[PATCH 0/4] Fix builds for MIPS64 N32

[PATCH 0/4] Fix builds for MIPS64 N32

[Dmitry Eremin-Solenikov]

Quite simple configuration (MIPS64 N32 image) fails to build (see below).
I have verified these patches in both single lib and in multilib cases.
Tri-lib images and SDK for MIPS build fine with these patches.

Current situation:

lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
NOTE: Started PRServer with DBfile:
/home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
15895
Loading cache: 100%
|########################################################################################################################|
ETA:  00:00:00
Loaded 1302 entries from dependency cache.
NOTE: Resolving any missing task queue dependencies
ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
/home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
RDEPENDS on or otherwise requires it)
NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
removing...
Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
unbuildable, removing...
Missing or unbuildable dependency chain was:
['packagegroup-cross-canadian-qemumips64n32',
'binutils-cross-canadian-mips64']
ERROR: Required build target 'core-image-base' has no buildable providers.
Missing or unbuildable dependency chain was: ['core-image-base',
'packagegroup-cross-canadian-qemumips64n32',
'binutils-cross-canadian-mips64']

[PATCH 1/4] arch-mips.inc: don't override TRANSLATED_TARGET_ARCH

[Dmitry Eremin-Solenikov]

Currently MIPS64 N32 is broken. There is internal disagreement
between TARGET_ARCH (which doesn't contain ABIEXTENSION) and
TRANSLATED_TARGET_ARCH (which contains ABIEXTENSION). ABI is already
encoded into the TARGET_OS. ARM tunes in the same situation override
neither the TARGET_ARCH nor the TRANSLATED_TARGET_ARCH. So let's drop
this override.

Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
---
 meta/conf/machine/include/mips/arch-mips.inc | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/meta/conf/machine/include/mips/arch-mips.inc b/meta/conf/machine/include/mips/arch-mips.inc
index c41fa5e..02626d3 100644
--- a/meta/conf/machine/include/mips/arch-mips.inc
+++ b/meta/conf/machine/include/mips/arch-mips.inc
@@ -100,7 +100,3 @@ TUNE_FEATURES_tune-mips64el-nf = "n64"
 BASE_LIB_tune-mips64el-nf = "lib64"
 MIPSPKGSFX_VARIANT_tune-mips64el-nf = "${TUNE_ARCH}"
 PACKAGE_EXTRA_ARCHS_tune-mips64el-nf = "mips64el-nf"
-
-# On mips we need to redefine this to include the ABIEXTENSION
-# we can avoid the python bit as there are no _ or - to translate
-TRANSLATED_TARGET_ARCH = "${TARGET_ARCH}${ABIEXTENSION}"
-- 
1.9.1

[PATCH 2/4] packagegroups-cross-canadian: don't override TRANSLATED_TARGET_ARCH

[Dmitry Eremin-Solenikov]

There is no point in overriding TRANSLATED_TARGET_ARCH in
packagegroups-cross-canadian recipe. The cross-canadian class sets the
PACKAGE_ARCH variable, thus allarch class (inherited through
packagegroup class) doesn't change variables.

Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
---
 meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb b/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
index e3b1c18..34e6e2b 100644
--- a/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
+++ b/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
@@ -2,9 +2,6 @@ SUMMARY = "Host SDK package for cross canadian toolchain"
 PN = "packagegroup-cross-canadian-${MACHINE}"
 LICENSE = "MIT"
 
-# Save TRANSLATED_TARGET_ARCH before allarch tramples it
-TRANSLATED_TARGET_ARCH = "${@d.getVar('TUNE_ARCH', True).replace('_', '-')}"
-
 inherit cross-canadian packagegroup
 
 PACKAGEGROUP_DISABLE_COMPLEMENTARY = "1"
-- 
1.9.1

[PATCH 3/4] cross.bbclass: override MULTIMACH_TARGET_SYS

[Dmitry Eremin-Solenikov]

Cross packages are built for BUILD_VENDOR/BUILD_OS, rather than
TARGET_VENDOR/TARGET_OS. E.g. there is no point in hardcoding the
target's LIBCEXTENSION and ABIEXTENSION into the STAMPDIR/WORKDIR
variables (set using MULTIMACH_TARGET_SYS).

Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
---
 meta/classes/cross.bbclass | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/meta/classes/cross.bbclass b/meta/classes/cross.bbclass
index 28fd211..5ee32b2 100644
--- a/meta/classes/cross.bbclass
+++ b/meta/classes/cross.bbclass
@@ -21,6 +21,8 @@ STAGING_DIR_HOST = "${STAGING_DIR}/${HOST_ARCH}${HOST_VENDOR}-${HOST_OS}"
 
 PACKAGE_ARCH = "${BUILD_ARCH}"
 
+MULTIMACH_TARGET_SYS = "${PACKAGE_ARCH}${BUILD_VENDOR}-${BUILD_OS}"
+
 export PKG_CONFIG_DIR = "${exec_prefix}/lib/pkgconfig"
 export PKG_CONFIG_SYSROOT_DIR = ""
 
-- 
1.9.1

[PATCH 4/4] glibc: don't override TUNE_CCARGS for MIPS

[Dmitry Eremin-Solenikov]

Overriding TUNE_CCARGS this way breaks MULTILIB setup for MIPS. This
override disables multilib handling of tunes for TUNE_CCARGS, thus
enforcing glibc's TUNE_CCARGS to the TUNE_CCARGS of main DEFAULTTUNE.

Glibc perfectly build without this override for both simple and multilib
cases.

Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
---
 meta/recipes-core/glibc/glibc_2.21.bb | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/meta/recipes-core/glibc/glibc_2.21.bb b/meta/recipes-core/glibc/glibc_2.21.bb
index 8197c29..52252b3 100644
--- a/meta/recipes-core/glibc/glibc_2.21.bb
+++ b/meta/recipes-core/glibc/glibc_2.21.bb
@@ -95,16 +95,6 @@ do_patch_append() {
     bb.build.exec_func('do_fix_readlib_c', d)
 }
 
-# for mips glibc now builds syscall tables for all abi's
-# so we make sure that we choose right march option which is
-# compatible with o32,n32 and n64 abi's
-# e.g. -march=mips32 is not compatible with n32 and n64 therefore
-# we filter it out in such case -march=from-abi which will be
-# mips1 when using o32 and mips3 when using n32/n64
-
-TUNE_CCARGS_mips := "${@oe_filter_out('-march=mips32', '${TUNE_CCARGS}', d)}"
-TUNE_CCARGS_mipsel := "${@oe_filter_out('-march=mips32', '${TUNE_CCARGS}', d)}"
-
 do_fix_readlib_c () {
 	sed -i -e 's#OECORE_KNOWN_INTERPRETER_NAMES#${EGLIBC_KNOWN_INTERPRETER_NAMES}#' ${S}/elf/readlib.c
 }
-- 
1.9.1

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Dmitry Eremin-Solenikov]

Hello,

2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
> I have verified these patches in both single lib and in multilib cases.
> Tri-lib images and SDK for MIPS build fine with these patches.

Any comments on this patchset? It would be nice to have N32 ubroken in
the current
master tree.

