(very) Brief overview of tnl module
[mesa.git] / src / mesa / tnl / NOTES
1 INTRODUCTION
2
3 A generic, configurable software implementation of GL transformation &
4 lighting.
5
6 This module provides an implementation of the routines required by the
7 'vtxfmt' mechanism of core mesa for tnl functionality in all
8 combinations of compile and execute modes.
9
10 Most current drivers use the tnl module exclusively to provide this
11 functionality, though there is an experimental alternate
12 implementation provided by the tnl_dd/t_dd_imm_* files which can
13 handle a small subset of GL states in execute mode only.
14
15
16 STATE
17
18 To create and destroy the module:
19
20 GLboolean _tnl_CreateContext( GLcontext *ctx );
21 void _tnl_DestroyContext( GLcontext *ctx );
22
23 The module is not active by default, and must be installed by calling
24 _tnl_Wakeup(). This function installs internal tnl functions into all
25 the vtxfmt dispatch hooks, thus taking over the task of transformation
26 and lighting entirely:
27
28 void _tnl_wakeup_exec( GLcontext *ctx );
29 void _tnl_wakeup_save_exec( GLcontext *ctx );
30
31
32 This module tracks state changes internally and maintains derived
33 values based on the current state. For this to work, the driver
34 ensure the following funciton is called whenever the state changes and
35 the swsetup module is 'awake':
36
37 void _tnl_InvalidateState( GLcontext *ctx, GLuint new_state );
38
39 There is no explicit call to put the tnl module to sleep. Simply
40 install other function pointers into all the vtxfmt dispatch slots,
41 and (optionally) cease calling _tnl_InvalidateState().
42
43 CUSTOMIZATION
44
45 The module provides customizability through several mechanisms. The
46 most important is by allowing drivers to specify the pipeline through
47 which vertex data is passed, including its eventual transfer to
48 rasterization hardware (or software).
49
50 The default pipeline is specified in t_pipeline.c, and is usually a
51 starting point for driver pipelines. Some drivers will remove a stage
52 where hardware provides support for the implemented operation (for
53 instance fog where per-pixel hardware fog is available, as in the dri
54 tdfx driver), or add stages to shortcircuit latter operations (for
55 example taking advantage of hardware support for strips and other
56 higher-level primitives (for example the radeon driver).
57
58 In addition, the following functions provide further tweaks:
59
60 extern void
61 _tnl_need_projected_coords( GLcontext *ctx, GLboolean flag );
62
63 - Direct the default vertex transformation stage to
64 produce/not produce projected clip coordinates.
65
66 extern void
67 _tnl_need_dlist_loopback( GLcontext *ctx, GLboolean flag );
68
69 - Direct the display list component of the tnl module to
70 replay display lists as 'glVertex' type calls, rather than
71 passing the display list data directly into the tnl pipeline
72 mechanism.
73
74 This allows display lists to be replayed by the tnl module
75 even when the module is not strictly active.
76
77
78 extern void
79 _tnl_need_dlist_norm_lengths( GLcontext *ctx, GLboolean flag );
80
81 - Direct the display list component to enable/disable caching
82 1/length values for display list normals. Doing so is
83 ususally helpful when lighting is performed in software, but
84 wasteful otherwise.
85
86
87 DRIVER INTERFACE
88
89 The module itself offers a minimal driver interface:
90
91 void (*RunPipeline)( GLcontext *ctx );
92
93 Normally this is set to _tnl_RunPipeline(), however the driver can use
94 this hook to wrap checks or other code around this call.
95
96 In addition, the driver interface for the default render pipeline
97 stage is housed in the tnl context struct (this could be cleaner).
98
99
100 RENDER DRIVER INTERFACE
101
102 See t_context.h for the definition and explanation of this.