Openembedded Core Discussions
 help / color / mirror / Atom feed
From: Mark Hatle <mark.hatle@windriver.com>
To: <openembedded-core@lists.openembedded.org>
Subject: Re: [PATCH 16/22] populate_sdk: Fix the introduction of sdknative packages
Date: Mon, 1 Aug 2011 09:47:20 -0500	[thread overview]
Message-ID: <4E36BC78.9060007@windriver.com> (raw)
In-Reply-To: <1312204015.2344.434.camel@rex>

On 8/1/11 8:06 AM, Richard Purdie wrote:
> On Mon, 2011-08-01 at 00:36 -0700, Saul Wold wrote:
>> From: Mark Hatle <mark.hatle@windriver.com>
>>
>> The sdknative packages were not being introduced into the list of available
>> packages properly.  Fix this by always introducing it and avoiding a sed of
>> one type to another.
>>
>> Also fix a number of minor issues with the RPM behavior when building sdk
>> images.
> 
> I'm afraid this change looks wrong to me. Firstly, all arch packages are
> "all", not "all-nativesdk". You're also cutting down the arch list to a
> single sdk entry rather than the intended package_archs list.
> 
> "all" == "all-nativesdk" as things stand might be wrong for various
> other reasons I can think of but it is what would currently be
> generated.

No, that's not what is happening at all.  The original sed was completely broken
and wouldn't always work.

The arch list being processed is the target arch list.. typically it's something
like:

all any noarch powerpc ppc603e qemuppc

Before the system was trying to do a search/replace on "HOST_ARCH" and replace
it with SDK_ARCH..  this is kind of insane.. because there is no guaranty that
HOST_ARCH will even be in the arch list.  So instead the patch simply adds
SDK_ARCH-nativesdk, and then processes the list as normal.. the result is:

x86_64-nativesdk all any noarch powerpc ppc603e qemuppc

before it would have been (assuming the sed even worked)

all any noarch x86_64-nativesdk ppc603e qemuppc

so any packages with arch of "powerpc" would have been missed and the install
failed..  (or if powerpc didn't exist, it would end up without x86_64-nativesdk
even in the list.. which is how I found the problem.)