>
> Current situation:
>
> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
> NOTE: Started PRServer with DBfile:
> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
> 15895
> Loading cache: 100%
> |########################################################################################################################|
> ETA:  00:00:00
> Loaded 1302 entries from dependency cache.
> NOTE: Resolving any missing task queue dependencies
> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
> RDEPENDS on or otherwise requires it)
> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
> removing...
> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
> unbuildable, removing...
> Missing or unbuildable dependency chain was:
> ['packagegroup-cross-canadian-qemumips64n32',
> 'binutils-cross-canadian-mips64']
> ERROR: Required build target 'core-image-base' has no buildable providers.
> Missing or unbuildable dependency chain was: ['core-image-base',
> 'packagegroup-cross-canadian-qemumips64n32',
> 'binutils-cross-canadian-mips64']
>



-- 
With best wishes
Dmitry

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Richard Purdie]

On Wed, 2015-07-08 at 15:00 +0300, Dmitry Eremin-Solenikov wrote:
> Hello,
> 
> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
> > Quite simple configuration (MIPS64 N32 image) fails to build (see below).
> > I have verified these patches in both single lib and in multilib cases.
> > Tri-lib images and SDK for MIPS build fine with these patches.
> 
> Any comments on this patchset? It would be nice to have N32 ubroken in
> the current
> master tree.

Mark, can you review these patches please?

Cheers,

Richard

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Mark Hatle]

On 7/8/15 7:00 AM, Dmitry Eremin-Solenikov wrote:
> Hello,
> 
> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
>> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
>> I have verified these patches in both single lib and in multilib cases.
>> Tri-lib images and SDK for MIPS build fine with these patches.
> 
> Any comments on this patchset? It would be nice to have N32 ubroken in
> the current
> master tree.

Sorry, I'm traveling right now and haven't had time to review this.  I should be
able to get back to the office and test it by the end of the week.

I visually reviewed the patches and I don't object to them going in -- but I'm
still not sure they are all correct.

--Mark

>>
>> Current situation:
>>
>> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
>> NOTE: Started PRServer with DBfile:
>> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
>> 15895
>> Loading cache: 100%
>> |########################################################################################################################|
>> ETA:  00:00:00
>> Loaded 1302 entries from dependency cache.
>> NOTE: Resolving any missing task queue dependencies
>> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
>> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
>> RDEPENDS on or otherwise requires it)
>> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
>> removing...
>> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
>> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
>> unbuildable, removing...
>> Missing or unbuildable dependency chain was:
>> ['packagegroup-cross-canadian-qemumips64n32',
>> 'binutils-cross-canadian-mips64']
>> ERROR: Required build target 'core-image-base' has no buildable providers.
>> Missing or unbuildable dependency chain was: ['core-image-base',
>> 'packagegroup-cross-canadian-qemumips64n32',
>> 'binutils-cross-canadian-mips64']
>>
> 
> 
> 

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Dmitry Eremin-Solenikov]

2015-07-08 15:42 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
> On 7/8/15 7:00 AM, Dmitry Eremin-Solenikov wrote:
>> Hello,
>>
>> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
>>> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
>>> I have verified these patches in both single lib and in multilib cases.
>>> Tri-lib images and SDK for MIPS build fine with these patches.
>>
>> Any comments on this patchset? It would be nice to have N32 ubroken in
>> the current
>> master tree.
>
> Sorry, I'm traveling right now and haven't had time to review this.  I should be
> able to get back to the office and test it by the end of the week.
>
> I visually reviewed the patches and I don't object to them going in -- but I'm
> still not sure they are all correct.

Re-ping. I'd like to receive a confirmation from you (and thus get the
patches merged)
or to discuss, what seems wrong to you.

>
> --Mark
>
>>>
>>> Current situation:
>>>
>>> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
>>> NOTE: Started PRServer with DBfile:
>>> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
>>> 15895
>>> Loading cache: 100%
>>> |########################################################################################################################|
>>> ETA:  00:00:00
>>> Loaded 1302 entries from dependency cache.
>>> NOTE: Resolving any missing task queue dependencies
>>> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
>>> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
>>> RDEPENDS on or otherwise requires it)
>>> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
>>> removing...
>>> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
>>> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
>>> unbuildable, removing...
>>> Missing or unbuildable dependency chain was:
>>> ['packagegroup-cross-canadian-qemumips64n32',
>>> 'binutils-cross-canadian-mips64']
>>> ERROR: Required build target 'core-image-base' has no buildable providers.
>>> Missing or unbuildable dependency chain was: ['core-image-base',
>>> 'packagegroup-cross-canadian-qemumips64n32',
>>> 'binutils-cross-canadian-mips64']
>>>
>>
>>
>>
>



-- 
With best wishes
Dmitry

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Mark Hatle]

Between customer escalations, other work and vacation.. I didn't happen.

I've started testing it this morning and should have results tomorrow.. (US time).

--Mark

On 7/20/15 7:11 AM, Dmitry Eremin-Solenikov wrote:
> 2015-07-08 15:42 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>> On 7/8/15 7:00 AM, Dmitry Eremin-Solenikov wrote:
>>> Hello,
>>>
>>> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
>>>> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
>>>> I have verified these patches in both single lib and in multilib cases.
>>>> Tri-lib images and SDK for MIPS build fine with these patches.
>>>
>>> Any comments on this patchset? It would be nice to have N32 ubroken in
>>> the current
>>> master tree.
>>
>> Sorry, I'm traveling right now and haven't had time to review this.  I should be
>> able to get back to the office and test it by the end of the week.
>>
>> I visually reviewed the patches and I don't object to them going in -- but I'm
>> still not sure they are all correct.
> 
> Re-ping. I'd like to receive a confirmation from you (and thus get the
> patches merged)
> or to discuss, what seems wrong to you.
> 
>>
>> --Mark
>>
>>>>
>>>> Current situation:
>>>>
>>>> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
>>>> NOTE: Started PRServer with DBfile:
>>>> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
>>>> 15895
>>>> Loading cache: 100%
>>>> |########################################################################################################################|
>>>> ETA:  00:00:00
>>>> Loaded 1302 entries from dependency cache.
>>>> NOTE: Resolving any missing task queue dependencies
>>>> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
>>>> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
>>>> RDEPENDS on or otherwise requires it)
>>>> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
>>>> removing...
>>>> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
>>>> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
>>>> unbuildable, removing...
>>>> Missing or unbuildable dependency chain was:
>>>> ['packagegroup-cross-canadian-qemumips64n32',
>>>> 'binutils-cross-canadian-mips64']
>>>> ERROR: Required build target 'core-image-base' has no buildable providers.
>>>> Missing or unbuildable dependency chain was: ['core-image-base',
>>>> 'packagegroup-cross-canadian-qemumips64n32',
>>>> 'binutils-cross-canadian-mips64']
>>>>
>>>
>>>
>>>
>>
> 
> 
> 

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Dmitry Eremin-Solenikov]

Hello,

2015-07-20 18:47 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
> Between customer escalations, other work and vacation.. I didn't happen.

No problem, really. I just wanted to be sure that the patches aren't going
to be lost.

>
> I've started testing it this morning and should have results tomorrow.. (US time).

Thank you very much!

