[PATCH v7] scripts/config: honor $KBUILD_OUTPUT by default

[PATCH v7] scripts/config: honor $KBUILD_OUTPUT by default

[Dmitry Voytik]

If $KBUILD_OUTPUT is set, use it by default if --file is not passed.

Fix the issue when the script silently updates a stale .config in
the root directory, while the user expects ${KBUILD_OUTPUT}/.config
to be updated.
Handle properly if the config file is a relative path.

The problem can be worked around with --file, but this is easy to miss,
and is inconsistent with how the Makefile and other tools work
(e.g., scripts/diffconfig).

How to reproduce the issue:
	make mrproper
	export KBUILD_OUTPUT=.out
	make defconfig
	./scripts/config \
	       -e CONFIG_UBSAN
Error reported:
	grep: .config: No such file or directory

How the patch was tested:
	make mrproper
	export KBUILD_OUTPUT=.out
	make defconfig
	grep "CONFIG_UBSAN=y" $KBUILD_OUTPUT/.config || echo "OK"

Test the basic usage:
	./scripts/config -e CONFIG_UBSAN
	grep "CONFIG_UBSAN=y" $KBUILD_OUTPUT/.config && echo "OK"

Test --refresh picks up correct config file by using $KBUILD_OUTPUT:
	./scripts/config -d CONFIG_UBSAN
	grep "CONFIG_UBSAN=y" $KBUILD_OUTPUT/.config && echo "FAIL"
	./scripts/config --refresh -e CONFIG_UBSAN
	grep "CONFIG_UBSAN=y" $KBUILD_OUTPUT/.config && echo "OK"

Test --file takes precedence over $KBUILD_OUTPUT and a relative
path is handled properly:
	./scripts/config -d CONFIG_UBSAN
	mkdir .out2
	cp .out/.config .out2/.my_config
	grep "CONFIG_UBSAN=y" .out2/.my_config  && echo "FAIL"
	./scripts/config --file .out2/.my_config --refresh \
	       -e CONFIG_UBSAN
	grep "CONFIG_UBSAN=y" .out2/.my_config && echo "OK"
	ls .out/.out2 && echo "FAIL"
	ls .out/.out && echo "FAIL"

Test --file can accept files with hyphen:
	./scripts/config -d CONFIG_UBSAN
	cp .out/.config -- -my_config
	grep "CONFIG_UBSAN=y" -- -my_config && echo "FAIL"
	./scripts/config  --refresh \
	       -e CONFIG_UBSAN --file -my_config
	grep "CONFIG_UBSAN=y" -- -my_config && echo "OK"

Signed-off-by: Dmitry Voytik <voytikd@gmail.com>
---
v7:
 * Adressed https://sashiko.dev/#/patchset/20261006142337.821859-1-voytikd%40gmail.com
v6:
 * Per Nathan Chancellor's suggestions, simplified
 * https://lore.kernel.org/lkml/179121581235.2275809.12565811683284940522.b4-review@b4/
v5:
 * Per Nathan Chancellor's suggestion, improved handling $FN path by
   realpath
 * https://lore.kernel.org/lkml/179097210878.309835.10523307262544578337.b4-review@b4/
 * Documented the new test scenario in the git commit message
v4:
 * Addressed Nicolas Schier's comments - dropped the change how '--refresh'
   is handled
 * Implemented conversion of $FN to the absolute path which helps with
   '--refresh' and relative paths
 * Documented testing steps in the commit message
v3:
 * Added an example in the commit message how to reproduce the issue
v2:
 * Addressed comments by sashiko:
   * https://sashiko.dev/#/patchset/20260920064350.17999-1-voytikd%40gmail.com
   * use basename when --refresh is used
   * remove echo which can be misleading when --file is passed
v1:
 * https://lore.kernel.org/all/20260920064350.17999-1-voytikd@gmail.com/
---
 scripts/config | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)

diff --git a/scripts/config b/scripts/config
index 32428ea909c2..1e29e449077c 100755
--- a/scripts/config
+++ b/scripts/config
@@ -130,13 +130,15 @@ on_exit() {
 trap on_exit EXIT
 
