| Message ID | 1362484056-9778-6-git-send-email-kyungsik.lee@lge.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Tue, Mar 05, 2013 at 08:47:36PM +0900, Kyungsik Lee wrote: > This patch makes x86 and arm kernels default to the LZ4-compressed > to test new LZ4 code in the linux-next. This is requested by > Andrew Morton. > > Signed-off-by: Kyungsik Lee <kyungsik.lee@lge.com> > --- > init/Kconfig | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/init/Kconfig b/init/Kconfig > index fc8eb1f..de3cb00 100644 > --- a/init/Kconfig > +++ b/init/Kconfig > @@ -138,7 +138,7 @@ config HAVE_KERNEL_LZ4 > > choice > prompt "Kernel compression mode" > - default KERNEL_GZIP > + default KERNEL_LZ4 Do I understand it correctly that when this thing gets committed, kernel builds on all systems missing the lz4 tool will fail and everybody should go running to get it so that everybody can build kernels again? WTF?
Hello, On 03/05/2013 12:47 PM, Kyungsik Lee wrote: > This patch makes x86 and arm kernels default to the LZ4-compressed > to test new LZ4 code in the linux-next. This is requested by > Andrew Morton. I do not think making this the default is good idea, because the lz4demo utility that you need to actually compress the kernel (used in patch 2) is not installed nor available by default on most systems, while gzip is. Considering that such a concern was already raised in your v2 patchset, I am surprised to see this. -- Florian -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
I'm ok with this only if the patch is clearly marked LINUS DO NOT MERGE. Kyungsik Lee <kyungsik.lee@lge.com> wrote: >This patch makes x86 and arm kernels default to the LZ4-compressed >to test new LZ4 code in the linux-next. This is requested by >Andrew Morton. > >Signed-off-by: Kyungsik Lee <kyungsik.lee@lge.com> >--- > init/Kconfig | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > >diff --git a/init/Kconfig b/init/Kconfig >index fc8eb1f..de3cb00 100644 >--- a/init/Kconfig >+++ b/init/Kconfig >@@ -138,7 +138,7 @@ config HAVE_KERNEL_LZ4 > > choice > prompt "Kernel compression mode" >- default KERNEL_GZIP >+ default KERNEL_LZ4 > depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA >|| HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 > help > The linux kernel is a kind of self-extracting executable.
On Tue, 05 Mar 2013 13:04:20 +0100 Florian Fainelli <florian@openwrt.org> wrote: > Hello, > > On 03/05/2013 12:47 PM, Kyungsik Lee wrote: > > This patch makes x86 and arm kernels default to the LZ4-compressed > > to test new LZ4 code in the linux-next. This is requested by > > Andrew Morton. > > I do not think making this the default is good idea, because the lz4demo > utility that you need to actually compress the kernel (used in patch 2) > is not installed nor available by default on most systems, while gzip > is. Yes, that's a showstopper even for linux-next. What a shame. It means this code will walk into mainline with practically zero third-party testing. Oh well, the risk is minor. If it breaks, people can switch back to gzip easily enough. -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 03/06/2013 12:08 AM, Andrew Morton wrote: > On Tue, 05 Mar 2013 13:04:20 +0100 Florian Fainelli <florian@openwrt.org> wrote: > >> Hello, >> >> On 03/05/2013 12:47 PM, Kyungsik Lee wrote: >>> This patch makes x86 and arm kernels default to the LZ4-compressed >>> to test new LZ4 code in the linux-next. This is requested by >>> Andrew Morton. >> >> I do not think making this the default is good idea, because the lz4demo >> utility that you need to actually compress the kernel (used in patch 2) >> is not installed nor available by default on most systems, while gzip >> is. > > Yes, that's a showstopper even for linux-next. What a shame. > > It means this code will walk into mainline with practically zero > third-party testing. Oh well, the risk is minor. If it breaks, people > can switch back to gzip easily enough. If we want to have more coverage, how about importing the lz4demo utility source code to tools/lz4/, and change scripts/Makefile.lib to point at it? Once the lz4 utilities have reached a broader audience and are available with most distributions, we could revert back to assuming this utility can be found in PATH. -- Florian -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, Mar 06, 2013 at 10:57:52AM +0100, Florian Fainelli wrote: > If we want to have more coverage, how about importing the lz4demo > utility source code to tools/lz4/ That looks like on overkill to me, besides I don't think that's the purpose of tools/. Maybe everyone who wants to test this, should install the tool and go wild without disturbing the default. Thanks.
diff --git a/init/Kconfig b/init/Kconfig index fc8eb1f..de3cb00 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -138,7 +138,7 @@ config HAVE_KERNEL_LZ4 choice prompt "Kernel compression mode" - default KERNEL_GZIP + default KERNEL_LZ4 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 help The linux kernel is a kind of self-extracting executable.
This patch makes x86 and arm kernels default to the LZ4-compressed to test new LZ4 code in the linux-next. This is requested by Andrew Morton. Signed-off-by: Kyungsik Lee <kyungsik.lee@lge.com> --- init/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)