Re: [OE-core] [kirkstone] binutils: stable 2.38 branch updates

[Randy MacLeod <randy.macleod@windriver.com>]

On 2022-09-13 12:59, Khem Raj wrote:
> On 9/13/22 2:43 AM, Sundeep KOKKONDA wrote:
>> Hello,
>> I am planning to take the gcc, glibc & binutils updates to Kirkstone. 
>> I found only the glibc-2.35 branch is having a few updates (given 
>> below).
>> /f8ad66a4ca nscd: Fix netlink cache invalidation if epoll is used [BZ 
>> #29415]/
>> /9e960717e1 Apply asm redirections in wchar.h before first use/
>> /577c2fc7f3 elf: Call __libc_early_init for reused namespaces (bug 
>> 29528)/
>> /83f1d9851e NEWS: Add entry for bug 28846/
>> /cbd8685e82 socket: Check lengths before advancing pointer in 
>> CMSG_NXTHDR/
>> /4bafc4001d alpha: Fix generic brk system call emulation in 
>> __brk_call (bug 29490)/
>> /37fd2ac665 stdlib: Fixup mbstowcs NULL __dst handling. [BZ #29279]/
>> /a1ec4157bc stdlib: Remove attr_write from mbstows if dst is NULL 
>> [BZ: 29265]/
>> /813a8d0171 Update syscall lists for Linux 5.19/
>> /e200127c6c riscv: Update rv64 libm test ulps/
>> /8c172a6cb0 dlfcn: Pass caller pointer to static dlopen 
>> implementation (bug 29446)/

Ideally you'd get that in this week since I think that 4.0.4 is being 
built on Monday, September 19th.

CCing Steve in case he missed this thread.


>> The binutils-2.38 has no new commits. And, gcc-11.x also not have any 
>> new release after 11.3.

That's good to know. I was just wondering if there was going to be 
gcc-11.4 so this info is timely.


>> In /master/ branch gcc_12.2, glibc_2.36 & binutils_2.39 versions are 
>> used. Kirkstone will continue to use the old versions?


To answer this more directly than Khem. ;-)

The Kirkstone branch will NOT update to the versions shown above. Users 
want a stable version

for the supported timeline of a release; some even may be surprised that 
we update from 11.2 to 11.3
even if it's all bug fixes.


>>
>
> the glibc fixes are good so go ahead bump the SRCREV to latest on 2.35 
> branch

Agreed.


>  major version backports are exceptions. We can consider doing mixin 
> layers if needed for say newer compiler.


Answered above.

Thanks Sundeep!

../Randy


>
>>
>>
>> Thanks,
>> Sundeep K.
>>
>>
>>
>>
>
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#170603): https://lists.openembedded.org/g/openembedded-core/message/170603
> Mute This Topic: https://lists.openembedded.org/mt/92739561/3616765
> Group Owner: openembedded-core+owner@lists.openembedded.org
> Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [randy.macleod@windriver.com]
> -=-=-=-=-=-=-=-=-=-=-=-
>

-- 
# Randy MacLeod
# Wind River Linux