summaryrefslogtreecommitdiff
path: root/src/mesa/drivers/dri/radeon
AgeCommit message (Collapse)Author
2005-12-29rv370 not rv350Dave Airlie
2005-12-29typoDave Airlie
2005-12-29add some card type ioctl supportDave Airlie
2005-12-28add some r300 informationDave Airlie
2005-12-23fixup driver for some api changesDave Airlie
2005-12-05update r300 drm minimum to 20Dave Airlie
2005-12-02Added PCI_CHIP_RV410_5E4B (Adam Kirchhoff)Brian Paul
2005-11-29If the DDX driver didn't reserve any video RAM for textures, fall back toMichel Dänzer
using only GART memory for textures instead of segfaulting in the texture management code. (Bug 5115)
2005-11-28check for driver_modes == NULL. Don't need driver_modes for EGL driverBrian Paul
2005-11-28rearrange some code to put in more logical order, misc clean-upsBrian Paul
2005-11-24use new _egl_api structBrian Paul
2005-11-23fixed wrong return value in radeonShowSurfaceMESA()Brian Paul
2005-11-02First step of Radeon DRI unification:Eric Anholt
- Makes all three drivers use the same screen structure and setup code, with a few ifdefs for the separate compilation to deal with symbols not being available to all drivers and the fact that we have no mechanism for dealing with different config options for different chip families in the same driver. These issues should be dealt with later. - Introduces IS_R[123]00_CLASS(radeonScreenPtr) macro for code for taking different paths depending on the general class of chipset. - Adds many new R300-class PCI IDs, though not all those listed in radeon_driver.c.
2005-10-31fix problems found with gcc 2.96 (bug 4934)Brian Paul
2005-10-29Bug #4901: Correct secondary color emission when lighting is turned off, testedEric Anholt
with seccolor on rv200 and r200. Submitted by: sroland
2005-10-27Bug #4900: Fix the non-fogcoord fog test on r100 (and rv200) at tcl_mode=1 byEric Anholt
flipping the signs on two factors. While this contradicts the DDK, it matches what r200 does, and it fixes the test.
2005-10-19s/memcpy/_mesa_memcpy/, s/FREE/_mesa_free/Brian Paul
2005-10-18Add _mesa_exec_malloc() and _mesa_exec_free() for allocatingKeith Whitwell
executable memory. Based on Thomas Hellstrom's patch. TODO: glapi.c also needs this, but cannot access this code.
2005-10-18on r200, don't emit fog coords when fog isn't actually enabled. Fix bug with ↵Roland Scheidegger
vtxfmt not causing a fallback when fog coords are active.
2005-10-17add GL_EXT_fog_coord support to radeon driver. No vtxfmt code (just uses ↵Roland Scheidegger
fallback) yet. May cause a tcl fallback if fog coord is used together with separate specular lighting. Fog factors are precomputed and then submitted instead of the raw coords (chip limitation, same as on r200).
2005-10-16fix some obvious typos in radeon_reg.h (only currently unused bits affected)Roland Scheidegger
2005-10-15enable cube maps on radeon (#2241 on bugzilla). No vtxfmt code yet (just ↵Roland Scheidegger
generates vfmt fallback). Code by Andreas Stenglein, some small adjustments by me.
2005-10-13enable 3rd texture unit on radeon (default still 2 enabled units). Disable ↵Roland Scheidegger
some multitexcoord codegen stuff noone understands to make it work. Replace most code testing explicitly for unit 0 and 1 with loops instead of adding test for unit 2, smaller/more readable code at the cost of maybe some slight performance hit. (Code provided by Andreas Stenglein, some adjustments by me.)
2005-10-13make sure texture matrix is updated often enough with texture rectanglesRoland Scheidegger
2005-10-11Minor driver cleanup. Remove unnecessery/unneded ↵Roland Scheidegger
radeon/r200AllocDmaRegionVerts function. No longer compile the radeon_compat stuff which was for drm 1.1 compatibility presumably but unused since about forever into the driver.
