Matthew Gretton-Dann

icon indicating a website link https://www.gretton-dann.org.uk/

icon company @ARM-Software icon location Currently Software Architect investigating deploying Arm enhancements to new devices The opinions expressed are my own views and not those of my employer.

Results 23 comments of Matthew Gretton-Dann

Use of `std::thread::hardware_concurrency()` does not take into account it may return `0` as the number of threads.

Looking at https://github.com/llvm/llvm-project/blob/main/libcxx/src/thread.cpp this may affect any platform which uses libc++ except for Windows or POSIX.

Add -v, -h command line options and handle 'unsigning' already unsigned binaries.

> Thanks for this contrib. I'll merge this in another branch because it breaks fat binary support. If PPC, i386, and ARM support return, then it would be a candidate...

brew install on mac

From a technical perspective the cask is ready for publishing from our own repo - @sj was handling the other parts of the process. Initially we'll have our own cask...

add individual state level admin files

> @matt-gretton-dann - this PR fell asleep - I've moved the USA files back into the top level. > > @dlaydon - I can't remember whether `[Population size]` for these...

add individual state level admin files

I have retargeted this to mainline as well (it was targeted at one of my own branches).

More robust parameter file handling

Things I would like to see in this PR: * Detection and warning of when we do have multiple values for a field * Detection and warning of when a...

Regression Test Heap Optimization: Remove Extinct Result Arrays

> This led me down a rabbit hole, but I think your past theory about all of the fprintf() calls being responsible for the slowdown are probably correct. I have...

Checksum fails on Cori. (Haswell/KNL DOE supercomputer)

So I agree the checksum regression test is not a perfect solution. As another example of the issue you highlight, we deliberately run it single-threaded so we don't get variance...

Memory leak in SetupModel.cpp

Thank you for your report. I agree there is an issue here. We will investigate improving this, and making the memory flow more obvious.

GNUmake based build.

Thank you for this. I will investigate whether we can use at least parts of this to configure the compiler, and improve the CMakeLists.txt file. However, I am cautious about...