[OE-core] [PATCH v2] valgrind: update from 3.13.0 to 3.14.0

Randy MacLeod randy.macleod at windriver.com
Mon Oct 22 21:38:17 UTC 2018


On 10/21/18 6:24 PM, Khem Raj wrote:
> On Sun, Oct 21, 2018 at 9:48 PM Randy MacLeod
> <randy.macleod at windriver.com> wrote:
>>
>> On 10/21/18 5:47 AM, Khem Raj wrote:
>>> fails ptest compile on musl/mips
>>> http://errors.yoctoproject.org/Errors/Details/197122/
>>
>> I have a fix that I'll send later tonight or tomorrow morning
>> once I've tested it a bit more. I'll paste it inline below in
>> case anyone wants it now.
>>
>> I did notice that the ptest build of valgrind takes a more than
>> 30 minutes!:
>>
>> 0: valgrind-3.14.0-r0 do_compile_ptest_base - 2306s (pid 21911)
>>
>> It's building:
>>     none/tests/mips32/msa_fpu.c
>> which is from a commit [1] that tries to cover
>> all MIPS SIMD architecture instructions. I'll
>> drop that test unless someone objects since it's overkill
>> for a ptest IMO.
> 
> I do not use MSA here so it might be ok to to save some time. but I am
> curious how it looks on real h/w
> 

I don't have a mips target at hand but I may decide
to dig one up and see if there's any value in the MSA test
at some point. Until then I'll leave the test as is.


Here are the valgrind bugs that I reported :

musl bug:
https://bugs.kde.org/show_bug.cgi?id=400162

assembler bug:
https://bugs.kde.org/show_bug.cgi?id=400164

I have update the oe-core patch status locally
with a link to these bugs IDs and I'll send that
once the:
   valgrind: Make local functions static to avoid assembler error
patch is merged.

-- 
# Randy MacLeod
# Wind River Linux



More information about the Openembedded-core mailing list