From: Arnd Bergmann <arnd@arndb.de>
To: Sam Ravnborg <sam@ravnborg.org>
Cc: lkml <linux-kernel@vger.kernel.org>,
linux-arch <linux-arch@vger.kernel.org>
Subject: Re: [PATCH 38/38] asm-generic: introduc Kbuild.generic
Date: Tue, 15 Jul 2014 09:03 +0200 [thread overview]
Message-ID: <4667634.oiTEQ4v5vj@wuerfel> (raw)
In-Reply-To: <1405350546-28467-38-git-send-email-sam@ravnborg.org>
On Monday 14 July 2014 17:09:06 Sam Ravnborg wrote:
> Kbuild.generic is for asm-generic files that are used
> by all architectures.
> So rather than adding the file to the individual Kbuild files
> it is enough to add it to this file
>
> Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
> Cc: Arnd Bergmann <arnd@arndb.de>
This has been suggested in the past, and I remember there being some
controversy around it, but I don't really mind it, so you have my
ok in principle. Two points though that we have to work out first:
> diff --git a/arch/alpha/include/asm/Kbuild b/arch/alpha/include/asm/Kbuild
> index 96e54be..e09f8e7 100644
> --- a/arch/alpha/include/asm/Kbuild
> +++ b/arch/alpha/include/asm/Kbuild
> @@ -1,9 +1,9 @@
> -
> +# Generic wrappers from include/asm-generic
> +include include/asm-generic/Kbuild.generic
We have the same kind of header for the uapi files, it's called
include/uapi/asm-generic/Kbuild.asm. I don't know why I picked that
name back then, Kbuild.generic seems more appropriate, but I think
either way it would be good to be consistent.
Can you either rename the existing file to Kbuild.generic, or call
the new one Kbuild.asm as well?
> diff --git a/include/asm-generic/Kbuild.generic b/include/asm-generic/Kbuild.generic
> new file mode 100644
> index 0000000..f44815b
> --- /dev/null
> +++ b/include/asm-generic/Kbuild.generic
> @@ -0,0 +1,6 @@
> +# List of generic header files that is used by all architectures
> +# This file is included by arch/*/include/asm/Kbuild
> +#
> +# Keep the list sorted alphabetically
> +
> +generic-y += mcs_spinlock.h
We have to make up our minds what we want from this list. At the moment,
you list only one file because that is the only file that is shared across
all architectures. The idea for this file is that architectures /can/
provide their own (according to the comments in this file), but no
architecture actually does. A few thoughts on this:
* Normally we just delete any header file that is common across all
architectures, and change the files including it (only one is present
in this case) not to include it. If we do that, your new file becomes
pointless.
* If one architecture actually starts providing its own version of this
file, we'd have to revert your patch first, which really defeats the
purpose of doing it in the first place.
* We could improve the situation if the definition of this list is
changed so we only generate the generic header files that are not
already provided by an architecture. If we do this, we can consolidate
a lot more here, which is nice, but the cost would be equivalent to
another level of indirection and a potential for subtle bugs if it's
not done very careful.
Arnd
next prev parent reply other threads:[~2014-07-15 7:03 UTC|newest]
Thread overview: 84+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-07-14 15:06 [RFC PATCH 0/38] clean-up uapi + asm-generic Kbuild files Sam Ravnborg
2014-07-14 15:08 ` [PATCH 01/38] cris: fix headers_install Sam Ravnborg
2014-07-14 15:08 ` [PATCH 02/38] openrisc: drop ucontext from list exported headers Sam Ravnborg
2014-07-14 15:08 ` [PATCH 03/38] hexagon: " Sam Ravnborg
2014-07-14 15:08 ` Sam Ravnborg
2014-07-28 16:19 ` rkuo
2014-07-14 15:08 ` [PATCH 04/38] score: drop irrelevant assignment from Kbuild file Sam Ravnborg
2014-07-15 13:57 ` Lennox Wu
2014-07-15 13:57 ` Lennox Wu
2014-07-14 15:08 ` [PATCH 05/38] tile: clean up Kbuild files Sam Ravnborg
2014-07-17 19:41 ` Chris Metcalf
2014-07-17 19:41 ` Chris Metcalf
