[PATCH v2 0/2] rust: reduce duplicate information

[PATCH v2 0/2] rust: reduce duplicate information

[Tamir Duberstein]

This series reuses `rust/Makefile`'s knowledge of `pin_init` and
`pin_init_internal`'s `cfg`s in `scripts/generate_rust_analzyer.py` and
reduces the ceremony needed to use that knowledge in the script.

Signed-off-by: Tamir Duberstein <tamird@kernel.org>
---
Changes in v2:
- Split into two patches. (Miguel Ojeda)
- CC kbuild. (Miguel Ojeda)
- Link to v1: https://patch.msgid.link/20260116-rust-analyzer-pin-init-duplication-v1-1-5e85c5824a02@kernel.org

---
Tamir Duberstein (2):
      rust: kbuild: extract configs for reuse
      scripts: generate_rust_analyzer.py: reduce cfg plumbing

 rust/Makefile                     | 32 +++++++++++++++++++++++---------
 scripts/generate_rust_analyzer.py | 13 +++++--------
 2 files changed, 28 insertions(+), 17 deletions(-)
---
base-commit: 24d479d26b25bce5faea3ddd9fa8f3a6c3129ea7
change-id: 20260116-rust-analyzer-pin-init-duplication-a22d05221771

Best regards,
--  
Tamir Duberstein <tamird@kernel.org>

[PATCH v2 1/2] rust: kbuild: extract configs for reuse

[Tamir Duberstein]

Introduce `pin_init{,_internal}-{cfgs,flags}` and pass these new cfgs to
`generate_rust_analyzer.py` to reduce duplicate information.

Signed-off-by: Tamir Duberstein <tamird@kernel.org>
---
 rust/Makefile | 32 +++++++++++++++++++++++---------
 1 file changed, 23 insertions(+), 9 deletions(-)

diff --git a/rust/Makefile b/rust/Makefile
index 5d357dce1704..2461c5835a7e 100644
--- a/rust/Makefile
+++ b/rust/Makefile
@@ -210,16 +210,30 @@ rustdoc-ffi: private is-kernel-object := y
 rustdoc-ffi: $(src)/ffi.rs rustdoc-core FORCE
 	+$(call if_changed,rustdoc)
 
+pin_init_internal-cfgs := \
+    kernel
+
+pin_init_internal-flags := \
+    $(call cfgs-to-flags,$(pin_init_internal-cfgs))
+
 rustdoc-pin_init_internal: private rustdoc_host = yes
-rustdoc-pin_init_internal: private rustc_target_flags = --cfg kernel \
+rustdoc-pin_init_internal: private rustc_target_flags = $(pin_init_internal-flags) \
     --extern proc_macro --crate-type proc-macro
 rustdoc-pin_init_internal: $(src)/pin-init/internal/src/lib.rs \
     rustdoc-clean FORCE
 	+$(call if_changed,rustdoc)
 
+pin_init-cfgs := \
+    kernel
+
+pin_init-flags := \
+    --extern pin_init_internal \
+    --extern macros \
+    $(call cfgs-to-flags,$(pin_init-cfgs))
+
 rustdoc-pin_init: private rustdoc_host = yes
-rustdoc-pin_init: private rustc_target_flags = --extern pin_init_internal \
-    --extern macros --extern alloc --cfg kernel --cfg feature=\"alloc\"
+rustdoc-pin_init: private rustc_target_flags = $(pin_init-flags) \
+    --extern alloc --cfg feature=\"alloc\"
 rustdoc-pin_init: $(src)/pin-init/src/lib.rs rustdoc-pin_init_internal \
     rustdoc-macros FORCE
 	+$(call if_changed,rustdoc)
@@ -272,14 +286,13 @@ rusttestlib-macros: $(src)/macros/lib.rs \
     rusttestlib-proc_macro2 rusttestlib-quote rusttestlib-syn FORCE
 	+$(call if_changed,rustc_test_library)
 
