[One oddity of the rpi2 context for the below is that the root file system is
on a fast USB SSD drive: it is really a fast USB 3.0 SSD drive just used on the
USB 2 bus via a USB 3.0 capable hub.]
I just did the following sort of sequence on an rpi2 for 11.0 -r301975
rebuilding itself where the .sh script used indicates to use WITH_META_MODE=yes
for the make command :
> #
> ~/sys_build_scripts.rpi2-host/make_rpi2_nodebug_clang_bootstrap-rpi2-host.sh
> -j 5 buildworld buildkernel
> #
> ~/sys_build_scripts.rpi2-host/make_rpi2_nodebug_clang_bootstrap-rpi2-host.sh
> -j 5 buildworld buildkernel
(Also the .sh runs script to log all the make output. The first build was
actually more involved than shown above, see later notes.)
The 2nd one took about 15 or 16 minutes to go through everything and, in
essence, discover that nothing needed to be rebuilt. This was (still) using
WITH_LIBSOFT= in the src.conf file. Also: WITH_CLANG_FULL= WITH_CLANG_EXTRAS=
WITH_LLDB= were in use. . .
> # more ~/src.configs/src.conf.rpi2-clang-bootstrap.rpi2-host
> TO_TYPE=armv6
> #
> KERNCONF=RPI2-NODBG
> TARGET=arm
> .if ${.MAKE.LEVEL} == 0
> TARGET_ARCH=${TO_TYPE}
> .export TARGET_ARCH
> .endif
> #
> #WITH_CROSS_COMPILER=
> WITH_SYSTEM_COMPILER=
> #
> #CPUTYPE=soft
> WITH_LIBSOFT=
> WITH_LIBCPLUSPLUS=
> WITH_BINUTILS_BOOTSTRAP=
> #WITHOUT_CLANG_BOOTSTRAP=
> WITH_CLANG=
> WITH_CLANG_IS_CC=
> WITH_CLANG_FULL=
> WITH_CLANG_EXTRAS=
> WITH_LLDB=
> #
> WITH_BOOT=
> WITHOUT_LIB32=
> #
> WITHOUT_ELFTOOLCHAIN_BOOTSTRAP=
> WITHOUT_GCC_BOOTSTRAP=
> WITHOUT_GCC=
> WITHOUT_GCC_IS_CC=
> WITHOUT_GNUCXX=
> #
> NO_WERROR=
> #WERROR=
> MALLOC_PRODUCTION=
> #
> WITH_DEBUG_FILES=
> #
> XCFLAGS+= -march=armv7-a -mcpu=cortex-a7
> XCXXFLAGS+= -march=armv7-a -mcpu=cortex-a7
> # There is no XCPPFLAGS but XCPP ets XCFLAGS content.
~/src.configs/make.conf was empty.
Other notes:
I've not had any problems with WITH_META_MODE=yes use for the self hosting 11.0
-r301975 rpi2 builds.
Going from not using WITH_META_MODE=yes to using WITH_META_MODE=yes in the
self-hosting context did not require a cleanworld between the builds. (Cross
build transitions in that direction do require such still.)
I do not report the WITH_META_MODE=yes build time for the first time because I
actually did more. I:
A) forced a root file system I/O failure/panic
(during what happened to be libxo building)
B) then started WITH_META_MODE=yes again
(after booting)
C) let it run until the incomplete file(s) from the failure caused it to stop
D) did a make clean for libxo
E) then started WITH_META_MODE=yes yet again
F) let it run to completion
The WITH_META_MODE=yes use appears to have worked just fine despite the abusive
nature of the test.
I will report that not using WITH_META_MODE=yes took between 15 hr and 16 hr
(no such odd testing sequence involved).
===
Mark Millard
markmi at dsl-only.net
_______________________________________________
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"
