[Bug 11736] New: KCONFIG_ALLCONFIG does not apply passed config (regression in 0b1c62934)

bugzilla at busybox.net bugzilla at busybox.net
Tue Mar 26 21:10:53 UTC 2019


https://bugs.busybox.net/show_bug.cgi?id=11736

            Bug ID: 11736
           Summary: KCONFIG_ALLCONFIG does not apply passed config
                    (regression in 0b1c62934)
           Product: Busybox
           Version: 1.33.x
          Hardware: All
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P5
         Component: Other
          Assignee: unassigned at busybox.net
          Reporter: ac at alexeicolin.com
                CC: busybox-cvs at busybox.net
  Target Milestone: ---

First reported in comment 3 in #10296:
https://bugs.busybox.net/show_bug.cgi?id=10296#c3

make allnoconfig KCONFIG_ALLCONFIG="configfile" is supposed to generate a
.config where a symbol is enabled if it or its dependee is enabled in
configfile. The code for this use case is in conf.c:

    case set_no:
        case set_mod:
        case set_yes:
        case set_random:
                name = getenv("KCONFIG_ALLCONFIG");
                if (name && !stat(name, &tmpstat)) {
                        conf_read_simple(name);
                        break;
                }

KCONFIG_ALLCONFIG allows overriding a default filename that would otherwise be
read (with the same purpose of setting symbols in .config):
{allno,allyes,...}.config.

To reproduce:
$ echo 'CONFIG_MKDIR=y' > shortconfig
$ make allnoconfig KCONFIG_ALLCONFIG="shortconfig"
$ grep MKDIR .config
# CONFIG_MKDIR is not set
but should be (and is after reverting the commit below)
CONFIG_MKDIR=y

This is a regression introduced in 0b1c62934215a08351a80977c7cf8e9346683a1e.

This regression breaks the miniconf [1,2,3,4] use case.

Is there a way to fix the original issue without breaking KCONFIG_ALLCONFIG?
Thank you.

[1] http://landley.net/hg/aboriginal/file/c8293b3ab81f/more/miniconfig.sh
[2] http://landley.net/aboriginal/FAQ.html#dev_miniconfig
[3] https://lwn.net/Articles/161086/
[4] http://lists.uclibc.org/pipermail/uclibc/2007-March/017612.html

-- 
You are receiving this mail because:
You are on the CC list for the bug.


More information about the busybox-cvs mailing list