From: Ulf Magnusson <ulfalizer@gmail.com>
To: yann.morin.1998@free.fr, linux-kbuild@vger.kernel.org
Cc: sam@ravnborg.org, zippel@linux-m68k.org,
nicolas.pitre@linaro.org, mmarek@suse.com, dirk@gouders.net,
yamada.masahiro@socionext.com, lacombar@gmail.com,
JBeulich@suse.com, linux-kernel@vger.kernel.org,
Ulf Magnusson <ulfalizer@gmail.com>
Subject: [PATCH 2/3] kconfig: Clarify expression rewriting
Date: Thu, 5 Oct 2017 14:01:14 +0200 [thread overview]
Message-ID: <1507204875-5021-3-git-send-email-ulfalizer@gmail.com> (raw)
In-Reply-To: <1507204875-5021-1-git-send-email-ulfalizer@gmail.com>
menu_finalize() is one of the more opaque parts of Kconfig, and I need
to make some changes to it to fix an issue related to modules. Add some
comments related to expression rewriting and dependency propagation as a
review aid. They will also help other people trying to understand the
code.
Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
---
| 52 ++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 52 insertions(+)
--git a/scripts/kconfig/menu.c b/scripts/kconfig/menu.c
index 8354dfa..94f192de 100644
--- a/scripts/kconfig/menu.c
+++ b/scripts/kconfig/menu.c
@@ -296,6 +296,11 @@ void menu_finalize(struct menu *parent)
sym = parent->sym;
if (parent->list) {
+ /*
+ * This menu node has children. We (recursively) process them
+ * and propagate parent dependencies before moving on.
+ */
+
if (sym && sym_is_choice(sym)) {
if (sym->type == S_UNKNOWN) {
/* find the first choice value to find out choice type */
@@ -319,24 +324,66 @@ void menu_finalize(struct menu *parent)
else
parentdep = parent->dep;
+ /* For each child menu node... */
for (menu = parent->list; menu; menu = menu->next) {
+ /*
+ * Propagate parent dependencies to the child menu
+ * node, also rewriting and simplifying expressions
+ */
basedep = expr_transform(menu->dep);
basedep = expr_alloc_and(expr_copy(parentdep), basedep);
basedep = expr_eliminate_dups(basedep);
menu->dep = basedep;
+
if (menu->sym)
+ /*
+ * Note: For symbols, all prompts are included
+ * too in the symbol's own property list
+ */
prop = menu->sym->prop;
else
+ /*
+ * For non-symbol menu nodes, we just need to
+ * handle the prompt
+ */
prop = menu->prompt;
+
+ /* For each property... */
for (; prop; prop = prop->next) {
if (prop->menu != menu)
+ /*
+ * Two possibilities:
+ *
+ * 1. The property lacks dependencies
+ * and so isn't location-specific,
+ * e.g. an 'option'
+ *
+ * 2. The property belongs to a symbol
+ * defined in multiple locations and
+ * is from some other location. It
+ * will be handled there in that
+ * case.
+ *
+ * Skip the property.
+ */
continue;
+
+ /*
+ * Propagate parent dependencies to the
+ * property's condition, rewriting and
+ * simplifying expressions at the same time
+ */
dep = expr_transform(prop->visible.expr);
dep = expr_alloc_and(expr_copy(basedep), dep);
dep = expr_eliminate_dups(dep);
if (menu->sym && menu->sym->type != S_TRISTATE)
dep = expr_trans_bool(dep);
prop->visible.expr = dep;
+
+ /*
+ * Handle selects and implies, which modify the
+ * dependencies of the selected/implied symbol
+ */
if (prop->type == P_SELECT) {
struct symbol *es = prop_get_symbol(prop);
es->rev_dep.expr = expr_alloc_or(es->rev_dep.expr,
@@ -348,6 +395,11 @@ void menu_finalize(struct menu *parent)
}
}
}
+
+ /*
+ * Recursively process children in the same fashion before
+ * moving on
+ */
for (menu = parent->list; menu; menu = menu->next)
menu_finalize(menu);
} else if (sym) {
--
2.7.4
next prev parent reply other threads:[~2017-10-05 12:02 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-05 12:01 [PATCH 0/3] kconfig: Make 'm' safe before modules symbol is defined Ulf Magnusson
2017-10-05 12:01 ` [PATCH 1/3] kconfig: Rename menu_check_dep() to rewrite_m() Ulf Magnusson
2017-10-05 12:01 ` Ulf Magnusson [this message]
2017-10-05 12:01 ` [PATCH 3/3] kconfig: Clean up modules handling and fix crash Ulf Magnusson
2017-12-14 23:27 ` [PATCH 0/3] kconfig: Make 'm' safe before modules symbol is defined Masahiro Yamada
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1507204875-5021-3-git-send-email-ulfalizer@gmail.com \
--to=ulfalizer@gmail.com \
--cc=JBeulich@suse.com \
--cc=dirk@gouders.net \
--cc=lacombar@gmail.com \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mmarek@suse.com \
--cc=nicolas.pitre@linaro.org \
--cc=sam@ravnborg.org \
--cc=yamada.masahiro@socionext.com \
--cc=yann.morin.1998@free.fr \
--cc=zippel@linux-m68k.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).