> All things considered, this issue needs more work.
> 
> Cheers,
> 
> Richard
> 
> 
> 
>> Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
>> ---
>>  meta/classes/package_rpm.bbclass      |   19 +++++++++----------
>>  meta/classes/populate_sdk_deb.bbclass |   11 ++---------
>>  meta/classes/populate_sdk_rpm.bbclass |   11 ++---------
>>  3 files changed, 13 insertions(+), 28 deletions(-)
>>
>> diff --git a/meta/classes/package_rpm.bbclass b/meta/classes/package_rpm.bbclass
>> index 0377100..856dfb0 100644
>> --- a/meta/classes/package_rpm.bbclass
>> +++ b/meta/classes/package_rpm.bbclass
>> @@ -32,14 +32,12 @@ package_update_index_rpm () {
>>  		eval archs=\${${archvar}}
>>  		packagedirs=""
>>  		packagedirs_sdk=""
>> -		for arch in $archs; do
>> -			sdkarch=`echo $arch | sed -e 's/${HOST_ARCH}/${SDK_ARCH}/'`
>> -			extension="-nativesdk"
>> -			if [ "$sdkarch" = "all" -o "$sdkarch" = "any" -o "$sdkarch" = "noarch" ]; then
>> -			    extension=""
>> -			fi
>> +		if [ "${SDK_ARCH}-nativesdk" != "-nativesdk" ]; then
>> +			packagedirs_sdk="${DEPLOY_DIR_RPM}/${SDK_ARCH}-nativesdk"
>> +		fi
>> +		for arch in $archs ; do
>>  			packagedirs="${DEPLOY_DIR_RPM}/$arch $packagedirs"
>> -			packagedirs_sdk="${DEPLOY_DIR_RPM}/$sdkarch$extension $packagedirs_sdk"
>> +			packagedirs_sdk="${DEPLOY_DIR_RPM}/$arch $packagedirs_sdk"
>>  
>>  			rm -rf ${DEPLOY_DIR_RPM}/$arch/solvedb
>>  			rm -rf ${DEPLOY_DIR_RPM}/$sdkarch$extension/solvedb
>> @@ -95,6 +93,7 @@ package_update_index_rpm () {
>>  #
>>  package_generate_rpm_conf () {
>>  	printf "_solve_dbpath " > ${RPMCONF_TARGET_BASE}.macro
>> +	printf "_solve_dbpath " > ${RPMCONF_HOST_BASE}.macro
>>  	o_colon_t=false
>>  	o_colon_h=false
>>  
>> @@ -238,7 +237,7 @@ package_install_internal_rpm () {
>>  					echo "Unable to find package $pkg ($ml_pkg)!"
>>  					exit 1
>>  				fi
>> -				echo $pkg_name >> ${IMAGE_ROOTFS}/install/install.manifest
>> +				echo $pkg_name >> ${target_rootfs}/install/install.manifest
>>  			done
>>  		fi
>>  	fi
>> @@ -258,7 +257,7 @@ package_install_internal_rpm () {
>>  				echo "Unable to find package $pkg ($ml_pkg)!"
>>  				exit 1
>>  			fi
>> -			echo $pkg_name >> ${IMAGE_ROOTFS}/install/install.manifest
>> +			echo $pkg_name >> ${target_rootfs}/install/install.manifest
>>  		done
>>  	fi
>>  
>> @@ -303,7 +302,7 @@ package_install_internal_rpm () {
>>  		# Dump the full set of recommends...
>>  		${RPM} --predefine "_rpmds_sysinfo_path ${target_rootfs}/etc/rpm/sysinfo" \
>>  			--predefine "_rpmrc_platform_path ${target_rootfs}/etc/rpm/platform" \
>> -			-D "_dbpath ${IMAGE_ROOTFS}/install" -D "`cat ${confbase}.macro`" \
>> +			-D "_dbpath ${target_rootfs}/install" -D "`cat ${confbase}.macro`" \
>>  			-D "__dbi_txn create nofsync private" \
>>  			-qa --qf "[%{RECOMMENDS}\n]" | sort -u > ${target_rootfs}/install/recommend
>>  		# Did we add more to the list?
>> diff --git a/meta/classes/populate_sdk_deb.bbclass b/meta/classes/populate_sdk_deb.bbclass
>> index a5b6384..9acc52d 100644
>> --- a/meta/classes/populate_sdk_deb.bbclass
>> +++ b/meta/classes/populate_sdk_deb.bbclass
>> @@ -34,16 +34,9 @@ fakeroot populate_sdk_deb () {
>>  	export INSTALL_ROOTFS_DEB="${SDK_OUTPUT}"
>>  	export INSTALL_BASEARCH_DEB="${SDK_ARCH}"
>>  	export INSTALL_PACKAGES_NORMAL_DEB="${TOOLCHAIN_HOST_TASK}"
>> -	INSTALL_ARCHS_DEB=""
>> +	INSTALL_ARCHS_DEB="${SDK_ARCH}-nativesdk"
>>  	for arch in ${PACKAGE_ARCHS}; do
>> -		sdkarch=`echo $arch | sed -e 's/${HOST_ARCH}/${SDK_ARCH}/'`
>> -		extension="-nativesdk"
>> -		if [ "$sdkarch" = "all" -o "$sdkarch" = "any" -o "$sdkarch" = "noarch" ]; then
>> -		    extension=""
>> -		fi
>> -		if [ -e ${DEPLOY_DIR_DEB}/$sdkarch$extension ]; then
>> -			INSTALL_ARCHS_DEB="$INSTALL_ARCHS_DEB $sdkarch$extension"
>> -		fi
>> +		INSTALL_ARCHS_DEB="$INSTALL_ARCHS_DEB $arch"
>>  	done
>>  	export INSTALL_ARCHS_DEB
>>  
>> diff --git a/meta/classes/populate_sdk_rpm.bbclass b/meta/classes/populate_sdk_rpm.bbclass
>> index da308c1..8196542 100644
>> --- a/meta/classes/populate_sdk_rpm.bbclass
>> +++ b/meta/classes/populate_sdk_rpm.bbclass
>> @@ -53,16 +53,9 @@ fakeroot populate_sdk_rpm () {
>>  	export INSTALL_PLATFORM_RPM="${SDK_ARCH}"
>>  	export INSTALL_CONFBASE_RPM="${RPMCONF_HOST_BASE}"
>>  	export INSTALL_PACKAGES_NORMAL_RPM="${TOOLCHAIN_HOST_TASK}"
>> -	export INSTALL_PLATFORM_EXTRA_RPM=""
>> +	export INSTALL_PLATFORM_EXTRA_RPM="${SDK_ARCH}-nativesdk"
>>  	for arch in ${PACKAGE_ARCHS}; do
>> -		sdkarch=`echo $arch | sed -e 's/${HOST_ARCH}/${SDK_ARCH}/'`
>> -		extension="-nativesdk"
>> -		if [ "$sdkarch" = "all" -o "$sdkarch" = "any" -o "$sdkarch" = "noarch" ]; then
>> -		    extension=""
>> -		fi
>> -		if [ -e ${DEPLOY_DIR_RPM}/$sdkarch$extension ]; then
>> -			INSTALL_PLATFORM_EXTRA_RPM="$sdkarch $INSTALL_PLATFORM_EXTRA_RPM"
>> -		fi
>> +		INSTALL_PLATFORM_EXTRA_RPM="$arch $INSTALL_PLATFORM_EXTRA_RPM"
>>  	done
>>  	export INSTALL_PLATFORM_EXTRA_RPM
>>  
> 
> 
> 
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core




