If not, set _OPSYS_SUPPORTS_SSP=no during bootstrap and in mk.conf.
Do SSP detection on "SunOS", and let mk/platform/SunOS.mk's default
"yes" be overridden in mk.conf.
No change to generated mk.conf on NetBSD 8 or CentOS 6. Fixes bootstrap
on Tribblix.
* Put indentation right to follow the existing style
* Correct wrong justification for need of pax on Arch: pax (and not PaX) is
only provided throw https://aur.archlinux.org/packages/pax/ and get low
support (currently broken)
This avoids errors when running from a read-only pkgsrc checkout as the
resulting packages cannot be written to the default PACKAGES directory. The
binary packages aren't useful anyway, as they are often built with reduced
configuration options due to the limited bootstrap environment.
Fixes issue reported by Julien Savard and others.
This is based on the decision The NetBSD Foundation made in 2008 to
do so, which was already applied to src.
This change has been applied to code which is likely not in other
repositories.
ok board@, reviewed by riastradh@
Both sed and awk in the Chrome OS base are permanently put into sandbox
mode, which makes them unsuitable for pkgsrc use. We were already building
nbawk; also add nbsed.
Since Chrome 64, /usr/bin/awk has been a version of mawk that is permanently
in restricted mode. Notably, this means it cannot run pipes, which pkgsrc
requires. Use nawk.
When bootstrap is run with --cwrappers=no (or yes, for that matter),
the value is written into mk.conf at the end of the bootstrap but not
used _during_ the bootstrap itself. Thus, when bootstrapping on Linux
with no builtin diffutils, cwrappers ends up indirectly depending on
itself, and there is no way around that. This commit fixes this.
Eric Brown: PR pkg/52567: Readme.MacOSX is out of date
Additional changes by me: don't mention that the file isn't README.OSX
due to OS X being a different name, because the new name macOS makes
the filename less confusing.
Since cwrappers was enabled by default on some platforms, it became
effectively impossible to bootstrap without cwrappers there.
With this change "--cwrappers" needs to be set to "auto", "yes" or "no"
explicitly when used.
No functional change intended without specifying "--cwrappers" (or with
"auto").
Submitted on tech-pkg@ with no objection.
release, state that instructions only apply to legacy release.
- State that only the compler in base is currently support.
- Update the list of tested releases.
indirectly via it being pulled in using USE_CWRAPPERS=auto, we need to
ensure that it is not built with -DPKG_PRESERVE as otherwise it interacts
badly with pbulk's pkg_delete test.
Substantially revise the ABI 32/64 discussion, separating the
close-in-time changes about the default ABI vs how it is encoded.
Thanks to jperkin for off-list clarifications.
Previously there were at least 5 different ways MACHINE_ARCH could be set,
some statically and some at run time, and in many cases these settings
differed, leading to issues at pkg_add time where there was conflict
between the setting encoded into the package and that used by pkg_install.
Instead, move to a single source of truth where the correct value based on
the host and the chosen (or default) ABI is determined in the bootstrap
script. The value can still be overridden in mk.conf if necessary, e.g.
for cross-compiling.
ABI is now set by default and if unset a default is calculated based on
MACHINE_ARCH. This fixes some OS, e.g. Linux, where the wrong default was
previously chosen.
As a result of the refactoring there is no need for LOWER_ARCH, with
references to it replaced by MACHINE_ARCH. SPARC_TARGET_ARCH is also
removed.
Switch to using arch(1) to detect the architecture of host
Set $machine_arch to x86_64 if on amd64
Pass $machine_arch to the devel/bmake build process
Reviewed by wiz@
Replace and simplify the versions section. Drop notion of
supported/deprecated. Explain why people run specific old versions.
Assert that there are no 10.4 pkgsrc users, and very few 10.5. Keep
notion that PRs for <= 10.5 without fixes can be closed without
fixing.
Explain Darwin vs OS X. Point to PureDarwin. Explain that pkgsrc for
Darwin is really for OSX and invite doc patches.
Make a few minor edits.
It remains for someone to explain more clearly the issue of
broken/missing SDKs in 10.11 and command line tools.
As discussed on tech-pkg, with input from jperkin and tron.
- Remove information about older versions of pkgsrc.
- Note that /usr/include is gone and that there are SDKs
- Add hazy explanation of the missing SDK/command-line-tools issue.
- Add 10.11 (but do not change deprecation status of anything).
- Clarify (but do not try to update for current reality) the 10.6/10.9
bulk build descriptions.
Initial patchset to add support by rodent@
Further adjustments made based on feedback by joerg@
Tested by myself with numerous bulkbuilds thanks to Patrick Wildt @ Bitrig
Reviewed by bsiegert@ joerg@ wiz@
This was added in 2004 and there have been no reports of anyone even
trying to use it in many many years. Searching on the web finds only
information from 2003.
Add Joyent bulk build links.
CVS: ----------------------------------------------------------------------
CVS: CVSROOT cvs.NetBSD.org:/cvsroot
CVS: please use "PR category/123" to have the commitmsg appended to PR 123
looking sufficiently like gcc. Inspect "cc" instead, and don't
believe it's __GNUC__ unless it's also not __clang__.
While here, if you manage to find and install PackageMaker.app, it
still works on 10.9 for --binary-macpkg.
From Sevan Janiyan.
Basically: 10.9's sed is broken, bootstrap works around this via nbsed
on 10.9, and therefore while a 10.6 boostrap's binaries will *run* on
10.9, one can't reliably build new packages.
(Thanks to jperkin@ for pointing this out.)
