| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Use "memory" on inline fence instructions to suggest to the compiler |   jsg | 2020-10-28 | 1 | -4/+4 |
| * | Re-arrange struct bus_dmamap members |   sf | 2017-06-04 | 1 | -2/+2 |
| * | add a BUS_DMA_64BIT flag to bus_dma on all our archs. |   dlg | 2017-05-08 | 1 | -1/+2 |
| * | Add bus_dmamem_alloc_range(9) to allow drivers to allocate DMA'able memory |   kettenis | 2015-01-24 | 1 | -2/+8 |
| * | It's been a quarter century: we can assume volatile is present with that name. |   guenther | 2014-03-29 | 1 | -4/+4 |
| * | Remove the scatter/gather dma implementation as it is no longer used. | kettenis | 2013-12-12 | 1 | -70/+1 |
| * | Add an interface to rebind AGP DMA mappings. To be used by the upcoming KMS | kettenis | 2013-03-17 | 1 | -1/+2 |
| * | the bus_space api implies that there's a bus endianness that it will | dlg | 2013-01-29 | 1 | -1/+12 |
| * | twist the amd64 bus_space implementation a bit. instead of using the tag | dlg | 2013-01-18 | 1 | -126/+223 |
| * | implement bus_space_foo_8 for amd64. it will panic on attempts to do 8 | dlg | 2013-01-15 | 1 | -61/+27 |
| * | Normalize sentinel. Use _MACHINE_*_H_ and _<ARCH>_*_H_ properly and consitently. |   pirofti | 2011-03-23 | 1 | -4/+4 |
| * | Make sure bus_dmamap_sync() always involves a function call, to prevent the | kettenis | 2010-09-06 | 1 | -3/+2 |
| * | On amd64, move the bus_dma buftype stuff that is only used by sg_dma |   oga | 2010-04-08 | 1 | -6/+27 |
| * | OpenBSD uses bus_space_copy_#, not bus_space_copy_region_#. Make sure only the |   miod | 2010-04-04 | 1 | -9/+4 |
| * | Get rid of the obsolet BUS_BARRIER_xxx constants for bus_space_barrier(), only | miod | 2009-07-30 | 1 | -4/+1 |
| * | Make sure all platforms understand the flags argument of bus_space_map() and | miod | 2009-07-26 | 1 | -2/+2 |
| * | on amd64, store a pointer to the loaded data and the type. This will | oga | 2009-06-07 | 1 | -1/+13 |
| * | When I changed the implementation of iommu and sg_dma over to use | oga | 2009-05-10 | 1 | -3/+3 |
| * | add a sg_dma backend for amd64 bus_dma. This is a lot more clever about | oga | 2009-04-21 | 1 | -1/+46 |
| * | Add a BUS_DMA_ZERO flag for bus_dmamem_alloc() to return zeroed memory. | oga | 2009-04-20 | 1 | -13/+14 |
| * | remove the _BUS_DMA_PRIVATE define from amd64 and i386. | oga | 2009-03-10 | 1 | -5/+1 |
| * | Mirroring the i386 commit just made. Add MD PMAP_NOCACHE flag to pmap, | oga | 2009-02-05 | 1 | -1/+2 |
| * | Remove the x86 and i386 prefixes to the bus_dma types. It's really quite | oga | 2008-12-03 | 1 | -9/+9 |
| * | same change as for amd64. Move bus_space_* functions from inlines and | oga | 2008-11-22 | 1 | -5/+1 |
| * | de-inline the bus_space_* functions in favour of real functions | oga | 2008-11-21 | 1 | -551/+63 |
| * | First pass at removing clauses 3 and 4 from NetBSD licenses. |   ray | 2008-06-26 | 1 | -8/+1 |
| * | All your memory belong to us. This has been in snaps for a while, |   weingart | 2008-06-10 | 1 | -2/+2 |
| * | fix the bus_space #define nightmare, so that amd64 and i386 are much more |   deraadt | 2007-11-16 | 1 | -62/+27 |
| * | Remove stale comments; prompted by millert. |   marco | 2007-09-01 | 1 | -2/+1 |
| * | Add vaddr. | marco | 2007-08-31 | 1 | -1/+14 |
| * | from PAE work: |   mickey | 2006-04-27 | 1 | -2/+2 |
| * | Add bus_space_alloc and _free. | marco | 2005-11-05 | 1 | -32/+11 |
| * | comment typo |   jason | 2005-05-25 | 1 | -2/+2 |
| * | provide _barrier method w/ {l,s,m}fence instructions | mickey | 2004-02-23 | 1 | -10/+13 |
| * | an amd64 arch support. | mickey | 2004-01-28 | 1 | -0/+1158 |