-rusttestlib-pin_init_internal: private rustc_target_flags = --cfg kernel \
+rusttestlib-pin_init_internal: private rustc_target_flags = $(pin_init_internal-flags) \
     --extern proc_macro
 rusttestlib-pin_init_internal: private rustc_test_library_proc = yes
 rusttestlib-pin_init_internal: $(src)/pin-init/internal/src/lib.rs FORCE
 	+$(call if_changed,rustc_test_library)
 
-rusttestlib-pin_init: private rustc_target_flags = --extern pin_init_internal \
-    --extern macros --cfg kernel
+rusttestlib-pin_init: private rustc_target_flags = $(pin_init-flags)
 rusttestlib-pin_init: $(src)/pin-init/src/lib.rs rusttestlib-macros \
     rusttestlib-pin_init_internal $(obj)/$(libpin_init_internal_name) FORCE
 	+$(call if_changed,rustc_test_library)
@@ -547,7 +560,7 @@ $(obj)/$(libmacros_name): $(src)/macros/lib.rs $(obj)/libproc_macro2.rlib \
     $(obj)/libquote.rlib $(obj)/libsyn.rlib FORCE
 	+$(call if_changed_dep,rustc_procmacro)
 
-$(obj)/$(libpin_init_internal_name): private rustc_target_flags = --cfg kernel
+$(obj)/$(libpin_init_internal_name): private rustc_target_flags = $(pin_init_internal-flags)
 $(obj)/$(libpin_init_internal_name): $(src)/pin-init/internal/src/lib.rs FORCE
 	+$(call if_changed_dep,rustc_procmacro)
 
@@ -570,6 +583,8 @@ rust-analyzer:
 		--cfgs='proc_macro2=$(proc_macro2-cfgs)' \
 		--cfgs='quote=$(quote-cfgs)' \
 		--cfgs='syn=$(syn-cfgs)' \
+		--cfgs='pin_init_internal=$(pin_init_internal-cfgs)' \
+		--cfgs='pin_init=$(pin_init-cfgs)' \
 		$(realpath $(srctree)) $(realpath $(objtree)) \
 		$(rustc_sysroot) $(RUST_LIB_SRC) $(if $(KBUILD_EXTMOD),$(srcroot)) \
 		> rust-project.json
@@ -642,8 +657,7 @@ $(obj)/compiler_builtins.o: $(src)/compiler_builtins.rs $(obj)/core.o FORCE
 	+$(call if_changed_rule,rustc_library)
 
 $(obj)/pin_init.o: private skip_gendwarfksyms = 1
-$(obj)/pin_init.o: private rustc_target_flags = --extern pin_init_internal \
-    --extern macros --cfg kernel
+$(obj)/pin_init.o: private rustc_target_flags = $(pin_init-flags)
 $(obj)/pin_init.o: $(src)/pin-init/src/lib.rs $(obj)/compiler_builtins.o \
     $(obj)/$(libpin_init_internal_name) $(obj)/$(libmacros_name) FORCE
 	+$(call if_changed_rule,rustc_library)

-- 
2.52.0

[PATCH v2 2/2] scripts: generate_rust_analyzer.py: reduce cfg plumbing

[Tamir Duberstein]

Centralize `cfg` lookup in `append_crate` to avoid having to do so for
each crate. Remove hardcoded `cfg`s for `pin-init{,-internal}` now that
these are passed from `rust/Makefile`.

Signed-off-by: Tamir Duberstein <tamird@kernel.org>
---
 scripts/generate_rust_analyzer.py | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/scripts/generate_rust_analyzer.py b/scripts/generate_rust_analyzer.py
index 147d0cc94068..b96d3cbe3df1 100755
--- a/scripts/generate_rust_analyzer.py
+++ b/scripts/generate_rust_analyzer.py
@@ -35,7 +35,9 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
     crates_indexes = {}
     crates_cfgs = args_crates_cfgs(cfgs)
 