>
> --Mark
>
> On 7/20/15 7:11 AM, Dmitry Eremin-Solenikov wrote:
>> 2015-07-08 15:42 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>> On 7/8/15 7:00 AM, Dmitry Eremin-Solenikov wrote:
>>>> Hello,
>>>>
>>>> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
>>>>> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
>>>>> I have verified these patches in both single lib and in multilib cases.
>>>>> Tri-lib images and SDK for MIPS build fine with these patches.
>>>>
>>>> Any comments on this patchset? It would be nice to have N32 ubroken in
>>>> the current
>>>> master tree.
>>>
>>> Sorry, I'm traveling right now and haven't had time to review this.  I should be
>>> able to get back to the office and test it by the end of the week.
>>>
>>> I visually reviewed the patches and I don't object to them going in -- but I'm
>>> still not sure they are all correct.
>>
>> Re-ping. I'd like to receive a confirmation from you (and thus get the
>> patches merged)
>> or to discuss, what seems wrong to you.
>>
>>>
>>> --Mark
>>>
>>>>>
>>>>> Current situation:
>>>>>
>>>>> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
>>>>> NOTE: Started PRServer with DBfile:
>>>>> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
>>>>> 15895
>>>>> Loading cache: 100%
>>>>> |########################################################################################################################|
>>>>> ETA:  00:00:00
>>>>> Loaded 1302 entries from dependency cache.
>>>>> NOTE: Resolving any missing task queue dependencies
>>>>> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
>>>>> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
>>>>> RDEPENDS on or otherwise requires it)
>>>>> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
>>>>> removing...
>>>>> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
>>>>> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
>>>>> unbuildable, removing...
>>>>> Missing or unbuildable dependency chain was:
>>>>> ['packagegroup-cross-canadian-qemumips64n32',
>>>>> 'binutils-cross-canadian-mips64']
>>>>> ERROR: Required build target 'core-image-base' has no buildable providers.
>>>>> Missing or unbuildable dependency chain was: ['core-image-base',
>>>>> 'packagegroup-cross-canadian-qemumips64n32',
>>>>> 'binutils-cross-canadian-mips64']
>>>>>
>>>>
>>>>
>>>>
>>>
>>
>>
>>
>



-- 
With best wishes
Dmitry

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Mark Hatle]

On 7/21/15 3:23 AM, Dmitry Eremin-Solenikov wrote:
> Hello,
> 
> 2015-07-20 18:47 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>> Between customer escalations, other work and vacation.. I didn't happen.
> 
> No problem, really. I just wanted to be sure that the patches aren't going
> to be lost.
> 
>>
>> I've started testing it this morning and should have results tomorrow.. (US time).
> 
> Thank you very much!

With master and the four patches I got an error trying to construct the SDK.

I'm still investigating the error:

Collected errors:
 * satisfy_dependencies_for: Cannot satisfy the following dependencies for
packagegroup-cross-canadian-qemumips64:
 *      gcc-cross-canadian-mips *       binutils-cross-canadian-mips *
gdb-cross-canadian-mips *
 * opkg_install_cmd: Cannot install package packagegroup-cross-canadian-qemumips64.


My configuration for this particular test was:

(oe-init-build-env stock w/ the following changes)

MACHINE = "qemumips64"

require conf/multilib.conf
DEFAULTTUNE = "mips64-n32"
MULTILIBS = "multilib:lib64 multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib64 = "mips64"
DEFAULTTUNE_virtclass-multilib-lib32 = "mips32r2"

# Set RPM_PREFER_ELF_ARCH to configure preferred ABI when using rpm packaging
# backend to generate a rootfs, choices are:
# 1: ELF32 wins
# 2: ELF64 wins
# 4: ELF64 N32 wins (for mips64 or mips64el only)
RPM_PREFER_ELF_ARCH = "4"


>>
>> --Mark
>>
>> On 7/20/15 7:11 AM, Dmitry Eremin-Solenikov wrote:
>>> 2015-07-08 15:42 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>>> On 7/8/15 7:00 AM, Dmitry Eremin-Solenikov wrote:
>>>>> Hello,
>>>>>
>>>>> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
>>>>>> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
>>>>>> I have verified these patches in both single lib and in multilib cases.
>>>>>> Tri-lib images and SDK for MIPS build fine with these patches.
>>>>>
>>>>> Any comments on this patchset? It would be nice to have N32 ubroken in
>>>>> the current
>>>>> master tree.
>>>>
>>>> Sorry, I'm traveling right now and haven't had time to review this.  I should be
>>>> able to get back to the office and test it by the end of the week.
>>>>
>>>> I visually reviewed the patches and I don't object to them going in -- but I'm
>>>> still not sure they are all correct.
>>>
>>> Re-ping. I'd like to receive a confirmation from you (and thus get the
>>> patches merged)
>>> or to discuss, what seems wrong to you.
>>>
>>>>
>>>> --Mark
>>>>
>>>>>>
>>>>>> Current situation:
>>>>>>
>>>>>> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
>>>>>> NOTE: Started PRServer with DBfile:
>>>>>> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
>>>>>> 15895
>>>>>> Loading cache: 100%
>>>>>> |########################################################################################################################|
>>>>>> ETA:  00:00:00
>>>>>> Loaded 1302 entries from dependency cache.
>>>>>> NOTE: Resolving any missing task queue dependencies
>>>>>> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
>>>>>> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
>>>>>> RDEPENDS on or otherwise requires it)
>>>>>> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
>>>>>> removing...
>>>>>> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
>>>>>> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
>>>>>> unbuildable, removing...
>>>>>> Missing or unbuildable dependency chain was:
>>>>>> ['packagegroup-cross-canadian-qemumips64n32',
>>>>>> 'binutils-cross-canadian-mips64']
>>>>>> ERROR: Required build target 'core-image-base' has no buildable providers.
>>>>>> Missing or unbuildable dependency chain was: ['core-image-base',
>>>>>> 'packagegroup-cross-canadian-qemumips64n32',
>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>>
>>
> 
> 
> 

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Mark Hatle]

On 7/21/15 9:31 AM, Mark Hatle wrote:
> On 7/21/15 3:23 AM, Dmitry Eremin-Solenikov wrote:
>> Hello,
>>
>> 2015-07-20 18:47 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>> Between customer escalations, other work and vacation.. I didn't happen.
>>
>> No problem, really. I just wanted to be sure that the patches aren't going
>> to be lost.
>>
>>>
>>> I've started testing it this morning and should have results tomorrow.. (US time).
>>
>> Thank you very much!
> 
> With master and the four patches I got an error trying to construct the SDK.
> 
> I'm still investigating the error:
> 
> Collected errors:
>  * satisfy_dependencies_for: Cannot satisfy the following dependencies for
> packagegroup-cross-canadian-qemumips64:
>  *      gcc-cross-canadian-mips *       binutils-cross-canadian-mips *
> gdb-cross-canadian-mips *
>  * opkg_install_cmd: Cannot install package packagegroup-cross-canadian-qemumips64.
> 
> 
> My configuration for this particular test was:
> 
> (oe-init-build-env stock w/ the following changes)

I've narrowed down the reproducer to just:

MACHINE = "qemumips64"
require conf/multilib.conf
DEFAULTTUNE = "mips64-n32"
MULTILIBS = ""

(If you remove the require conf/multilib.conf then things work.)

I'm going to continue to investigate.

--Mark

> MACHINE = "qemumips64"
> 
> require conf/multilib.conf
> DEFAULTTUNE = "mips64-n32"
> MULTILIBS = "multilib:lib64 multilib:lib32"
> DEFAULTTUNE_virtclass-multilib-lib64 = "mips64"
> DEFAULTTUNE_virtclass-multilib-lib32 = "mips32r2"
> 
> # Set RPM_PREFER_ELF_ARCH to configure preferred ABI when using rpm packaging
> # backend to generate a rootfs, choices are:
> # 1: ELF32 wins
> # 2: ELF64 wins
> # 4: ELF64 N32 wins (for mips64 or mips64el only)
> RPM_PREFER_ELF_ARCH = "4"
> 
> 
>>>
>>> --Mark
>>>
>>> On 7/20/15 7:11 AM, Dmitry Eremin-Solenikov wrote:
>>>> 2015-07-08 15:42 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>>>> On 7/8/15 7:00 AM, Dmitry Eremin-Solenikov wrote:
>>>>>> Hello,
>>>>>>
>>>>>> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
>>>>>>> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
>>>>>>> I have verified these patches in both single lib and in multilib cases.
>>>>>>> Tri-lib images and SDK for MIPS build fine with these patches.
>>>>>>
>>>>>> Any comments on this patchset? It would be nice to have N32 ubroken in
>>>>>> the current
>>>>>> master tree.
>>>>>
>>>>> Sorry, I'm traveling right now and haven't had time to review this.  I should be
>>>>> able to get back to the office and test it by the end of the week.
>>>>>
>>>>> I visually reviewed the patches and I don't object to them going in -- but I'm
>>>>> still not sure they are all correct.
>>>>
>>>> Re-ping. I'd like to receive a confirmation from you (and thus get the
>>>> patches merged)
>>>> or to discuss, what seems wrong to you.
>>>>
>>>>>
>>>>> --Mark
>>>>>
>>>>>>>
>>>>>>> Current situation:
>>>>>>>
>>>>>>> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
>>>>>>> NOTE: Started PRServer with DBfile:
>>>>>>> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
>>>>>>> 15895
>>>>>>> Loading cache: 100%
>>>>>>> |########################################################################################################################|
>>>>>>> ETA:  00:00:00
>>>>>>> Loaded 1302 entries from dependency cache.
>>>>>>> NOTE: Resolving any missing task queue dependencies
>>>>>>> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
>>>>>>> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
>>>>>>> RDEPENDS on or otherwise requires it)
>>>>>>> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
>>>>>>> removing...
>>>>>>> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
>>>>>>> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
>>>>>>> unbuildable, removing...
>>>>>>> Missing or unbuildable dependency chain was:
>>>>>>> ['packagegroup-cross-canadian-qemumips64n32',
>>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>> ERROR: Required build target 'core-image-base' has no buildable providers.
>>>>>>> Missing or unbuildable dependency chain was: ['core-image-base',
>>>>>>> 'packagegroup-cross-canadian-qemumips64n32',
>>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>
>>
>>
>>
> 

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Dmitry Eremin-Solenikov]

2015-07-21 18:26 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
> On 7/21/15 9:31 AM, Mark Hatle wrote:
>> On 7/21/15 3:23 AM, Dmitry Eremin-Solenikov wrote:
>>> Hello,
>>>
>>> 2015-07-20 18:47 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>>> Between customer escalations, other work and vacation.. I didn't happen.
>>>
>>> No problem, really. I just wanted to be sure that the patches aren't going
>>> to be lost.
>>>
>>>>
>>>> I've started testing it this morning and should have results tomorrow.. (US time).
>>>
>>> Thank you very much!
>>
>> With master and the four patches I got an error trying to construct the SDK.
>>
>> I'm still investigating the error:
>>
>> Collected errors:
>>  * satisfy_dependencies_for: Cannot satisfy the following dependencies for
>> packagegroup-cross-canadian-qemumips64:
>>  *      gcc-cross-canadian-mips *       binutils-cross-canadian-mips *
>> gdb-cross-canadian-mips *
>>  * opkg_install_cmd: Cannot install package packagegroup-cross-canadian-qemumips64.
>>
>>
>> My configuration for this particular test was:
>>
>> (oe-init-build-env stock w/ the following changes)
>
> I've narrowed down the reproducer to just:
>
> MACHINE = "qemumips64"
> require conf/multilib.conf
> DEFAULTTUNE = "mips64-n32"
> MULTILIBS = ""
>
> (If you remove the require conf/multilib.conf then things work.)
>
> I'm going to continue to investigate.

I fail to reproduce this on my side. What is your exact bitbake
command that fails?
All my attempts to build an SDK/toolchains/etc succeeded.

>
> --Mark
>
>> MACHINE = "qemumips64"
>>
>> require conf/multilib.conf
>> DEFAULTTUNE = "mips64-n32"
>> MULTILIBS = "multilib:lib64 multilib:lib32"
>> DEFAULTTUNE_virtclass-multilib-lib64 = "mips64"
>> DEFAULTTUNE_virtclass-multilib-lib32 = "mips32r2"
>>
>> # Set RPM_PREFER_ELF_ARCH to configure preferred ABI when using rpm packaging
>> # backend to generate a rootfs, choices are:
>> # 1: ELF32 wins
>> # 2: ELF64 wins
>> # 4: ELF64 N32 wins (for mips64 or mips64el only)
>> RPM_PREFER_ELF_ARCH = "4"
>>
>>
>>>>
>>>> --Mark
>>>>
>>>> On 7/20/15 7:11 AM, Dmitry Eremin-Solenikov wrote:
>>>>> 2015-07-08 15:42 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>>>>> On 7/8/15 7:00 AM, Dmitry Eremin-Solenikov wrote:
>>>>>>> Hello,
>>>>>>>
>>>>>>> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
>>>>>>>> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
>>>>>>>> I have verified these patches in both single lib and in multilib cases.
>>>>>>>> Tri-lib images and SDK for MIPS build fine with these patches.
>>>>>>>
>>>>>>> Any comments on this patchset? It would be nice to have N32 ubroken in
>>>>>>> the current
>>>>>>> master tree.
>>>>>>
>>>>>> Sorry, I'm traveling right now and haven't had time to review this.  I should be
>>>>>> able to get back to the office and test it by the end of the week.
>>>>>>
>>>>>> I visually reviewed the patches and I don't object to them going in -- but I'm
>>>>>> still not sure they are all correct.
>>>>>
>>>>> Re-ping. I'd like to receive a confirmation from you (and thus get the
>>>>> patches merged)
>>>>> or to discuss, what seems wrong to you.
>>>>>
>>>>>>
>>>>>> --Mark
>>>>>>
>>>>>>>>
>>>>>>>> Current situation:
>>>>>>>>
>>>>>>>> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
>>>>>>>> NOTE: Started PRServer with DBfile:
>>>>>>>> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
>>>>>>>> 15895
>>>>>>>> Loading cache: 100%
>>>>>>>> |########################################################################################################################|
>>>>>>>> ETA:  00:00:00
>>>>>>>> Loaded 1302 entries from dependency cache.
>>>>>>>> NOTE: Resolving any missing task queue dependencies
>>>>>>>> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
>>>>>>>> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
>>>>>>>> RDEPENDS on or otherwise requires it)
>>>>>>>> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
>>>>>>>> removing...
>>>>>>>> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
>>>>>>>> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
>>>>>>>> unbuildable, removing...
>>>>>>>> Missing or unbuildable dependency chain was:
>>>>>>>> ['packagegroup-cross-canadian-qemumips64n32',
>>>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>>> ERROR: Required build target 'core-image-base' has no buildable providers.
>>>>>>>> Missing or unbuildable dependency chain was: ['core-image-base',
>>>>>>>> 'packagegroup-cross-canadian-qemumips64n32',
>>>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>>
>>
>



