Intel's Newest Software Effort For Achieving Greater Performance: Thin Layout Optimizer

[phoronix]

Phoronix: Intel's Newest Software Effort For Achieving Greater Performance: Thin Layout Optimizer

Intel's software team is today sharing their newest innovation for achieving greater performance on Linux systems: the Thin Layout Optimizer. Intel's Thin Layout Optimizer is inspired by the likes of the Meta/LLVM BOLT optimizer and Google's Propeller but aims to be much easier to use while still delivering measurable performance gains for optimized binaries...

https://www.phoronix.com/news/Intel-Thin-Layout-Optimizer

[JEBjames]

Michael

Typo

"Binary Optimization Laoyout Tool" should be "layout"

[Kjell]

Benchmarks?

[V1tol]

Meanwhile mainstream distros struggle to provide x86-64-v3 and LTO builds -_-

[kylew77]

I wonder how this affects security? Modern binaries have some randomness in their layout to make it harder to exploit them if bugged. Would this put code into a predictable pattern then?

[dragorth]

I wonder how this affects security? Modern binaries have some randomness in their layout to make it harder to exploit them if bugged. Would this put code into a predictable pattern then?

Does that happen at the linker or loader level. I thought code was loaded by the linked parts, just moving that code around in the file isn't changing that.

[arjan_intel]

I wonder how this affects security? Modern binaries have some randomness in their layout to make it harder to exploit them if bugged. Would this put code into a predictable pattern then?

The randomization you talk about is on a whole library level ... this is about the order inside the library and does not interfere with the ASLR randomization of the whole library .. it coexists completely

[vancha]

Those are such good looking machines...

[joebonrichie]

That workflow looks significantly easier than BOLT, although you basically have to compile your own binutils unless they get those gnu linker patches upstreamed. So it probably evens out.

A bummer for me personally is that AMD CPUs do not support LBR (perf -j flag) until zen3 AFAIK. I personally have a zen2 CPU so I have to rely on

llvm-bolt -instrument

to experiment with post-link optimizers. LBR is also unavailable in some environments such as VMs AFAIU.

[ms178]

arjan_intel Some blog posts with how-to's for normies like me to use that tool to optimize the Linux Kernel or Mesa would be highly appreciated.