From: Michal Marek <mmarek@suse.cz>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>,
Andrew Morton <akpm@linux-foundation.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>
Subject: Re: [RFC PATCH] Don't reset timestamps in include/generated if not needed
Date: Wed, 11 Mar 2015 11:01:01 +0100 [thread overview]
Message-ID: <20150311100101.GA30787@sepie.suse.cz> (raw)
In-Reply-To: <CA+55aFx368q6kAuNV_7zskFdyh2yxYfpm77Yak5WQiexin_d9A@mail.gmail.com>
On Sun, Mar 08, 2015 at 04:08:06PM -0700, Linus Torvalds wrote:
> On Tue, Mar 3, 2015 at 5:07 PM, Valdis Kletnieks
> <Valdis.Kletnieks@vt.edu> wrote:
> >
> > Kbuild regenerates bounds.h and asm-offsets.h, resetting the timestamps
> > and forcing rebuilds even if the contents haven't changed. Add a bit of
> > shell magic to only replace the file if the contents have in fact changed,
> > which should speed up git bisects and similar.
> > ...
> > RFC because I can't wrap my head around why this wasn't done ages ago.
> > If I'm missing something something obvious, please apply a cluestick. :)
We do not regenerate the files *always*, but only if bounds.c /
asm-offsets.c or one of the headers they include change. So the rule is
not any worse than the rule for compiling a regular .c file into an .o
file. You can use ccache to avoid repeated rebuilds.
However, I can see that especially bounds.h ends up being included in
many places, so cutting its dependencies helps in some cases.
> > Lightly tested - if I rm one of those two files, it gets rebuilt. If I
> > insert some whitespace, it gets replaced. If I don't touch it, the datestamp
> > doesn't change.
>
> I don't think this is wrong, but I'd really prefer to do the whole
> move-if-changed thing as a separate rule, and a separate command.
>
> IOW, could we please have something that generates the "build.h" file
> by separately creating a new temporary file, and then having the
> traditional kind of move-if-changed definition
>
> define move-if-changed
> if ! cmp -s $(1) $(2); then mv -f $(1) $(2); else rm -f $(1); fi
> endef
We already have it and it is called "filechk." Valdis, can you check if
the below patch works equally well for you?
diff --git a/Kbuild b/Kbuild
index ab8ded9..47d0630 100644
--- a/Kbuild
+++ b/Kbuild
@@ -13,8 +13,9 @@ define sed-y
s:->::; p;}"
endef
-quiet_cmd_offsets = GEN $@
-define cmd_offsets
+# Use filechk to avoid rebuilds when a header changes, but the resulting file
+# does not
+define filechk_offsets
(set -e; \
echo "#ifndef $2"; \
echo "#define $2"; \
@@ -24,9 +25,9 @@ define cmd_offsets
echo " * This file was generated by Kbuild"; \
echo " */"; \
echo ""; \
- sed -ne $(sed-y) $<; \
+ sed -ne $(sed-y); \
echo ""; \
- echo "#endif" ) > $@
+ echo "#endif" )
endef
#####
@@ -44,7 +45,7 @@ kernel/bounds.s: kernel/bounds.c FORCE
$(obj)/$(bounds-file): kernel/bounds.s Kbuild
$(Q)mkdir -p $(dir $@)
- $(call cmd,offsets,__LINUX_BOUNDS_H__)
+ $(call filechk,offsets,__LINUX_BOUNDS_H__)
#####
# 2) Generate asm-offsets.h
@@ -63,7 +64,7 @@ arch/$(SRCARCH)/kernel/asm-offsets.s: arch/$(SRCARCH)/kernel/asm-offsets.c \
$(call if_changed_dep,cc_s_c)
$(obj)/$(offsets-file): arch/$(SRCARCH)/kernel/asm-offsets.s Kbuild
- $(call cmd,offsets,__ASM_OFFSETS_H__)
+ $(call filechk,offsets,__ASM_OFFSETS_H__)
#####
# 3) Check for missing system calls
next prev parent reply other threads:[~2015-03-11 10:01 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <15447.1425431256@turing-police.cc.vt.edu>
2015-03-08 23:08 ` [RFC PATCH] Don't reset timestamps in include/generated if not needed Linus Torvalds
2015-03-11 10:01 ` Michal Marek [this message]
2015-03-13 4:59 ` Masahiro Yamada
2015-03-13 10:00 ` Michal Marek
2015-03-24 15:36 ` [PATCH] kbuild: " Michal Marek
2015-03-24 16:08 ` Valdis.Kletnieks
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=20150311100101.GA30787@sepie.suse.cz \
--to=mmarek@suse.cz \
--cc=Valdis.Kletnieks@vt.edu \
--cc=akpm@linux-foundation.org \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=torvalds@linux-foundation.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