[Ltrace-devel] [PATCH 0/1] Move mipsel to mips

[Sedat Dilek]

On Tue, Feb 19, 2013 at 4:41 PM, Sedat Dilek <sedat.dilek at gmail.com> wrote:
> On Tue, Feb 19, 2013 at 4:37 PM, Sedat Dilek <sedat.dilek at gmail.com> wrote:
>> On Tue, Feb 19, 2013 at 1:25 PM, Petr Machata <pmachata at redhat.com> wrote:
>>> Sedat Dilek <sedat.dilek at gmail.com> writes:
>>>
>>>> On Sat, Feb 16, 2013 at 3:58 PM, Oliver Spornitz <ospornit at rz-online.de> wrote:
>>>>> I applied the three patches to the branch ospornit/move_mipsel on top
>>>>> of the current master.
>>>>
>>>> Cherry-picking the 3 patches to ltrace-0.7 branch worked nicely!
>>>> Petr, can you apply (cherry-pick) them after merging into master?
>>>
>>> This is a cleanup that shouldn't have any visible effect for users.
>>> I don't consider it appropriate for 0.7.x.
>>>
>>
>> The first thing I did not understand when looking at the Freetz ltrace
>> package was WTF-arch=mips or =mipsel.
>> So there was this symlink from mipsel->mips in the source-code.
>> If you look at the kernel-archives you see arch=mips.
>> That's why I changed in my local Freetz GIT repo to:
>>
>>                 ARCH="$(KERNEL_ARCH)" \
>>
>> So this stuff might be a "cleanup" in your eyes, but for me this is to
>> correct what $ARCH really should be for ltrace in the MIPS world.
>>
>> As the 3 patches can be easily cherry-picked, there is nothing
>> speaking against  to apply them.
>>
>> If your philosphy/policy is to only include real BUGFIXES into a
>> ltrace-x.y release, please document this!
>>
>> Furthermore, I highly appreciate that you put a NOTE in the
>> commit-changelog of patches taken from ltrace-upstream to any BUGFIX
>> release. This is common usage for example stable Linux-kernels. So
>> people can follow what's happening.
>>
>> EXAMPLE from [1]:
>>
>> "commit 1de63d60cd5b0d33a812efa455d5933bf1564a51 upstream."
>>
>> BTW, when are you planning a v0.8.0 release?
>>
>> With latest ltrace-upstream GIT I get this:
>>
>> configure: updating cache
>> /home/wearefam/src/freetz/freetz-git/source/target-mipsel_uClibc-0.9.32.1/config.cache
>> configure: creating ./config.status
>> config.status: error: cannot find input file: `Makefile.in'
>>
>> ERROR: Build failed.
>> make: *** [source/target-mipsel_uClibc-0.9.32.1/ltrace-0.7.90+gitbc0de43/.configured]
>> Error 1
>>
>> I must check if this is a local problem or comes from ltrace.
>>
>
> Damn, it was a local problem:
>
> $ rm -f ./source/target-mipsel_uClibc-0.9.32.1/config.cache
>

NO, it seems MIPSEL is not built correctly.
I switched to a 7390 target and do NOT get this build-error.

$ egrep -i 'mips|^FREETZ_TYPE_FON_WLAN' .config*
.config:FREETZ_TYPE_FON_WLAN_7390=y
.config:FREETZ_TARGET_ARCH="mips"
.config:FREETZ_TARGET_CROSS="mips-linux-uclibc-"
.config:FREETZ_KERNEL_CROSS="mips-unknown-linux-gnu-"
.config.old:FREETZ_TYPE_FON_WLAN_7170=y
.config.old:FREETZ_TARGET_ARCH="mipsel"
.config.old:FREETZ_TARGET_CROSS="mipsel-linux-uclibc-"
.config.old:FREETZ_KERNEL_CROSS="mipsel-unknown-linux-gnu-"

Oliver did you test for a MIPSEL target, too?

- Sedat -

> - Sedat -
>
>
>> - Sedat -
>>
>> [1] http://git.kernel.org/?p=linux/kernel/git/stable/linux-stable.git;a=commitdiff;h=a7a878d102299bec638233e6a8ab679ce89a1c24
>>
>>> Thanks,
>>> PM