-- 
With best wishes
Dmitry

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Dmitry Eremin-Solenikov]

2015-07-21 18:26 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
> On 7/21/15 9:31 AM, Mark Hatle wrote:
>> On 7/21/15 3:23 AM, Dmitry Eremin-Solenikov wrote:
>>> Hello,
>>>
>>> 2015-07-20 18:47 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>>> Between customer escalations, other work and vacation.. I didn't happen.
>>>
>>> No problem, really. I just wanted to be sure that the patches aren't going
>>> to be lost.
>>>
>>>>
>>>> I've started testing it this morning and should have results tomorrow.. (US time).
>>>
>>> Thank you very much!
>>
>> With master and the four patches I got an error trying to construct the SDK.
>>
>> I'm still investigating the error:
>>
>> Collected errors:
>>  * satisfy_dependencies_for: Cannot satisfy the following dependencies for
>> packagegroup-cross-canadian-qemumips64:
>>  *      gcc-cross-canadian-mips *       binutils-cross-canadian-mips *
>> gdb-cross-canadian-mips *
>>  * opkg_install_cmd: Cannot install package packagegroup-cross-canadian-qemumips64.
>>
>>
>> My configuration for this particular test was:
>>
>> (oe-init-build-env stock w/ the following changes)
>
> I've narrowed down the reproducer to just:
>
> MACHINE = "qemumips64"
> require conf/multilib.conf
> DEFAULTTUNE = "mips64-n32"
> MULTILIBS = ""
>
> (If you remove the require conf/multilib.conf then things work.)
>
> I'm going to continue to investigate.

Any progress on your side? As I failed to reproduce the issue on my side,
could you please specify:
1) what is the exact bitbake command that failed?
2) what additional layers do you have enabled in your build?
3) is it a cleanroom build, or you have an sstate cache populated with previous
   build results?
4) any additional information that I can use to reproduce the issue on my side?

>
> --Mark
>
>> MACHINE = "qemumips64"
>>
>> require conf/multilib.conf
>> DEFAULTTUNE = "mips64-n32"
>> MULTILIBS = "multilib:lib64 multilib:lib32"
>> DEFAULTTUNE_virtclass-multilib-lib64 = "mips64"
>> DEFAULTTUNE_virtclass-multilib-lib32 = "mips32r2"
>>
>> # Set RPM_PREFER_ELF_ARCH to configure preferred ABI when using rpm packaging
>> # backend to generate a rootfs, choices are:
>> # 1: ELF32 wins
>> # 2: ELF64 wins
>> # 4: ELF64 N32 wins (for mips64 or mips64el only)
>> RPM_PREFER_ELF_ARCH = "4"
>>
>>
>>>>
>>>> --Mark
>>>>
>>>> On 7/20/15 7:11 AM, Dmitry Eremin-Solenikov wrote:
>>>>> 2015-07-08 15:42 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>>>>> On 7/8/15 7:00 AM, Dmitry Eremin-Solenikov wrote:
>>>>>>> Hello,
>>>>>>>
>>>>>>> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
>>>>>>>> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
>>>>>>>> I have verified these patches in both single lib and in multilib cases.
>>>>>>>> Tri-lib images and SDK for MIPS build fine with these patches.
>>>>>>>
>>>>>>> Any comments on this patchset? It would be nice to have N32 ubroken in
>>>>>>> the current
>>>>>>> master tree.
>>>>>>
>>>>>> Sorry, I'm traveling right now and haven't had time to review this.  I should be
>>>>>> able to get back to the office and test it by the end of the week.
>>>>>>
>>>>>> I visually reviewed the patches and I don't object to them going in -- but I'm
>>>>>> still not sure they are all correct.
>>>>>
>>>>> Re-ping. I'd like to receive a confirmation from you (and thus get the
>>>>> patches merged)
>>>>> or to discuss, what seems wrong to you.
>>>>>
>>>>>>
>>>>>> --Mark
>>>>>>
>>>>>>>>
>>>>>>>> Current situation:
>>>>>>>>
>>>>>>>> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
>>>>>>>> NOTE: Started PRServer with DBfile:
>>>>>>>> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
>>>>>>>> 15895
>>>>>>>> Loading cache: 100%
>>>>>>>> |########################################################################################################################|
>>>>>>>> ETA:  00:00:00
>>>>>>>> Loaded 1302 entries from dependency cache.
>>>>>>>> NOTE: Resolving any missing task queue dependencies
>>>>>>>> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
>>>>>>>> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
>>>>>>>> RDEPENDS on or otherwise requires it)
>>>>>>>> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
>>>>>>>> removing...
>>>>>>>> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
>>>>>>>> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
>>>>>>>> unbuildable, removing...
>>>>>>>> Missing or unbuildable dependency chain was:
>>>>>>>> ['packagegroup-cross-canadian-qemumips64n32',
>>>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>>> ERROR: Required build target 'core-image-base' has no buildable providers.
>>>>>>>> Missing or unbuildable dependency chain was: ['core-image-base',
>>>>>>>> 'packagegroup-cross-canadian-qemumips64n32',
>>>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>>
>>
>



-- 
With best wishes
Dmitry

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Mark Hatle]

On 7/28/15 12:26 PM, Dmitry Eremin-Solenikov wrote:
> 2015-07-21 18:26 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>> On 7/21/15 9:31 AM, Mark Hatle wrote:
>>> On 7/21/15 3:23 AM, Dmitry Eremin-Solenikov wrote:
>>>> Hello,
>>>>
>>>> 2015-07-20 18:47 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>>>> Between customer escalations, other work and vacation.. I didn't happen.
>>>>
>>>> No problem, really. I just wanted to be sure that the patches aren't going
>>>> to be lost.
>>>>
>>>>>
>>>>> I've started testing it this morning and should have results tomorrow.. (US time).
>>>>
>>>> Thank you very much!
>>>
>>> With master and the four patches I got an error trying to construct the SDK.
>>>
>>> I'm still investigating the error:
>>>
>>> Collected errors:
>>>  * satisfy_dependencies_for: Cannot satisfy the following dependencies for
>>> packagegroup-cross-canadian-qemumips64:
>>>  *      gcc-cross-canadian-mips *       binutils-cross-canadian-mips *
>>> gdb-cross-canadian-mips *
>>>  * opkg_install_cmd: Cannot install package packagegroup-cross-canadian-qemumips64.
>>>
>>>
>>> My configuration for this particular test was:
>>>
>>> (oe-init-build-env stock w/ the following changes)
>>
>> I've narrowed down the reproducer to just:
>>
>> MACHINE = "qemumips64"
>> require conf/multilib.conf
>> DEFAULTTUNE = "mips64-n32"
>> MULTILIBS = ""
>>
>> (If you remove the require conf/multilib.conf then things work.)
>>
>> I'm going to continue to investigate.
> 
> Any progress on your side? As I failed to reproduce the issue on my side,
> could you please specify:
> 1) what is the exact bitbake command that failed?
> 2) what additional layers do you have enabled in your build?
> 3) is it a cleanroom build, or you have an sstate cache populated with previous
>    build results?
> 4) any additional information that I can use to reproduce the issue on my side?

Recent check-ins have changed some things.  The compile side was working for a
while, but not the compilation side (using it).

I've been working with RP on IRC, and he's come up with some additional patches.
 (including this set of 4..)

