| /linux/Documentation/admin-guide/hw-vuln/ |
| H A D | processor_mmio_stale_data.rst | 2 Processor MMIO Stale Data Vulnerabilities
5 Processor MMIO Stale Data Vulnerabilities are a class of memory-mapped I/O
12 stale data into core fill buffers where the data can subsequently be inferred
21 Propagators are operations that result in stale data being copied or moved from
22 one microarchitectural buffer or register to another. Processor MMIO Stale Data
23 Vulnerabilities are operations that may result in stale data being directly
27 Fill Buffer Stale Data Propagator (FBSDP)
29 Stale data may propagate from fill buffers (FB) into the non-coherent portion
31 does not make stale data architecturally visible. Stale data must be propagated
34 Sideband Stale Data Propagator (SSDP)
[all …]
|
| H A D | gather_data_sampling.rst | 13 instruction that is transiently executed encounters a fault, stale data from
16 infer stale data using typical side channel techniques like cache timing
19 The attacker uses gather instructions to infer the stale vector register data.
29 Without mitigation, GDS can infer stale data across virtually all
|
| /linux/tools/testing/selftests/net/ |
| H A D | test_neigh.sh | 82 run_cmd "ip -n $ns1 neigh add $ip_addr lladdr $mac nud stale dev veth0 extern_valid"
103 run_cmd "ip -n $ns1 neigh add $ip_addr lladdr $mac nud stale dev veth0 extern_valid use"
112 run_cmd "ip -n $ns1 neigh add $ip_addr lladdr $mac nud stale dev veth0"
113 run_cmd "ip -n $ns1 neigh replace $ip_addr lladdr $mac nud stale dev veth0 extern_valid"
116 run_cmd "ip -n $ns1 neigh replace $ip_addr lladdr $mac nud stale dev veth0"
127 run_cmd "ip -n $ns1 neigh add $ip_addr lladdr $mac nud stale dev veth0 extern_valid"
128 run_cmd "ip -n $ns1 neigh replace $ip_addr lladdr $mac nud stale dev veth0 extern_valid managed"
138 run_cmd "ip -n $ns1 neigh add $ip_addr lladdr $mac nud stale dev veth0 extern_valid"
149 run_cmd "ip -n $ns1 neigh add $ip_addr lladdr $mac nud stale dev veth0 extern_valid"
162 run_cmd "ip -n $ns1 neigh add $ip_addr lladdr $mac nud stale dev veth0 extern_valid"
[all …]
|
| /linux/drivers/gpu/drm/i915/gt/ |
| H A D | intel_gt_buffer_pool.c | 40 struct intel_gt_buffer_pool_node *node, *stale = NULL; in pool_free_older_than() local
68 node->free = stale; in pool_free_older_than()
69 stale = node; in pool_free_older_than()
80 while ((node = stale)) { in pool_free_older_than()
81 stale = stale->free; in pool_free_older_than()
|
| /linux/scripts/ |
| H A D | remove-stale-files | 8 # running 'make clean'. Then, the stale generated files are left over, often
11 # Also, 'git status' shows such stale build artifacts as untracked files.
20 # yard. Stale files stay in this file for a while (for some release cycles?),
|
| /linux/drivers/md/bcache/ |
| H A D | extents.c | 15 * counter. Garbage collection is used to remove stale pointers.
85 return "stale"; in bch_ptr_status()
540 unsigned int i, stale; in bch_extent_bad() local
552 stale = ptr_stale(b->c, k, i); in bch_extent_bad()
554 if (stale && KEY_DIRTY(k)) { in bch_extent_bad()
556 pr_info("stale dirty pointer, stale %u, key: %s\n", in bch_extent_bad()
557 stale, buf); in bch_extent_bad()
560 btree_bug_on(stale > BUCKET_GC_GEN_MAX, b, in bch_extent_bad()
561 "key too stale: %i, need_gc %u", in bch_extent_bad()
562 stale, b->c->need_gc); in bch_extent_bad()
[all …]
|
| /linux/fs/xfs/ |
| H A D | xfs_buf_item.c | 193 * regions, we do not relog them in stale buffers. This has the effect of
213 * The buffer is stale, so all we need to log is the buf log in xfs_buf_item_size()
313 * The buffer is stale, so all we need to log in xfs_buf_item_format_segment()
443 * For a stale BLI, process all the necessary completions that must be
505 int stale = bip->bli_flags & XFS_BLI_STALE; in xfs_buf_item_unpin() local
526 if (stale) { in xfs_buf_item_unpin()
531 * stale so we own both lock and reference exclusively here. We in xfs_buf_item_unpin()
578 * been marked stale, we could end up stalling until someone else in xfs_buf_item_push()
579 * issues a log force to unpin the stale buffer. Check for the in xfs_buf_item_push()
637 * We should never get here with a stale BLI via that path as in xfs_buf_item_put()
[all …]
|
| H A D | xfs_buf.c | 82 * When we mark a buffer stale, we remove the buffer from the LRU and clear the
87 * This prevents build-up of stale buffers on the LRU.
99 * flush this buffer to disk now that it is stale. The delwri queue has in xfs_buf_stale()
380 * in the cache is stale and the transaction that made in _xfs_buf_obj_cmp()
381 * it stale has not yet committed. i.e. we are in _xfs_buf_obj_cmp()
385 * Note: If we're scanning for incore buffers to stale, don't in _xfs_buf_obj_cmp()
386 * complain if we find non-stale buffers. in _xfs_buf_obj_cmp()
444 * if the buffer is stale, clear all the external state associated with in xfs_buf_find_lock()
721 * mark the buffer stale. This ensures that anyone who has a current in xfs_buf_read_map()
771 /* Let the actual reader deal with stale buffers. */ in xfs_buf_readahead_map()
[all …]
|
| /linux/drivers/md/ |
| H A D | dm-ps-historical-service-time.c | 9 * Marks paths stale if they have not finished within hst *
10 * num_paths. If a path is stale and unused, we will send a single
14 * multipath device is unused. If a path is stale and in use, limit the
346 u64 *out, u64 *stale) in hst_fill_compare() argument
353 *stale = pi->stale_after; in hst_fill_compare()
392 * If an unloaded path is stale, choose it. If both paths are unloaded, in hst_compare()
393 * choose path that is the most stale. in hst_compare()
419 /* In the case that the 'winner' is stale, limit to equal usage. */ in hst_compare()
|
| /linux/fs/xfs/scrub/ |
| H A D | dirtree.c | 294 dl->stale = true; in xchk_dirpath_revalidate()
381 if (dl->stale) { in xchk_dirpath_step_up()
481 if (dl->stale) { in xchk_dirpath_step_up()
582 if (!error && dl->stale) in xchk_dirpath_walk_upwards()
670 * path cannot be stale. in xchk_dirpath_is_stale()
710 if (dl->stale || dl->aborted) in xchk_dirtree_live_update()
720 dl->stale = true; in xchk_dirtree_live_update()
750 dl->stale = false; in xchk_dirtree_reset()
840 ASSERT(dl->stale); in xchk_dirtree_find_paths_to_root()
848 } while (dl->stale); in xchk_dirtree_find_paths_to_root()
[all …]
|
| H A D | dirtree_repair.c | 213 * if the scan results have become stale.
425 if (dl->stale) { in xrep_dirtree_unlink()
550 * failed or the scan data are now stale. This keeps things simple for in xrep_dirtree_delete_path()
557 if (!error && dl->stale) in xrep_dirtree_delete_path()
698 if (dl->stale) { in xrep_dirtree_adopt()
754 * failed or the scan data are now stale. This keeps things simple for in xrep_dirtree_move_to_orphanage()
761 if (!error && dl->stale) in xrep_dirtree_move_to_orphanage()
819 if (!dl->stale) { in xrep_dirtree()
|
| /linux/include/linux/ |
| H A D | errno.h | 20 #define EOPENSTALE 518 /* open found a stale dentry */
26 #define EBADCOOKIE 523 /* Cookie is stale */
|
| /linux/arch/powerpc/mm/nohash/ |
| H A D | mmu_context.c | 18 * - Implement flush_tlb_mm() by making the context stale and picking
20 * - More aggressively clear stale map bits and maybe find some way to
94 * the stale map as we can just flush the local CPU
122 /* Mark it stale on all CPUs that used this mm. For threaded in steal_context_smp()
280 /* If that context got marked stale on this CPU, then flush the in switch_mmu_context()
|
| /linux/scripts/package/ |
| H A D | buildtar | 58 # outcome because it may pick up a stale file remaining in the build tree.
69 # Please note the following code may copy a stale file.
91 # Please note the following code may copy a stale file.
|
| /linux/tools/testing/selftests/net/netfilter/ |
| H A D | conntrack_sctp_collision.sh | 5 # 1. Stale INIT_ACK capture:
14 # 2. Stale INIT capture:
91 "Expect to drop stale INIT/INIT_ACK chunks"
|
| /linux/drivers/dax/ |
| H A D | fsdev.c | 137 * module is unloaded -- dispatch through a stale or freed in fsdev_clear_pgmap_ops()
184 * Clear any stale folio state from pages in the given range.
