Home
last modified time | relevance | path

Searched refs:performance (Results 1 – 200 of 385) sorted by relevance

12

/linux-4.1.27/drivers/acpi/
Dprocessor_perflib.c96 if (!pr || !pr->performance) in acpi_processor_ppc_notifier()
101 if (ppc >= pr->performance->state_count) in acpi_processor_ppc_notifier()
105 pr->performance->states[ppc]. in acpi_processor_ppc_notifier()
200 if (!pr || !pr->performance || !pr->performance->state_count) in acpi_processor_get_bios_limit()
202 *limit = pr->performance->states[pr->performance_platform_limit]. in acpi_processor_get_bios_limit()
263 memcpy(&pr->performance->control_register, obj.buffer.pointer, in acpi_processor_get_performance_control()
280 memcpy(&pr->performance->status_register, obj.buffer.pointer, in acpi_processor_get_performance_control()
352 pr->performance->state_count = pss->package.count; in acpi_processor_get_performance_states()
353 pr->performance->states = in acpi_processor_get_performance_states()
356 if (!pr->performance->states) { in acpi_processor_get_performance_states()
[all …]
Dprocessor_throttling.c711 state, pr->throttling.states[state].performance)); in acpi_processor_get_throttling_fadt()
959 pr->throttling.states[i].performance = 1000 - step * i; in acpi_processor_get_fadt_info()
1032 (pr->throttling.states[state].performance ? pr-> in acpi_processor_set_throttling_fadt()
1033 throttling.states[state].performance / 10 : 0))); in acpi_processor_set_throttling_fadt()
/linux-4.1.27/drivers/staging/lustre/lustre/include/
Dobd_cksum.h69 unsigned int performance = 0, tmp; in cksum_type_pack() local
74 if (tmp > performance) { in cksum_type_pack()
75 performance = tmp; in cksum_type_pack()
81 if (tmp > performance) { in cksum_type_pack()
82 performance = tmp; in cksum_type_pack()
88 if (tmp > performance) { in cksum_type_pack()
89 performance = tmp; in cksum_type_pack()
/linux-4.1.27/tools/power/cpupower/bench/
DREADME-BENCH7 - Identify worst case performance loss when doing dynamic frequency
12 - Identify cpufreq related performance regressions between kernels
18 - Power saving related regressions (In fact as better the performance
28 For that purpose, it compares the performance governor to a configured
56 takes on this machine and needs to be run in a loop using the performance
58 Then the above test runs are processed using the performance governor
61 on full performance and you get the overall performance loss.
80 trigger of the cpufreq-bench, you will see no performance loss (compare with
84 will always see 50% loads and you get worst performance impact never
/linux-4.1.27/drivers/xen/
Dxen-acpi-processor.c151 dst_states = kcalloc(_pr->performance->state_count, in xen_copy_pss_data()
156 dst_perf->state_count = _pr->performance->state_count; in xen_copy_pss_data()
157 for (i = 0; i < _pr->performance->state_count; i++) { in xen_copy_pss_data()
159 memcpy(&(dst_states[i]), &(_pr->performance->states[i]), in xen_copy_pss_data()
175 dst->shared_type = _pr->performance->shared_type; in xen_copy_psd_data()
177 pdomain = &(_pr->performance->domain_info); in xen_copy_psd_data()
226 xen_copy_pct_data(&(_pr->performance->control_register), in push_pxx_to_hypervisor()
228 xen_copy_pct_data(&(_pr->performance->status_register), in push_pxx_to_hypervisor()
253 perf = _pr->performance; in push_pxx_to_hypervisor()
286 if (_pr->performance && _pr->performance->states) in upload_pm_data()
[all …]
/linux-4.1.27/include/acpi/
Dprocessor.h145 u16 performance; member
185 u8 performance:1; member
208 struct acpi_processor_performance *performance; member
227 __percpu *performance);
230 *performance, unsigned int cpu);
233 *performance,
Dacpi_bus.h301 u8 performance; /* % Performance ( " ) */ member
367 struct acpi_device_perf performance; member
/linux-4.1.27/Documentation/cpu-freq/
Dintel-pstate.txt29 the driver stated as a percentage of the available performance. The
30 available (P states) performance may be reduced by the no_turbo
35 performance level.
40 turbo_pct: displays the percentage of the total performance that
50 performance levels. The idea that frequency can be set to a single
Dboost.txt23 in technical documentation "Core performance boost". In Linux we use
29 Though the idea is to just give better performance without any user
36 single thread performance can vary. By explicitly disabling the boost
Dpcc-cpufreq.txt48 performance (ie: frequency) between the platform firmware and the OS.
82 When PCC mode is enabled, the platform will not expose processor performance
88 computes the required performance for each processor based on server workload.
91 responsible for delivering the requested performance.