So I think the set is fine, it's up to the other pieces that need to mesh with
this to produce the working SDK.

--Mark

>>
>> --Mark
>>
>>> MACHINE = "qemumips64"
>>>
>>> require conf/multilib.conf
>>> DEFAULTTUNE = "mips64-n32"
>>> MULTILIBS = "multilib:lib64 multilib:lib32"
>>> DEFAULTTUNE_virtclass-multilib-lib64 = "mips64"
>>> DEFAULTTUNE_virtclass-multilib-lib32 = "mips32r2"
>>>
>>> # Set RPM_PREFER_ELF_ARCH to configure preferred ABI when using rpm packaging
>>> # backend to generate a rootfs, choices are:
>>> # 1: ELF32 wins
>>> # 2: ELF64 wins
>>> # 4: ELF64 N32 wins (for mips64 or mips64el only)
>>> RPM_PREFER_ELF_ARCH = "4"
>>>
>>>
>>>>>
>>>>> --Mark
>>>>>
>>>>> On 7/20/15 7:11 AM, Dmitry Eremin-Solenikov wrote:
>>>>>> 2015-07-08 15:42 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>>>>>> On 7/8/15 7:00 AM, Dmitry Eremin-Solenikov wrote:
>>>>>>>> Hello,
>>>>>>>>
>>>>>>>> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
>>>>>>>>> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
>>>>>>>>> I have verified these patches in both single lib and in multilib cases.
>>>>>>>>> Tri-lib images and SDK for MIPS build fine with these patches.
>>>>>>>>
>>>>>>>> Any comments on this patchset? It would be nice to have N32 ubroken in
>>>>>>>> the current
>>>>>>>> master tree.
>>>>>>>
>>>>>>> Sorry, I'm traveling right now and haven't had time to review this.  I should be
>>>>>>> able to get back to the office and test it by the end of the week.
>>>>>>>
>>>>>>> I visually reviewed the patches and I don't object to them going in -- but I'm
>>>>>>> still not sure they are all correct.
>>>>>>
>>>>>> Re-ping. I'd like to receive a confirmation from you (and thus get the
>>>>>> patches merged)
>>>>>> or to discuss, what seems wrong to you.
>>>>>>
>>>>>>>
>>>>>>> --Mark
>>>>>>>
>>>>>>>>>
>>>>>>>>> Current situation:
>>>>>>>>>
>>>>>>>>> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
>>>>>>>>> NOTE: Started PRServer with DBfile:
>>>>>>>>> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
>>>>>>>>> 15895
>>>>>>>>> Loading cache: 100%
>>>>>>>>> |########################################################################################################################|
>>>>>>>>> ETA:  00:00:00
>>>>>>>>> Loaded 1302 entries from dependency cache.
>>>>>>>>> NOTE: Resolving any missing task queue dependencies
>>>>>>>>> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
>>>>>>>>> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
>>>>>>>>> RDEPENDS on or otherwise requires it)
>>>>>>>>> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
>>>>>>>>> removing...
>>>>>>>>> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
>>>>>>>>> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
>>>>>>>>> unbuildable, removing...
>>>>>>>>> Missing or unbuildable dependency chain was:
>>>>>>>>> ['packagegroup-cross-canadian-qemumips64n32',
>>>>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>>>> ERROR: Required build target 'core-image-base' has no buildable providers.
>>>>>>>>> Missing or unbuildable dependency chain was: ['core-image-base',
>>>>>>>>> 'packagegroup-cross-canadian-qemumips64n32',
>>>>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>
>>
> 
> 
> 

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Dmitry Eremin-Solenikov]

2015-07-29 17:58 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
> On 7/28/15 12:26 PM, Dmitry Eremin-Solenikov wrote:
>> 2015-07-21 18:26 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>> On 7/21/15 9:31 AM, Mark Hatle wrote:
>>>> On 7/21/15 3:23 AM, Dmitry Eremin-Solenikov wrote:
>>>>> Hello,
>>>>>
>>>>> 2015-07-20 18:47 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>>>>> Between customer escalations, other work and vacation.. I didn't happen.
>>>>>
>>>>> No problem, really. I just wanted to be sure that the patches aren't going
>>>>> to be lost.
>>>>>
>>>>>>
>>>>>> I've started testing it this morning and should have results tomorrow.. (US time).
>>>>>
>>>>> Thank you very much!
>>>>
>>>> With master and the four patches I got an error trying to construct the SDK.
>>>>
>>>> I'm still investigating the error:
>>>>
>>>> Collected errors:
>>>>  * satisfy_dependencies_for: Cannot satisfy the following dependencies for
>>>> packagegroup-cross-canadian-qemumips64:
>>>>  *      gcc-cross-canadian-mips *       binutils-cross-canadian-mips *
>>>> gdb-cross-canadian-mips *
>>>>  * opkg_install_cmd: Cannot install package packagegroup-cross-canadian-qemumips64.
>>>>
>>>>
>>>> My configuration for this particular test was:
>>>>
>>>> (oe-init-build-env stock w/ the following changes)
>>>
>>> I've narrowed down the reproducer to just:
>>>
>>> MACHINE = "qemumips64"
>>> require conf/multilib.conf
>>> DEFAULTTUNE = "mips64-n32"
>>> MULTILIBS = ""
>>>
>>> (If you remove the require conf/multilib.conf then things work.)
>>>
>>> I'm going to continue to investigate.
>>
>> Any progress on your side? As I failed to reproduce the issue on my side,
>> could you please specify:
>> 1) what is the exact bitbake command that failed?
>> 2) what additional layers do you have enabled in your build?
>> 3) is it a cleanroom build, or you have an sstate cache populated with previous
>>    build results?
>> 4) any additional information that I can use to reproduce the issue on my side?
>
> Recent check-ins have changed some things.  The compile side was working for a
> while, but not the compilation side (using it).
>
> I've been working with RP on IRC, and he's come up with some additional patches.
>  (including this set of 4..)
>
> So I think the set is fine, it's up to the other pieces that need to mesh with
> this to produce the working SDK.

Ok. So what would be the plan then? If you need any additional tests
from my side,
I can provide them.

