Re: Is anyone still using gcc 4.1 on master?
On 11/8/2011 7:10 AM, joris dedieu wrote: 2011/11/8 Juan Francisco Cantero Hurtado: On 11/07/2011 10:50 PM, Samuel J. Greear wrote: Our C++ dependencies would not be that difficult to overcome and I do not see why the system compiler should necessarily have to support pkgsrc directly. What C++ software or dependencies has DragonFly? I'm curious. At least devd(8), Joris -- Juan Francisco Cantero Hurtado http://juanfra.info groff and gold linker are two more. Tangential to the discussion of the lack of stated and current project goals, it's not a stated goal that DragonFly have a C-only base, nor that the virtue of simplicity is more valued over performance to the point of eliminating useful functionality. As Samuel alluded, that's what he thinks is best. I'm in the other camp and actually favor a system compiler capable of more languages. John
Re: Is anyone still using gcc 4.1 on master?
On Mon, Nov 7, 2011 at 9:04 PM, Juan Francisco Cantero Hurtado wrote: > On 11/07/2011 10:50 PM, Samuel J. Greear wrote: >> >> Our C++ dependencies would not be that difficult to overcome and I >> do not see why the system compiler should necessarily have to >> support pkgsrc directly. > > What C++ software or dependencies has DragonFly? I'm curious. > > Answered in the context of, "What C++ dependencies does DragonFly have and how hard would it be to remove them?" devd(8), groff(1) and binutils gold, we have a C binutils as well though so that is not a hard requirement. mdoc is already staged up to replace groff, someone just needs to put in the labor to get it done. devd is 1200-1300 lines of C++, only part of which is "classy", it should be a straightforward port to C. If someone really wanted to rid DragonFly of its C++ dependencies, it would be fairly easy to do so given how things sit right now. Sam
Re: Is anyone still using gcc 4.1 on master?
2011/11/8 Juan Francisco Cantero Hurtado : > On 11/07/2011 10:50 PM, Samuel J. Greear wrote: >> >> Our C++ dependencies would not be that difficult to overcome and I >> do not see why the system compiler should necessarily have to >> support pkgsrc directly. > > What C++ software or dependencies has DragonFly? I'm curious. At least devd(8), Joris > > > -- > Juan Francisco Cantero Hurtado http://juanfra.info >
Re: Is anyone still using gcc 4.1 on master?
On 11/07/2011 10:50 PM, Samuel J. Greear wrote: Our C++ dependencies would not be that difficult to overcome and I do not see why the system compiler should necessarily have to support pkgsrc directly. What C++ software or dependencies has DragonFly? I'm curious. -- Juan Francisco Cantero Hurtado http://juanfra.info
Re: Is anyone still using gcc 4.1 on master?
On Mon, Nov 7, 2011 at 12:51 PM, John Marino wrote: > On 11/7/2011 8:03 PM, Samuel J. Greear wrote: >>> >>> pcc is not a candidate. >> >> Sorry, I disagree, although I understand if you aren't going to be the >> one to port it. >> >> Sam > > I don't understand that sentence. Are you saying you or somebody else is > going to port pcc into base? A compiler that don't do c++? > Our C++ dependencies would not be that difficult to overcome and I do not see why the system compiler should necessarily have to support pkgsrc directly. If you account for the entire history of DragonFly there have been long periods where the compiler was not updated, and not for lack of need but for lack of knowledge and/or manpower. _I_ feel that the best approach for the overall long-term health of the project would be to kill off our (very minimal) C++ dependencies and use a small, simple and easy to reason about C compiler that most or all of the C developers which DragonFly attracts can work on and fix. There are numerous options for distributing/bootsrapping GCC, clang or $othercompiler to support pkgsrc. I do not expect everyone to agree, but I do not think the position of the project as a whole is against someone working in support of pcc and I wanted to make sure that your comment was not construed as being the concrete position of the project. Sam
Re: Is anyone still using gcc 4.1 on master?
On 11/7/2011 8:03 PM, Samuel J. Greear wrote: pcc is not a candidate. Sorry, I disagree, although I understand if you aren't going to be the one to port it. Sam I don't understand that sentence. Are you saying you or somebody else is going to port pcc into base? A compiler that don't do c++?
Re: Is anyone still using gcc 4.1 on master?
> pcc is not a candidate. Sorry, I disagree, although I understand if you aren't going to be the one to port it. Sam
Re: Is anyone still using gcc 4.1 on master?
On Mon, Nov 7, 2011 at 2:05 PM, Venkatesh Srinivas wrote: > kaffe (from pkgsrc) seems to need gcc 4.1 to build correctly... > > -- vs; > I would like to second to that. gcc 4.1 is needed to build wip/jdk16 since kaffe can be built only by using gcc 4.1 at the moment. (Many thanks to Francois Tigeot for this!) We critically depend on wip/jdk16, so dropping 4.1 would be very bad news for us. On x86_64, wip/jdk16 is the only way to get java since last time I checked, there is no linux.ko and hence can't use lang/sun-jdk6. Peeter (aka must) --
Re: Is anyone still using gcc 4.1 on master?
On 11/7/2011 3:05 PM, Venkatesh Srinivas wrote: > kaffe (from pkgsrc) seems to need gcc 4.1 to build correctly... > > -- vs; I would think that's either a problem with kaffe or a problem with the gcc44 compiler. If it's the latter, the gcc44 compiler should be fixed rather than use kaffe as a reason to maintain gcc41.
Re: Is anyone still using gcc 4.1 on master?
kaffe (from pkgsrc) seems to need gcc 4.1 to build correctly... -- vs;
Re: Is anyone still using gcc 4.1 on master?
On 11/7/2011 8:59 AM, elekktrett...@exemail.com.au wrote: > Is it time to get rid of it? It seems like a waste of space/compile time. > > Maybe replace it with clang-3? > > Is anyone using GCC 4.1 over 4.4? > > Petr > 1. space, we have plenty 2. compile time: you can skip the compilation by adding "NO_GCC41=yes" in the /etc/make.conf file 3. Yes, gcc 4.1 will eventually be replaced Candidates are gcc 4.6 (more likely) and clang/llvm. pcc is not a candidate. Both candidate are non-trivial to port, and will take significant time. Even after it gets in base (and gcc 4.1 removed) gcc 4.4 will likely be the primary compiler until the newer one is proven. John
Re: Is anyone still using gcc 4.1 on master?
It is always good to have an older gcc version, too, especially given that other BSDs stick to them. Importing clang is not an option unless you are volunteering. Cheers, Alex On 07/11/11 07:59, elekktrett...@exemail.com.au wrote: > Is it time to get rid of it? It seems like a waste of space/compile time. > > Maybe replace it with clang-3? > > Is anyone using GCC 4.1 over 4.4? > > Petr >
Is anyone still using gcc 4.1 on master?
Is it time to get rid of it? It seems like a waste of space/compile time. Maybe replace it with clang-3? Is anyone using GCC 4.1 over 4.4? Petr
