playX
playX
That's not a problem! I'm already switching to SMI + heap allocate everything other so I can not use native overflow catching mechanism for overflow checks (at least on 32...
Thank you for your answer! I've got another question: is there any chances to get `patchpoint` like function/intrinsic or patchable jumps and calls? I use code patching heavily in my...
Thanks for answer! > Right now multi-threading support doesn't really work correctly Multi-threading is not important for me because I will use BEAM-like concurrency where each process(green thread) will have...
@Techcable I actually decided to do STW in my GC impl so the question is: should `basic_safepoint` impl actually handle STW or I should look at something else? UPD: I...
I just don't know where to send this message so I'll just send it there :P @Techcable , there are MMTk rewrite in Rust and it should be available in...
I've implemented concurrent copying in Rust but it was based on C4: [cgc](https://github.com/playXE/cgc). It's object header is quite large as it has 2 words: pointer to vtable and pointer to...
Thanks for answer! My question was about panicking in safepoint_recurse! macro. I'm still not moving from mostly-precise collector in starlight since support for precise collection is quite hard to do...
Maybe do TLAB allocation? Some of [Dora](https://github.com/dinfuehr/dora/) GCs and some JVM GCs use it. What Dora does is that it allocates 32KB memory for TLAB and if object fits into...
What I've thought of is to write rustc plugin which will invoke rustc_codegen_cranelift on specified function and then save Cranelift IR in some of data sections to be interpreted and...
I have seen ykrustc, yes. It uses Intel hardware tracing which is not portable across CPUs. I belive that interpreting Cranelift IR is much easier and more portable way of...