linux.git/scripts/Makefile.btf, branch v6.18.21

kbuild, bpf: Enable --btf_features=attributes

2025-04-15T18:17:50+00:00

pahole v1.30 has a BTF encoding feature for arbitrary attributes, used
in particular for tagging bpf_arena_alloc_pages and
bpf_arena_free_pages BPF kfuncs [1][2].

Enable it for the kernel build.

[1] https://lore.kernel.org/bpf/20250130201239.1429648-1-ihor.solodrai@linux.dev/
[2] https://lore.kernel.org/bpf/20250228194654.1022535-1-ihor.solodrai@linux.dev/

Signed-off-by: Ihor Solodrai 
Signed-off-by: Martin KaFai Lau 
Tested-by: Eduard Zingerman 
Link: https://patch.msgid.link/20250414185918.538195-1-ihor.solodrai@linux.dev

kbuild, bpf: Correct pahole version that supports distilled base btf feature

2025-02-24T22:14:52+00:00

pahole commit [0] of supporting distilled base btf feature released on
pahole v1.28 rather than v1.26. So let's correct this.

Signed-off-by: Pu Lehui 
Signed-off-by: Andrii Nakryiko 
Link: https://git.kernel.org/pub/scm/devel/pahole/pahole.git/commit/?id=c7b1f6a29ba1 [0]
Link: https://lore.kernel.org/bpf/20250219063113.706600-1-pulehui@huaweicloud.com

kbuild,bpf: Pass make jobs' value to pahole

2024-11-12T04:00:21+00:00

Pass the value of make's -j/--jobs argument to pahole, to avoid out of
memory errors and make pahole respect the "jobs" value of make.

On systems with little memory but many cores, invoking pahole using -j
without argument potentially creates too many pahole instances,
causing an out-of-memory situation. Instead, we should pass make's
"jobs" value as an argument to pahole's -j, which is likely configured
to be (much) lower than the actual core count on such systems.

If make was invoked without -j, either via cmdline or MAKEFLAGS, then
JOBS will be simply empty, resulting in the existing behavior, as
expected.

Signed-off-by: Florian Schmaus 
Signed-off-by: Andrii Nakryiko 
Tested-by: Holger Hoffstätte 
Link: https://lore.kernel.org/bpf/20241102100452.793970-1-flo@geekplace.eu

kbuild,bpf: Add module-specific pahole flags for distilled base BTF

2024-06-21T21:45:07+00:00

Support creation of module BTF along with distilled base BTF;
the latter is stored in a .BTF.base ELF section and supplements
split BTF references to base BTF with information about base types,
allowing for later relocation of split BTF with a (possibly
changed) base.  resolve_btfids detects the presence of a .BTF.base
section and will use it instead of the base BTF it is passed in
BTF id resolution.

Modules will be built with a distilled .BTF.base section for external
module build, i.e.

make -C. -M=path2/module

...while in-tree module build as part of a normal kernel build will
not generate distilled base BTF; this is because in-tree modules
change with the kernel and do not require BTF relocation for the
running vmlinux.

Signed-off-by: Alan Maguire 
Signed-off-by: Andrii Nakryiko 
Reviewed-by: Eduard Zingerman 
Link: https://lore.kernel.org/bpf/20240620091733.1967885-6-alan.maguire@oracle.com

kbuild: bpf: Tell pahole to DECL_TAG kfuncs

2024-06-12T18:01:30+00:00

With [0], pahole can now discover kfuncs and inject DECL_TAG
into BTF. With this commit, we will start shipping said DECL_TAGs
to downstream consumers if pahole supports it.

This is useful for feature probing kfuncs as well as generating
compilable prototypes. This is particularly important as kfuncs
do not have stable ABI.

[0]: https://git.kernel.org/pub/scm/devel/pahole/pahole.git/commit/?id=72e88f29c6f7e14201756e65bd66157427a61aaf

Signed-off-by: Daniel Xu 
Link: https://lore.kernel.org/r/324aac5c627bddb80d9968c30df6382846994cc8.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov

kbuild, bpf: Use test-ge check for v1.25-only pahole

2024-05-18T17:46:16+00:00

There is no need to set the pahole v1.25-only flags in an
"ifeq" version clause; we are already in a <= v1.25 branch
of "ifeq", so that combined with a "test-ge" v1.25 ensures the
flags will be applied for v1.25 only.

Suggested-by: Masahiro Yamada 
Signed-off-by: Alan Maguire 
Signed-off-by: Andrii Nakryiko 
Link: https://lore.kernel.org/bpf/20240514162716.2448265-1-alan.maguire@oracle.com
Signed-off-by: Alexei Starovoitov

kbuild,bpf: Switch to using --btf_features for pahole v1.26 and later

2024-05-09T21:44:35+00:00

The btf_features list can be used for pahole v1.26 and later -
it is useful because if a feature is not yet implemented it will
not exit with a failure message.  This will allow us to add feature
requests to the pahole options without having to check pahole versions
in future; if the version of pahole supports the feature it will be
added.

Signed-off-by: Alan Maguire 
Signed-off-by: Andrii Nakryiko 
Tested-by: Eduard Zingerman 
Acked-by: Andrii Nakryiko 
Link: https://lore.kernel.org/bpf/20240507135514.490467-1-alan.maguire@oracle.com

kbuild: avoid too many execution of scripts/pahole-flags.sh

2023-10-28T12:10:08+00:00

scripts/pahole-flags.sh is executed so many times.

You can confirm it, as follows:

  $ cat <> scripts/pahole-flags.sh
  > echo "scripts/pahole-flags.sh was executed" >&2
  > EOF

  $ make -s
  scripts/pahole-flags.sh was executed
  scripts/pahole-flags.sh was executed
  scripts/pahole-flags.sh was executed
  scripts/pahole-flags.sh was executed
  scripts/pahole-flags.sh was executed
    [ lots of repeated lines... ]

This scripts is executed more than 20 times during the kernel build
because PAHOLE_FLAGS is a recursively expanded variable and exported
to sub-processes.

With GNU Make >= 4.4, it is executed more than 60 times because
exported variables are also passed to other $(shell ) invocations.
Without careful coding, it is known to cause an exponential fork
explosion. [1]

The use of $(shell ) in an exported recursive variable is likely wrong
because $(shell ) is always evaluated due to the 'export' keyword, and
the evaluation can occur multiple times by the nature of recursive
variables.

Convert the shell script to a Makefile, which is included only when
CONFIG_DEBUG_INFO_BTF=y.

[1]: https://savannah.gnu.org/bugs/index.php?64746

Signed-off-by: Masahiro Yamada 
Reviewed-by: Alan Maguire 
Tested-by: Alan Maguire 
Reviewed-by: Nicolas Schier 
Tested-by: Miguel Ojeda 
Acked-by: Miguel Ojeda 
Acked-by: Jiri Olsa 
Reviewed-by: Martin Rodriguez Reboredo