1 |
I think you're getting a bit confused |
2 |
|
3 |
libsupc++ is the default now, from GNU |
4 |
|
5 |
libcxxabi is the bloated runtime from Apple |
6 |
|
7 |
libcxxrt is the faster c++ runtime, PathScale+David Chisnall, which |
8 |
PathScale and FreeBSD use by default. We don't need a version number |
9 |
because it's pretty much rock solid stable for a while. |
10 |
I'd encourage you to consider libcxxrt for at least the code size and |
11 |
performance reasons. Build it and you'll see. Locally my unoptimized |
12 |
libcxxrt.so is like 88K. How much is your libcxxabi (static and |
13 |
shared) |
14 |
|
15 |
88K /opt/enzo-2016-06-26/lib/6.0.983/x8664/64/libcxxrt.so |
16 |
140K /opt/enzo-2016-06-26/lib/6.0.983/x8664/64/libcxxrt.a |
17 |
// This seems larger than I remember and I need to check why. |
18 |
|
19 |
https://github.com/pathscale/libcxxrt |
20 |
|
21 |
On Fri, Aug 19, 2016 at 10:40 AM, Lei Zhang <zhanglei.april@×××××.com> wrote: |
22 |
> 2016-08-19 10:07 GMT+08:00 <cbergstrom@×××××××××.com>: |
23 |
>> That seems a lot like what we've already done. I guess a GSOC student is working on the libcxxabi piece. |
24 |
> |
25 |
> I am that GSoC student :) |
26 |
> |
27 |
> I'm currently trying to push libc++abi to replace libcxxrt as the |
28 |
> default runtime: https://github.com/gentoo/gentoo/pull/2048 |
29 |
> |
30 |
> The reason is I think libc++abi blends in more naturally with other |
31 |
> LLVM components, and it has a clear version number as opposed to |
32 |
> libcxxrt. |
33 |
> |
34 |
>> The only advantage to using our runtime, libcxxrt, is performance and code size. |
35 |
> |
36 |
> Honestly I don't know what essential difference these two libs have; I |
37 |
> can't find any decent comparison of them on the internet. Do you have |
38 |
> some real numbers to show the difference in performance and code size? |
39 |
> |
40 |
> |
41 |
> Lei |
42 |
> |