[LEDE-DEV] Release planning

John Crispin john at phrozen.org
Thu Dec 22 07:29:36 EST 2016

On 22/12/2016 13:19, Koen Vandeputte wrote:
> On 2016-12-21 20:13, Jo-Philipp Wich wrote:
>> - Are there any outstanding changes?
>>    Is there important changes we should wait for before branching the
>>    release? Is there pending stuff in the staging trees which should
>>    absolutely go into the first release?
> Bump musl to a newer head? (or better to backport some fixes?)

i thinkt hat bumping the libc shortly before the release is a really bad
idea. it might fix some problems but can introduce some regressions

> I see a lot of fixes since 1.1.15 on important parts which could
> influence stability as a lot of packages depend on musl:

backporting would be an option. could you provide a patch adding these
fixes ?


> http://git.musl-libc.org/cgit/musl/log/
> Some examples:
>     fix build regression on archs with variable page size
>     fix swprintf internal buffer state and error handling
>     fix integer overflow of tm_year in __secs_to_tm
>     fix minor problem in previous strtod non-nearest rounding bug fix
>     fix strtod int optimization in non-nearest rounding mode
>     fix strtod and strtof rounding with many trailing zeros
>     fix gratuitous undefined behavior in strptime
>     fix float formatting of some exact halfway cases
>     add pthread_setname_np (useful for debugging)
>     fix clock_nanosleep error case
>     math: fix pow signed shift ub
>     fix integer overflows and uncaught EOVERFLOW in printf core
>     fix integer overflow in float printf needed-precision computation
>     fix regexec with haystack strings longer than INT_MAX
>     fix missing integer overflow checks in regexec buffer size computations
>     fix strftime %y for negative tm_year
>     fix getservby*_r result pointer value on error
>     fix undefined behavior in sched.h cpu_set_t usage
>     add missing *_unlocked and wcsftime_l prototypes to wchar.h
>     fix if_indextoname error case
>     fix ifru_data and ifcu_buf types in net/if.h
>     fix printf regression with alt-form octal, zero flag, and field width
>     fix FFSYNC by changing it to O_SYNC
>     verify that ttyname refers to the same file as the fd
>     fix pread/pwrite syscall calling convention on sh
>     fix regression in tcsetattr on all mips archs
>     fix asctime day/month names not to vary by locale
> Please share your view on this one :-)
> Thanks

More information about the openwrt-adm mailing list