[gentoo-catalyst] [PATCH 2/3] arch/ppc.toml: don't force -mabi=elfv2 on power subarches.

-mabi=elfv2 is default on ppc64le in gcc/glibc toolchain for all powerpc64le-*-linux-gnu CHOSTS, so we don't need to pass it. forcing -mabi=elfv2 on ppc64(be) is quite risky, as it's not standard configuration and not officialy supported by glibc at all. according to glibc upstream it should be a

[gentoo-catalyst] [PATCH 3/3] arch/ppc.toml: add power9 subarches

Signed-off-by: Georgy Yakovlev --- arch/ppc.toml | 10 ++ 1 file changed, 10 insertions(+) diff --git a/arch/ppc.toml b/arch/ppc.toml index 6f74789d..c8623cca 100644 --- a/arch/ppc.toml +++ b/arch/ppc.toml @@ -42,6 +42,16 @@ COMMON_FLAGS = "-O2 -pipe -mcpu=power8 -mtune=power8" CHOST =

[gentoo-catalyst] [PATCH 1/3] arch/ppc.toml: add CHOST based subarches

if stage spec sets chost value, code in catalyst/base/stagebase.py#L76 will derive subarch value from the frontmost part of CHOST triple. if such subarch is not defined in toml file, build will fail. Add generic copy of base subarches defined in CHOST format. Signed-off-by: Georgy Yakovlev ---

[gentoo-catalyst] [PATCH 05/37] catalyst: Sanitize variable kernel name before giving it to bash

Signed-off-by: Matt Turner --- catalyst/base/stagebase.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/catalyst/base/stagebase.py b/catalyst/base/stagebase.py index 1ccc9f04..ffcc211c 100644 --- a/catalyst/base/stagebase.py +++ b/catalyst/base/stagebase.py @@ -1504,7

[gentoo-catalyst] [PATCH 09/37] targets: Remove useless sleep 0

Signed-off-by: Matt Turner --- targets/support/netboot-final.sh | 12 1 file changed, 12 deletions(-) diff --git a/targets/support/netboot-final.sh b/targets/support/netboot-final.sh index f71e4099..7a26e0f4 100755 --- a/targets/support/netboot-final.sh +++

[gentoo-catalyst] [PATCH 07/37] targets: Consolidate evals using filtered_kname

Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh index d150a530..b0583785 100755 --- a/targets/support/kmerge.sh +++ b/targets/support/kmerge.sh @@ -65,9 +65,6

[gentoo-catalyst] [PATCH 10/37] targets: Remove trivial build_kernel() function

Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 7 ++- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh index b4c3963a..bcc83cd6 100755 --- a/targets/support/kmerge.sh +++ b/targets/support/kmerge.sh @@ -111,10

[gentoo-catalyst] [PATCH 02/37] catalyst: Drop outdated comment

We don't use os.system anymore. We pass self.env to Popen, so the comment isn't accurate. Signed-off-by: Matt Turner --- catalyst/base/stagebase.py | 5 - 1 file changed, 5 deletions(-) diff --git a/catalyst/base/stagebase.py b/catalyst/base/stagebase.py index 48157837..532f7133 100644 ---

[gentoo-catalyst] [PATCH 17/37] targets: Delete some more dead code

Dead since 2015. Fixes: 1c7687f6 ("update-modules doesn't exists anymore.") Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 7 --- 1 file changed, 7 deletions(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh index d512954b..911d7cb3 100755 ---

[gentoo-catalyst] [PATCH 06/37] targets: Fix filtered_kname

A follow-on commit is going to remove this code, but I thought it was still valuable to fix the immediate problem for documentation purporses. Broken since it was first added in 2006. Fixes: 6aa64bf7 ("filter kname so - and . work") Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 2

[gentoo-catalyst] [PATCH 11/37] targets: Remove some disabled debugging messages

Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 5 + 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh index bcc83cd6..3d7f697a 100755 --- a/targets/support/kmerge.sh +++ b/targets/support/kmerge.sh @@ -66,8 +66,7 @@

[gentoo-catalyst] [PATCH 16/37] targets: Remove a bunch of stray newlines

Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 7 --- 1 file changed, 7 deletions(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh index aec4d71e..d512954b 100755 --- a/targets/support/kmerge.sh +++ b/targets/support/kmerge.sh @@ -128,7 +128,6 @@ fi if [

[gentoo-catalyst] [PATCH 01/37] catalyst: Use early return to unindent code

Signed-off-by: Matt Turner --- catalyst/base/stagebase.py | 521 +++-- 1 file changed, 269 insertions(+), 252 deletions(-) diff --git a/catalyst/base/stagebase.py b/catalyst/base/stagebase.py index 2e313bd8..48157837 100644 --- a/catalyst/base/stagebase.py +++

