llvm.org GIT mirror llvm / daec1e0
[llvm-mca][docs] clarify how the quality of the perf report is affected by the quality of the scheduling models. Differential Revision: https://reviews.llvm.org/D63556 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@363830 91177308-0d34-0410-b5e6-96231b3b80d8 Andrea Di Biagio 27 days ago
1 changed file(s) with 11 addition(s) and 0 deletion(s). Raw diff Collapse all Expand all
3636 .. code-block:: bash
3737
3838 $ clang foo.c -O2 -target x86_64-unknown-unknown -mllvm -x86-asm-syntax=intel -S -o - | llvm-mca -mcpu=btver2
39
40 Scheduling models are not just used to compute instruction latencies and
41 throughput, but also to understand what processor resources are available
42 and how to simulate them.
43
44 By design, the quality of the analysis conducted by :program:`llvm-mca` is
45 inevitably affected by the quality of the scheduling models in LLVM.
46
47 If you see that the performance report is not accurate for a processor,
48 please `file a bug `_
49 against the appropriate backend.
3950
4051 OPTIONS
4152 -------