Home
last modified time | relevance | path

Searched refs:nice (Results 1 – 25 of 138) sorted by relevance

123456

/linux-6.3-rc2/Documentation/translations/zh_CN/scheduler/
A Dsched-nice-design.rst11 调度器nice值设计
16 Linux的nice级别总是非常脆弱，人们持续不断地缠着我们，让nice +19的任务占用
51 nice +19级别运行数量颇多的应用程序）。
63 nice级别的行为取决于 _绝对的_ nice级别，而nice应用程序接口本身从根本上
66 int nice(int inc);
71 注意，“inc”是相对当前nice级别而言的，类似bash的“nice”命令等工具是这个
74 在旧的调度器中，举例来说，如果你以nice +1启动一个任务，并以nice +2启动
78 第三个关于Linux nice级别支持的抱怨是，负数nice级别“不够有力”，以很多人
87 更一致的nice +19支持：在新的调度器中，nice +19的任务得到一个HZ无关的
90 为了解决第二个抱怨（nice各级别不一致），新调度器令调用nice(1)对各任务的
[all …]
A Dsched-design-CFS.rst46 CFS剩下的其它设计，一般脱离了这个简单的概念，附加的设计包括nice级别，多处理，以及各种
92 CFS调度器处理nice级别和SCHED_BATCH的能力比之前的原始调度器更强：两种类型的工作负载
110 - SCHED_IDLE：它比nice 19更弱，不过它不是真正的idle定时器调度器，因为要避免给机器
A Dindex.rst29 sched-nice-design
A Dsched-energy.rst248 b. 所有任务，不管它们的nice值是多大，都应该被提供了足够多的CPU算力。
254 又回到了在CPU密集的条件下基于负载的算法做负载均衡。这更好地尊重了任务的nice值。
/linux-6.3-rc2/Documentation/scheduler/
A Dsched-nice-design.rst6 nice-levels implementation in the new Linux scheduler.
9 pestered us to make nice +19 tasks use up much less CPU time.
19 rule so that nice +19 level would be _exactly_ 1 jiffy. To better
34 -*----------------------------------*-----> [nice level]
49 people were running number crunching apps at nice +19.)
66 nice level as well, while the nice API itself is fundamentally
69 int nice(int inc);
75 bash's "nice" command mirror this relative API.
79 depend on the nice level of the parent shell - if it was at nice -10 the
101 scheduler, running a nice +10 and a nice 11 task has the same CPU
[all …]
A Dindex.rst19 sched-nice-design
/linux-6.3-rc2/kernel/sched/
A Dautogroup.c231 int proc_sched_autogroup_set_nice(struct task_struct *p, int nice) in proc_sched_autogroup_set_nice() argument
238 if (nice < MIN_NICE || nice > MAX_NICE) in proc_sched_autogroup_set_nice()
241 err = security_task_setnice(current, nice); in proc_sched_autogroup_set_nice()
245 if (nice < 0 && !can_nice(current, nice)) in proc_sched_autogroup_set_nice()
255 idx = array_index_nospec(nice + 20, 40); in proc_sched_autogroup_set_nice()
261 ag->nice = nice; in proc_sched_autogroup_set_nice()
277 seq_printf(m, "/autogroup-%ld nice %d\n", ag->id, ag->nice); in proc_sched_autogroup_show_task()
A Dautogroup.h17 int nice; member
/linux-6.3-rc2/include/linux/sched/
A Dprio.h26 #define NICE_TO_PRIO(nice) ((nice) + DEFAULT_PRIO) argument
32 static inline long nice_to_rlimit(long nice) in nice_to_rlimit() argument
34 return (MAX_NICE - nice + 1); in nice_to_rlimit()
A Dautogroup.h18 extern int proc_sched_autogroup_set_nice(struct task_struct *p, int nice);
/linux-6.3-rc2/fs/proc/
A Dstat.c111 u64 user, nice, system, idle, iowait, irq, softirq, steal; in show_stat() local
118 user = nice = system = idle = iowait = in show_stat()
132 nice += cpustat[CPUTIME_NICE]; in show_stat()
