[llvm-bugs] [Bug 35784] New: [meta][x86] Scheduler Driven Optimizations
via llvm-bugs
llvm-bugs at lists.llvm.org
Sun Dec 31 10:21:46 PST 2017
https://bugs.llvm.org/show_bug.cgi?id=35784
Bug ID: 35784
Summary: [meta][x86] Scheduler Driven Optimizations
Product: libraries
Version: trunk
Hardware: PC
OS: Windows NT
Status: NEW
Severity: enhancement
Priority: P
Component: Backend: X86
Assignee: unassignedbugs at nondot.org
Reporter: llvm-dev at redking.me.uk
CC: andrea.dibiagio at gmail.com,
andrew.v.tischenko at gmail.com, craig.topper at gmail.com,
gadi.haber at intel.com, llvm-bugs at lists.llvm.org,
spatel+llvm at rotateright.com, zvi.rackover at intel.com
Blocks: 32325
Split off from Bug #26183.
With the scheduler models maturing we should be investigating ways to better
select between MC optimizations based on scheduler characteristics, register
pressure etc.
Bug #26183 mentioned several X86 slow/fast features that could be replaced with
scheduler costings:
FeatureSlowBTMem
FeatureSlowSHLD
FeatureSlowUAMem16
FeatureSlowUAMem32
FeatureSlowDivide32
FeatureSlowDivide64
Some of these affect (older) CPUs that currently don't have a accurate
scheduler models - taking numbers from Agner is probably enough to be useful?
Additionally, the use of when to combine multiple shuffles to a single variable
shuffle mask keeps coming up. As well as the actual instruction cost, the cost
of loading the mask needs to be accurately modelled: register pressure? can it
be hoisted? etc.
Referenced Bugs:
https://bugs.llvm.org/show_bug.cgi?id=32325
[Bug 32325] [META][X86] Improve implementation and use of X86 scheduler models
--
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20171231/1e7aa4a3/attachment.html>
More information about the llvm-bugs
mailing list