| Age | Commit message (Expand) | Author |
| 2022-07-17 | agp(4): Use bus_space_barrier, not membar_producer. | riastradh |
| 2019-11-10 | in many device attach paths, allocate memory with M_WAITOK instead of M_NOWAIT | chs |
| 2018-08-27 | Restore gtt page table control register on resume. | riastradh |
| 2016-05-01 | kern/50453: Separate Pineview chipset from i915-family chipset support. | nonaka |
| 2015-10-21 | Fix typo | christos |
| 2015-10-21 | CID 1328433: Anding with 0 and comparing != 0 will not work very will. | christos |
| 2015-10-13 | Fix mapping Intel graphics device registers. | riastradh |
| 2015-04-05 | Don't use the video BIOS memory as an i915 flush page! | riastradh |
| 2015-03-06 | Pass cache-related flags through to the GTT on pre-SNB devices. | riastradh |
| 2015-02-26 | Include <sys/atomic.h> for membar_producer. | riastradh |
| 2015-02-26 | Issue a write barrier after updating the GTT. | riastradh |
| 2014-08-24 | Use the right BAR for MMIO register sizing on i810. | riastradh |
| 2014-08-22 | Revert to mapping the whole GTTMMADR region all at once for now. | riastradh |
| 2014-07-25 | Mark isc __diagused in agp_i810_unbind_memory. | riastradh |
| 2014-07-23 | Tweak style, make it compile...oops. | riastradh |
| 2014-07-23 | For the unbind code to work the bind code must set the offset. | riastradh |
| 2014-07-23 | Also unbind the right offset. | riastradh |
| 2014-07-23 | Bind i810 dcache pages at the requested offset, not at VA start. | riastradh |
| 2014-07-01 | Tweak debug printf directives. | riastradh |
| 2014-06-27 | Collect `#ifdef AGP_DEBUG printf(...)' into local DPRINTF macro. | riastradh |
| 2014-06-25 | No, that should be unsigned. | riastradh |
| 2014-06-25 | Fix integer type mismatch in debug printf. | riastradh |
| 2014-06-12 | Get GTT size from PGTBL_CTL on G4X chipsets. | riastradh |
| 2014-06-12 | Fix MMIO registers sizes for pre-Ironlake chipsets. | riastradh |
| 2014-06-12 | Extend agp_generic_bind_memory with bounds to check. | riastradh |
| 2014-06-12 | Fix type in cast too, after changing type of gtt_off. | riastradh |
| 2014-06-12 | Check bounds in agp_i810_borrow. | riastradh |
| 2014-06-12 | Use correct type for gtt_off, an offset relative to a bus address. | riastradh |
| 2014-06-12 | Ensure we map all the MMIO registers. | riastradh |
| 2014-06-12 | Giving a 0 gtt_size for G4X causes a panic when you try to bus map 0 size. | christos |
| 2014-06-11 | Fix fencepost. | riastradh |
| 2014-06-11 | Fix various mistakes in yesterday's commit. | riastradh |
| 2014-06-11 | Ignore result of agp_i810_unbind_page in agp_i810_unbind_memory. | riastradh |
| 2014-06-11 | Back out previous mistake. | riastradh |
| 2014-06-11 | Add missing bus_dmamap_destroy. | riastradh |
| 2014-06-11 | Add missing bus_dmamap_destroy. | riastradh |
| 2014-06-11 | Check bounds more carefully to avoid integer overflow. | riastradh |
| 2014-06-11 | Don't trust the MSAC register after all; use the GMADR BAR size. | riastradh |
| 2014-06-11 | Fix sense of conditional for GTTMMADR size. | riastradh |
| 2014-06-10 | Rework agp_i810 memory binding. | riastradh |
| 2014-06-10 | Another round of weed-whacking for agp_i810. | riastradh |
| 2014-05-28 | Cosmetic changes require testing too...oops. | riastradh |
| 2014-05-28 | Cosmetics: long lines, needless variables, PCI parameter verbiage. | riastradh |
| 2014-05-28 | Rationalize agp_i810 aprint output. Fix agp_i810_init error branches. | riastradh |
| 2014-05-28 | Oops -- gtt_off is relative to mmadr, not absolute. | riastradh |
| 2014-05-27 | Suppress uninitialized variable warnings in older and dumber GCC. | riastradh |
| 2014-05-27 | Rework agp_i810 attachment code a little. | riastradh |
| 2014-05-26 | Use correct address for 64-bit flush page config register. | riastradh |
| 2014-05-26 | Avoid xcall(9) while cold. | riastradh |
| 2014-05-24 | Handle 64-bit paddr registers delicately for non-PAE i386. | riastradh |