public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Aurelien Jarno <aurelien@aurel32.net>
To: libc-alpha@sourceware.org
Cc: Aurelien Jarno <aurelien@aurel32.net>
Subject: [PATCH] nscd_stat.c: make the build reproducible
Date: Tue, 08 Mar 2016 17:06:00 -0000	[thread overview]
Message-ID: <1457456791-17402-1-git-send-email-aurelien@aurel32.net> (raw)

nscd_stat.c uses the __DATE__ and __TIME__ macros to make sure the
client and the server use the same format. This prevents reproducible
builds and fails to build with -Werror=date-time. In addition another
build of the same file a bit later does not necessary imply a change
in the format.

Instead compute a checksum of the file in the Makefile and pass it to
the preprocessor with the -D option. Use the md5sum command limited to
20 characters to avoid changing the structure size.

Changelog:
	* math/nscd (CPPFLAGS-nscd_stat.c): Append -DNSCD_STAT_C_CHECKSUM=...
	with the checksum of nscd_stat.c computed with md5sum.
	* nscd/nscd_stat.c (compilation): Use NSCD_STAT_C_CHECKSUM instead of
	__DATE__ " " __TIME__.
---
 ChangeLog        | 7 +++++++
 nscd/Makefile    | 4 ++++
 nscd/nscd_stat.c | 2 +-
 3 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/ChangeLog b/ChangeLog
index 289d578..8f0b24f 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,10 @@
+2016-03-08  Aurelien Jarno  <aurelien@aurel32.net>
+
+	* math/nscd (CPPFLAGS-nscd_stat.c): Append -DNSCD_STAT_C_CHECKSUM=...
+	with the checksum of nscd_stat.c computed with md5sum.
+	* nscd/nscd_stat.c (compilation): Use NSCD_STAT_C_CHECKSUM instead of
+	__DATE__ " " __TIME__.
+
 2016-03-08  H.J. Lu  <hongjiu.lu@intel.com>
 
 	[BZ #19783]
diff --git a/nscd/Makefile b/nscd/Makefile
index 50bad32..2678c6b 100644
--- a/nscd/Makefile
+++ b/nscd/Makefile
@@ -90,6 +90,10 @@ ifeq (yesyes,$(have-fpie)$(build-shared))
 LDFLAGS-nscd = -Wl,-z,now
 endif
 
+# Compute a checksum of the file to verify that the sender and the
+# receiver are using the same format.
+CPPFLAGS-nscd_stat.c += -DNSCD_STAT_C_CHECKSUM='"$(shell md5sum $< | cut -c1-20)"'
+
 # Set libof-nscd.
 cpp-srcs-left := $(nscd-modules)
 lib := nscd
diff --git a/nscd/nscd_stat.c b/nscd/nscd_stat.c
index f34c352..fa294ca 100644
--- a/nscd/nscd_stat.c
+++ b/nscd/nscd_stat.c
@@ -37,7 +37,7 @@
 
 
 /* We use this to make sure the receiver is the same.  */
-static const char compilation[21] = __DATE__ " " __TIME__;
+static const char compilation[21] = NSCD_STAT_C_CHECKSUM;
 
 /* Statistic data for one database.  */
 struct dbstat
-- 
2.7.0

             reply	other threads:[~2016-03-08 17:06 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-03-08 17:06 Aurelien Jarno [this message]
2016-03-08 23:37 ` Mike Frysinger
2016-03-09  7:54   ` Aurelien Jarno
2016-03-09 22:30     ` Mike Frysinger
2016-07-28 19:33       ` Florian Weimer
2016-07-29 13:30         ` Ludovic Courtès
2016-08-01  4:52         ` Mike Frysinger
2016-11-04 17:54           ` Ximin Luo
2016-11-04 18:14             ` Ximin Luo

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=1457456791-17402-1-git-send-email-aurelien@aurel32.net \
    --to=aurelien@aurel32.net \
    --cc=libc-alpha@sourceware.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 read-only IMAP folder(s) and NNTP newsgroup(s).