 FN=.config
+
+
 CMDS=()
 while [[ $# -gt 0 ]]; do
 	if [ "$1" = "--file" ]; then
 		if [ "$2" = "" ]; then
 			usage
 		fi
-		FN="$2"
+		FN=$(realpath -q -- "$2" || printf -- "%s\n" "$2")
 		shift 2
 	else
 		CMDS+=("$1")
@@ -144,6 +146,16 @@ while [[ $# -gt 0 ]]; do
 	fi
 done
 
+# ignore $KBUILD_OUTPUT if "--file" option is present
+if [[ "$FN" == ".config" && -n "$KBUILD_OUTPUT" ]]; then
+	if [[ ! -d "$KBUILD_OUTPUT" ]]; then
+		echo "KBUILD_OUTPUT is set but does not exist or is not a " >&2
+		echo "directory. Create it before running $myname." >&2
+		exit 1
+	fi
+	FN=$(realpath -- "$KBUILD_OUTPUT")/"$FN"
+fi
+
 set -- "${CMDS[@]}"
 if [ "$1" = "" ] ; then
 	usage
@@ -200,7 +212,7 @@ while [ "$1" != "" ] ; do
 		;;
 
 	--state|-s)
-		if grep -q "# ${CONFIG_}$ARG is not set" $FN ; then
+		if grep -q "# ${CONFIG_}$ARG is not set" "$FN" ; then
 			echo n
 		else
 			V="$(grep "^${CONFIG_}$ARG=" $FN)"
@@ -229,7 +241,7 @@ while [ "$1" != "" ] ; do
 		;;
 
 	--refresh)
-		yes "" | make oldconfig KCONFIG_CONFIG=$FN
+		yes "" | make oldconfig KCONFIG_CONFIG="$FN"
 		;;
 
 	*)
-- 
2.55.0

Re: [PATCH v7] scripts/config: honor $KBUILD_OUTPUT by default

[Nathan Chancellor]

Hi Dmitry,

On 2026-10-06 18:11 +0200, Dmitry Voytik wrote:
> If $KBUILD_OUTPUT is set, use it by default if --file is not passed.
> 
> Fix the issue when the script silently updates a stale .config in
> the root directory, while the user expects ${KBUILD_OUTPUT}/.config
> to be updated.
> Handle properly if the config file is a relative path.
> 
> The problem can be worked around with --file, but this is easy to miss,
> and is inconsistent with how the Makefile and other tools work
> (e.g., scripts/diffconfig).

I was finally able to sit down and look at this a little bit more. I
think we can avoid the use of realpath altogether, which would avoid
most of what Sashiko is complaining about. I tested the following diff
with your examples and they appear to pass without any issues. Please
give it a shot. I think we can drop most of the reproduction and
examples in the commit message. If you want to include them for others,
please put it below the '---' where you have the changelog, as I don't
think it needs to end up in git history.

diff --git a/scripts/config b/scripts/config
index 32428ea909c2..497551538433 100755
--- a/scripts/config
+++ b/scripts/config
@@ -144,6 +144,22 @@ while [[ $# -gt 0 ]]; do
 	fi
 done
 
+# If no '--file' was specified, use KBUILD_OUTPUT as the base if it exists
+if [[ "$FN" = ".config" && -n "$KBUILD_OUTPUT" ]]; then
+	if [[ ! -d "$KBUILD_OUTPUT" ]]; then
+		echo "KBUILD_OUTPUT is set ('$KBUILD_OUTPUT') but does not exist or is not a directory." >&2
+		echo "Create it before running $myname or use '--file' to specify configuration file." >&2
+		exit 1
+	fi
+	FN="$KBUILD_OUTPUT/$FN"
+fi
+# Turn FN into an absolute path if it is relative so that '--refresh' works on
+# the correct file. This does not use realpath, as $FN may not exist yet.
+case "$FN" in
+	/*) ;;
+	*) FN="$PWD/$FN" ;;
+esac
+
 set -- "${CMDS[@]}"
 if [ "$1" = "" ] ; then
 	usage
@@ -200,10 +216,10 @@ while [ "$1" != "" ] ; do
 		;;
 
 	--state|-s)
-		if grep -q "# ${CONFIG_}$ARG is not set" $FN ; then
+		if grep -q "# ${CONFIG_}$ARG is not set" "$FN" ; then
 			echo n
 		else
-			V="$(grep "^${CONFIG_}$ARG=" $FN)"
+			V="$(grep "^${CONFIG_}$ARG=" "$FN")"
 			if [ $? != 0 ] ; then
 				echo undef
 			else
@@ -229,7 +245,7 @@ while [ "$1" != "" ] ; do
 		;;
 
