Home
last modified time | relevance | path

Searched refs:batching (Results 1 – 12 of 12) sorted by relevance

/linux-6.1.9/block/
Dkyber-iosched.c187 unsigned int batching; member
507 khd->batching = 0; in kyber_init_hctx()
777 khd->batching++; in kyber_dispatch_cur_domain()
790 khd->batching++; in kyber_dispatch_cur_domain()
817 if (khd->batching < kyber_batch_size[khd->cur_domain]) { in kyber_dispatch_request()
832 khd->batching = 0; in kyber_dispatch_request()
984 seq_printf(m, "%u\n", khd->batching); in kyber_batching_show()
Dmq-deadline.c93 unsigned int batching; /* number of sequential requests made */ member
447 if (rq && dd->batching < dd->fifo_batch) in __dd_dispatch_request()
513 dd->batching = 0; in __dd_dispatch_request()
522 dd->batching++; in __dd_dispatch_request()
1059 seq_printf(m, "%u\n", dd->batching); in deadline_batching_show()
/linux-6.1.9/Documentation/trace/
Devents-kmem.rst66 callers should be batching their activities.
/linux-6.1.9/tools/memory-model/Documentation/
Dsimple.txt64 single-threaded grace-period processing is use of batching, where all
67 it more efficient. Nor is RCU unique: Similar batching optimizations
/linux-6.1.9/Documentation/RCU/Design/Expedited-Grace-Periods/
DExpedited-Grace-Periods.rst258 This batching is controlled by a sequence counter named
497 batching, so that a single grace-period operation can serve numerous
520 permits much higher degrees of batching, and thus much lower per-request
/linux-6.1.9/Documentation/networking/
Dkcm.rst245 Message batching
/linux-6.1.9/Documentation/scheduler/
Dsched-design-CFS.rst100 "server" (i.e., good batching) workloads. It defaults to a setting suitable
/linux-6.1.9/Documentation/RCU/
DRTFP.txt36 this paper helped inspire the update-side batching used in the later
2382 RCU updates, RCU grace-period batching, update overhead,
2663 RCU updates, RCU grace-period batching, update overhead,
DwhatisRCU.rst362 implementations of the RCU infrastructure make heavy use of batching in
/linux-6.1.9/Documentation/RCU/Design/Requirements/
DRequirements.rst1217 synchronize_rcu() are required to use batching optimizations so that
1560 requirement is another factor driving batching of grace periods, but it
1577 complete more quickly, but at the cost of restricting RCU's batching
2314 must *decrease* the per-operation overhead, witness the batching
/linux-6.1.9/drivers/scsi/aic7xxx/
Daic79xx.seq365 * our batching and round-robin selection scheme
/linux-6.1.9/Documentation/filesystems/
Dxfs-delayed-logging-design.rst357 in memory - batching them, if you like - to minimise the impact of the log IO on