-    def append_crate(display_name, root_module, deps, cfg=[], is_workspace_member=True, is_proc_macro=False, edition="2021"):
+    def append_crate(display_name, root_module, deps, cfg=None, is_workspace_member=True, is_proc_macro=False, edition="2021"):
+        if cfg is None:
+            cfg = crates_cfgs.get(display_name, [])
         crate = {
             "display_name": display_name,
             "root_module": str(root_module),
@@ -60,7 +62,7 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
     def append_sysroot_crate(
         display_name,
         deps,
-        cfg=[],
+        cfg=None,
         edition="2021",
     ):
         append_crate(
@@ -75,7 +77,7 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
     # NB: sysroot crates reexport items from one another so setting up our transitive dependencies
     # here is important for ensuring that rust-analyzer can resolve symbols. The sources of truth
     # for this dependency graph are `(sysroot_src / crate / "Cargo.toml" for crate in crates)`.
-    append_sysroot_crate("core", [], cfg=crates_cfgs.get("core", []), edition=core_edition)
+    append_sysroot_crate("core", [], edition=core_edition)
     append_sysroot_crate("alloc", ["core"])
     append_sysroot_crate("std", ["alloc", "core"])
     append_sysroot_crate("proc_macro", ["core", "std"])
@@ -90,21 +92,18 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
         "proc_macro2",
         srctree / "rust" / "proc-macro2" / "lib.rs",
         ["core", "alloc", "std", "proc_macro"],
-        cfg=crates_cfgs["proc_macro2"],
     )
 
     append_crate(
         "quote",
         srctree / "rust" / "quote" / "lib.rs",
         ["alloc", "proc_macro", "proc_macro2"],
-        cfg=crates_cfgs["quote"],
     )
 
     append_crate(
         "syn",
         srctree / "rust" / "syn" / "lib.rs",
         ["proc_macro", "proc_macro2", "quote"],
-        cfg=crates_cfgs["syn"],
     )
 
     append_crate(
@@ -124,7 +123,6 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
         "pin_init_internal",
         srctree / "rust" / "pin-init" / "internal" / "src" / "lib.rs",
         [],
-        cfg=["kernel"],
         is_proc_macro=True,
     )
 
@@ -132,7 +130,6 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
         "pin_init",
         srctree / "rust" / "pin-init" / "src" / "lib.rs",
         ["core", "pin_init_internal", "macros"],
-        cfg=["kernel"],
     )
 
     append_crate(

-- 
2.52.0

Re: [PATCH v2 0/2] rust: reduce duplicate information

[Benno Lossin]

On Tue Jan 20, 2026 at 5:10 PM CET, Tamir Duberstein wrote:
> This series reuses `rust/Makefile`'s knowledge of `pin_init` and
> `pin_init_internal`'s `cfg`s in `scripts/generate_rust_analzyer.py` and
> reduces the ceremony needed to use that knowledge in the script.

This is a cool idea, I don't maintain the makefile code, so this doesn't
mean much, but it's on my radar :)

Acked-by: Benno Lossin <lossin@kernel.org>

Cheers,
Benno

> Signed-off-by: Tamir Duberstein <tamird@kernel.org>
> ---
> Changes in v2:
> - Split into two patches. (Miguel Ojeda)
> - CC kbuild. (Miguel Ojeda)
> - Link to v1: https://patch.msgid.link/20260116-rust-analyzer-pin-init-duplication-v1-1-5e85c5824a02@kernel.org
>
> ---
> Tamir Duberstein (2):
>       rust: kbuild: extract configs for reuse
>       scripts: generate_rust_analyzer.py: reduce cfg plumbing
>
>  rust/Makefile                     | 32 +++++++++++++++++++++++---------
>  scripts/generate_rust_analyzer.py | 13 +++++--------
>  2 files changed, 28 insertions(+), 17 deletions(-)
> ---
> base-commit: 24d479d26b25bce5faea3ddd9fa8f3a6c3129ea7
> change-id: 20260116-rust-analyzer-pin-init-duplication-a22d05221771
>
> Best regards,
> --  
> Tamir Duberstein <tamird@kernel.org>

Re: [PATCH v2 2/2] scripts: generate_rust_analyzer.py: reduce cfg plumbing

[Jesung Yang]

On Wed Jan 21, 2026 at 1:10 AM KST, Tamir Duberstein wrote:
> Centralize `cfg` lookup in `append_crate` to avoid having to do so for
> each crate. Remove hardcoded `cfg`s for `pin-init{,-internal}` now that
> these are passed from `rust/Makefile`.
>
> Signed-off-by: Tamir Duberstein <tamird@kernel.org>
> ---
>  scripts/generate_rust_analyzer.py | 13 +++++--------
>  1 file changed, 5 insertions(+), 8 deletions(-)
>
> diff --git a/scripts/generate_rust_analyzer.py b/scripts/generate_rust_analyzer.py
> index 147d0cc94068..b96d3cbe3df1 100755
> --- a/scripts/generate_rust_analyzer.py
> +++ b/scripts/generate_rust_analyzer.py
> @@ -35,7 +35,9 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
>      crates_indexes = {}
>      crates_cfgs = args_crates_cfgs(cfgs)
>  
> -    def append_crate(display_name, root_module, deps, cfg=[], is_workspace_member=True, is_proc_macro=False, edition="2021"):
> +    def append_crate(display_name, root_module, deps, cfg=None, is_workspace_member=True, is_proc_macro=False, edition="2021"):
> +        if cfg is None:
> +            cfg = crates_cfgs.get(display_name, [])

Could we add a brief comment explaining how the behavior of
`append_crate` changes according to the `cfg` parameter? Since `None`
and an empty list have different effects, documenting that distinction
would make the intended behavior clearer.

This would also help later when we add proper Python docstrings.

Thanks!

Best regards,
Jesung

>          crate = {
>              "display_name": display_name,
>              "root_module": str(root_module),
> @@ -60,7 +62,7 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
>      def append_sysroot_crate(
>          display_name,
>          deps,
> -        cfg=[],
> +        cfg=None,
>          edition="2021",
>      ):
>          append_crate(
> @@ -75,7 +77,7 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
>      # NB: sysroot crates reexport items from one another so setting up our transitive dependencies
>      # here is important for ensuring that rust-analyzer can resolve symbols. The sources of truth
>      # for this dependency graph are `(sysroot_src / crate / "Cargo.toml" for crate in crates)`.
> -    append_sysroot_crate("core", [], cfg=crates_cfgs.get("core", []), edition=core_edition)
> +    append_sysroot_crate("core", [], edition=core_edition)
>      append_sysroot_crate("alloc", ["core"])
>      append_sysroot_crate("std", ["alloc", "core"])
>      append_sysroot_crate("proc_macro", ["core", "std"])
> @@ -90,21 +92,18 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
>          "proc_macro2",
>          srctree / "rust" / "proc-macro2" / "lib.rs",
>          ["core", "alloc", "std", "proc_macro"],
> -        cfg=crates_cfgs["proc_macro2"],
>      )
>  
>      append_crate(
>          "quote",
>          srctree / "rust" / "quote" / "lib.rs",
>          ["alloc", "proc_macro", "proc_macro2"],
> -        cfg=crates_cfgs["quote"],
>      )
>  
>      append_crate(
>          "syn",
>          srctree / "rust" / "syn" / "lib.rs",
>          ["proc_macro", "proc_macro2", "quote"],
> -        cfg=crates_cfgs["syn"],
>      )
>  
>      append_crate(
> @@ -124,7 +123,6 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
>          "pin_init_internal",
>          srctree / "rust" / "pin-init" / "internal" / "src" / "lib.rs",
>          [],
> -        cfg=["kernel"],
>          is_proc_macro=True,
>      )
>  
> @@ -132,7 +130,6 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
>          "pin_init",
>          srctree / "rust" / "pin-init" / "src" / "lib.rs",
>          ["core", "pin_init_internal", "macros"],
> -        cfg=["kernel"],
>      )
>  
>      append_crate(

Re: [PATCH v2 2/2] scripts: generate_rust_analyzer.py: reduce cfg plumbing

[Tamir Duberstein]

