efl/legacy/ecore/src/lib/ecore/ecore_poll.c

354 lines
11 KiB
C
Raw Normal View History

e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
#include "ecore_private.h"
#include "Ecore.h"
static Ecore_Timer *timer = NULL;
static int min_interval = -1;
static int interval_incr = 0;
static int at_tick = 0;
static int just_added_poller = 0;
static int poller_delete_count = 0;
static int poller_walking = 0;
static double poll_interval = 0.125;
static double poll_cur_interval = 0.0;
static double last_tick = 0.0;
static Ecore_Poller *pollers[16] =
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
{
NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL
};
static unsigned short poller_counters[16] =
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
{
0,0,0,0,0,0,0,0,
0,0,0,0,0,0,0,0
};
static void _ecore_poller_next_tick_eval(void);
static int _ecore_poller_cb_timer(void *data);
static void
_ecore_poller_next_tick_eval(void)
{
int i;
double interval;
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
min_interval = -1;
for (i = 0; i < 15; i++)
{
if (pollers[i])
{
min_interval = i;
break;
}
}
if (min_interval < 0)
{
/* no pollers */
if (timer)
{
ecore_timer_del(timer);
timer = NULL;
}
return;
}
interval_incr = (1 << min_interval);
interval = interval_incr * poll_interval;
/* we are at the tick callback - so no need to do inter-tick adjustments
* so we can fasttrack this as t -= last_tick in theory is 0.0 (though
* in practice it will be a very very very small value. also the tick
* callback will adjust the timer interval at the end anyway */
if (at_tick)
{
if (!timer)
timer = ecore_timer_add(interval, _ecore_poller_cb_timer, NULL);
}
else
{
double t;
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
if (!timer)
timer = ecore_timer_add(interval, _ecore_poller_cb_timer, NULL);
else
{
t = ecore_time_get();
if (interval != poll_cur_interval)
{
t -= last_tick; /* time since we last ticked */
/* delete the timer and reset it to tick off in the new
* time interval. at the tick this will be adjusted */
ecore_timer_del(timer);
timer = ecore_timer_add(interval - t,
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
_ecore_poller_cb_timer, NULL);
}
}
}
poll_cur_interval = interval;
}
static int
2008-10-05 11:33:35 -07:00
_ecore_poller_cb_timer(void *data __UNUSED__)
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
{
int i;
Ecore_List2 *l;
Ecore_Poller *poller;
int changes = 0;
at_tick++;
last_tick = ecore_time_get();
/* we have 16 counters - each incriments every time the poller counter
* "ticks". it incriments by the minimum interval (which can be 1, 2, 4,
* 7, 16 etc. up to 32768) */
for (i = 0; i < 15; i++)
{
poller_counters[i] += interval_incr;
/* wrap back to 0 if we exceed out loop count for the counter */
if (poller_counters[i] >= (1 << i)) poller_counters[i] = 0;
}
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
just_added_poller = 0;
/* walk the pollers now */
poller_walking++;
for (i = 0; i < 15; i++)
{
/* if the counter is @ 0 - this means that counter "went off" this
* tick interval, so run all pollers hooked to that counter */
if (poller_counters[i] == 0)
{
for (l = (Ecore_List2 *)pollers[i]; l; l = l->next)
{
poller = (Ecore_Poller *)l;
if (!poller->delete_me)
{
if (!poller->func(poller->data))
{
if (!poller->delete_me)
{
poller->delete_me = 1;
poller_delete_count++;
}
}
}
}
}
}
poller_walking--;
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
/* handle deletes afterwards */
if (poller_delete_count > 0)
{
/* FIXME: walk all pollers and remove deleted ones */
for (i = 0; i < 15; i++)
{
for (l = (Ecore_List2 *)pollers[i]; l;)
{
poller = (Ecore_Poller *)l;
l = l->next;
if (poller->delete_me)
{
pollers[poller->ibit] = _ecore_list2_remove(pollers[poller->ibit], poller);
free(poller);
poller_delete_count--;
changes++;
if (poller_delete_count <= 0) break;
}
}
if (poller_delete_count <= 0) break;
}
}
/* if we deleted or added any pollers, then we need to re-evaluate our
* minimum poll interval */
if ((changes > 0) || (just_added_poller > 0))
_ecore_poller_next_tick_eval();
just_added_poller = 0;
poller_delete_count = 0;
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
at_tick--;
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
/* if the timer was deleted then there is no point returning 1 - ambiguous
* if we do as it im plies "keep running me" but we have been deleted
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
* anyway */
if (!timer) return 0;
/* adjust interval */
ecore_timer_interval_set(timer, poll_cur_interval);
return 1;
}
/**
* @defgroup Ecore_Poll_Group Ecore Poll Functions
*
* These functions are for the need to poll information, but provide a shared
* abstracted API to pool such polling to minimise wakeup and ensure all the
* polling happens in as few spots as possible areound a core poll interval.
