| /linux/drivers/gpu/drm/i915/ |
| H A D | Kconfig.profile | 16 int "Timeout for unsignaled foreign fences (ms, jiffy granularity)"
56 int "Preempt timeout (ms, jiffy granularity)"
74 int "Preempt timeout for compute engines (ms, jiffy granularity)"
124 int "Scheduling quantum for userspace batches (ms, jiffy granularity)"
|
| /linux/init/ |
| H A D | calibrate.c | 31 * loops per jiffy directly, instead of guessing it using delay().
58 * will not do. As we don't really know whether jiffy switch in calibrate_delay_direct()
63 * 1. pre_start <- When we are sure that jiffy switch hasn't happened in calibrate_delay_direct()
64 * 2. check jiffy switch in calibrate_delay_direct()
65 * 3. start <- timer value before or after jiffy switch in calibrate_delay_direct()
66 * 4. post_start <- When we are sure that jiffy switch has happened in calibrate_delay_direct()
|
| /linux/Documentation/timers/ |
| H A D | timekeeping.rst | 115 implementation is not provided, the system jiffy counter will be used as
135 jiffy frequency for the architecture. This will affect scheduling accuracy
165 loop to delay a certain number of jiffy fractions using a "lpj" (loops per
166 jiffy) value, calibrated on boot.
|
| H A D | highres.rst | 235 handler has not to deal with an eventually stale jiffy value.
|
| /linux/Documentation/scheduler/ |
| H A D | sched-nice-design.rst | 19 rule so that nice +19 level would be _exactly_ 1 jiffy. To better
44 HZ=1000 it caused 1 jiffy to be 1 msec, which meant 0.1% CPU usage which
|
| /linux/net/netfilter/ |
| H A D | xt_limit.c | 37 `prev' tracks the last packet hit: you gain one credit per jiffy.
45 credits per jiffy). We want to allow a rate as low as 1 per day
|
| /linux/arch/arm/mach-shmobile/ |
| H A D | timer.c | 31 * Calculate a worst-case loops-per-jiffy value in shmobile_init_delay()
|
| /linux/Documentation/translations/zh_CN/scheduler/ |
| H A D | sched-nice-design.rst | 25 的级别 _正好_ 是1 jiffy。为了让大家更好地理解它，时间片的图会是这样的（质量
|
| /linux/kernel/trace/ |
| H A D | trace_clock.c | 53 * This is based on cpu_clock(), which will allow at most ~1 jiffy of
|
| /linux/drivers/accessibility/speakup/ |
| H A D | synth.c | 67 jiffy_delta = spk_get_var(JIFFY); in _spk_do_catch_up()
433 { JIFFY, .u.n = {NULL, 50, 20, 200, 0, 0, NULL } },
|
| H A D | spk_types.h | 44 DELAY, TRIGGER, JIFFY, FULL, /* all timers must be together */ enumerator
|
| H A D | speakup_apollo.c | 150 jiffy_delta = spk_get_var(JIFFY); in do_catch_up()
|
| H A D | speakup_decext.c | 164 jiffy_delta = spk_get_var(JIFFY); in do_catch_up()
|
| H A D | speakup_keypc.c | 186 jiffy_delta = spk_get_var(JIFFY); in do_catch_up()
|
| H A D | speakup_acntpc.c | 184 jiffy_delta = spk_get_var(JIFFY); in do_catch_up()
|
| H A D | varhandlers.c | 24 { "jiffy_delta", JIFFY, VAR_TIME, NULL, NULL },
|
| H A D | speakup_dectlk.c | 225 jiffy_delta = spk_get_var(JIFFY); in do_catch_up()
|
| /linux/arch/arm/plat-orion/ |
| H A D | time.c | 55 * Number of timer ticks per jiffy.
|
| /linux/Documentation/ABI/testing/ |
| H A D | sysfs-class-devfreq | 101 less than 1 jiffy, it is considered to be 0, which means
|
| /linux/Documentation/filesystems/ |
| H A D | orangefs.rst | 546 The timeout values d_time and getattr_time are jiffy based, and the
547 code is designed to avoid the jiffy-wrap problem::
|
| /linux/drivers/gpu/drm/v3d/ |
| H A D | v3d_bo.c | 301 /* Asked to wait beyond the jiffy/scheduler precision? */ in v3d_wait_bo_ioctl()
|
| /linux/arch/x86/kernel/cpu/mce/ |
| H A D | dev-mcelog.c | 310 * so do it a jiffy or two later everywhere. in mce_chrdev_write()
|
| /linux/arch/m68k/q40/ |
| H A D | q40ints.c | 109 int ql_ticks; /* 200Hz ticks since last jiffy */
|
| /linux/drivers/watchdog/ |
| H A D | shwdt.c | 45 * resolution of a jiffy is enough to trigger the overflow. For things like
|
| /linux/Documentation/filesystems/caching/ |
| H A D | cachefiles.rst | 458 Each step is 1 jiffy in size. The JIFS column indicates the particular
459 jiffy range covered, and the SECS field the equivalent number of seconds.
|