On Tue, Jan 20, 2026 at 8:16 PM Jesung Yang <y.j3ms.n@gmail.com> wrote:
>
> On Wed Jan 21, 2026 at 1:10 AM KST, Tamir Duberstein wrote:
> > Centralize `cfg` lookup in `append_crate` to avoid having to do so for
> > each crate. Remove hardcoded `cfg`s for `pin-init{,-internal}` now that
> > these are passed from `rust/Makefile`.
> >
> > Signed-off-by: Tamir Duberstein <tamird@kernel.org>
> > ---
> >  scripts/generate_rust_analyzer.py | 13 +++++--------
> >  1 file changed, 5 insertions(+), 8 deletions(-)
> >
> > diff --git a/scripts/generate_rust_analyzer.py b/scripts/generate_rust_analyzer.py
> > index 147d0cc94068..b96d3cbe3df1 100755
> > --- a/scripts/generate_rust_analyzer.py
> > +++ b/scripts/generate_rust_analyzer.py
> > @@ -35,7 +35,9 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
> >      crates_indexes = {}
> >      crates_cfgs = args_crates_cfgs(cfgs)
> >
> > -    def append_crate(display_name, root_module, deps, cfg=[], is_workspace_member=True, is_proc_macro=False, edition="2021"):
> > +    def append_crate(display_name, root_module, deps, cfg=None, is_workspace_member=True, is_proc_macro=False, edition="2021"):
> > +        if cfg is None:
> > +            cfg = crates_cfgs.get(display_name, [])
>
> Could we add a brief comment explaining how the behavior of
> `append_crate` changes according to the `cfg` parameter? Since `None`
> and an empty list have different effects, documenting that distinction
> would make the intended behavior clearer.

I don't think this is necessary - it won't age well when we change the
entire API surface here in the next version of the series[0] that adds
type annotations.

Link: https://lore.kernel.org/all/20250424-rust-analyzer-host-v6-0-40e67fe5c38a@gmail.com/
[0]

> This would also help later when we add proper Python docstrings.

Same as above: a lot of change is coming. Is this worth considering right now?

>
> Thanks!
>
> Best regards,
> Jesung
>
> >          crate = {
> >              "display_name": display_name,
> >              "root_module": str(root_module),
> > @@ -60,7 +62,7 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
> >      def append_sysroot_crate(
> >          display_name,
> >          deps,
> > -        cfg=[],
> > +        cfg=None,
> >          edition="2021",
> >      ):
> >          append_crate(
> > @@ -75,7 +77,7 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
> >      # NB: sysroot crates reexport items from one another so setting up our transitive dependencies
> >      # here is important for ensuring that rust-analyzer can resolve symbols. The sources of truth
> >      # for this dependency graph are `(sysroot_src / crate / "Cargo.toml" for crate in crates)`.
> > -    append_sysroot_crate("core", [], cfg=crates_cfgs.get("core", []), edition=core_edition)
> > +    append_sysroot_crate("core", [], edition=core_edition)
> >      append_sysroot_crate("alloc", ["core"])
> >      append_sysroot_crate("std", ["alloc", "core"])
> >      append_sysroot_crate("proc_macro", ["core", "std"])
> > @@ -90,21 +92,18 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
> >          "proc_macro2",
> >          srctree / "rust" / "proc-macro2" / "lib.rs",
> >          ["core", "alloc", "std", "proc_macro"],
> > -        cfg=crates_cfgs["proc_macro2"],
> >      )
> >
> >      append_crate(
> >          "quote",
> >          srctree / "rust" / "quote" / "lib.rs",
> >          ["alloc", "proc_macro", "proc_macro2"],
> > -        cfg=crates_cfgs["quote"],
> >      )
> >
> >      append_crate(
> >          "syn",
> >          srctree / "rust" / "syn" / "lib.rs",
> >          ["proc_macro", "proc_macro2", "quote"],
> > -        cfg=crates_cfgs["syn"],
> >      )
> >
> >      append_crate(
> > @@ -124,7 +123,6 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
> >          "pin_init_internal",
> >          srctree / "rust" / "pin-init" / "internal" / "src" / "lib.rs",
> >          [],
> > -        cfg=["kernel"],
> >          is_proc_macro=True,
> >      )
> >
> > @@ -132,7 +130,6 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
> >          "pin_init",
> >          srctree / "rust" / "pin-init" / "src" / "lib.rs",
> >          ["core", "pin_init_internal", "macros"],
> > -        cfg=["kernel"],
> >      )
> >
> >      append_crate(
>