[gentoo-catalyst] [PATCH 03/37] catalyst: Fix obvious logic error

Broken since 2005. I don't even know what the author could have been thinking when they changed this. Fixes: c06264e3 ("Initial import of Catalyst 2.0.0") Signed-off-by: Matt Turner --- catalyst/base/stagebase.py | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git

[gentoo-catalyst] [PATCH 08/37] targets: Remove filtered_kname

Now that the Python code passes the sanitize name, we don't need this. Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 14 +- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh index b0583785..b4c3963a

[gentoo-catalyst] [PATCH 04/37] catalyst: Add and use sanitize_name() function

Will be used in an additional place in a follow-on commit. Signed-off-by: Matt Turner --- catalyst/base/stagebase.py | 10 +- catalyst/support.py| 9 + 2 files changed, 14 insertions(+), 5 deletions(-) diff --git a/catalyst/base/stagebase.py

[gentoo-catalyst] [PATCH 12/37] targets: Remove some unnecessary exports

Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 2 -- 1 file changed, 2 deletions(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh index 3d7f697a..0e72e98b 100755 --- a/targets/support/kmerge.sh +++ b/targets/support/kmerge.sh @@ -67,8 +67,6 @@ setup_gk_args() {

[gentoo-catalyst] [PATCH 25/37] targets: Use full 'sys-kernel/genkernel' package name

Signed-off-by: Matt Turner --- targets/support/pre-kmerge.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/targets/support/pre-kmerge.sh b/targets/support/pre-kmerge.sh index 165dfd51..a92c436c 100755 --- a/targets/support/pre-kmerge.sh +++ b/targets/support/pre-kmerge.sh

[gentoo-catalyst] [PATCH 24/37] targets: Rewrite hppa pre-kmerge check

Signed-off-by: Matt Turner --- targets/support/pre-kmerge.sh | 43 ++- 1 file changed, 17 insertions(+), 26 deletions(-) diff --git a/targets/support/pre-kmerge.sh b/targets/support/pre-kmerge.sh index dd51ae9a..165dfd51 100755 --- a/targets/support/pre-kmerge.sh

[gentoo-catalyst] [PATCH 18/37] targets: Remove bizarre make_destpath() call

The commit that added it says it did so in order to "export ROOT=/ prior to running genkernel". Why that was necessary is unclear. I can find no evidence that genkernel needs (or needed) ROOT to be set, and why it would need ROOT=/ seems even more doubtful. Fixes: 8fb885ff ("Added an additional

[gentoo-catalyst] [PATCH 20/37] targets: Remove wrong make.conf munging

With pkgcache fixed for kernel sources in a prior commit, I noticed that if a livecd-stage2 built three kernels, the first would emerge normally and build a binpkg, the second would emerge and fail to use the binpkg produced earlier, and the third would use a previously produced binpkg. I

[gentoo-catalyst] [PATCH 36/37] targets: Remove unnecessary CONFIG_PROTECT assignment

CONFIG_PROTECT is set in run_merge, making this assignment and export unnecessary. Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 1 - 1 file changed, 1 deletion(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh index 972feb99..5e284183 100755 ---

[gentoo-catalyst] [PATCH 30/37] targets: Remove post-kmerge

The find ... touch modules.dep command doesn't seem necessary anymore, and this is all the script still does. Additionally confusing, pre-kmerge runs only once while post-kmerge runs once for each kernel. Signed-off-by: Matt Turner --- catalyst/base/stagebase.py | 4

[gentoo-catalyst] [PATCH 23/37] targets: Default to sys-kernel/gentoo-sources directly

Using virtual/linux-sources was a nice idea, but leads to multiple failures in practice. For example, we use 'portageq contents' later, and the virtual provides no files. We could handle the indirection with 'portageq expand_virtual' in some cases but that requires the virtual to be installed on

[gentoo-catalyst] [PATCH 31/37] targets: Remove unnecessary unset PACKAGES

Dead code since 2008. Fixes: b79d06a3 ("Remove temporary package listing code, ...") Signed-off-by: Matt Turner --- targets/support/chroot-functions.sh | 1 - 1 file changed, 1 deletion(-) diff --git a/targets/support/chroot-functions.sh b/targets/support/chroot-functions.sh index

[gentoo-catalyst] [PATCH 19/37] targets: Remove unnecessary unset USE

Dead code since 2008. Fixes: 454c1682 ("Rather than using the environment, we write out our USE to make.conf ...") Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 1 - 1 file changed, 1 deletion(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh index

[gentoo-catalyst] [PATCH 35/37] targets: Reimplement kerncache support

I cannot see how the code added in commit fed3b45b could have possibly worked, and in practice it did not. An earlier commit (beb92087) deleted most or all of the uses of the *_MATCH variables but left the code that set them in place. Fixes: beb92087 ("Added patches from Joshua Kinard ") Fixes:

[gentoo-catalyst] [PATCH 34/37] targets: Merge setup_gk_args() and genkernel_compile()

These both just do a bunch of argument concatenation and are not separate functions in any meaningful way. Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 6 +- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh

[gentoo-catalyst] [PATCH 32/37] targets: Delete never-used packages.txt code

Added in 2008 with a commit message saying Added code to livecd-stage1 to generate a package list, which will be used in livecd-stage2 with the upcoming livedvd livecd/type to fetch the distfiles automatically. which sounds like a nice plan, but until it's implemented we

[gentoo-catalyst] [PATCH 29/37] targets: Drop warning about no cdtar

Most platforms use grub now and don't need a cdtar. Also use double brackets while we're here. Signed-off-by: Matt Turner --- targets/support/bootloader-setup.sh | 6 +- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/targets/support/bootloader-setup.sh

[gentoo-catalyst] [PATCH 28/37] targets: Move create_handbook_icon() to its use

And use <<- so we can indent the heredoc properly. Signed-off-by: Matt Turner --- targets/support/chroot-functions.sh | 14 -- targets/support/livecdfs-update.sh | 15 +++ 2 files changed, 15 insertions(+), 14 deletions(-) diff --git

[gentoo-catalyst] [PATCH 21/37] targets: Use USE=symlink when emerging kernel sources

Just use the thing that exists instead of rolling our own... Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 12 +++- 1 file changed, 3 insertions(+), 9 deletions(-) diff --git a/targets/support/kmerge.sh b/targets/support/kmerge.sh index 1209e330..7eb18313 100755 ---

[gentoo-catalyst] [PATCH 33/37] targets: Remove kernelpkgs.txt generation

Evidently this was to be consumed by the Gentoo Linux Installer, which saw its last commit in 2008. Signed-off-by: Matt Turner --- targets/support/kmerge.sh | 10 -- targets/support/livecdfs-update.sh | 11 --- 2 files changed, 21 deletions(-) diff --git

[gentoo-catalyst] [PATCH 27/37] targets: Inline run_default_funcs() function

Signed-off-by: Matt Turner --- targets/support/chroot-functions.sh | 17 ++--- 1 file changed, 6 insertions(+), 11 deletions(-) diff --git a/targets/support/chroot-functions.sh b/targets/support/chroot-functions.sh index b531eb6a..307a9042 100755 ---

[gentoo-catalyst] [PATCH 14/37] targets: Remove some comments whose meanings have been lost

Signed-off-by: Matt Turner --- targets/livecd-stage2/controller.sh | 3 --- targets/netboot/controller.sh | 3 --- 2 files changed, 6 deletions(-) diff --git a/targets/livecd-stage2/controller.sh b/targets/livecd-stage2/controller.sh index 566571e6..fe078eed 100755 ---

[gentoo-catalyst] [PATCH 13/37] targets: Remove more clst_ prefixes

clst_* prefixed variables are created by setup_environment(), which these variables are not. Signed-off-by: Matt Turner --- targets/embedded/controller.sh | 2 +- targets/livecd-stage2/controller.sh | 4 +- targets/netboot/controller.sh | 4 +- targets/stage4/controller.sh

[gentoo-catalyst] [PATCH 26/37] targets: Explicitly set RUN_DEFAULT_FUNCS="yes"

For documentation purposes. Signed-off-by: Matt Turner --- targets/support/pre-kmerge.sh | 2 ++ 1 file changed, 2 insertions(+) diff --git a/targets/support/pre-kmerge.sh b/targets/support/pre-kmerge.sh index a92c436c..0dfc835e 100755 --- a/targets/support/pre-kmerge.sh +++

Re: [gentoo-catalyst] [PATCH 02/37] catalyst: Drop outdated comment

On Tue, 20 Oct 2020 17:23:09 -0700 Matt Turner wrote: > We don't use os.system anymore. We pass self.env to Popen, so the > comment isn't accurate. > > Signed-off-by: Matt Turner > --- > catalyst/base/stagebase.py | 5 - > 1 file changed, 5 deletions(-) > > diff --git

Re: [gentoo-catalyst] [PATCH 02/37] catalyst: Drop outdated comment

On Tue, Oct 20, 2020 at 8:02 PM Brian Dolbec wrote: > > On Tue, 20 Oct 2020 17:23:09 -0700 > Matt Turner wrote: > > > We don't use os.system anymore. We pass self.env to Popen, so the > > comment isn't accurate. > > > > Signed-off-by: Matt Turner > > --- > > catalyst/base/stagebase.py | 5