Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>It's best not to use <code>gettimeofday()</code> or <code>currentTimeMillis()</code> on a mobile device. These return "wall clock" time, which can jump forward or backward suddenly if the network updates the time.</p> <p>Use the monotonic clock instead for performance measurements -- <code>System.nanoTime</code>() or <code>clock_gettime()</code> with <code>CLOCK_MONOTONIC</code>. Note this returns a <code>struct timespec</code> rather than a <code>struct timeval</code>; primary difference is that the clock resolution is nanoseconds rather than microseconds.</p> <pre><code>int64_t getTimeNsec() { struct timespec now; clock_gettime(CLOCK_MONOTONIC, &amp;now); return (int64_t) now.tv_sec*1000000000LL + now.tv_nsec; } </code></pre> <p>In addition to wall-clock time you may be interested in per-thread CPU time; see <a href="https://stackoverflow.com/questions/3564161/thread-performance-in-android/3570077#3570077">Thread Performance in Android</a>.</p>
    singulars
    1. This table or related slice is empty.
    1. POHow to obtain computation time in NDK
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USCommunity
    1. USfadden
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. COHow frequently is the wall clock time updated? I obtain very different time values between gettimeofday() and clock_gettime(CLOCK_PROCESS_CPUTIME_ID) methods. The second one seems more stable, but it's result is a larger time then the first one, instead I expected a smaller time!
      singulars
      1. PO
      1. USAlessandro Gaietta
    2. COYou can tell how often the wall clock timer is updated by calling it as fast as you can in a loop and subtracting the results the first time the value changes. It varies from device to device and release to release. The PROCESS_CPUTIME clock only advances when your process does work; if it's asleep, the clock doesn't change. (I'd recommend CLOCK_THREAD_CPUTIME_ID for performance work, since it only advances when the current thread uses the CPU.) The clocks are not tied to each other, and may not start at zero.
      singulars
      1. PO
      1. USfadden
    3. COWhat I mean is that by measuring with clock_gettime(CLOCK_PROCESS_CPUTIME_ID) (or THREAD, same results) on a small set of operations (a few milliseconds) I get a time equal to gettimeofday() or a simple clock(). Instead, in long cycles I get a longer time with the first one, and the same, shorter, time with gettimeofday() and clock(). NOTE: in all cases I measure time between an end point and a start point, the result of clock_gettime() is corrected if nanoseconds difference < 0 (seconds++). clock() is translated dividing by CLOCKS_PER_SEC.
      singulars
      1. PO
      1. USAlessandro Gaietta
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload