| /linux/tools/perf/Documentation/ |
| H A D | perf-c2c.txt | 20 you to track down the cacheline contentions.
95 Specify sorting fields for single cacheline display.
141 Group the detection of shared cacheline events into double cacheline
143 feature, which causes cacheline sharing to behave like the cacheline
155 The perf c2c record command setup options related to HITM cacheline analysis
214 - sort all the data based on the cacheline address
215 - store access details for each cacheline
221 2) offsets details for each cacheline
[all...] |
| H A D | perf-amd-ibs.txt | 197 cacheline analyser tool. Both of them internally uses IBS Op PMU on AMD.
|
| H A D | perf-mem.txt | 114 - dcacheline: the cacheline the data address is on at the time of the sample
|
| H A D | perf-report.txt | 190 - dcacheline: the cacheline the data address is on at the time of the sample
|
| /linux/drivers/gpu/drm/i915/gt/ |
| H A D | intel_ring.h | 110 #define cacheline(a) round_down(a, CACHELINE_BYTES) in assert_ring_tail_valid() macro
111 GEM_BUG_ON(cacheline(tail) == cacheline(head) && tail < head); in assert_ring_tail_valid()
112 #undef cacheline in assert_ring_tail_valid()
|
| H A D | selftest_timeline.c | 97 unsigned long cacheline; in __mock_hwsp_timeline() local
110 cacheline = hwsp_cacheline(tl); in __mock_hwsp_timeline()
111 err = radix_tree_insert(&state->cachelines, cacheline, tl); in __mock_hwsp_timeline()
115 cacheline); in __mock_hwsp_timeline()
|
| /linux/drivers/md/bcache/ |
| H A D | bset.c | 526 unsigned int cacheline, in cacheline_to_bkey() argument
529 return ((void *) t->data) + cacheline * BSET_CACHELINE + offset * 8; in cacheline_to_bkey()
538 unsigned int cacheline, in bkey_to_cacheline_offset() argument
541 return (u64 *) k - (u64 *) cacheline_to_bkey(t, cacheline, 0); in bkey_to_cacheline_offset()
558 static struct bkey *table_to_bkey(struct bset_tree *t, unsigned int cacheline) in table_to_bkey() argument
560 return cacheline_to_bkey(t, cacheline, t->prev[cacheline]); in table_to_bkey()
694 unsigned int j, cacheline = 1; in bch_bset_build_written_tree() local
715 while (bkey_to_cacheline(t, k) < cacheline) { in bch_bset_build_written_tree()
721 t->tree[j].m = bkey_to_cacheline_offset(t, cacheline++, k); in bch_bset_build_written_tree()
|
| /linux/Documentation/edac/ |
| H A D | memory_repair.rst | 55 sparing has cacheline/row/bank/rank sparing granularities. For example, in
72 cacheline. Row sparing is provided as an alternative to PPR sparing functions
132 for this operation, cacheline/row/bank/rank sparing, vary in terms of the
|
| /linux/Documentation/translations/zh_CN/locking/ |
| H A D | mutex-design.rst | 60 （cacheline bouncing）这种昂贵的开销。一个类MCS锁是为实现睡眠锁的
|
| /linux/Documentation/admin-guide/hw-vuln/ |
| H A D | indirect-target-selection.rst | 8 of indirect branches and RETs located in the lower half of a cacheline.
58 in the lower half of the cacheline are vulnerable to ITS, the basic idea behind
64 second half of the cacheline. Not all retpoline sites are patched to thunks, if
|
| /linux/Documentation/arch/sparc/ |
| H A D | adi.rst | 35 size is same as cacheline size which is 64 bytes. A task that sets ADI
103 the corresponding cacheline, a memory corruption trap occurs. By
123 the corresponding cacheline, a memory corruption trap occurs. If
|
| /linux/arch/sparc/kernel/ |
| H A D | prom_irqtrans.c | 356 static unsigned char cacheline[64] in tomatillo_wsync_handler() local
367 "i" (FPRS_FEF), "r" (&cacheline[0]), in tomatillo_wsync_handler()
|
| H A D | cherrs.S | 203 sub %g1, %g2, %g1 ! Move down 1 cacheline
215 subcc %g1, %g2, %g1 ! Next cacheline
|
| /linux/arch/parisc/kernel/ |
| H A D | perf_asm.S | 132 ; Cacheline start (32-byte cacheline)
145 ; Cacheline start (32-byte cacheline)
|
| /linux/Documentation/locking/ |
| H A D | mutex-design.rst | 55 cacheline bouncing that common test-and-set spinlock implementations
|
| /linux/Documentation/ABI/testing/ |
| H A D | sysfs-edac-memory-repair | 45 - cacheline-sparing
|
| /linux/Documentation/ |
| H A D | atomic_t.txt | 358 loop body. As a result there is no guarantee what so ever the cacheline
|
| /linux/Documentation/driver-api/ |
| H A D | edac.rst | 46 lockstep is enabled, the cacheline is doubled, but it generally brings
|
| /linux/security/ |
| H A D | Kconfig.hardening | 330 best effort at restricting randomization to cacheline-sized
|
| /linux/Documentation/mm/ |
| H A D | multigen_lru.rst | 191 promotes hot pages. If the scan was done cacheline efficiently, it
|
| /linux/tools/perf/util/ |
| H A D | Build | 16 perf-util-y += cacheline.o
|
| /linux/Documentation/networking/device_drivers/ethernet/amazon/ |
| H A D | ena.rst | 28 and CPU cacheline optimized data placement.
|
| /linux/drivers/scsi/aic7xxx/ |
| H A D | aic7xxx.seq | 754 * We fetch a "cacheline aligned" and sized amount of data
758 * cacheline size is unknown.
795 * If the ending address is on a cacheline boundary,
|
| H A D | aic7xxx.reg | 1436 * Partial transfer past cacheline end to be
|
| H A D | aic79xx.seq | 1523 * We fetch a "cacheline aligned" and sized amount of data
1527 * cacheline size is unknown.
|