
Note: this is a weak but possibly sufficient step towards not sending
multiple results from the same WU to one user.
- Scheduler: before updating a result, reread it from the database.
What's in shared memory may be way out of date.
E.g. it may no longer be in server state UNSENT.
Hopefully this fixes the missing-input-file bug!!
sched/
handle_request.C

16
sched/feeder.C
View File

@ -349,12 +349,16 @@ int main(int argc, char** argv) {
}
}
// // Call lock_file after fork(), because file locks are not always inherited
// if (lock_file(LOCKFILE)) {
// log_messages.printf(SchedMessages::NORMAL, "Another copy of feeder is already running\n");
// exit(1);
// }
// write_pid_file(PIDFILE);
#if 0
// Call lock_file after fork(), because file locks are not always inherited
//
if (lock_file(LOCKFILE)) {
log_messages.printf(SchedMessages::NORMAL, "Another copy of feeder is already running\n");
exit(1);
}
write_pid_file(PIDFILE);
#endif
log_messages.printf(SchedMessages::NORMAL, "Starting\n");
retval = destroy_shmem(config.shmem_key);

80
sched/handle_request.C
View File

@ -653,6 +653,18 @@ static int update_wu_transition_time(WORKUNIT wu, time_t x) {
return 0;
}
// return true iff a result for same WU is already being sent
//
static bool already_in_reply(WU_RESULT& wu_result, SCHEDULER_REPLY& reply) {
unsigned int i;
for (i=0; i<reply.results.size(); i++) {
if (wu_result.workunit.id == reply.results[i].workunitid) {
return true;
}
}
return false;
}
// Make a pass through the wu/results array, sending work.
// If "infeasible_only" is true, send only results that were
// previously infeasible for some host
@ -664,7 +676,8 @@ static void scan_work_array(
) {
int i, retval;
WORKUNIT wu;
DB_RESULT result, result_copy;
DB_RESULT result;
double wu_seconds_filled;
for (i=0; i<ss.nwu_results && seconds_to_fill>0; i++) {
WU_RESULT& wu_result = ss.wu_results[i];
@ -674,12 +687,18 @@ static void scan_work_array(
if (!wu_result.present) {
continue;
}
if (infeasible_only && wu_result.infeasible_count==0) {
continue;
}
wu = wu_result.workunit;
double wu_seconds_filled = estimate_duration(wu, reply.host);
// don't send if we're already sending a result for same WU
//
if (already_in_reply(wu_result, reply)) {
continue;
}
wu = wu_result.workunit;
if (!wu_is_feasible(wu, reply.host)) {
log_messages.printf(
@ -690,33 +709,44 @@ static void scan_work_array(
continue;
}
// XXX FIXME: result should be re-read from database here and checked
// that server_state is still UNSENT -- if no longer `UNSENT'
// (perhaps because the WU had an unrecoverable error), then skip it.
// Should handle_request or feeder take care of removing it?
result = wu_result.result;
// The following will fail if e.g. there's no app_version
// for the client's platform.
// Treat the same as the WU being infeasible
//
retval = add_wu_to_reply(wu, reply, platform, ss);
if (retval) {
wu_result.infeasible_count++;
continue;
}
result = wu_result.result;
// mark slot as empty AFTER we've copied out of it
//
wu_result.present = false;
log_messages.printf(
SchedMessages::NORMAL,
"[HOST#%d] Sending [RESULT#%d %s] (fills %d seconds)\n",
reply.host.id, result.id, result.name, int(wu_seconds_filled)
);
// reread result from DB, make sure it's still unsent
// TODO: from here to update() should be a transaction
//
retval = result.lookup_id(result.id);
if (retval) continue;
if (result.server_state != RESULT_SERVER_STATE_UNSENT) continue;
// update the result in DB
//
result.server_state = RESULT_SERVER_STATE_IN_PROGRESS;
result.hostid = reply.host.id;
result.sent_time = time(0);
result.report_deadline = result.sent_time + wu.delay_bound;
result.update();
wu_seconds_filled = estimate_duration(wu, reply.host);
log_messages.printf(
SchedMessages::NORMAL,
"[HOST#%d] Sending [RESULT#%d %s] (fills %d seconds)\n",
reply.host.id, result.id, result.name, int(wu_seconds_filled)
);
retval = update_wu_transition_time(wu, result.report_deadline);
if (retval) {
log_messages.printf(
@ -725,19 +755,23 @@ static void scan_work_array(
);
}
// copy the result so we don't overwrite its XML fields
// The following overwrites the result's xml_doc field.
// But that's OK cuz we're done with DB updates
//
result_copy = result;
retval = insert_name_tags(result_copy, wu);
retval = insert_name_tags(result, wu);
if (retval) {
log_messages.printf(SchedMessages::CRITICAL, "send_work: can't insert name tags\n");
log_messages.printf(
SchedMessages::CRITICAL, "send_work: can't insert name tags\n"
);
}
retval = insert_deadline_tag(result_copy);
retval = insert_deadline_tag(result);
if (retval) {
log_messages.printf(SchedMessages::CRITICAL, "send_work: can't insert deadline tag\n");
log_messages.printf(
SchedMessages::CRITICAL,
"send_work: can't insert deadline tag\n"
);
}
reply.insert_result(result_copy);
reply.insert_result(result);
seconds_to_fill -= wu_seconds_filled;