 	--refresh)
-		yes "" | make oldconfig KCONFIG_CONFIG=$FN
+		yes "" | make oldconfig KCONFIG_CONFIG="$FN"
 		;;
 
 	*)

-- 
Cheers,
Nathan

Re: [PATCH v7] scripts/config: honor $KBUILD_OUTPUT by default

[Nicolas Schier]

On Thu, Nov 12, 2026 at 09:59:06PM -0700, Nathan Chancellor wrote:
> Hi Dmitry,
> 
> On 2026-10-06 18:11 +0200, Dmitry Voytik wrote:
> > If $KBUILD_OUTPUT is set, use it by default if --file is not passed.
> > 
> > Fix the issue when the script silently updates a stale .config in
> > the root directory, while the user expects ${KBUILD_OUTPUT}/.config
> > to be updated.
> > Handle properly if the config file is a relative path.
> > 
> > The problem can be worked around with --file, but this is easy to miss,
> > and is inconsistent with how the Makefile and other tools work
> > (e.g., scripts/diffconfig).
> 
> I was finally able to sit down and look at this a little bit more. I
> think we can avoid the use of realpath altogether, which would avoid
> most of what Sashiko is complaining about. I tested the following diff
> with your examples and they appear to pass without any issues. Please
> give it a shot.

Thanks, somehow the sashiko mail missed my inbox.

realpath on Linux uses '-E' behaviour by default, which is what we want:
check existence of all directory path components of the given path.
Adding the POSIX '-E' option explicitly should solve the non-Linux
issues.

As a follow-up, I would like to add support for $KCONFIG_CONFIG as well.

> I think we can drop most of the reproduction and
> examples in the commit message. If you want to include them for others,
> please put it below the '---' where you have the changelog, as I don't
> think it needs to end up in git history.

+1

But still thanks for providing the test examples, it was and is indeed
helpful for my follow-up tests.

> 
> diff --git a/scripts/config b/scripts/config
> index 32428ea909c2..497551538433 100755
> --- a/scripts/config
> +++ b/scripts/config
> @@ -144,6 +144,22 @@ while [[ $# -gt 0 ]]; do
>  	fi
>  done
>  
> +# If no '--file' was specified, use KBUILD_OUTPUT as the base if it exists
> +if [[ "$FN" = ".config" && -n "$KBUILD_OUTPUT" ]]; then
> +	if [[ ! -d "$KBUILD_OUTPUT" ]]; then
> +		echo "KBUILD_OUTPUT is set ('$KBUILD_OUTPUT') but does not exist or is not a directory." >&2
> +		echo "Create it before running $myname or use '--file' to specify configuration file." >&2
> +		exit 1
> +	fi
> +	FN="$KBUILD_OUTPUT/$FN"
> +fi
> +# Turn FN into an absolute path if it is relative so that '--refresh' works on
> +# the correct file. This does not use realpath, as $FN may not exist yet.
> +case "$FN" in
> +	/*) ;;
> +	*) FN="$PWD/$FN" ;;
> +esac
> +
>  set -- "${CMDS[@]}"
>  if [ "$1" = "" ] ; then
>  	usage
> @@ -200,10 +216,10 @@ while [ "$1" != "" ] ; do
>  		;;
>  
>  	--state|-s)
> -		if grep -q "# ${CONFIG_}$ARG is not set" $FN ; then
> +		if grep -q "# ${CONFIG_}$ARG is not set" "$FN" ; then
>  			echo n
>  		else
> -			V="$(grep "^${CONFIG_}$ARG=" $FN)"
> +			V="$(grep "^${CONFIG_}$ARG=" "$FN")"
>  			if [ $? != 0 ] ; then
>  				echo undef
>  			else
> @@ -229,7 +245,7 @@ while [ "$1" != "" ] ; do
>  		;;
>  
>  	--refresh)
> -		yes "" | make oldconfig KCONFIG_CONFIG=$FN
> +		yes "" | make oldconfig KCONFIG_CONFIG="$FN"
>  		;;
>  
>  	*)
> 
> -- 
> Cheers,
> Nathan
> 

Looks good to me, thanks.

Reviewed-by: Nicolas Schier <nsc@fjasle.eu>


-- 
Nicolas