Re: [PATCH v2 2/2] scripts: generate_rust_analyzer.py: reduce cfg plumbing

[Jesung Yang]

On Wed Jan 21, 2026 at 11:13 PM KST, Tamir Duberstein wrote:
> On Tue, Jan 20, 2026 at 8:16 PM Jesung Yang <y.j3ms.n@gmail.com> wrote:
>> On Wed Jan 21, 2026 at 1:10 AM KST, Tamir Duberstein wrote:
>> > Centralize `cfg` lookup in `append_crate` to avoid having to do so for
>> > each crate. Remove hardcoded `cfg`s for `pin-init{,-internal}` now that
>> > these are passed from `rust/Makefile`.
>> >
>> > Signed-off-by: Tamir Duberstein <tamird@kernel.org>
>> > ---
>> >  scripts/generate_rust_analyzer.py | 13 +++++--------
>> >  1 file changed, 5 insertions(+), 8 deletions(-)
>> >
>> > diff --git a/scripts/generate_rust_analyzer.py b/scripts/generate_rust_analyzer.py
>> > index 147d0cc94068..b96d3cbe3df1 100755
>> > --- a/scripts/generate_rust_analyzer.py
>> > +++ b/scripts/generate_rust_analyzer.py
>> > @@ -35,7 +35,9 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
>> >      crates_indexes = {}
>> >      crates_cfgs = args_crates_cfgs(cfgs)
>> >
>> > -    def append_crate(display_name, root_module, deps, cfg=[], is_workspace_member=True, is_proc_macro=False, edition="2021"):
>> > +    def append_crate(display_name, root_module, deps, cfg=None, is_workspace_member=True, is_proc_macro=False, edition="2021"):
>> > +        if cfg is None:
>> > +            cfg = crates_cfgs.get(display_name, [])
>>
>> Could we add a brief comment explaining how the behavior of
>> `append_crate` changes according to the `cfg` parameter? Since `None`
>> and an empty list have different effects, documenting that distinction
>> would make the intended behavior clearer.
>
> I don't think this is necessary - it won't age well when we change the
> entire API surface here in the next version of the series[0] that adds
> type annotations.
>
> Link: https://lore.kernel.org/all/20250424-rust-analyzer-host-v6-0-40e67fe5c38a@gmail.com/
> [0]
>
>> This would also help later when we add proper Python docstrings.
>
> Same as above: a lot of change is coming. Is this worth considering right now?

Fair point, thanks for the context.

Reviewed-by: Jesung Yang <y.j3ms.n@gmail.com>

By the way, could you share your plan for upcoming major changes? It
would help me prepare and align my series if I know the intended
direction. I currently have these on my radar:

  [1] https://lore.kernel.org/all/20250424-rust-analyzer-host-v6-0-40e67fe5c38a@gmail.com/
  [2] https://lore.kernel.org/rust-for-linux/20260109-ra-fix-primitive-v2-0-249852a4145a@gmail.com/
  [3] https://lore.kernel.org/rust-for-linux/20260120-ra-fix-v1-0-829e4e92818c@nvidia.com/

I assume you would prefer merging [1] first? Please let me know if I'm
missing something.

Thanks!

Best regards,
Jesung

Re: [PATCH v2 2/2] scripts: generate_rust_analyzer.py: reduce cfg plumbing

[Tamir Duberstein]

