On Thu, Feb 2, 2017 at 5:39 PM, Ken Moffat <[email protected]> wrote:
> On Thu, Feb 02, 2017 at 06:23:57PM +0000, Ken Moffat wrote: > > In a fresh system, ghostscript's 'make so' fails - several warnings > > which seem to be unchanged from my last build in November, but then > > > > In file included from ./psi/dxmain.c:38:0: > > ./psi/iapi.h:215:1: note: expected 'char **' but argument is of type > 'const char **' > > gsapi_get_default_device_list(void *lib, char **list, int *listlen); > > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > make[2]: Leaving directory '/scratch/working/ghostscript-9.20' > > make[1]: Leaving directory '/scratch/working/ghostscript-9.20' > > > > I'm way out of my depth on this, with no idea what changed. Any > > suggestions, pretty please ? > > > > Sat scratching my head, looked all through the log from the > beginning of 'make so' but couldn't see any errors only warnings. > Added some instrumentation to shout $? to my log after each command, > gave it anther try and this time it sailed through. > > Odd, but works for me. > > > Meanwhile, I looked at Arch - they don't seem to make so - and then > > at Fedora : nothing there for this problem, but a whole string of > > CVE fixes - some in the last few days, some earlier - for > > CVE-2016-7976/7977/7978/7979/8602/9601 although not all applied in > > that order. > > > > I am tempted to apply all of these as security fixes (assuming, for > > the moment, that they do apply), but it doesn't look as if they will > > solve my FTBFS so I'm stalled. > > > I intended to post this to -dev, maybe I'll take that there or maybe > I'll just let it lie for the moment - it's not as if details of the > issues are at Mitre yet, although redhat and debian seem to know > about them. > > That is because they are embargoed. That's unfortunately the *corporate* nature of Linux now (Thanks Novell, Red Hat, Microsoft (they have a huge part now), and Debian). They don't want us to know about those problems until it is too late, or the embargo expires, or both. Any time that Red Hat or Debian get their hands on a CVE assignment, they never give MITRE any information. There's an uber-secret list that I want to be a part of, but we have to have a place to hide issues from Users first, among a website detailing security advisories and many other requirements.
-- http://lists.linuxfromscratch.org/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
