Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
llvm-epi-0.8
Manage
Activity
Members
Code
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Deploy
Releases
Model registry
Analyze
Contributor analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
This is an archived project. Repository and other project resources are read-only.
Show more breadcrumbs
Roger Ferrer
llvm-epi-0.8
Commits
4acfad7c
Commit
4acfad7c
authored
14 years ago
by
Anton Korobeynikov
Browse files
Options
Downloads
Patches
Plain Diff
Add some A8-based approximation for instructions with unknown cycle times
llvm-svn: 100669
parent
4fb6a66c
No related branches found
Branches containing commit
No related tags found
No related merge requests found
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
llvm/lib/Target/ARM/ARMScheduleV7.td
+52
-0
52 additions, 0 deletions
llvm/lib/Target/ARM/ARMScheduleV7.td
with
52 additions
and
0 deletions
llvm/lib/Target/ARM/ARMScheduleV7.td
+
52
−
0
View file @
4acfad7c
...
...
@@ -1025,6 +1025,58 @@ def CortexA9Itineraries : ProcessorItineraries<[
InstrStage<1, [FU_Pipe0, FU_Pipe1]>,
InstrStage<1, [FU_NPipe]>], [3]>,
//
// Double-register Permute Move
InstrItinData<IIC_VMOVD, [InstrStage<1, [FU_DRegsN], 0, Required>,
// FIXME: all latencies are arbitrary, no information is available
InstrStage<3, [FU_DRegsVFP], 0, Reserved>,
InstrStage<1, [FU_Pipe0, FU_Pipe1]>,
InstrStage<1, [FU_NLSPipe]>], [2, 1]>,
//
// Quad-register Permute Move
// Result written in N2, but that is relative to the last cycle of multicycle,
// so we use 3 for those cases
InstrItinData<IIC_VMOVQ, [InstrStage<1, [FU_DRegsN], 0, Required>,
// FIXME: all latencies are arbitrary, no information is available
InstrStage<4, [FU_DRegsVFP], 0, Reserved>,
InstrStage<1, [FU_Pipe0, FU_Pipe1]>,
InstrStage<2, [FU_NPipe]>], [3, 1]>,
//
// Integer to Single-precision Move
InstrItinData<IIC_VMOVIS , [InstrStage<1, [FU_DRegsN], 0, Required>,
// FIXME: all latencies are arbitrary, no information is available
InstrStage<3, [FU_DRegsVFP], 0, Reserved>,
InstrStage<1, [FU_Pipe0, FU_Pipe1]>,
InstrStage<1, [FU_NPipe]>], [2, 1]>,
//
// Integer to Double-precision Move
InstrItinData<IIC_VMOVID , [InstrStage<1, [FU_DRegsN], 0, Required>,
// FIXME: all latencies are arbitrary, no information is available
InstrStage<3, [FU_DRegsVFP], 0, Reserved>,
InstrStage<1, [FU_Pipe0, FU_Pipe1]>,
InstrStage<1, [FU_NPipe]>], [2, 1, 1]>,
//
// Single-precision to Integer Move
InstrItinData<IIC_VMOVSI , [InstrStage<1, [FU_DRegsN], 0, Required>,
// FIXME: all latencies are arbitrary, no information is available
InstrStage<3, [FU_DRegsVFP], 0, Reserved>,
InstrStage<1, [FU_Pipe0, FU_Pipe1]>,
InstrStage<1, [FU_NPipe]>], [2, 1]>,
//
// Double-precision to Integer Move
InstrItinData<IIC_VMOVDI , [InstrStage<1, [FU_DRegsN], 0, Required>,
// FIXME: all latencies are arbitrary, no information is available
InstrStage<3, [FU_DRegsVFP], 0, Reserved>,
InstrStage<1, [FU_Pipe0, FU_Pipe1]>,
InstrStage<1, [FU_NPipe]>], [2, 2, 1]>,
//
// Integer to Lane Move
InstrItinData<IIC_VMOVISL , [InstrStage<1, [FU_DRegsN], 0, Required>,
// FIXME: all latencies are arbitrary, no information is available
InstrStage<4, [FU_DRegsVFP], 0, Reserved>,
InstrStage<1, [FU_Pipe0, FU_Pipe1]>,
InstrStage<2, [FU_NPipe]>], [3, 1, 1]>,
//
// Double-register FP Unary
InstrItinData<IIC_VUNAD, [InstrStage<1, [FU_DRegsN], 0, Required>,
// Extra latency cycles since wbck is 6 cycles
...
...
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment