Age | Commit message (Expand) | Author | Files | Lines |
2003-04-10 | Add the PCI relevent part of the new DRM API patch, i.e, no AGP/scattermach64-0-0-5-branch | Jose Fonseca | 5 | -2/+274 |
2003-01-24 | Add scissor regs to dump_engine_info. | Leif Delgass | 2 | -0/+13 |
2002-12-06 | DRM Makefile dependency fix (Allen Barnett) | Leif Delgass | 1 | -1/+1 |
2002-11-10 | Verify client buffer size. Use unsigned format in debug output. | Leif Delgass | 1 | -3/+8 |
2002-11-10 | Update the vertex buffer IOCTL interface. | Jose Fonseca | 2 | -34/+9 |
2002-11-10 | Discard entire buffer if verify fails -- eliminates possible lockups with | Leif Delgass | 1 | -32/+43 |
2002-11-10 | Use userspace memory for the vertex buffers - DRM bits. Note that the old | Jose Fonseca | 1 | -82/+35 |
2002-10-31 | Revamped copy_and_verify to use linux user memory access routines to cope | Jose Fonseca | 1 | -29/+36 |
2002-10-30 | Verify against exact setup engine register address ranges, as I've | Leif Delgass | 1 | -9/+3 |
2002-10-30 | Fix reg. range verify. SETUP_CNTL (1_C1) should only be written by the drm | Leif Delgass | 1 | -1/+1 |
2002-10-29 | Support (disabled by default for now) for copying and verifying the client | Jose Fonseca | 1 | -2/+107 |
2002-10-20 | Changes to the output macros to eliminate gcc 3.2 warnings regarding the | Jose Fonseca | 1 | -5/+6 |
2002-09-25 | Fix bug in mach64_find_pending_buf_entry. Need another level of indirection | Leif Delgass | 1 | -7/+12 |
2002-09-07 | Endianess fix for textures with mmio mode | Leif Delgass | 1 | -4/+1 |
2002-09-04 | Frame-rate throttling - new GETPARAM ioctl, green performance box for waits | Leif Delgass | 5 | -722/+590 |
2002-08-23 | add clearbit from linux asm for alpha to mach64_clear_dma_eol. | Leif Delgass | 1 | -0/+16 |
2002-08-22 | add mach64 to Makefile.kernel, merge in some drm build fixes from the | Leif Delgass | 8 | -17/+31 |
2002-08-18 | Revert bus master test to using pattern registers. On some GTPros, the test | Leif Delgass | 1 | -33/+52 |
2002-08-16 | Print error messages before cleanup. | Leif Delgass | 1 | -10/+10 |
2002-08-12 | change DRM_INFO -> DRM_ERROR on failure to set pci dma mask. | Leif Delgass | 1 | -1/+1 |
2002-08-12 | Use pci_alloc_consistent instead of pci_pool_alloc for descriptor ring and | Leif Delgass | 2 | -45/+31 |
2002-08-09 | Don't debug print the entire DMA test buffer. | Leif Delgass | 1 | -1/+5 |
2002-08-09 | Add author credits for Jose and myself to various source files, GL vendor | Leif Delgass | 1 | -1/+1 |
2002-07-28 | Use unsigned long for handles passed to DRM as other drivers do. | Leif Delgass | 2 | -9/+9 |
2002-07-14 | Remove the duplicate GUI_ACTIVE check (spotted by Leif) | Jose Fonseca | 1 | -6/+5 |
2002-07-14 | Changed mach64_ring_idle to mesure the time for the ring head advances | Jose Fonseca | 1 | -5/+16 |
2002-07-12 | - Subpixel offsets for viewport coordinates -- fixes almost all glean | Leif Delgass | 2 | -4/+21 |
2002-07-10 | DRM portion of commit for cliprects with vertex buffers | Leif Delgass | 3 | -82/+222 |
2002-07-08 | Don't emit setup for swaps if there are no cliprects. | Leif Delgass | 1 | -0/+3 |
2002-07-08 | Don't send empty buffer in _dispatch_clear if there are no cliprects. | Leif Delgass | 1 | -1/+4 |
2002-07-06 | use MACH64_INTERRUPTS define in private struct. Code formatting cleanup. | Leif Delgass | 1 | -13/+30 |
2002-07-06 | Disable unimplemented MMIO mode (this should be implemented or the option | Leif Delgass | 2 | -19/+51 |
2002-07-06 | Commit several small changes to the DRM: | Jose Fonseca | 2 | -60/+64 |
2002-06-27 | Merged in DRI trunk from 2002-06-26 (incl. Mesa 4.0.3 and drmCommand). | Leif Delgass | 37 | -906/+1264 |
2002-06-26 | Less verbose ring and buffer dump routines in the DRM to avoid overflowmach64-0-0-4-branch | Jose Fonseca | 1 | -27/+53 |
2002-06-21 | Fixed the endian problems in mach64_clear_bit in non-x86 architectures and | Jose Fonseca | 1 | -28/+15 |
2002-06-19 | Use of atomic bit ops in ADVANCE_RING() to lock the BUS when writing to the | Jose Fonseca | 2 | -27/+57 |
2002-06-15 | Use of le32_to_cpu for the descriptor data and other small fixes in | Jose Fonseca | 1 | -11/+14 |
2002-06-13 | verify blit buffer size | Leif Delgass | 1 | -0/+4 |
2002-06-12 | Better structuring of UPDATE_RING_HEAD and ring related code. | Jose Fonseca | 2 | -104/+213 |
2002-06-11 | Fixed the bug which caused some lockups: UPDATE_RING_HEAD could (and was) | Jose Fonseca | 2 | -55/+44 |
2002-06-08 | First half of commit (DRM) for blits using BM_HOSTDATA. This elminates the | Leif Delgass | 3 | -10/+84 |
2002-06-08 | Added mach64_dump_ring - a function to dump the active portion of the ring | Jose Fonseca | 2 | -0/+74 |
2002-06-07 | Protect the GETBUFPTR and GETBUFADDR macros with parenthesis so that they | Jose Fonseca | 1 | -2/+2 |
2002-06-06 | Fix bug in UPDATE_RING_HEAD when calculating head_addr. Also some code | Leif Delgass | 2 | -62/+34 |
2002-06-06 | The bus master operation is now always checked on the wait loops, fixing | Jose Fonseca | 2 | -175/+69 |
2002-06-04 | Don't release all buffers in the pending list in do_dma_idle for the | Leif Delgass | 1 | -1/+3 |
2002-06-01 | Code clean up, added RING macros, initial try at emitting descriptors to | Leif Delgass | 4 | -307/+476 |
2002-05-29 | Do not add commands to the end of each buffer to disable busmastering, so | Jose Fonseca | 1 | -6/+8 |
2002-05-28 | Initialize the freelists before doing any check to the X server data, to | Jose Fonseca | 1 | -4/+5 |