David Spickett

> Right now, it exists in the unexplored space between "using Windows APIs" (fgets in this case being a wrapper around ReadConsoleInput) and "using VT". It's not something we're likely...

> Checks are already covered by single stage and 2 stage bots. What is the reason to not add this onto the 2 stage bots? I can think of a...

Looks ok to me, X86 and ARM already do this. > One point we noticed is that it could end up turning "aes+sha2" into "crypto" and "crypto" back into "sha2+aes+sha3+sm4",...

We already merge: ``` unclean unclean ``` And ``` clean clean ``` As the properties matched. Now we want to merge: ``` clean unclean ``` Assuming unclean merges into clean,...

I'm going to see if it's feasible to modify the request that we collapse into, to address Mehdi's question. And to be clear, Galina has the deciding approval here.

> I'm going to see if it's feasible to modify the request that we collapse into, to address Mehdi's question. This is the buildbot function that calls the callback I'm...

One sentence answer based on current buildbot behaviour: "The result of a request collapse retains the properties of the request being collapsed into, if that request is unclean we would...

> In the mean time I staged the patch to see it in action. I've cleared the queue for one of our affected builders, https://lab.llvm.org/staging/#/builders/75, and will see how the...

Sorry I was away for a long weekend. Linaro's bots also seemed to have no problems with this, didn't break our non-clean builders either. I will merge this now.

Maybe we do document this but I looked in the triage guide and it wasn't there, I don't recall it anywhere else. Tell me if so, we can move/link one...