>
> --Mark
>
>>>
>>> --Mark
>>>
>>>> MACHINE = "qemumips64"
>>>>
>>>> require conf/multilib.conf
>>>> DEFAULTTUNE = "mips64-n32"
>>>> MULTILIBS = "multilib:lib64 multilib:lib32"
>>>> DEFAULTTUNE_virtclass-multilib-lib64 = "mips64"
>>>> DEFAULTTUNE_virtclass-multilib-lib32 = "mips32r2"
>>>>
>>>> # Set RPM_PREFER_ELF_ARCH to configure preferred ABI when using rpm packaging
>>>> # backend to generate a rootfs, choices are:
>>>> # 1: ELF32 wins
>>>> # 2: ELF64 wins
>>>> # 4: ELF64 N32 wins (for mips64 or mips64el only)
>>>> RPM_PREFER_ELF_ARCH = "4"
>>>>
>>>>
>>>>>>
>>>>>> --Mark
>>>>>>
>>>>>> On 7/20/15 7:11 AM, Dmitry Eremin-Solenikov wrote:
>>>>>>> 2015-07-08 15:42 GMT+03:00 Mark Hatle <mark.hatle@windriver.com>:
>>>>>>>> On 7/8/15 7:00 AM, Dmitry Eremin-Solenikov wrote:
>>>>>>>>> Hello,
>>>>>>>>>
>>>>>>>>> 2015-07-01 17:02 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
>>>>>>>>>> Quite simple configuration (MIPS64 N32 image) fails to build (see below).
>>>>>>>>>> I have verified these patches in both single lib and in multilib cases.
>>>>>>>>>> Tri-lib images and SDK for MIPS build fine with these patches.
>>>>>>>>>
>>>>>>>>> Any comments on this patchset? It would be nice to have N32 ubroken in
>>>>>>>>> the current
>>>>>>>>> master tree.
>>>>>>>>
>>>>>>>> Sorry, I'm traveling right now and haven't had time to review this.  I should be
>>>>>>>> able to get back to the office and test it by the end of the week.
>>>>>>>>
>>>>>>>> I visually reviewed the patches and I don't object to them going in -- but I'm
>>>>>>>> still not sure they are all correct.
>>>>>>>
>>>>>>> Re-ping. I'd like to receive a confirmation from you (and thus get the
>>>>>>> patches merged)
>>>>>>> or to discuss, what seems wrong to you.
>>>>>>>
>>>>>>>>
>>>>>>>> --Mark
>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Current situation:
>>>>>>>>>>
>>>>>>>>>> lumag@nexs:~/OE$ MACHINE=qemumips64n32 bitbake core-image-base
>>>>>>>>>> NOTE: Started PRServer with DBfile:
>>>>>>>>>> /home/lumag/OE/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 46391, PID:
>>>>>>>>>> 15895
>>>>>>>>>> Loading cache: 100%
>>>>>>>>>> |########################################################################################################################|
>>>>>>>>>> ETA:  00:00:00
>>>>>>>>>> Loaded 1302 entries from dependency cache.
>>>>>>>>>> NOTE: Resolving any missing task queue dependencies
>>>>>>>>>> ERROR: Nothing RPROVIDES 'binutils-cross-canadian-mips64' (but
>>>>>>>>>> /home/lumag/OE/sources/openembedded-core/meta/recipes-core/packagegroups/packagegroup-cross-canadian.bb
>>>>>>>>>> RDEPENDS on or otherwise requires it)
>>>>>>>>>> NOTE: Runtime target 'binutils-cross-canadian-mips64' is unbuildable,
>>>>>>>>>> removing...
>>>>>>>>>> Missing or unbuildable dependency chain was: ['binutils-cross-canadian-mips64']
>>>>>>>>>> NOTE: Runtime target 'packagegroup-cross-canadian-qemumips64n32' is
>>>>>>>>>> unbuildable, removing...
>>>>>>>>>> Missing or unbuildable dependency chain was:
>>>>>>>>>> ['packagegroup-cross-canadian-qemumips64n32',
>>>>>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>>>>> ERROR: Required build target 'core-image-base' has no buildable providers.
>>>>>>>>>> Missing or unbuildable dependency chain was: ['core-image-base',
>>>>>>>>>> 'packagegroup-cross-canadian-qemumips64n32',
>>>>>>>>>> 'binutils-cross-canadian-mips64']
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>
>>
>>
>



-- 
With best wishes
Dmitry

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Richard Purdie]

On Wed, 2015-07-29 at 18:00 +0300, Dmitry Eremin-Solenikov wrote:
> > Recent check-ins have changed some things.  The compile side was working for a
> > while, but not the compilation side (using it).
> >
> > I've been working with RP on IRC, and he's come up with some additional patches.
> >  (including this set of 4..)
> >
> > So I think the set is fine, it's up to the other pieces that need to mesh with
> > this to produce the working SDK.
> 
> Ok. So what would be the plan then? If you need any additional tests
> from my side,
> I can provide them.

Sorry for the delays on this. I found several issues and ended up with
some quite subtle gcc changes to make all this work. I've put both our
changes into master-next. The local.conf testing setup I've been using
is:

MACHINE = "qemumips64"
require conf/multilib.conf
DEFAULTTUNE = "mips64-n32"
MULTILIBS = "multilib:lib64 multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib64 = "mips64"
DEFAULTTUNE_virtclass-multilib-lib32 = "mips32r2"

then the build/test commands:

bitbake core-image-minimal -c populate_sdk
bitbake core-image-minimal -c testsdk

The nice thing is the latter automated test part now passes for this
configuration (and its tests all three multilibs).

What remains is to run these changes through the autobuilder and make
sure nothing else breaks as a result of this patchset. I'm about to do
some business travel which means my ability to debug any further issues
will be limited, compiling toolchains on the laptop isn't realistic. I
may therefore need help in ironing out any further wrinkles or we'll
have to wait until I get back in a couple of weeks.

Cheers,

Richard

Re: [PATCH 0/4] Fix builds for MIPS64 N32

[Dmitry Eremin-Solenikov]

2015-07-31 13:59 GMT+03:00 Richard Purdie <richard.purdie@linuxfoundation.org>:
> On Wed, 2015-07-29 at 18:00 +0300, Dmitry Eremin-Solenikov wrote:
>> > Recent check-ins have changed some things.  The compile side was working for a
>> > while, but not the compilation side (using it).
>> >
>> > I've been working with RP on IRC, and he's come up with some additional patches.
>> >  (including this set of 4..)
>> >
>> > So I think the set is fine, it's up to the other pieces that need to mesh with
>> > this to produce the working SDK.
>>
>> Ok. So what would be the plan then? If you need any additional tests
>> from my side,
>> I can provide them.
>
> Sorry for the delays on this. I found several issues and ended up with
> some quite subtle gcc changes to make all this work. I've put both our
> changes into master-next. The local.conf testing setup I've been using
> is:
>
> MACHINE = "qemumips64"
> require conf/multilib.conf
> DEFAULTTUNE = "mips64-n32"
> MULTILIBS = "multilib:lib64 multilib:lib32"
> DEFAULTTUNE_virtclass-multilib-lib64 = "mips64"
> DEFAULTTUNE_virtclass-multilib-lib32 = "mips32r2"
>
> then the build/test commands:
>
> bitbake core-image-minimal -c populate_sdk
> bitbake core-image-minimal -c testsdk
>
> The nice thing is the latter automated test part now passes for this
> configuration (and its tests all three multilibs).
>
> What remains is to run these changes through the autobuilder and make
> sure nothing else breaks as a result of this patchset. I'm about to do
> some business travel which means my ability to debug any further issues
> will be limited, compiling toolchains on the laptop isn't realistic. I
> may therefore need help in ironing out any further wrinkles or we'll
> have to wait until I get back in a couple of weeks.

I should be able to test your changes in a more or less timely manner,
so feel free to ping me if the need arises.

-- 
With best wishes
Dmitry

Re: [PATCH 4/4] glibc: don't override TUNE_CCARGS for MIPS

[Khem Raj]

[-- Attachment #1: Type: text/plain, Size: 2021 bytes --]


> On Jul 1, 2015, at 7:02 AM, Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> wrote:
> 
> Overriding TUNE_CCARGS this way breaks MULTILIB setup for MIPS. This
> override disables multilib handling of tunes for TUNE_CCARGS, thus
> enforcing glibc's TUNE_CCARGS to the TUNE_CCARGS of main DEFAULTTUNE.
> 
> Glibc perfectly build without this override for both simple and multilib
> cases.
> 

this patch has been applied however we should reopen it.
This has to be fixed differently. its not a build time issue but a runtime one. So find
alternative way of passing these options to glibc build to accommodate your changes.

> Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
> ---
> meta/recipes-core/glibc/glibc_2.21.bb | 10 ----------
> 1 file changed, 10 deletions(-)
> 
> diff --git a/meta/recipes-core/glibc/glibc_2.21.bb b/meta/recipes-core/glibc/glibc_2.21.bb
> index 8197c29..52252b3 100644
> --- a/meta/recipes-core/glibc/glibc_2.21.bb
> +++ b/meta/recipes-core/glibc/glibc_2.21.bb
> @@ -95,16 +95,6 @@ do_patch_append() {
>     bb.build.exec_func('do_fix_readlib_c', d)
> }
> 
> -# for mips glibc now builds syscall tables for all abi's
> -# so we make sure that we choose right march option which is
> -# compatible with o32,n32 and n64 abi's
> -# e.g. -march=mips32 is not compatible with n32 and n64 therefore
> -# we filter it out in such case -march=from-abi which will be
> -# mips1 when using o32 and mips3 when using n32/n64
> -
> -TUNE_CCARGS_mips := "${@oe_filter_out('-march=mips32', '${TUNE_CCARGS}', d)}"
> -TUNE_CCARGS_mipsel := "${@oe_filter_out('-march=mips32', '${TUNE_CCARGS}', d)}"
> -
> do_fix_readlib_c () {
> 	sed -i -e 's#OECORE_KNOWN_INTERPRETER_NAMES#${EGLIBC_KNOWN_INTERPRETER_NAMES}#' ${S}/elf/readlib.c
> }
> --
> 1.9.1
> 
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core


[-- Attachment #2: Message signed with OpenPGP using GPGMail --]
[-- Type: application/pgp-signature, Size: 211 bytes --]

Re: [PATCH 4/4] glibc: don't override TUNE_CCARGS for MIPS

[Dmitry Eremin-Solenikov]

2015-08-15 22:34 GMT+03:00 Khem Raj <raj.khem@gmail.com>:
>
>> On Jul 1, 2015, at 7:02 AM, Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> wrote:
>>
>> Overriding TUNE_CCARGS this way breaks MULTILIB setup for MIPS. This
>> override disables multilib handling of tunes for TUNE_CCARGS, thus
>> enforcing glibc's TUNE_CCARGS to the TUNE_CCARGS of main DEFAULTTUNE.
>>
>> Glibc perfectly build without this override for both simple and multilib
>> cases.
>>
>
> this patch has been applied however we should reopen it.
> This has to be fixed differently. its not a build time issue but a runtime one. So find
> alternative way of passing these options to glibc build to accommodate your changes.

What is the issue that you  are facing with the current OE-Core?


-- 
With best wishes
Dmitry

Re: [PATCH 4/4] glibc: don't override TUNE_CCARGS for MIPS

[Khem Raj]

On Sat, Aug 15, 2015 at 4:15 PM, Dmitry Eremin-Solenikov
<dbaryshkov@gmail.com> wrote:
> 2015-08-15 22:34 GMT+03:00 Khem Raj <raj.khem@gmail.com>:
>>
>>> On Jul 1, 2015, at 7:02 AM, Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> wrote:
>>>
>>> Overriding TUNE_CCARGS this way breaks MULTILIB setup for MIPS. This
>>> override disables multilib handling of tunes for TUNE_CCARGS, thus
>>> enforcing glibc's TUNE_CCARGS to the TUNE_CCARGS of main DEFAULTTUNE.
>>>
>>> Glibc perfectly build without this override for both simple and multilib
>>> cases.
>>>
>>
>> this patch has been applied however we should reopen it.
>> This has to be fixed differently. its not a build time issue but a runtime one. So find
>> alternative way of passing these options to glibc build to accommodate your changes.
>
> What is the issue that you  are facing with the current OE-Core?
>

systemd remount segfaults in glibc. The problems will be more where it
will use wrong syscall headers.

>
> --
> With best wishes
> Dmitry

Re: [PATCH 4/4] glibc: don't override TUNE_CCARGS for MIPS

[Dmitry Eremin-Solenikov]

2015-08-16 3:29 GMT+03:00 Khem Raj <raj.khem@gmail.com>:
> On Sat, Aug 15, 2015 at 4:15 PM, Dmitry Eremin-Solenikov
> <dbaryshkov@gmail.com> wrote:
>> 2015-08-15 22:34 GMT+03:00 Khem Raj <raj.khem@gmail.com>:
>>>
>>>> On Jul 1, 2015, at 7:02 AM, Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> wrote:
>>>>
>>>> Overriding TUNE_CCARGS this way breaks MULTILIB setup for MIPS. This
>>>> override disables multilib handling of tunes for TUNE_CCARGS, thus
>>>> enforcing glibc's TUNE_CCARGS to the TUNE_CCARGS of main DEFAULTTUNE.
>>>>
>>>> Glibc perfectly build without this override for both simple and multilib
>>>> cases.
>>>>
>>>
>>> this patch has been applied however we should reopen it.
>>> This has to be fixed differently. its not a build time issue but a runtime one. So find
>>> alternative way of passing these options to glibc build to accommodate your changes.
>>
>> What is the issue that you  are facing with the current OE-Core?
>>
>
> systemd remount segfaults in glibc. The problems will be more where it
> will use wrong syscall headers.

Hmm. N32 and N64 packages seem to work for me. O32 programs segfault
with illegal
instruction. I'll take a look.

-- 
With best wishes
Dmitry

Re: [PATCH 4/4] glibc: don't override TUNE_CCARGS for MIPS

[Dmitry Eremin-Solenikov]

2015-08-17 18:30 GMT+03:00 Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>:
> 2015-08-16 3:29 GMT+03:00 Khem Raj <raj.khem@gmail.com>:
>> On Sat, Aug 15, 2015 at 4:15 PM, Dmitry Eremin-Solenikov
>> <dbaryshkov@gmail.com> wrote:
>>> 2015-08-15 22:34 GMT+03:00 Khem Raj <raj.khem@gmail.com>:
>>>>
>>>>> On Jul 1, 2015, at 7:02 AM, Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> wrote:
>>>>>
>>>>> Overriding TUNE_CCARGS this way breaks MULTILIB setup for MIPS. This
>>>>> override disables multilib handling of tunes for TUNE_CCARGS, thus
>>>>> enforcing glibc's TUNE_CCARGS to the TUNE_CCARGS of main DEFAULTTUNE.
>>>>>
>>>>> Glibc perfectly build without this override for both simple and multilib
>>>>> cases.
>>>>>
>>>>
>>>> this patch has been applied however we should reopen it.
>>>> This has to be fixed differently. its not a build time issue but a runtime one. So find
>>>> alternative way of passing these options to glibc build to accommodate your changes.
>>>
>>> What is the issue that you  are facing with the current OE-Core?
>>>
>>
>> systemd remount segfaults in glibc. The problems will be more where it
>> will use wrong syscall headers.
>
> Hmm. N32 and N64 packages seem to work for me. O32 programs segfault
> with illegal
> instruction. I'll take a look.

It looks like something is broken wrt. MIPS64 kernel or QEMU. Because N64 kernel
causes segfaults with plain O32 rootfs. O32 kernel + O32 rootfs work w/o
any issues on emulated mips32. Emulated MIPS64 boots  32-bit kernel just fine,
but then the rootfs segfaults (?).

-- 
With best wishes
Dmitry