| Commit message (Expand) | Author | Age | Files | Lines |
* | Introduce bus_dmamap_load_mbuf_sg(). Instead of taking a callback arg, this | scottl | 2005-01-07 | 1 | -13/+48 |
* | Begin all license/copyright comments with /*- | imp | 2005-01-05 | 1 | -1/+1 |
* | Don't flag alignment constraints as a reason for bouncing. This fixes the | scottl | 2004-11-29 | 1 | -1/+1 |
* | Remove an extra #include | scottl | 2004-11-21 | 1 | -1/+0 |
* | Consolidate all of the bounce tests into the BUS_DMA_COULD_BOUNCE flag. | scottl | 2004-11-21 | 1 | -29/+45 |
* | Revert part of rev 1.56. Tag boundaries are handled by splitting segments, | scottl | 2004-11-19 | 1 | -9/+5 |
* | MFi386 rev 1.63-1.64: | scottl | 2004-11-10 | 1 | -37/+142 |
* | MFi386: | peter | 2004-11-05 | 1 | -28/+79 |
* | Fix a problem with tag->boundary inheritence that has existed since day one | scottl | 2004-09-08 | 1 | -5/+5 |
* | Sync with i386 - dont count needed bounce pages if loading a buffer that | peter | 2004-08-16 | 1 | -1/+1 |
* | MFi386: 1.57: transparently respect alignment/boundary tags | peter | 2004-07-08 | 1 | -11/+47 |
* | Now that contigfree() does not require Giant, don't grab it in busdma. | scottl | 2004-03-13 | 1 | -2/+0 |
* | MFi386 rev 1.51 by scottl: make dflt_lock() always panic. | peter | 2003-09-22 | 1 | -4/+0 |
* | MFi386 rev 1.53 by scottl: Allocate the S/G list in the tag, not on | peter | 2003-09-22 | 1 | -27/+32 |
* | MFi386 rev 1.55 by sam: remove unused #define BUS_DMAMAP_NSEGS | peter | 2003-09-22 | 1 | -2/+0 |
* | In _bus_dmamap_load_buffer(), only count the number of bounce pages needed if | scottl | 2003-08-04 | 1 | -1/+2 |
* | - Introduce a new busdma flag BUS_DMA_ZERO to request for zero'ed | mux | 2003-07-27 | 1 | -8/+11 |
* | Sync more things with other backends. | mux | 2003-07-01 | 1 | -12/+10 |
* | Honor the boundary of the busdma tag when allocating bounce pages. | mux | 2003-07-01 | 1 | -1/+1 |
* | Mega busdma API commit. | scottl | 2003-07-01 | 1 | -6/+55 |
* | Catch amd64 up with the pending busdma async callback locking. Though this | scottl | 2003-06-28 | 1 | -0/+5 |
* | Bring back bus_dmasync_op_t. It is now a typedef to an int, though the | scottl | 2003-05-27 | 1 | -1/+1 |
* | De-orbit bus_dmamem_alloc_size(). It's a hack and was never used anyways. | scottl | 2003-05-26 | 1 | -26/+9 |
* | style(9) | mux | 2003-04-15 | 1 | -2/+2 |
* | Restore delayed load support for the resource shortage case. | simokawa | 2003-04-14 | 1 | -3/+27 |
* | * Use _bus_dmamap_load_buffer() and respect maxsegsz in bus_dmamap_load(). | simokawa | 2003-04-14 | 1 | -119/+30 |
* | Change the operation parameter of bus_dmamap_sync() from an | mux | 2003-04-10 | 1 | -11/+4 |
* | Add support for bounce buffers to _bus_dmamap_load_buffer, which is the | jake | 2003-04-07 | 1 | -20/+63 |
* | - Add vm_paddr_t, a physical address type. This is required for systems | jake | 2003-03-25 | 1 | -13/+17 |
* | Use atomic operations to increment and decrement the refcount | mux | 2003-03-20 | 1 | -4/+4 |
* | - Lock down the bounce pages structures. We use the same locking scheme | mux | 2003-03-17 | 1 | -26/+34 |
* | Grab Giant around calls to contigmalloc() and contigfree() so | mux | 2003-03-13 | 1 | -1/+8 |
* | Correctly set BUS_SPACE_MAXSIZE in all the busdma backends. | mux | 2003-02-26 | 1 | -1/+1 |
* | Back out M_* changes, per decision of the TRB. | imp | 2003-02-19 | 1 | -2/+2 |
* | Fix a problem in bus_dmamap_load_{mbuf,uio} when the first mbuf or the first | harti | 2003-02-04 | 1 | -11/+16 |
* | Consolidate MIN/MAX macros into one place (param.h). | alfred | 2003-02-02 | 1 | -2/+0 |
* | Implement bus_dmamem_alloc_size() and bus_dmamem_free_size() as | scottl | 2003-01-29 | 1 | -8/+26 |
* | Remove M_TRYWAIT/M_WAITOK/M_WAIT. Callers should use 0. | alfred | 2003-01-21 | 1 | -2/+2 |
* | New bus_dma interfaces for use by crypto device drivers: | sam | 2002-10-04 | 1 | -0/+206 |
* | Don't use "NULL" when "0" is really meant. | archie | 2002-08-21 | 1 | -1/+1 |
* | Clean up: | alfred | 2002-04-19 | 1 | -4/+10 |
* | Change the preemption code for software interrupt thread schedules and | jhb | 2002-01-05 | 1 | -1/+1 |
* | Teach bus_dmamem_free() about contigfree(). This is a bit of a hack, | wpaul | 2001-08-15 | 1 | -0/+3 |
* | With Alfred's permission, remove vm_mtx in favor of a fine-grained approach | dillon | 2001-07-04 | 1 | -0/+1 |
* | Introduce a global lock for the vm subsystem (vm_mtx). | alfred | 2001-05-19 | 1 | -0/+2 |
* | - Catch up to the new swi API changes: | jhb | 2001-02-09 | 1 | -1/+1 |
* | Convert more malloc+bzero to malloc+M_ZERO. | dwmalone | 2000-12-08 | 1 | -9/+7 |
* | - Overhaul the software interrupt code to use interrupt threads for each | jhb | 2000-10-25 | 1 | -1/+3 |
* | Back out the previous change to the queue(3) interface. | jake | 2000-05-26 | 1 | -5/+5 |
* | Change the way that the queue(3) structures are declared; don't assume that | jake | 2000-05-23 | 1 | -5/+5 |