2014-07-14 15:08 ` [PATCH 06/38] x86: clean-up Kbuild file - drop unused assignments Sam Ravnborg
2014-07-14 22:50 ` H. Peter Anvin
2014-07-14 15:08 ` [PATCH 07/38] Delete unused Kbuild files in include/* Sam Ravnborg
2014-07-15 9:06 ` James Hogan
2014-07-15 9:19 ` Sam Ravnborg
2014-07-14 15:08 ` [PATCH 08/38] kbuild: drop support for non uapi Kbuild files in headers_install Sam Ravnborg
2014-07-14 15:08 ` Sam Ravnborg
2014-07-14 15:08 ` [PATCH 09/38] uapi: check for optional files which are generated Sam Ravnborg
2014-07-14 15:08 ` [PATCH 10/38] uapi: drop obsolete logic for generated headers Sam Ravnborg
2014-07-14 15:08 ` [PATCH 11/38] alpha: clean-up uapi Kbuild file Sam Ravnborg
2014-07-14 15:08 ` [PATCH 12/38] arm: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 13/38] arm64: " Sam Ravnborg
2014-07-14 15:08 ` Sam Ravnborg
2014-07-14 17:02 ` Catalin Marinas
2014-07-15 10:14 ` James Hogan
2014-07-14 15:08 ` [PATCH 14/38] avr32: " Sam Ravnborg
2014-07-14 17:53 ` Hans-Christian Egtvedt
2014-07-14 15:08 ` [PATCH 15/38] blackfin: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 16/38] c6x: " Sam Ravnborg
2014-07-14 15:08 ` Sam Ravnborg
2014-07-14 20:40 ` Mark Salter
2014-07-14 15:08 ` [PATCH 17/38] cris: " Sam Ravnborg
2014-07-14 15:08 ` Sam Ravnborg
2014-08-11 5:35 ` Mikael Starvik
2014-09-03 11:23 ` Jesper Nilsson
2014-09-03 11:23 ` Jesper Nilsson
2014-07-14 15:08 ` [PATCH 18/38] frv: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 19/38] hexagon: " Sam Ravnborg
2014-07-14 15:08 ` Sam Ravnborg
2014-07-28 16:19 ` rkuo
2014-07-14 15:08 ` [PATCH 20/38] ia64: " Sam Ravnborg
2014-07-14 15:08 ` Sam Ravnborg
2014-07-14 15:08 ` [PATCH 21/38] m32r: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 22/38] m68k: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 23/38] metag: " Sam Ravnborg
[not found] ` <1405350546-28467-23-git-send-email-sam-uyr5N9Q2VtJg9hUCZPvPmw@public.gmane.org>
2014-07-15 9:22 ` James Hogan
2014-07-15 9:22 ` James Hogan
2014-07-14 15:08 ` [PATCH 24/38] microblaze: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 25/38] mips: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 26/38] mn10300: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 27/38] openrisc: " Sam Ravnborg
2014-07-14 15:08 ` Sam Ravnborg
2014-07-14 15:08 ` [PATCH 28/38] parisc: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 29/38] powerpc: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 30/38] s390: " Sam Ravnborg
2014-07-14 15:08 ` [PATCH 31/38] score: " Sam Ravnborg
2014-07-14 15:08 ` Sam Ravnborg
2014-07-18 15:19 ` Lennox Wu
2014-07-18 15:19 ` Lennox Wu
2014-07-14 15:09 ` [PATCH 32/38] sh: " Sam Ravnborg
2014-07-14 15:09 ` [PATCH 33/38] sparc: " Sam Ravnborg
2014-07-14 15:09 ` Sam Ravnborg
2014-07-14 20:02 ` David Miller
2014-07-14 15:09 ` [PATCH 34/38] tile: " Sam Ravnborg
2014-07-17 19:44 ` Chris Metcalf
2014-07-17 19:44 ` Chris Metcalf
2014-07-14 15:09 ` [PATCH 35/38] unicore32: " Sam Ravnborg
2014-07-15 10:53 ` James Hogan
2014-07-15 11:07 ` Sam Ravnborg
2014-07-14 15:09 ` [PATCH 36/38] x86: " Sam Ravnborg
2014-07-14 15:09 ` Sam Ravnborg
2014-07-14 22:52 ` H. Peter Anvin
2014-07-14 15:09 ` [PATCH 37/38] xtensa: " Sam Ravnborg
2014-07-14 15:09 ` [PATCH 38/38] asm-generic: introduc Kbuild.generic Sam Ravnborg
2014-07-14 15:09 ` Sam Ravnborg
2014-07-15 7:03 ` Arnd Bergmann [this message]
2014-07-17 10:38 ` Sam Ravnborg
2014-07-17 10:50 ` Arnd Bergmann
2014-08-11 5:35 ` [PATCH 01/38] cris: fix headers_install Mikael Starvik
2014-09-03 11:22 ` Jesper Nilsson
2014-07-14 15:34 ` [RFC PATCH 0/38] clean-up uapi + asm-generic Kbuild files Arnd Bergmann
2014-07-14 22:14 ` Sam Ravnborg
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=4667634.oiTEQ4v5vj@wuerfel \
--to=arnd@arndb.de \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=sam@ravnborg.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