  reply	other threads:[~2011-08-01 14:51 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-08-01  7:35 [CONSOLIDATED PATCH 00/22] Fixes Multi-Lib & M3 Saul Wold
2011-08-01  7:35 ` [PATCH 01/22] meta-environment: set correct *MULTIMACH_TARGET_SYS Saul Wold
2011-08-01  7:35 ` [PATCH 02/22] glibc: deleted Saul Wold
2011-08-01  7:35 ` [PATCH 03/22] libzypp: Fix variable substitution problem in do_archgen step Saul Wold
2011-08-01  7:35 ` [PATCH 04/22] docbook-sgml-dtd-native: Write a bak file for shared state Saul Wold
2011-08-01  7:35 ` [PATCH 05/22] docbook-dsssl-stylesheets-native: " Saul Wold
2011-08-01  7:35 ` [PATCH 06/22] openjade-native: " Saul Wold
2011-08-01  7:35 ` [PATCH 07/22] sgml-common-native: Read a catalog backup to repopulate sgml-docbook.cat Saul Wold
2011-08-01  7:36 ` [PATCH 08/22] tune/arch-powerpc64: Fix typo with 64-bit TUNE_CCARGS handling Saul Wold
2011-08-01  7:36 ` [PATCH 09/22] tune/arch-powerpc64: Remove support for soft-float from ppc64 Saul Wold
2011-08-01  7:36 ` [PATCH 10/22] tune/arch-powerpc: Use fpu-soft to set PPCPKGSFX_FPU Saul Wold
2011-08-01 13:03   ` Richard Purdie
2011-08-01 14:13     ` Kumar Gala
2011-08-01 14:50       ` Mark Hatle
2011-08-01 15:57         ` Kumar Gala
2011-08-02 18:31     ` Kumar Gala
2011-08-02 22:36       ` Richard Purdie
2011-08-01  7:36 ` [PATCH 11/22] tune-ppce500mc: Adjust PowerPC e500mc tune file to correctly set hard fpu Saul Wold
2011-08-01  7:36 ` [PATCH 12/22] sanity.bbclass: Add sanity check that TUNE_PKGARCH appears in PACKAGE_ARCHS Saul Wold
2011-08-01  7:36 ` [PATCH 13/22] sanity: Dont throw a warning if TUNE_PKGARCH is in PACKAGE_ARCHS twice Saul Wold
2011-08-01 13:03   ` Richard Purdie
2011-08-01 13:28     ` Koen Kooi
2011-08-01 14:11     ` Kumar Gala
2011-08-01 17:34       ` Khem Raj
2011-08-01 18:51         ` Kumar Gala
2011-08-01 18:59           ` Khem Raj
2011-08-02 13:39             ` Richard Purdie
2011-08-01  7:36 ` [PATCH 14/22] libzypp: Bump PR Saul Wold
2011-08-01  7:36 ` [PATCH 15/22] tune/powerpc: Cleanup setting of TUNE_PKGARCH Saul Wold
2011-08-01  9:06   ` Richard Purdie
2011-08-01  7:36 ` [PATCH 16/22] populate_sdk: Fix the introduction of sdknative packages Saul Wold
2011-08-01 13:06   ` Richard Purdie
2011-08-01 14:47     ` Mark Hatle [this message]
2011-08-01  7:36 ` [PATCH 17/22] bitbake.conf: Add default tune-${DEFAULTTUNE} override Saul Wold
2011-08-01  9:00   ` Richard Purdie
2011-08-01  9:09   ` Phil Blundell
2011-08-01 14:36     ` Mark Hatle
2011-08-01  7:36 ` [PATCH 18/22] rootfs_rpm.bbclass: Fix multilib configurations to use DEFAULTTUNE Saul Wold
2011-08-01  7:36 ` [PATCH 19/22] package_{deb, rpm, ipk}.bbclass: fix 'lingusa' typo Saul Wold
2011-08-01  7:36 ` [PATCH 20/22] arch-armv7a.inc: fix armv7a-vfp-neon -> armv7a compat case Saul Wold
2011-08-01 13:10   ` Richard Purdie
2011-08-01 13:35     ` Koen Kooi
2011-08-01  7:36 ` [PATCH 21/22] external-*toolchain: remove glibc-utils provide Saul Wold
2011-08-01 13:11   ` Richard Purdie
2011-08-01  7:36 ` [PATCH 22/22] feature-arm-thumb: Take ARM_INSTRUCTION_SET into account to decide thumb mode Saul Wold
2011-08-01 13:14   ` Richard Purdie
2011-08-01 17:46     ` Khem Raj
2011-08-05 10:59     ` Phil Blundell
2011-08-05 11:39       ` Richard Purdie
2011-08-05 14:34         ` Phil Blundell
2011-08-05 15:23           ` Khem Raj
2011-08-05 16:23             ` Richard Purdie
2011-08-01 13:20 ` [CONSOLIDATED PATCH 00/22] Fixes Multi-Lib & M3 Richard Purdie

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=4E36BC78.9060007@windriver.com \
    --to=mark.hatle@windriver.com \
    --cc=openembedded-core@lists.openembedded.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