On Sun, 9 Dec 2012, Andreas Hansson wrote:
On Dec. 8, 2012, 2:05 p.m., Nilay Vaish wrote:
Are you saying it would be necessary for the users to have protoc
installed on their systems just to compile gem5 and use it?
If that's the case, I am opposed to this change. I think this should
be optional and not required, like the way we make use of google-perftools.
That is indeed the case, and it is essentially impossible to make it
optional like the tcmalloc that is just a matter of linking with a
library or not. With protobuf there is a fair bit of code involved in
reading/writing the traces that depend both on having protoc and
libprotobuf available.
I don't believe this. It should be possible to segregate the code and
compile only if the compiler and the library are present.
May I ask why you see this as a problem? Any Unix/Linux/OSx system from
the last 5+ years has these dependencies as packages (and I bet they
work smoother than e.g. swig :-)
I doubt that this statement is true. I am using RHEL 6 and it does not
have protoc installed. I am guessing that would mean people who use Fedora
will not have it installed either. Ubuntu (12.10) has it. I don't know
about Mint or other distributions.
Andreas, as you might have noticed, a portion of the traffic on gem5-users
mailing list is about getting gem5 to compile and run. We are already
moving to GCC 4.4, which will be a problem for users, in my belief. I am
ready to live with that. Having one more dependency will lead to further
problems. I have seen enough mails where users are not even ready to sort
out simple compilation problems related to GCC. Having one more dependency
might mean that we lose some of our users. This may not affect ARM, but
one of the reasons why I am involved with gem5 is that I would like people
in architecture to be using the simulator that I have contributed to.
--
Nilay
_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev
