Is GDC a solid option versus LDC?
From what I know, most people (if not all), seem to prefer LDC over GDC. When it comes to C, I always prefer GCC as GCC's backend compiles code faster and also produces code will run faster than LLVM (even if the differences are small). When it comes to D, GDC seems to compile code faster in my machine (AMD Ryzen 5 2400G) than LDC about 3-4 times faster! However, I'm really skeptical as to why people seem to prefer LDC and I wonder if GDC is not as stable and reliable as LDC or if most code that uses recent code doesn't work in GDC. I have also heard that LDC get support for new things faster than GDC. Is that true? And if yes, then for how much difference are we talking about? Any personal experiences would be much appreciated.
Re: Is GDC a solid option versus LDC?
On Friday, 4 March 2022 at 15:17:49 UTC, rikki cattermole wrote: GDC has a few issues that LDC simply doesn't have. For instance GCC can only depend on a binary of the last version of GCC (my knowledge of that is what I've heard from Iain). What this means is its stuck using a frontend that is based upon a version that was still C++ and that is ol. Many years old. The version you are probably using was released back in 2017, more or less. Yes some patches have been back ported (2.076 is the version, but even that was D so they had to have been rewritten into C++). Right now GDC is close to getting the frontend written in D, which from what I hear should be next release. At which point it'll keep up. Oh, sucks to hear that. Can I get some more info (links?) on that one? Also do you know where is the current development source of GDC?
Re: Is GDC a solid option versus LDC?
On Friday, 4 March 2022 at 15:26:38 UTC, H. S. Teoh wrote: Both GDC and LDC are good compilers. I wouldn't hesitate to use either if performance is important in what I'm doing. The only difference is that GDC, being a part of the GCC project, is tied to the GCC release cycle, so it tends to lag behind DMD and LDC in terms of the latest language developments. LDC releases on its own schedule, and as of a year or two ago, has been closely tracking DMD releases, so it is very up-to-date in terms of language development. Being practically on par with DMD means that I can easily switch between DMD and LDC latest releases without having to worry about my code being incompatible with either. (Though generally that doesn't really happen that much, and hardly at all if you aren't always keeping tabs on the latest language developments.) If your code isn't reliant on the bleeding edge, though, GDC is definitely not inferior to LDC in any way. T Thank you! I was wondering, Is staying so much up to date really impossible here? Other than ImportC (which will still be completed at some point), D has already support for 1 million features already so will this be a big problem? I see this becoming a smaller problem with each release.
Re: Is GDC a solid option versus LDC?
On Friday, 4 March 2022 at 15:59:33 UTC, Iain Buclaw wrote: announcement: https://forum.dlang.org/post/nuphsvwkdlygdcxai...@forum.dlang.org mainline: https://gcc.gnu.org/git.html gh-mirror: https://github.com/gcc-mirror/gcc/ Thanks a lot!
error: matching constraint references invalid operand number
Hello! After I've been convinced on using, GDC as my go to
compiler when I need optimization (or other than x86_x64 targets)
thanks to the replays of my previous posts, I now try to compiler
my project using GDC but I'm having an error. I'm trying to use
inline assembly (GCC syntax) but it won't compile (compiled
without any problems with LDC). The inline assembly block
specifically is part of the following function:
```d
i32 sys_clock_nanosleep(clock_id clock, i32 flags, timespec* req,
timespec* rem) {
i32 ret_code;
asm {
"syscall"
: "=a" (ret_code)
: "a" (230), "D" (clock), "S" (flags), "d" (req), "r10" (rem)
: "memory", "rcx", "r11";
}
return ret_code;
}
```
When I try to compile, I'm getting the following error message:
```
source/time.d: In function ‘sys_clock_nanosleep’:
source/time.d:132:31: error: matching constraint references
invalid operand number
132 | : "memory", "rcx", "r11";
```
Any ideas?
Re: error: matching constraint references invalid operand number
On Friday, 4 March 2022 at 23:33:12 UTC, Iain Buclaw wrote:
GCC doesn't have constraints for registers %r8 .. %15. As D
doesn't have register variables either, you'll have to set-up
r10 in the instruction string.
```
asm {
"mov %[rem], %%r10; syscall"
: "=a" (ret_code)
: "a" (230), "D" (clock), "S" (flags), "d" (req), [rem] "r"
(rem)
: "memory", "rcx", "r10", "r11";
}
```
Thank you! This will do the trick! Do you happen to know if D
plans to add "registers" in the future or if it has been
suggested but was rejected?
Also, how can I see the D specific options for GDC? Thinks like
how to set "versions", options like the "-fno-druntime" and other
specific D stuff as I'll I'm seeing in the "--help" message are
options that are general and the output is in general the same as
GCC.
Re: error: matching constraint references invalid operand number
On Saturday, 5 March 2022 at 17:31:17 UTC, Iain Buclaw wrote: With the option "--help=d" Alright! This will do the trick! Thanks a lot for your time and thanks a lot for your amazing work in GDC in general!
Re: Is GDC a solid option versus LDC?
On Friday, 4 March 2022 at 17:13:23 UTC, H. S. Teoh wrote: As I said, if you're not relying on the latest-and-greatest bleeding edge, GDC is a perfectly fine compiler. Iain has already said the next release will be self-hosting, so we can finally move on from the ancient C++-based compiler to the latest self-hosted one. Once that's done, GDC should keep up with language developments at a much better pace than before. T Thank you! I do rely in a small D subset so I will be fine. Have an amazing day ;)
