Date: Tue, 25 Dec 2012 16:03:38 +0100 From: Matthew Rezny <mrezny@hexaneinc.com> To: Kubilay Kocak <koobs.freebsd@gmail.com> Cc: freebsd-ports@freebsd.org Subject: Re: multimedia/x264 workaround is broken Message-ID: <BF0E52DE-C0C1-430D-889E-E032D0D2B8BD@hexaneinc.com> In-Reply-To: <50D9B0C6.1080009@gmail.com> References: <B00F7B07-CA3D-408E-A626-C39512DEACDD@hexaneinc.com> <50D92254.9060700@gmail.com> <73EA428D-4614-4938-893F-DCD5A3EE8007@hexaneinc.com> <50D9B0C6.1080009@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 25 Dec, 2012, at 14:57, Kubilay Kocak wrote: > On 26/12/2012 12:43 AM, Matthew Rezny wrote: >> On 25 Dec, 2012, at 4:49, Kubilay Kocak wrote: >>=20 >>> On 25/12/2012 1:55 PM, Matthew Rezny wrote: >>>=20 >>>> The immediate question is, what was the original error that >>>> mandated the workaround and does that error still occur with >>>> current version of Clang? If the latter is no, can we get rid >>>> of the temporary workaround? >>>=20 >>> This may help provide some answers: >>>=20 >>> http://www.freebsd.org/cgi/query-pr.cgi?pr=3Dports/171542 >>>=20 >> Thank you for the suggestion. However, the topic of that PR is >> disabling PGO to build with Clang. I had already turned off the PGO >> option before the first build attempt, and the problem I faced was >> the insistence upon calling the non-existent GCC regardless of >> chosen port options. >=20 > I referenced the PR as an additional information source, as I wasn't > entirely sure what issue you were facing without logs to reference. I > should have been more explicit than 'may help' and will do my best > next time. >=20 > Ta, >=20 > Koobs >=20 The log was not included because it was nothing but a single useful line = of information, the attempt to invoke gcc, following two screens worth = of boilerplate. In case I was unclear before, the problem I found is that the workaround = makes it always call the assumed-present base gcc. There are two = failures, never triggering a build of gcc from ports even when setting = the option that should require it (GCC4.4+), and still attempting to = call base gcc rather than the ports version had any gcc port been = installed. Both those errors stem from a workaround that does not even seem = necessary. The port appears to build fine with Clang.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?BF0E52DE-C0C1-430D-889E-E032D0D2B8BD>