192 * are stale and must be cleared along with compound state.
324 /* Drop fsdev's pgmap->ops/owner on unbind so no stale ops survive. */ in fsdev_dax_probe()
330 * Clear any stale compound folio state left over from a previous in fsdev_dax_probe()
|
| /linux/drivers/gpu/drm/i915/gem/selftests/ |
| H A D | mock_context.c | 31 spin_lock_init(&ctx->stale.lock); in mock_context()
32 INIT_LIST_HEAD(&ctx->stale.engines); in mock_context()
|
| /linux/net/mptcp/ |
| H A D | pm.c | 268 struct mptcp_subflow_context *subflow, *stale = NULL, *same_id = NULL; in mptcp_pm_addr_send_ack_avoid_list() local
281 if (unlikely(subflow->stale)) { in mptcp_pm_addr_send_ack_avoid_list()
282 if (!stale) in mptcp_pm_addr_send_ack_avoid_list()
283 stale = subflow; in mptcp_pm_addr_send_ack_avoid_list()
295 else if (stale) in mptcp_pm_addr_send_ack_avoid_list()
296 subflow = stale; in mptcp_pm_addr_send_ack_avoid_list()
359 if (unlikely(subflow->stale)) { in mptcp_adjust_add_addr_timeout()
1080 if (subflow->stale || !stale_loss_cnt || subflow->stale_count <= stale_loss_cnt) in mptcp_pm_subflow_chk_stale()
1091 subflow->stale = 1; in mptcp_pm_subflow_chk_stale()
1118 /* No data outstanding at TCP level? not stale */ in mptcp_pm_chk_stale()
|
| /linux/net/netfilter/ipvs/ |
| H A D | ip_vs_lblc.c | 12 * collect stale entries of 24+ hours when
55 * It is for garbage collection of stale IPVS lblc entries,
66 * in a half hour, do a full expiration check to collect stale
103 struct timer_list periodic_timer; /* collect stale entries */
283 * It is used to collect stale entries when the number of entries
|
| /linux/tools/testing/selftests/filesystems/fuse/ |
| H A D | fuse_acl_cache_test.c | 13 * the ACL leaves the stale kernel entry in place, and the next getxattr returns
31 * Buggy: cache hit, returns stale ACL_A (28 bytes). FAIL.
325 * Buggy kernel: cache hit -> stale ACL_A (28 bytes), count stays 3. in TEST_F()
336 TH_LOG("step 5 BUG: stale ACL_A (%zd bytes) from kernel cache " in TEST_F()
|
| /linux/drivers/virt/coco/guest/ |
| H A D | tsm-mr.c | 30 * will be called only when a live MR is being read and the cache is stale
61 * @ctx->in_sync indicates if the MR cache is stale. It is a global in tm_digest_read()
112 /* mark MR cache stale */ in tm_digest_write()
|
| /linux/Documentation/arch/x86/ |
| H A D | mds.rst | 29 deallocated it can retain the stale data of the preceding operations which
37 contain stale data from a previous operation which can be forwarded to
193 that stale data from the idling CPU from spilling to the Hyper-Thread
|
| /linux/drivers/net/ethernet/intel/ice/ |
| H A D | ice_ptp.c | 483 if (test_and_clear_bit(idx, tx->stale)) in ice_ptp_complete_tx_single_tstamp()
517 * 1) check that the timestamp request is not stale
521 * 5) check if the timestamp is stale, and discard if so
545 * this case, software will set the stale bit for any outstanding Tx
644 if (test_and_clear_bit(idx, tx->stale)) in ice_ptp_process_tx_tstamp()
702 unsigned long *in_use, *stale; in ice_ptp_alloc_tx_tracker() local
707 stale = bitmap_zalloc(tx->len, GFP_KERNEL); in ice_ptp_alloc_tx_tracker()
709 if (!tstamps || !in_use || !stale) { in ice_ptp_alloc_tx_tracker()
712 bitmap_free(stale); in ice_ptp_alloc_tx_tracker()
719 tx->stale = stale; in ice_ptp_alloc_tx_tracker()
[all …]
|
| /linux/arch/arc/include/asm/ |
| H A D | mmu_context.h | 141 * it to CPUs where a task once ran-on, could cause stale TLB entry in switch_mm()
146 * via any existing stale TLB entries. in switch_mm()
|
| /linux/Documentation/scsi/ |
| H A D | hpsa.rst | 80 and not be disrupted in any way by stale commands or other stale state
|