118 to deliver the requested processor performance
Dgovernors.txt49 -- they're the already existing "powersave" and "performance" which
87 The CPUfreq governor "performance" sets the CPU statically to the
165 speed due to high load. This improves performance by reducing the overhead
182 the performance of the workload running on a CPU will change when
/linux-4.1.27/tools/perf/Documentation/
Dperf-kvm.txt23 a performance counter profile of guest os in realtime
26 'perf kvm record <command>' to record the performance counter profile
39 'perf kvm report' to display the performance counter profile information
42 'perf kvm diff' to displays the performance difference amongst two perf.data
51 'perf kvm stat <command>' to run a command and gather performance counter
76 Collect host side performance profile.
78 Collect guest side performance profile.
Dperf-bench.txt53 Memory access performance.
67 Suite for evaluating performance of scheduler and IPC mechanisms.
137 Suite for evaluating performance of simple memory copy in various ways.
169 Suite for evaluating performance of simple memory set in various ways.
Dperf.txt33 that provide a framework for all things performance analysis. It
Dperf-sched.txt31 it a number of times, measuring its performance.)
Dperf-stat.txt6 perf-stat - Run a command and gather performance counter statistics
16 This command runs a command and gathers performance counter statistics
/linux-4.1.27/kernel/
DKconfig.hz23 with lots of processors that may show reduced performance if
29 250 Hz is a good compromise choice allowing server performance
37 300 Hz is a good compromise choice allowing server performance
/linux-4.1.27/Documentation/networking/
Ds2io.txt48 significant performance improvement on certain platforms(SGI Altix,
52 (IA64, Xeon) resulting in noticeable performance improvement(up to 7%
92 good performance.
99 Transmit performance:
120 Receive performance:
125 b. Use 2-buffer mode. This results in large performance boost on
Dcxgb.txt35 settings to achieve the highest performance during various types of network
77 thereby reducing system CPU utilization and enhancing performance.
100 parameters for "performance tuning" an what value to use. You may or may not
278 recommends the first option for best performance with this bug:
292 increasing the Data Length to 2k bytes for increased performance. If you
301 and/or unstable performance.
303 If running on an SMP system and taking performance measurements, it
309 CPUs will have a significant difference in performance measurements.
311 order to achieve the best performance for your system.
Dpktgen.txt25 Tuning NIC for max performance
34 A larger TX ring can improve pktgen's performance, while it can hurt
41 ring-buffers for various performance reasons, and packets stalling
Dvxge.txt50 Resulting in noticeable performance improvement (up to 7% on certain
Deql.txt227 DSLIP. I did find a good tip from LinuxNET:Billy for PPP performance:
339 Although you may already done this performance testing, here
358 The following performance measurements were derived from a set
Di40e.txt95 An excellent article on performance tuning can be found at:
DPLIP.txt78 indicate that there isn't a noticeable performance drop when using IRQ-less
80 There is a performance drop on the machine hosting the driver.
De1000.txt98 all traffic types,but lacking in small packet performance and latency.
267 This has a possible negative performance impact, which may be
414 Degradation in throughput performance may be observed in some Jumbo frames
424 poor performance or loss of link.
Ddl2k.txt213 Jumbo frame usually improve the performance
234 congestion usually lower performance of
Dipvlan.txt59 puts the NIC in promiscous mode and degraded performance is a concern.
Digb.txt58 poor performance or loss of link.
Dnetlink_mmap.txt54 copied to the ring on transmission, nullifying most of the performance gains.
118 performance negatively, in order to avoid this the nm_frame_size parameter
Dixgb.txt177 tuning your network performance.
195 echo "configuring network performance , edit this file to change the interface
373 Degradation in throughput performance may be observed in some Jumbo frames
Dtcp.txt49 needs to maintain fairness and performance. Please review current
/linux-4.1.27/Documentation/scsi/
Dlink_power_management_policy.txt8 sacrifice some performance due to increased latency
12 the controller to have performance be a priority
DFlashPoint.txt44 produces high performance disk array (RAID) controllers, and complementary
49 enhance CPU utilization, optimize I/O performance, and ensure data security
76 particularly well suited for a high performance multitasking operating
118 However, BusLogic does remain committed to providing a high performance
149 embedded system. I am presently working on some performance testing and
Dufs.txt32 * Optimized performance:
33 For UFS version 1.0 and 1.1 the target performance is as follows,
DBusLogic.txt19 BusLogic, Inc. designed and manufactured a variety of high performance SCSI
40 to achieve the full performance that BusLogic SCSI Host Adapters and modern
42 be depended upon for high performance mission critical applications. All of
43 the major performance features can be configured from the Linux kernel command
45 tune driver performance and error recovery to their particular needs.
64 readily achieve, and the Linux community has available high performance host
139 device or logical unit, and can improve I/O performance substantially. In
141 performance, and scatter/gather I/O can support as many segments as can be
Dtmscsim.txt180 performance. You're likely to crash your disks.)
227 performance.
338 The defaults (7,0,31,15,3,1) are aggressive to allow good performance. You
339 can use tmscsim=7,0,31,63,4,0 for maximum performance, if your SCSI chain
/linux-4.1.27/Documentation/ABI/testing/
Dsysfs-bus-event_source-devices-events15 Description: Generic performance monitoring events
17 A collection of performance monitoring events that may be
33 Description: Per-pmu performance monitoring events specific to the running system
37 performance monitoring event supported by the <pmu>. The name
Dsysfs-devices-mmc7 area can help to improve the card performance. If the feature
18 area can help to improve the card performance. If the feature
Dsysfs-driver-samsung-laptop5 Description: Some Samsung laptops have different "performance levels"
10 will show the current performance level. Writing to the
Dsysfs-driver-genwqe46 Used for performance and utilization measurements.
52 Used for performance and utilization measurements.
Dsysfs-class-scsi_host23 stack for a performance advantage when possible. A value of
Dsysfs-devices-system-cpu157 power/performance results for platforms using acpi-cpufreq.
218 the driver stated as a percentage of the available performance.
221 the driver stated as a percentage of the available performance.
/linux-4.1.27/fs/squashfs/
DKconfig50 Doing so can significantly improve performance because
62 decompression performance and CPU and memory usage.
77 poor performance on parallel I/O workloads when using multiple CPU
81 using this option may improve overall I/O performance.
91 poor performance on parallel I/O workloads when using multiple CPU
177 This, however, gives poor performance on MTD NAND devices where
182 performance for some file access patterns (e.g. sequential
/linux-4.1.27/Documentation/mmc/
Dmmc-async-req.txt8 transfer, the DMA preparation overhead would not affect the MMC performance.
24 performance gain is 5% for large writes and 10% on large reads on a L2 cache
27 in parallel with the transfer performance won't be affected.
60 the performance loss. A way to optimize for this is to split the current
/linux-4.1.27/Documentation/virtual/
Dparavirt_ops.txt26 are called frequently and are performance critical. The overhead is
32 them are very performance critical.
/linux-4.1.27/fs/ext4/
DKconfig15 and a number of other features to improve performance and speed
20 filesystem; while there will be some performance gains from
22 performance gains will require enabling ext4 features in the
/linux-4.1.27/arch/parisc/kernel/
Dperf_asm.S41 ; Enable the performance counters
56 mtctl %r26,ccr ; turn on performance coprocessor
57 pmenb ; enable performance monitor
61 mtctl %r26,ccr ; turn off performance coprocessor
82 mtctl %r26,ccr ; turn on performance coprocessor
83 pmdis ; disable performance monitor
87 mtctl %r26,ccr ; turn off performance coprocessor
/linux-4.1.27/drivers/cpufreq/
DKconfig55 startup. If in doubt, select 'performance'.
58 bool "performance"
61 Use the CPUFreq governor 'performance' as default. This sets
92 driver. Fallback governor will be the performance governor.
104 driver. Fallback governor will be the performance governor.
108 tristate "'performance' governor"
/linux-4.1.27/drivers/staging/lustre/lustre/
DKconfig26 performance computing. Source code of both kernel space and user space
49 expensive checks that may have a performance impact.
/linux-4.1.27/drivers/amba/
DKconfig11 which controls AHB bus master arbitration and some performance
/linux-4.1.27/fs/ext3/
DKconfig38 performance. The use of "data=writeback" can cause
41 "data=ordered" mode can also result in major performance
47 If you have been historically happy with ext3's performance,
/linux-4.1.27/Documentation/devicetree/bindings/devfreq/event/
Dexynos-ppmu.txt5 each IP. PPMU provides the primitive values to get performance data. These
7 use to analyze system performance, to make behaviors visible and to count
/linux-4.1.27/Documentation/infiniband/
Dipoib.txt45 datagrams, TCP segments, etc and increases the performance for large
89 The performance impact of this option is negligible, so it
96 performance, because it adds tests to the fast path.
/linux-4.1.27/arch/parisc/
DKconfig.debug11 data. This option may have a slight performance impact because a
DKconfig213 performance, a page size of 16KB is recommended. For best
219 16KB For best performance
220 64KB For best performance, might give more overhead.
/linux-4.1.27/drivers/mmc/card/
DKconfig41 send large requests in order to get high performance. Many
48 a big performance gain at the cost of up to 64 KiB of
/linux-4.1.27/Documentation/devicetree/bindings/arc/
Dpct.txt3 The ARC700 can be configured with a pipeline performance monitor for counting
/linux-4.1.27/fs/cifs/
DKconfig27 useful performance and security features (see the description
53 These additional statistics may have a minor effect on performance
56 Unless you are a developer or are doing network performance analysis
147 messages in some error paths, slowing performance. This
187 In general SMB3 enables better performance, security
/linux-4.1.27/Documentation/filesystems/
Dext3.txt57 will hurt performance, but it's good for data-safety.
61 performance.
70 safe to use, at some performance penalty. If
72 disabling barriers may safely improve performance.
170 typically provide the best ext3 performance.
Docfs2.txt73 will hurt performance, but it's good for data-safety.
77 performance.
101 nodes, which gains high performance at risk of getting
Dext4.txt61 - When comparing performance with other filesystems, it's always
92 * journal checksumming for robustness, performance
119 The big performance win will come with mballoc, delalloc and flex_bg
179 will hurt performance, but it's good for data-safety.
183 performance.
192 safe to use, at some performance penalty. If
194 disabling barriers may safely improve performance.
341 minimizes the impact on the system performance
361 performance, it is off by default.
381 cause severe performance problems or even
[all …]
Dceph.txt5 performance, reliability, and scalability.
39 storage to significantly improve performance for common workloads. In
131 complex code, sacrificing performance without affecting correctness,
Dxfs.txt5 XFS is a high performance journaling filesystem which originated
9 Btrees (directories, extents, free space) to aid both performance
66 machine images, but may have a performance impact.
70 mount option because the performance impact of this option
136 systems, then it may be reduced at some cost to performance
Djfs.txt21 is to allow for higher performance when restoring a volume
Dudf.txt19 performance due to very poor read-modify-write support supplied internally
Dnilfs2.txt68 write performance for overwriting.
117 performance penalty or wear of flash device
Dpath-lookup.txt21 the lock hold time significantly and affects performance in large SMP machines.
134 hlist-->| N-+-------->| N-+-> impose a measurable performance overhead
287 these cases is fundamental for performance and scalability because blocking
333 or stored into. The result is massive improvements in performance and
Df2fs.txt51 and in order to enhance the performance, it should eliminate or relax the update
80 - Enlarge the random write area for better performance, but provide the high
141 time as much as possible, even though normal performance
543 for devices showing very good sequential write performance, since free segments
/linux-4.1.27/Documentation/ABI/stable/
Dsysfs-acpi-pmprofile6 power management (and performance) requirement expectations
/linux-4.1.27/Documentation/arm/
Dmem_alignment13 a high performance cost. It better be rare.
18 mode isn't recommended for performance reasons (just think about the
Dkernel_mode_neon.txt19 code that runs in kernel mode. However, for performance reasons, the NEON/VFP
50 For reasons of performance and simplicity, it was decided that there shall be no
DIXP4xx90 cases the performance hit is acceptable. In addition, you cannot
94 By default, the direct method is used for performance reasons. If
/linux-4.1.27/Documentation/filesystems/pohmelfs/
Ddesign_notes.txt22 * Very high performance.
49 Only a few non-performance critical operations use the write-through cache and
70 Crypto performance heavily depends on the number of crypto threads, which asynchronously perform
/linux-4.1.27/Documentation/x86/
D00-INDEX14 - how to use x86 Memory Type Range Registers to increase performance
Dtlb.txt57 performance counters and 'perf stat', like this:
/linux-4.1.27/Documentation/powerpc/
Dcpu_features.txt24 several paths that are performance-critical and would suffer if an array
26 performance penalty but still allow for runtime (rather than compile-time) CPU
/linux-4.1.27/drivers/scsi/arm/
DKconfig18 This is a feature of SCSI-2 which improves performance: the host
30 In general, this improves performance; however some SCSI devices
/linux-4.1.27/arch/powerpc/platforms/ps3/
DKconfig46 system will have optimal runtime performance.
154 This support is required to use the logical performance monitor
157 If you intend to use the advanced performance monitoring and
/linux-4.1.27/Documentation/ABI/removed/
Dvideo13948 performance issues in its first generation. Any video1394 user had
/linux-4.1.27/drivers/power/avs/
DKconfig9 (temperature depending performance).
/linux-4.1.27/Documentation/filesystems/nfs/
Dknfsd-stats.txt69 values may indicate a performance limitation.
75 configuring more nfsd threads will probably improve the performance
118 improve the performance of the workload. If this patten is not
129 affecting performance. Unfortunately, it's only a clue and not
Drpc-server-gss.txt37 performance-sensitive. The latter is simpler and needs to be very fast.
/linux-4.1.27/Documentation/device-mapper/
Ddm-service-time.txt9 of in-flight I/Os on a path with the performance value of the path.
10 The performance value is a relative throughput value among all paths
Dcache.txt7 It aims to improve performance of a block device (eg, a spindle) by
14 closely with the virtual memory system should give better performance.
151 Policy hints affect performance, not correctness.
200 back cache block contents later for performance reasons,
211 the policy we currently think is giving best all round performance.
Ddm-log.txt34 performance. This method can also be used if no storage device is
Dpersistent-data.txt47 the same transaction so performance is reasonable. The commit method
Ddm-crypt.txt74 encryption threads to a single thread degrades performance
Dthin-provisioning.txt15 lookup tables, and so performance was O(depth). This new
26 - Improve performance by storing the metadata on SSD.
33 feedback. Different use cases will have different performance
/linux-4.1.27/fs/cramfs/
DKconfig20 in terms of performance and features.
/linux-4.1.27/mm/
DKconfig25 decreased performance over SPARSEMEM. If unsure between
40 can have degraded performance from the extra overhead that
56 performance benefits, along with decreased code complexity,
247 used within a guest, thus imposing performance penalties associated
417 This feature can improve computing performance to certain
442 performance improvement benefit to the applications using
475 in a negligible performance hit.
491 compare-against-NULL resulting in a negligible performance hit
561 reads, can also improve workload performance.
/linux-4.1.27/Documentation/crypto/
Ddescore-readme.txt34 This second release includes a number of performance enhancements for
46 performance comparison to other available des code which i could
68 so despite the author's inclusion of some of the performance
99 performance. his code takes 26 sparc instructions to compute one
165 portable (the ifdef's are performance enhancements) and it is faster
197 OPTIONAL performance optimizations
203 these are just performance tweaks.
212 be able to get performance equal to assembly-coding, except that:
/linux-4.1.27/arch/x86/xen/
DKconfig47 Enabling this option may incur a significant performance overhead.
/linux-4.1.27/fs/ceph/
DKconfig12 scalable file system designed to provide high performance,
/linux-4.1.27/drivers/mtd/spi-nor/
DKconfig15 on the usage this feature may provide performance gain in comparison
/linux-4.1.27/Documentation/
Dunaligned-memory-access.txt51 transparently, but there is usually a significant performance cost.
54 at significant cost to performance.
64 platforms and will cause performance problems on others.
131 the extra instructions obviously cause a loss in performance compared to the
225 terms of performance.
Dpercpu-rw-semaphore.txt9 is bouncing between L1 caches of the cores, causing performance
Dbasic_profiling.txt41 a 0Hz CPU, APIC was not on. Be aware that idle=poll may mean a performance
Dstatic-keys.txt9 performance-sensitive fast-path kernel code, via a GCC feature and a code
137 Tracepoints are disabled by default, and can be placed in performance critical
247 performance improvement. Testing done on 3.3.0-rc2:
Dlocal_ops.txt21 counters. They minimize the performance cost of standard atomic operations by
28 and for various performance monitoring counters.
Dthis_cpu_ops.txt34 exceptional situations and may impact performance and/or correctness
116 Per cpu variables are used for performance reasons. Bouncing cache
328 missing local cache line of a per cpu area, its performance and hence
Dio-mapping.txt82 in a significant performance penalty.
Dbcache.txt9 It's designed around the performance characteristics of SSDs - it only allocates
136 - Bad write performance
138 If write performance is not what you expected, you probably wanted to be
145 - Bad performance, or traffic not going to the SSD that you'd expect
Dkmemleak.txt182 The main drawback is the reduced performance of memory allocation and
185 intended for debugging purposes where the performance might not be the
DDMA-attributes.txt39 buffered to improve performance.
/linux-4.1.27/Documentation/vm/
Dfrontswap.txt2 In some environments, dramatic performance savings may be obtained because
43 in swap device writes is lost (and also a non-trivial performance advantage)
67 When a workload starts swapping, performance falls through the floor.
68 Frontswap significantly increases performance in many such workloads by
88 low while providing a significant performance improvement (25%+)
113 conditions allow), thus mitigating the potentially awful performance impact
120 2) Sure there may be performance advantages in some situations, but
Dzswap.txt7 significant performance improvement if reads from the compressed cache are
17     performance impact of swapping.
Dcleancache.txt100 Cleancache provides a significant performance benefit to many workloads
196 5) What's the performance impact?
199 Briefly, performance gains can be significant on most workloads,
230 - To maximize performance, all pages fetched from the FS should
237 hook to get best performance for some backends.
262 time, but have insignificant performance impact when cleancache remains
Dremap_file_pages.txt25 One side effect of emulation (apart from performance) is that user can hit
Dslub.txt8 an impact on overall system performance which may make a bug more
110 Getting more performance
113 To some degree SLUB's performance is limited by the need to take the
Dpage_migration44 performance of the process does not sink dramatically. Also the pages
52 Processes will run with similar performance after migration.
Dpage_owner.txt33 would not affect to allocation performance. Following is the kernel's
Dsplit_page_table_lock82 To avoid increasing size of struct page and have best performance, we use a
/linux-4.1.27/Documentation/ia64/
DREADME36 o Hardly any performance tuning has been done. Obvious targets
/linux-4.1.27/drivers/net/ethernet/sfc/
DKconfig37 features, allowing accelerated network performance in
/linux-4.1.27/Documentation/ide/
DChangeLog.ide-tape.1995-20027 * overall Linux performance and eventually failed as
12 * and performance doesn't seem to be much affected
85 * boosted performance by a factor of 2. Pipelined
101 * without any decrease in performance.
113 * unit, making performance almost independent of the
/linux-4.1.27/arch/ia64/
DKconfig157 under a hypervisor, potentially improving performance significantly
270 performance, a page size of 8KB or 16KB is recommended. For best
277 8KB For best IA-64 performance
278 16KB For best IA-64 performance
360 performance hit.
492 Selects whether support for the IA-64 performance monitor hardware
DKconfig.debug38 translation performance. Normally you want the VHPT active but you
/linux-4.1.27/drivers/atm/
DKconfig59 generally using printks, but still has some impact on performance.
174 generally using printks, but still has some impact on performance.
231 inactive, they will have only a modest impact on performance.
269 inactive, they will have only a modest impact on performance.
293 inactive, they will have only a modest impact on performance.
326 inactive, they will have only a modest impact on performance.
/linux-4.1.27/arch/metag/
DKconfig.soc50 raised on unaligned accesses. This feature does cost performance
/linux-4.1.27/arch/microblaze/
DKconfig.platform17 shifts will be necessary. This might result in bad performance
/linux-4.1.27/drivers/staging/unisys/Documentation/
Dproc-entries.txt30 performance, by setting the number of cycles we wait before going idle
34 /proc/uislib/smart_wakeup: This entry is used to tune performance, by
/linux-4.1.27/fs/xfs/
DKconfig8 XFS is a high performance journaling filesystem which originated
12 Btrees (directories, extents, free space) to aid both performance
/linux-4.1.27/Documentation/cdrom/
Dpacket-writing.txt53 follow the specification, but suffer bad performance problems if the
69 necessary. However, using the pktcdvd driver can improve performance
/linux-4.1.27/Documentation/fb/
Dudlfb.txt19 can support surprisingly high resolutions with good performance for
109 more stable, and higher performance.
155 performance tests to start and finish in a very short
Ds3fb.txt71 * better values for some magic registers (performance issues)
Dintelfb.txt67 to occur in bursts which can significantly increase performance.
/linux-4.1.27/Documentation/filesystems/cifs/
Dcifs.txt16 performance safe distributed caching (oplock), optional packet
DTODO99 3) Additional performance testing and optimization using iozone and similar -
DCHANGES193 time to a larger value which may hurt performance in some cases).
210 mounted to windows. Add support for POSIX MkDir (slight performance
295 Use fsuid (fsgid) more consistently instead of uid (gid). Improve performance
316 to be smaller at first (but increasing) so large write performance performance
323 file copy performance on lightly loaded networks. Fix oops in dnotify
349 Add writepage performance improvements. Fix path name conversions
601 This allows files that differ only in case and improves performance of file
851 lookup information for directory inodes, improving performance,
/linux-4.1.27/security/integrity/
DKconfig27 Different keyrings improves search performance, but also allow
/linux-4.1.27/Documentation/x86/i386/
DIO-APIC.txt34 none of those IRQ sources is performance-critical.
76 necessity, PCI IRQs can be shared at will, but it's a good for performance
/linux-4.1.27/drivers/infiniband/ulp/ipoib/
DKconfig48 turned off, this debugging code will have some performance
/linux-4.1.27/lib/
DKconfig.kasan17 ~x3 performance slowdown.
/linux-4.1.27/Documentation/arm/nwfpe/
DNOTES17 then no conversion would occur. This has performance considerations. The
/linux-4.1.27/net/caif/
DKconfig30 Be aware that doing this will impact performance.
/linux-4.1.27/drivers/net/ethernet/adi/
DKconfig40 To get maximum network performance, you should use L1 memory as rx/tx
/linux-4.1.27/Documentation/arm64/
Dlegacy_instructions.txt28 performance, but at the loss of ability to gather runtime statistics
/linux-4.1.27/Documentation/arm/OMAP/
Domap_pm112 DSPBridge expresses target DSP performance levels in terms of OPP IDs.
113 CPUFreq expresses target MPU performance levels in terms of MPU
/linux-4.1.27/Documentation/trace/
Devents-kmem.txt48 the per-CPU allocator (high performance) or the buddy allocator.
53 impairs performance by disabling interrupts, dirtying cache lines between
Dtracepoint-analysis.txt146 performance analyst to do it by hand. In the event that the discrete event
147 occurrences are useful to the performance analyst, then perf can be used.
/linux-4.1.27/Documentation/hwmon/
Dacpi_power_meter42 result in reduced performance.
/linux-4.1.27/crypto/
DKconfig360 gain performance compared with software implementation.
391 and gain better performance as compared with the table implementation.
601 the data lanes filled to get performance benefit. If the data
681 still having decent performance on 32-bit processors.
722 demonstrates excellent performance. Rijndael's operations are
744 demonstrates excellent performance. Rijndael's operations are
766 demonstrates excellent performance. Rijndael's operations are
796 demonstrates excellent performance. Rijndael's operations are
825 demonstrates excellent performance. Rijndael's operations are
1104 an algorithm optimized for 64-bit processors with good performance
/linux-4.1.27/arch/blackfin/
DKconfig685 still be able to read it (such as for performance monitoring), but
714 CYCLES performance registers.
724 but you can safely use the CYCLES performance registers
734 safely use the CYCLES performance registers on a Blackfin
1064 This can give a nice increase in performance, but beware of
1071 performance is worse than Write Back.
1081 This can give a nice increase in performance, but beware of
1088 performance is worse than Write Back.
1118 memory they do not own. This comes at a performance penalty
DKconfig.debug165 has a great impact on performance, (an interrupt every 16 change of
232 Say N here to disable that check to improve the performance.
/linux-4.1.27/drivers/thermal/int340x_thermal/
Dint3403_thermal.c36 u64 performance; member
/linux-4.1.27/fs/ocfs2/
DKconfig76 performance of the filesystem.
/linux-4.1.27/Documentation/isdn/
DREADME.concap13 to suggest changes that improve usability or performance of the
220 performance. However, typically the concap interface will be used by
255 protocol is configured. Thus, eliminating the performance penalty
/linux-4.1.27/arch/mips/include/asm/mach-cavium-octeon/
Dkernel-entry-init.h40 # First clear off CvmCtl[IPPCI] bit and move the performance
/linux-4.1.27/arch/mips/cavium-octeon/
DKconfig9 workarounds will cause a slight decrease in performance on
/linux-4.1.27/Documentation/scheduler/
Dsched-arch.txt17 Unlocked context switches introduce only a very minor performance
/linux-4.1.27/Documentation/mic/
Dmic_overview.txt29 single threaded performance for the host compared to MIC, the ability of
/linux-4.1.27/fs/f2fs/
DKconfig73 If you want to improve the performance, say N.
/linux-4.1.27/Documentation/x86/x86_64/
Dboot-options.txt101 to get slightly better performance in multiprocessor benchmarks. It also
102 makes some profiling using performance counters more accurate.
104 CPUs) this option has no performance advantage over the normal idle loop.
/linux-4.1.27/Documentation/usb/
Dehci.txt146 overall system load. Latency is also a performance concern.
190 When drivers don't do that, their performance results will show it.
209 TBD: Interrupt and ISO transfer performance issues. Those periodic
Diuu_phoenix.txt49 For example to have best performance than a simple clockmode=3, try this:
/linux-4.1.27/Documentation/thermal/
Dintel_powerclamp.txt45 Test/Analysis has been made in the areas of power, performance,
223 This section describes the general performance data collected on
256 achieve up to 40% better performance per watt. (measured by a spin
/linux-4.1.27/net/atm/
DKconfig56 enhancing overall network performance.
/linux-4.1.27/arch/x86/
DMakefile_32.cpu24 # They make zero difference whatsosever to performance at this time.
/linux-4.1.27/arch/hexagon/
DKconfig33 performance and low power across a wide variety of applications.
/linux-4.1.27/drivers/staging/lustre/
DREADME.txt15 in parallel to maximize performance.
/linux-4.1.27/fs/btrfs/
DKconfig80 performance, or export extra information via sysfs.
/linux-4.1.27/arch/mips/
DKconfig1343 performance.
1393 performance.
1692 but may decrease the performance of the system so this option should
2001 Using 8kB page size will result in higher performance kernel at
2010 Using 16kB page size will result in higher performance kernel at
2019 Using 32kB page size will result in higher performance kernel at
2028 Using 64kB page size will result in higher performance kernel at
2087 SiByte Linux port. Seems to give a small performance benefit.
2416 Access). This option improves performance on systems with more
2430 bool "Enable hardware performance counter support for perf events"
[all …]
/linux-4.1.27/arch/c6x/
DKconfig149 Say N here to disable that check to improve the performance.
/linux-4.1.27/arch/sh/
DKconfig.debug29 performance hit. Most users should say N.
/linux-4.1.27/Documentation/devicetree/bindings/arm/
Dfw-cfg.txt23 arrays, bundled together only for better performance. The bytes constituting
/linux-4.1.27/drivers/scsi/aic7xxx/
DKconfig.aic79xx30 the default is set to 32. Higher values may result in higher performance
DKconfig.aic7xxx35 the default is set to 32. Higher values may result in higher performance
/linux-4.1.27/Documentation/m68k/
DREADME.buddha185 performance of the card, since this doesn't happen very
193 mentioning). You could think of a small performance boost
/linux-4.1.27/Documentation/sound/alsa/
Dhdspm.txt24 scaled. (Only important for low performance boards).
99 hwdep-interface. Since it could be a performance problem always
/linux-4.1.27/Documentation/blockdev/
DREADME.DAC96019 Mylex, Inc. designs and manufactures a variety of high performance PCI RAID
54 The DAC960 RAID controllers are supported solely as high performance RAID
60 as a SCSI driver. The DAC960 driver is designed for as high a performance as
62 performance devices. The DAC960 driver includes extensive error logging and
248 For maximum performance and the most efficient E2FSCK performance, it is
489 the logical drives in a performance degraded mode until the failed drive is
/linux-4.1.27/arch/x86/math-emu/
DREADME60 (2) Divide has been tuned to get reasonable performance. The algorithm
66 (3) The sqrt function has been tuned to get good performance. It is based
165 upon instruction mix. Relative performance is best for the instructions
194 The performance under Linux is improved by the use of look-ahead code.
/linux-4.1.27/arch/arm64/
DKconfig.debug88 performance is more important than memory pressure.
/linux-4.1.27/Documentation/ptp/
Dptp.txt38 single system, in order to allow performance comparisons.
/linux-4.1.27/Documentation/sysctl/
Dvm.txt210 Use of this file can cause performance problems. Since it discards cached
579 be forced to incur a performance penalty in OOM conditions when the
773 performance impact. Reclaim code needs to take various locks to find freeable
799 memory would cause a measurable performance reduction. The page allocator
806 throttle the process. This may decrease the performance of a single process
808 anymore but it preserve the memory on other nodes so that the performance
/linux-4.1.27/Documentation/timers/
Dtimers-howto.txt92 delay / performance for your specific code path. Exact
Dhrtimers.txt94 used in various performance-critical areas of e.g. memory management and
132 to avoid the performance impact of 64-bit multiplications and divisions
Dtimekeeping.txt151 On SMP systems, it is crucial for performance that sched_clock() can be called
152 independently on each CPU without any synchronization performance hits.
/linux-4.1.27/drivers/net/ethernet/chelsio/
DKconfig29 performance tuning is in <file:Documentation/networking/cxgb.txt>.
/linux-4.1.27/arch/arm/crypto/
DKconfig88 demonstrates excellent performance. Rijndael's operations are
Dsha1-armv4-large.S13 @ Size/performance trade-off
15 @ impl size in bytes comp cycles[*] measured performance
31 @ performance is affected by prologue and epilogue overhead,
/linux-4.1.27/Documentation/locking/
Drt-mutex.txt63 Pending-ownership handling is a performance optimization:
Dlockstat.txt10 Because things like lock contention can severely impact performance.
/linux-4.1.27/arch/unicore32/
DKconfig162 here, which has a severe performance impact. This is necessary for
/linux-4.1.27/Documentation/sound/alsa/soc/
Doverview.txt91 clocking.txt: ASoC clocking for best power performance.
/linux-4.1.27/tools/perf/
Ddesign.txt95 * Generalized performance counter event types, used by the hw_event.event_id
128 * does not support performance counters. These counters measure various
450 If your architecture does not have hardware performance metrics, you can
/linux-4.1.27/arch/
DKconfig57 Certain performance-sensitive kernel code, such as trace points,
293 on a struct page for better performance. However selecting this
326 For best performance, an arch should use seccomp_phase1 and
/linux-4.1.27/Documentation/mtd/
Dnand_ecc.txt286 And of course the performance might depend on alignment, but I expect
288 otherwise that should be fixed to get maximum performance).
561 (using time to measure the performance). After this iteration I was back
656 Changed the code but again this slightly degrades performance. Tried all
673 million iterations. So basically I've improved the performance by a
/linux-4.1.27/arch/powerpc/platforms/8xx/
DKconfig127 performance hit.
/linux-4.1.27/Documentation/video4linux/bttv/
DREADME.quirks80 parking, thus lowering arbitration performance. The Bt879 drivers must
/linux-4.1.27/Documentation/block/
Dioprio.txt26 to be more directly mappable to performance, by passing in a wanted data
/linux-4.1.27/kernel/power/
DKconfig280 better performance thanks to cache locality; unfortunately,
287 lower power usage at the cost of small performance overhead.
/linux-4.1.27/Documentation/cgroups/
Dunified-hierarchy.txt341 system, but also impacts system performance due to overreclaim, to
355 well, resulting in much better overall workload performance.
373 lead to gradual performance degradation. The user can monitor this
375 gives acceptable performance is found.
/linux-4.1.27/drivers/md/
DKconfig114 of redundancy and performance.
291 dm-cache attempts to improve performance of a block device by
292 moving frequently used data to a smaller, higher performance
/linux-4.1.27/drivers/net/ethernet/marvell/
DKconfig83 with better performance and more complete ethtool support.
/linux-4.1.27/Documentation/parisc/
Dregisters66 F (performance monitor) 0
/linux-4.1.27/arch/arm/mach-omap2/
DKconfig260 SDRAM performance at lower CORE OPPs. There are relatively few
/linux-4.1.27/Documentation/video4linux/
Dsi476x.txt19 useful for EoL testing/RF performance estimation, parameters that have
/linux-4.1.27/arch/nios2/
DKconfig94 emulated in software if you say Y here, which has a performance
/linux-4.1.27/Documentation/RCU/
Drcu.txt23 on modern CPUs is what gives RCU its performance advantages
/linux-4.1.27/drivers/ide/
DKconfig355 speeds and to configure the chip to optimum performance.
380 optimum performance.
389 and to configure the chip to optimum performance.
503 configure the chip to optimum performance.
638 configure the chip to optimum performance.

12