154 seq_put_decimal_ull(p, " ", nsec_to_clock_t(nice)); in show_stat()
173 nice = cpustat[CPUTIME_NICE]; in show_stat()
184 seq_put_decimal_ull(p, " ", nsec_to_clock_t(nice)); in show_stat()
/linux-6.3-rc2/drivers/gpu/drm/omapdrm/
A Domap_dmm_tiler.c1029 int w = (map_width(xdiv, x0, x1) - strlen(nice)) / 2; in text_map()
1032 while (*nice) in text_map()
1033 *p++ = *nice++; in text_map()
1040 sprintf(nice, "%dK", tcm_sizeof(*a) * 4); in map_1d_info()
1045 if (strlen(nice) < map_width(xdiv, a->p0.x, 256 - 1)) in map_1d_info()
1046 text_map(map, xdiv, nice, a->p0.y / ydiv, in map_1d_info()
1048 else if (strlen(nice) < map_width(xdiv, 0, a->p1.x)) in map_1d_info()
1049 text_map(map, xdiv, nice, a->p1.y / ydiv, in map_1d_info()
1075 char nice[128]; in tiler_map_show() local
1116 map_2d_info(map, xdiv, ydiv, nice, in tiler_map_show()
[all …]
/linux-6.3-rc2/Documentation/admin-guide/namespaces/
A Dresource-control.rst9 users programs to play nice this problems becomes more acute.
14 memory user's they don't trust to play nice can use.
/linux-6.3-rc2/Documentation/block/
A Dioprio.rst10 priorities are supported for reads on files. This enables users to io nice
33 to the cpu nice levels just more coarsely implemented. 0 is the highest
34 BE prio level, 7 is the lowest. The mapping between cpu nice level and io
35 nice level is determined as: io_nice = (cpu_nice + 20) / 5.
/linux-6.3-rc2/drivers/crypto/qat/qat_common/
A Dicp_qat_hw_20_comp.h70 __u16 nice; member
105 QAT_FIELD_SET(val32, csr.nice, in ICP_QAT_FW_COMP_20_BUILD_CONFIG_UPPER()
A Dadf_gen4_dc.c41 hw_comp_upper_csr.nice = ICP_QAT_HW_COMP_20_CONFIG_CSR_NICE_PARAM_DEFAULT_VAL; in qat_comp_build_deflate()
/linux-6.3-rc2/Documentation/fb/
A Dpvr2fb.rst10 * It provides a nice large console (128 cols + 48 lines with 1024x768)
30 SUN12x22 font which is very nice at high resolutions.
A Dgxfb.rst13 * It provides a nice large console (128 cols + 48 lines with 1024x768)
A Dlxfb.rst14 * It provides a nice large console (128 cols + 48 lines with 1024x768)
/linux-6.3-rc2/drivers/parport/
A DTODO-parport9 a) ECP support would be nice. This can only work if both the port and
/linux-6.3-rc2/drivers/leds/
A DTODO67 It would be also nice to have useful listing mode -- name, type,
74 ethernet interface would be nice.
/linux-6.3-rc2/Documentation/cpu-freq/
A Dindex.rst10 fly. This is a nice method to save battery power, because the lower
/linux-6.3-rc2/Documentation/admin-guide/aoe/
A Dtodo.rst9 not been observed, but it would be nice to eliminate any potential for
/linux-6.3-rc2/Documentation/translations/zh_CN/admin-guide/
A Dcpu-load.rst12 avg-cpu: %user %nice %system %iowait %steal %idle
/linux-6.3-rc2/Documentation/translations/zh_TW/admin-guide/
A Dcpu-load.rst18 avg-cpu: %user %nice %system %iowait %steal %idle

Completed in 24 milliseconds

123456