On Wed, Jan 21, 2026 at 6:12 PM Jesung Yang <y.j3ms.n@gmail.com> wrote:
>
> On Wed Jan 21, 2026 at 11:13 PM KST, Tamir Duberstein wrote:
> > On Tue, Jan 20, 2026 at 8:16 PM Jesung Yang <y.j3ms.n@gmail.com> wrote:
> >> On Wed Jan 21, 2026 at 1:10 AM KST, Tamir Duberstein wrote:
> >> > Centralize `cfg` lookup in `append_crate` to avoid having to do so for
> >> > each crate. Remove hardcoded `cfg`s for `pin-init{,-internal}` now that
> >> > these are passed from `rust/Makefile`.
> >> >
> >> > Signed-off-by: Tamir Duberstein <tamird@kernel.org>
> >> > ---
> >> >  scripts/generate_rust_analyzer.py | 13 +++++--------
> >> >  1 file changed, 5 insertions(+), 8 deletions(-)
> >> >
> >> > diff --git a/scripts/generate_rust_analyzer.py b/scripts/generate_rust_analyzer.py
> >> > index 147d0cc94068..b96d3cbe3df1 100755
> >> > --- a/scripts/generate_rust_analyzer.py
> >> > +++ b/scripts/generate_rust_analyzer.py
> >> > @@ -35,7 +35,9 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs, core_edit
> >> >      crates_indexes = {}
> >> >      crates_cfgs = args_crates_cfgs(cfgs)
> >> >
> >> > -    def append_crate(display_name, root_module, deps, cfg=[], is_workspace_member=True, is_proc_macro=False, edition="2021"):
> >> > +    def append_crate(display_name, root_module, deps, cfg=None, is_workspace_member=True, is_proc_macro=False, edition="2021"):
> >> > +        if cfg is None:
> >> > +            cfg = crates_cfgs.get(display_name, [])
> >>
> >> Could we add a brief comment explaining how the behavior of
> >> `append_crate` changes according to the `cfg` parameter? Since `None`
> >> and an empty list have different effects, documenting that distinction
> >> would make the intended behavior clearer.
> >
> > I don't think this is necessary - it won't age well when we change the
> > entire API surface here in the next version of the series[0] that adds
> > type annotations.
> >
> > Link: https://lore.kernel.org/all/20250424-rust-analyzer-host-v6-0-40e67fe5c38a@gmail.com/
> > [0]
> >
> >> This would also help later when we add proper Python docstrings.
> >
> > Same as above: a lot of change is coming. Is this worth considering right now?
>
> Fair point, thanks for the context.
>
> Reviewed-by: Jesung Yang <y.j3ms.n@gmail.com>
>
> By the way, could you share your plan for upcoming major changes? It
> would help me prepare and align my series if I know the intended
> direction. I currently have these on my radar:
>
>   [1] https://lore.kernel.org/all/20250424-rust-analyzer-host-v6-0-40e67fe5c38a@gmail.com/
>   [2] https://lore.kernel.org/rust-for-linux/20260109-ra-fix-primitive-v2-0-249852a4145a@gmail.com/
>   [3] https://lore.kernel.org/rust-for-linux/20260120-ra-fix-v1-0-829e4e92818c@nvidia.com/
>
> I assume you would prefer merging [1] first? Please let me know if I'm
> missing something.

Yeah, these are the big changes I'm currently aware of. I'm going to
look for ways to split [1] into smaller parts. The main thing I'd like
to get in ASAP is type annotations as maintenance becomes much easier
following that.

>
> Thanks!
>
> Best regards,
> Jesung

Re: [PATCH v2 1/2] rust: kbuild: extract configs for reuse

[Miguel Ojeda]

On Tue, Jan 20, 2026 at 5:10 PM Tamir Duberstein <tamird@kernel.org> wrote:
>
> Introduce `pin_init{,_internal}-{cfgs,flags}` and pass these new cfgs to
> `generate_rust_analyzer.py` to reduce duplicate information.
>
> Signed-off-by: Tamir Duberstein <tamird@kernel.org>

Looks good -- my suggestion was to move the two rust-analyzer lines
here (`--cfgs=`) also into the second patch, i.e. here it would just
be about factoring out the flags, and then in the second one you pass
them to `rust-analyzer` and use the information.

Doing it here doesn't hurt, since it will be unused on the script
side, so it is also fine, but I thought it would be more
understandable the other way around.

Anyway, I could perhaps take this one with or without the two lines,
and then next cycle you can apply to rust-analyzer the other one.

Thanks!

Cheers,
Miguel