Skip to contentSkip to conversation

kconfig: tests: remove randconfig test for choice in choice

PATCH New Masahiro Yamada 1 participant 2 messages Apr 24, 2020

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 kconfig: tests: remove randconfig test for choice in choice

Nesting choice statements does not make any sense.

Commit df8df5e4bc37 ("usb: get rid of 'choice' for legacy gadget drivers") got rid of the only user.

I will make it a syntax error. Remove the test in advance.

---

 .../kconfig/tests/rand_nested_choice/Kconfig  | 35 -------------------
 .../tests/rand_nested_choice/__init__.py      | 17 ---------
 .../tests/rand_nested_choice/expected_stdout0 |  2 --
 .../tests/rand_nested_choice/expected_stdout1 |  4 ---
 .../tests/rand_nested_choice/expected_stdout2 |  5 ---
 5 files changed, 63 deletions(-)
 delete mode 100644 scripts/kconfig/tests/rand_nested_choice/Kconfig
 delete mode 100644 scripts/kconfig/tests/rand_nested_choice/__init__.py
 delete mode 100644 scripts/kconfig/tests/rand_nested_choice/expected_stdout0
 delete mode 100644 scripts/kconfig/tests/rand_nested_choice/expected_stdout1
 delete mode 100644 scripts/kconfig/tests/rand_nested_choice/expected_stdout2
scripts/kconfig/tests/rand_nested_choice/Kconfig deleted +0 -35
deleted file mode 100644
index 8350de7f732b..000000000000
@@ -1,35 +0,0 @@
-# SPDX-License-Identifier: GPL-2.0
-
-choice
-	prompt "choice"
-
-config A
-	bool "A"
-
-config B
-	bool "B"
-
-if B
-choice
-	prompt "sub choice"
-
-config C
-	bool "C"
-
-config D
-	bool "D"
-
-if D
-choice
-	prompt "subsub choice"
-
-config E
-	bool "E"
-
-endchoice
-endif # D
-
-endchoice
-endif # B
-
-endchoice
scripts/kconfig/tests/rand_nested_choice/__init__.py deleted +0 -17
deleted file mode 100644
index 9e4b2db53581..000000000000
@@ -1,17 +0,0 @@
-# SPDX-License-Identifier: GPL-2.0
-"""
-Set random values recursively in nested choices.
-
-Kconfig can create a choice-in-choice structure by using 'if' statement.
-randconfig should correctly set random choice values.
-
-Related Linux commit: 3b9a19e08960e5cdad5253998637653e592a3c29
-"""
-
-
-def test(conf):
-    for i in range(20):
-        assert conf.randconfig() == 0
-        assert (conf.config_contains('expected_stdout0') or
-                conf.config_contains('expected_stdout1') or
-                conf.config_contains('expected_stdout2'))
scripts/kconfig/tests/rand_nested_choice/expected_stdout0 deleted +0 -2
deleted file mode 100644
index 05450f3d4eb5..000000000000
@@ -1,2 +0,0 @@
-CONFIG_A=y
-# CONFIG_B is not set
scripts/kconfig/tests/rand_nested_choice/expected_stdout1 deleted +0 -4
deleted file mode 100644
index 37ab29584157..000000000000
@@ -1,4 +0,0 @@
-# CONFIG_A is not set
-CONFIG_B=y
-CONFIG_C=y
-# CONFIG_D is not set
scripts/kconfig/tests/rand_nested_choice/expected_stdout2 deleted +0 -5
deleted file mode 100644
index 849ff47e9848..000000000000
@@ -1,5 +0,0 @@
-# CONFIG_A is not set
-CONFIG_B=y
-# CONFIG_C is not set
-CONFIG_D=y
-CONFIG_E=y
Signature · 1 line
2.25.1
#1 lore raw

PATCH 2/2 kconfig: allow only 'config', 'comment', and 'if' statements inside choice

The statement block surrounded by 'if' ... 'endif' is reduced into if_stmt, which is accepted in the choice context. Therefore, you can do anything within a choice block by wrapping 'if y' ... 'end'.

For example, you can create a menu inside a choice.

[Test Code]

---------------->8----------------
choice
prompt "choice"

config A
bool "A"

config B
bool "B"

if y

menu "strange menu"

config C
bool "C"

endmenu

endif

endchoice
---------------->8----------------

I want to make such a weird structure a syntax error.

In fact, the USB gadget Kconfig file used nested choice for no good reason until commit df8df5e4bc37 ("usb: get rid of 'choice' for legacy gadget drivers") killed it.

I think the 'source' statement inside choice is on the fence. It is at least gramatically sensible as long as the included file contains only bool/tristate config statements. However, it makes the code unreadable, and in the end, people forget the fact that the file is included from the choice block. Commit 10e5e6c24963 ("usb: gadget: move choice ... endchoice to legacy/Kconfig") got rid of the only user.

Going forward, you can only use 'config', 'comment', and 'if' inside choice blocks. This also recursively applies to 'if' blocks inside choice blocks.

---

 scripts/kconfig/parser.y | 30 ++++++++++++++++--------------
 1 file changed, 16 insertions(+), 14 deletions(-)
scripts/kconfig/parser.y +16 -14
index 708b6c4b13ca..190f1117f35a 100644
@@ -119,20 +119,24 @@ mainmenu_stmt: T_MAINMENU T_WORD_QUOTE T_EOL
 
 stmt_list:
 	  /* empty */
-	| stmt_list common_stmt
+	| stmt_list assignment_stmt
 	| stmt_list choice_stmt
+	| stmt_list comment_stmt
+	| stmt_list config_stmt
+	| stmt_list if_stmt
 	| stmt_list menu_stmt
+	| stmt_list menuconfig_stmt
+	| stmt_list source_stmt
 	| stmt_list T_WORD error T_EOL	{ zconf_error("unknown statement \"%s\"", $2); }
 	| stmt_list error T_EOL		{ zconf_error("invalid statement"); }
 ;
 
-common_stmt:
-	  if_stmt
-	| comment_stmt
-	| config_stmt
-	| menuconfig_stmt
-	| source_stmt
-	| assignment_stmt
+stmt_list_in_choice:
+	  /* empty */
+	| stmt_list_in_choice comment_stmt
+	| stmt_list_in_choice config_stmt
+	| stmt_list_in_choice if_stmt_in_choice
+	| stmt_list_in_choice error T_EOL	{ zconf_error("invalid statement"); }
 ;
 
 /* config/menuconfig entry */
@@ -254,7 +258,7 @@ choice_end: end
 	}
 };
 
-choice_stmt: choice_entry choice_block choice_end
+choice_stmt: choice_entry stmt_list_in_choice choice_end
 ;
 
 choice_option_list:
@@ -305,11 +309,6 @@ default:
 	| T_DEF_BOOL		{ $$ = S_BOOLEAN; }
 	| T_DEF_TRISTATE	{ $$ = S_TRISTATE; }
 
-choice_block:
-	  /* empty */
-	| choice_block common_stmt
-;
-
 /* if entry */
 
 if_entry: T_IF expr T_EOL
@@ -331,6 +330,9 @@ if_end: end
 if_stmt: if_entry stmt_list if_end
 ;
 
+if_stmt_in_choice: if_entry stmt_list_in_choice if_end
+;
+
 /* menu entry */
 
 menu: T_MENU T_WORD_QUOTE T_EOL
Signature · 1 line
2.25.1