* For now only 1 core poller type is supprted: ECORE_POLLER_CORE
*/
/**
* Sets the time between ticks (in seconds) for the given ticker clock.
* @param type The ticker type to adjust
* @param poll_time The time (in seconds) between ticks of the clock
* @ingroup Ecore_Poller_Group
*
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
* This will adjust the time between ticks of the given ticker type defined
* by @p type to the time period defined by @p poll_time.
*/
EAPI void
2008-10-05 11:33:35 -07:00
ecore_poller_poll_interval_set(Ecore_Poller_Type type __UNUSED__, double poll_time)
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
{
poll_interval = poll_time;
_ecore_poller_next_tick_eval();
}
/**
* Gets the time between ticks (in seconds) for the fiven ticker clock.
* @param type The ticker type to query
* @return The time in seconds between ticks of the ticker clock
* @ingroup Ecore_Poller_Group
*
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
* This will get the time between ticks of the specifider ticker clock.
*/
EAPI double
2008-10-05 11:33:35 -07:00
ecore_poller_poll_interval_get(Ecore_Poller_Type type __UNUSED__)
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
{
return poll_interval;
}
/**
* Creates a poller to call the given function at a particular tick interval.
* @param type The ticker type to attach the poller to
* @param interval The poll interval
* @param func The given function. If @p func returns 1, the poller is
* rescheduled for the next tick interval.
* @param data Data to pass to @p func when it is called.
* @return A poller object on success. @c NULL on failure.
* @ingroup Ecore_Poller_Group
*
* This function adds a poller callback that is to be called regularly
* along with all other poller callbacks so the pollers are synchronized with
* all other pollers running off the same poller type and at the same tick
* interval. This should be used for polling things when polling is desired
* or required, and you do not have specific requirements on the exact times
* to poll and want to avoid extra process wakeups for polling. This will
* save power as the CPU has more of a chance to go into a low power state
* the longer it is asleep for, so this should be used if you are at all
* power conscious.
*
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
* The @p type parameter defines the poller tick type (there is a virtual
* clock ticking all the time - though ecore avoids making it tick when
* there will not be any work to do at that tick point). There is only one
* ticker at the moment - that is ECORE_POLLER_CORE. This is here for future
* expansion if multiple clocks with different frequencies are really required.
* The default time between ticks for the ECORE_POLLER_CORE ticker is 0.125
* seconds.
*
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
* The @p interval is the number of ticker ticks that will pass by in between
* invocations of the @p func callback. This must be between 1 and 32768
* inclusive, and must be a power of 2 (i.e. 1, 2, 4, 8, 16, ... 16384, 32768).
* If it is 1, then the function will be called every tick. if it is 2, then it
* will be called every 2nd tick, if it is 8, then every 8th tick etc. Exactly
* which tick is undefined, as only the interval between calls can be defined.
* Ecore will endeavour to keep pollers synchronised and to call as many in
* 1 wakeup event as possible.
*
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
* This function adds a poller and returns its handle on success and NULL on
* failure. The function @p func will be called at tick intervals described
* above. The function will be passed the @p data pointer as its parameter.
*
* When the poller @p func is called, it must return a value of either
* 1 (or ECORE_CALLBACK_RENEW) or 0 (or ECORE_CALLBACK_CANCEL). If it
* returns 1, it will be called again at the next tick, or if it returns
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
* 0 it will be deleted automatically making any references/handles for it
* invalid.
*/
EAPI Ecore_Poller *
2008-10-05 11:33:35 -07:00
ecore_poller_add(Ecore_Poller_Type type __UNUSED__, int interval, int (*func) (void *data), const void *data)
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
{
Ecore_Poller *poller;
int ibit;
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
if (!func) return NULL;
if (interval < 1) interval = 1;
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
poller = calloc(1, sizeof(Ecore_Poller));
if (!poller) return NULL;
ECORE_MAGIC_SET(poller, ECORE_MAGIC_POLLER);
/* interval MUST be a power of 2, so enforce it */
if (interval < 1) interval = 1;
ibit = -1;
while (interval != 0)
{
ibit++;
interval >>= 1;
}
/* only allow up to 32768 - i.e. ibit == 15, so limit it */
if (ibit > 15) ibit = 15;
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
poller->ibit = ibit;
poller->func = func;
poller->data = (void *)data;
pollers[poller->ibit] = _ecore_list2_prepend(pollers[poller->ibit], poller);
if (poller_walking)
just_added_poller++;
else
_ecore_poller_next_tick_eval();
return poller;
}
/**
* Delete the specified poller from the timer list.
