Deprecating and removing the MBlaze backend

Doesn't seem to get a lot of love since most of the commits in the
last 3 years have been maintenance. I guess it doesn't take a whole
lot of maintenance either, but...

cc'ing Wesley since he seems to be the last guy to commit to it.

Thoughts?

-eric

Chandler brought up removing it back in February but Rogelio Serrano said he could maintain it and Jeff Fifield from Xilinx was supposed to check if someone could help.

If no one has stepped up in the past 5 months, then I don't see an issue with removing it.

Micah

So this came up a few months ago and I briefly tried to drum up support within Xilinx to maintain/improve the MBlaze backend. I also put it on my todo list to starting doing some of this myself if needed. It now sounds like I should increase the priority of these tasks. My request is to wait another couple of months before moving to deprecate and remove. If I can’t generate any activity by then, go for it – I completely understand that a backend without a maintainer no good.

Thanks,
Jeff

Chandler brought up removing it back in February but Rogelio Serrano said he could maintain it and Jeff Fifield from Xilinx was supposed to check if someone could help.

Most of my customers changed their minds and dropped mblaze. One stuck to gcc and the rest moved to another arch and toolchain. I’m Sorry guys.

I say that we drop it. If someone steps up to start maintaining it, they can begin by resurrecting it from SVN.

-Chris

I say that we drop it. If someone steps up to start maintaining it, they can begin by resurrecting it from SVN.

Patches attached.

Cheers,
Rafael

llvm.patch.bz2 (74.9 KB)

clang.patch.bz2 (3.56 KB)

Haven't looked at them, but the idea is preapproved and ... if you
break it you buy it :wink:

-eric

I say that we drop it. If someone steps up to start maintaining it, they can begin by resurrecting it from SVN.

Patches attached.

Do it!

-Chris