2005-10-08remove code dealing with drm versions older than 1.3. The driver requires ↵Roland Scheidegger
(and requests) 1.3 since quite some time already to work at all.
2005-10-05fix typo in radeon_maos_arrays.c from last commit (code currently unused, ↵Roland Scheidegger
that path is compile-tested only).
2005-10-05remove unneeded parameter from radeonUploadTexMatrix.Roland Scheidegger
2005-10-05Remove the tcl fallback for texture rectangle (by manipulating the texture ↵Roland Scheidegger
matrix) (tested with texrect). Enable texgen for r/q coordinates (tested with projtex). Fix projected texcoords when an app uses TexCoord3x and the texture matrix to save on vertex size (fixes ut2k3 shadow projectors in tcl mode). From texgenmix, all cases with all texgen or no texgen work, with the exception of texgen enabled for s/t only, this one works with hw tcl, but not with vtxfmt (suspect issues with vtxfmt), the mixed cases do not work (which is expected, and should be rare in practice), with the exception of the first one which hits a tcl fallback.
2005-10-05In gl_texture_image replace IntFormat with InternalFormat and Format withBrian Paul
_BaseFormat to be consistant with gl_renderbuffer.
2005-09-26remove identity check for texgen planes (which was just plain wrong).Roland Scheidegger
2005-09-26fix projective texturing for (swtcl) texture rectangles. Fix wrong argument ↵Roland Scheidegger
order for texgen/texmat matrix multiplication.
2005-09-14updated commentBrian Paul
2005-09-14make some functions staticBrian Paul
2005-09-14Instead of calling _mesa_ResizeBuffersMESA() in the Viewport function,Brian Paul
call driUpdateFramebufferSize() when we detect changes in window size/ position in radeonGetLock().
2005-09-13Replace ctx->Driver.StencilOp/Func/Mask() functions withBrian Paul
ctx->Driver.Stencil*Separate() functions.
2005-09-13just some reformattingBrian Paul
2005-09-10add the new drm packets needed for ATI_fragment_shader to the sanity code of ↵Roland Scheidegger
the r200 and radeon driver
2005-09-04Simplify radeonSetCliprects(). Remove the 'mode' parameter.Brian Paul
2005-09-04Further clean-up and simplification of the span-related functions.Brian Paul
2005-09-04Add new void *addr and __DRIdrawablePrivate parameters toBrian Paul
driNewRenderbuffer().
2005-09-04Fix UpdatePageFlipping() function. Always call driFlipRenderbuffers()Brian Paul
with pfCurrentPage value, in case it's initially 1 instead of 0.
2005-09-04replace GET_SRC_PTR, GET_DST_PTR with GET_PTRBrian Paul
2005-09-03use new InitStencil/DepthPointer functionsBrian Paul
2005-09-02set depthHasSurface for stencil renderbufferBrian Paul
2005-09-01fix page flipping breakageBrian Paul
2005-09-01Finish up some of the gl_renderbuffer work.Brian Paul
Use driRenderbuffer's offset, pitch fields in the span routines. Remove the SetBuffer driver function. Consolidate the code for setting CTX_RB3D_COLOROFFSET and CTX_RB3D_COLORPITCH state in new radeonUpdateDrawBuffer() function.
2005-09-01s/__inline/INLINE/Brian Paul
2005-08-31remove redundant ctx->Driver fallback assignments that were already done in ↵Brian Paul
_mesa_init_driver_functions()
2005-08-25Attempt to fix the issue reported by Alan Grimes with DRM_RADEON_TEXTUREEric Anholt
erroring out when it shouldn't. The errno could be changed by usleep() between the ioctl call and the loop check, if a signal was received. This could cause an EAGAIN return from the DRM_RADEON_TEXTURE ioctl to not loop again. Instead of checking errno, check thevalue of ret itself, since it is a saved (and sign-flipped) copy of errno from the ioctl call.