* @param poller The poller to delete.
* @return The data pointer set for the timer when @ref ecore_poller_add was
* called. @c NULL is returned if the function is unsuccessful.
* @ingroup Ecore_Poller_Group
*
* Note: @p poller must be a valid handle. If the poller function has already
* returned 0, the handle is no longer valid (and does not need to be delete).
*/
EAPI void *
ecore_poller_del(Ecore_Poller *poller)
{
void *data;
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
if (!ECORE_MAGIC_CHECK(poller, ECORE_MAGIC_POLLER))
{
ECORE_MAGIC_FAIL(poller, ECORE_MAGIC_POLLER,
"ecore_poller_del");
return NULL;
}
/* we are walking the poller list - a bad idea to remove from it while
* walking it, so just flag it as delete_me and come back to it after
* the loop has finished */
if (poller_walking > 0)
{
poller_delete_count++;
poller->delete_me = 1;
return poller->data;
}
/* not in loop so safe - delete immediately */
data = poller->data;
pollers[poller->ibit] = _ecore_list2_remove(pollers[poller->ibit], poller);
free(poller);
_ecore_poller_next_tick_eval();
return data;
}
void
_ecore_poller_shutdown(void)
{
int i;
Ecore_List2 *l;
Ecore_Poller *poller;
e: 1. configure/build changes to allow cross-compiling painlessly 2. pager module namespace changes - this was still dirty afdter the namespace cleanup, so clean it up 3. add a powersave subsystem - doesnt have an "automatic" way to turn on and off right now, this i think is best provided by modules (that do things like monitor acpi status's (eg close lid of laptop), AC power status etc. etc. this allows e to nicely defer "power" expensive actions to avoid disk spinups etc. 4. move to use the new ecore poller system - discussed long ago as part of power management/saving issues. now it exists 5. add a canvas idle flush call that helsp cope with the new shm greedy software x11 engine stuff 6. use the new powersave subsystem where appropriate 7. fix non-zeroed/initted memory access in e_fm_main 8. fix mem leak for e menus 9. remove ipc handlers for changed/removed config values 10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps config 11. fix up timer/poll happienss of cursor idle stuff 12. remove avoid damage from popups for now - causing problems 13. change battery and temp readouts to b e shorter so they fit 14. pager can emit signals on focus change for mini-windows now 15. temperature module now uses a slave process and uses stdin/out to talk to it and get output - this makes e smoother as in my expereicne i found getting the temp on my laptop actually took like 200ms so e "hang" for 200ms while reading the acpi files - so now the subprocess does it and just writesa back to e when it gets it. ecore: 1. add ecore_pollers. see the documentation on them in doxygen comments :) 2. fix timers to only go off when they have to - bug there that made e's select time out a LOT more than it needed to. defensive coding hid the problem. now fixed. e should be much more power friendly now. 3. formatting/niceness in ecore_exe stuff 4. some comments on comments with SIGIO ideas vs. select 5. add call to be able to add an idle enterer at the start of the list of them, not just the end (as has been the default) 6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in all canvases - and to do it right 7. if argb destination - set the shape EVENT shape (to mask out events in transparent regions much like shape does withotu translucency) 8. in ecore_x add support for the event shape evas: 1. fix cache to work properly and not just always fill up (as it seemed to like to think cahce useage dropped below 0 when it didnt and thus just over-fill) 2. software x11 engine now ONLY uses shm segments - no ximages over the socket. this ximage hack was there to avoid the 2 round trips involved in setting up an shm image - now i mitigated that wih an shm image cache pool. it keeps shm images around and repurposes them for new update regions if appropriate. this means many fewer shm creates (about 1/100th the number) and since we recycle the memory less 0 memory page filling by the kernel - in the end, i recorded about a 10-20% speedup over the old software x11 engine. simple tests i have seen up to 120% speedups. idle flush now does something - it frees all the cached shm segments. it has a hard-coded limit of 4mb worth of shm segments (or 32 segments - whichever comes first) to keep around. once can never complain much about speedups methinks :). also evas will defer sync until the NEXT frame is written - this means evas can calculate the next frame of data while x dma's/copies the images to the screen at the same time (if you hve a dual core or multi-cpu machnike or your xserver is able to use DMA to copy image data to the screen/video ram then this should see a decent speedup). SVN revision: 33448
2008-01-10 23:33:57 -08:00
for (i = 0; i < 15; i++)
{
for (l = (Ecore_List2 *)pollers[i]; l;)
{
poller = (Ecore_Poller *)l;
l = l->next;
pollers[poller->ibit] = _ecore_list2_remove(pollers[poller->ibit], poller);
free(poller);
}
}
}