Home
last modified time | relevance | path

Searched refs:batched (Results 1 – 15 of 15) sorted by relevance

/linux/drivers/net/ethernet/intel/i40e/
H A Di40e_xsk.c553 u32 batched, leftover, i; in i40e_fill_tx_hw_ring() local
555 batched = nb_pkts & ~(PKTS_PER_BATCH - 1); in i40e_fill_tx_hw_ring()
557 for (i = 0; i < batched; i += PKTS_PER_BATCH) in i40e_fill_tx_hw_ring()
559 for (i = batched; i < batched + leftover; i++) in i40e_fill_tx_hw_ring()
/linux/drivers/net/ethernet/intel/ice/
H A Dice_xsk.c879 u32 batched, leftover, i; in ice_fill_tx_hw_ring() local
881 batched = ALIGN_DOWN(nb_pkts, PKTS_PER_BATCH); in ice_fill_tx_hw_ring()
883 for (i = 0; i < batched; i += PKTS_PER_BATCH) in ice_fill_tx_hw_ring()
885 for (; i < batched + leftover; i++) in ice_fill_tx_hw_ring()
/linux/include/net/libeth/
H A Dxdp.h527 u32 this, batched, off = 0; in libeth_xdp_tx_xmit_bulk() local
544 batched = ALIGN_DOWN(this, LIBETH_XDP_TX_BATCH); in libeth_xdp_tx_xmit_bulk()
546 for ( ; i < off + batched; i += LIBETH_XDP_TX_BATCH) { in libeth_xdp_tx_xmit_bulk()
555 if (batched < this) { in libeth_xdp_tx_xmit_bulk()
/linux/Documentation/networking/
H A Dcan_ucan_protocol.rst186 ``ucan_message_in`` values. If multiple messages are batched in a USB
256 ucan_message_out`` values. If multiple messages are batched into one
H A Daf_xdp.rst753 Zero-copy drivers usually use the batched APIs for Rx and Tx
/linux/Documentation/ABI/testing/
H A Dsysfs-kernel-iommu_groups45 DMA-FQ As above, but using batched invalidation to lazily
H A Dsysfs-fs-f2fs87 Description: Controls the dirty page count condition for batched sequential
/linux/Documentation/mm/
H A Darch_pgtable_helpers.rst69 | ptep_get_and_clear_full | Clears and returns PTE (batched PTE unmap) |
/linux/Documentation/bpf/
H A Dmap_hash.rst138 undesirable. It is recommended to use batched lookup if there is going
H A Dringbuf.rst139 efficient batched notifications. Default self-balancing strategy, though,
/linux/Documentation/filesystems/
H A Djournalling.rst78 journal. Since transactions can't be nested/batched across differing
/linux/drivers/iommu/
H A DKconfig125 DMA-mapped pages, but with "lazy" batched TLB invalidation. This
/linux/include/linux/
H A Dmmzone.h556 int batched; member
/linux/mm/
H A Dvmscan.c3323 walk->batched++; in update_batch_size()
3335 walk->batched = 0; in reset_batch_size()
3785 if (need_resched() || walk->batched >= MAX_LRU_BATCH) { in walk_pud_range()
3832 if (walk->batched) { in walk_mm()
4778 if (walk && walk->batched) { in evict_folios()
/linux/Documentation/
H A Dmemory-barriers.txt2830 memory or I/O hardware that can do batched accesses of adjacent locations,